Characterization of a Stable Form of Tryptophan Hydroxylase from the Human Parasite Schistosoma mansoni *

A cDNA (Schistosoma mansonitryptophan hydroxylase; SmTPH) encoding a protein homologous to tryptophan hydroxylase, the enzyme that catalyzes the rate-limiting step in the biosynthesis of serotonin, was cloned from the human parasite Schistosoma mansoni. Bacterial expression of SmTPH as a histidine fusion protein produced soluble active enzyme, which was purified to apparent homogeneity and a final specific activity of 0.17 μmol/min/mg of protein. The purified enzyme was found to be a tetramer of approximately 240 kDa with a subunit size of 58 kDa. Several of the biochemical and kinetic properties of SmTPH were similar to those of mammalian tryptophan hydroxylase. Unlike the mammalian enzyme, however, SmTPH was found to be stable at 37 °C, itst 1 2 being nearly 23 times higher than that of a similarly expressed rabbit tryptophan hydroxylase. A semiquantitative reverse transcription polymerase chain reaction showed that the level of SmTPH mRNA in a larval stage of the parasite (cercaria) is 2.5 times higher than in adult S. mansoni, suggesting possible differences in the level of enzyme expression between the two developmental stages. This study demonstrates for the first time the presence of a functional tryptophan hydroxylase in a parasitic helminth and further suggests that the parasites are capable of synthesizing serotonin endogenously.

Tryptophan hydroxylase (TPH 1 ; tryptophan 5-monooxygenase; EC 1.14. 16.4) catalyzes the hydroxylation of L-tryptophan to 5-hydroxy-L-tryptophan . This reaction is the first and rate-limiting step in the biosynthesis of the monoamine neurotransmitter, serotonin (5-hydroxytryptamine; 5-HT) (reviewed in Ref. 1). TPH belongs to a family of aromatic amino acid hydroxylases that also includes the catecholamine biosynthetic enzyme, tyrosine hydroxylase (TH; EC 1.14.16.2) and phenylalanine hydroxylase (PAH; EC 1.14.6.1) (for reviews, see Refs. [2][3][4]. The three enzymes catalyze similar hydroxylation reactions and share a distinctive requirement for tetrahydrobiopterin (BH 4 ) and non-heme ferrous iron as cofactors (2). Studies of TPH have been hampered by the extreme instability of the enzyme (5)(6)(7)(8)(9)(10)(11) and the difficulty in obtaining high levels of purified active enzyme from either native or heterologous sources (10 -17). Sequence analyses of TPH cDNAs derived from mammalian and other vertebrate species (13, 18 -23) revealed that TPH shares high overall sequence homology with the other two members of the hydroxylase family (1,2). More recently, deletion mutagenesis studies identified three main functional regions of TPH, including a conserved central core that comprises the catalytic domain (15,16,21,24,25), a C-terminal intersubunit binding region responsible for the formation of enzyme tetramers (26,27), and a divergent N-terminal end. The latter is predicted to have a regulatory function (15,24,28) and may contribute to the instability of TPH (25) Serotonin, a well known neuroactive agent of the mammalian central nervous system and periphery (29), has been identified in every invertebrate phylum thus far investigated (30). In parasitic flatworms (platyhelminths), including the human bloodfluke, Schistosoma mansoni, 5-HT acts as an important regulator of motor activity and carbohydrate metabolism (for a review, see Ref. 31) and as such is critical for the survival of the parasite in the host. Immunofluorescence and histochemical studies have localized 5-HT in the central and peripheral nervous systems of the worm, as well as holdfast structures, body musculature, and reproductive structures (31). Earlier studies of 5-HT biosynthesis in parasitic helminths reported conflicting results. The failure to demonstrate TPH activity in crude tissue extracts of S. mansoni led some researchers to conclude that TPH is absent in this animal and that the parasite depends entirely on the host for a source of serotonin (32)(33)(34)(35)(36). Although other authors have challenged these studies and presented preliminary evidence of 5-HT biosynthesis in related parasites (37)(38)(39), the question of whether TPH is present or absent in parasitic worms, in particular S. mansoni, remains largely unresolved.
Here we report the cDNA cloning, purification, and functional characterization of tryptophan hydroxylase from S. mansoni (SmTPH). This study provides the first evidence that S. mansoni, and probably related parasites, possess the endogenous capability for de novo synthesis of 5-HT. When expressed in Escherichia coli, the purified SmTPH was highly active and, in contrast to the mammalian enzyme, very stable during purification and storage. SmTPH represents a potentially useful model for detailed biochemical studies of TPH structure and function. activated charcoal were from Fisher.  H]-L-Tryptophan was from Amersham Pharmacia Biotech. (6R)-5,6,7,8-tetrahydrobiopterin (BH 4 ) and dopamine were from Research Biochemicals International. Aprotinin, leupeptin, phenylmethylsulfonyl fluoride, and catalase were from Roche Molecular Biochemicals. All other chemicals were of the highest purity and quality from available commercial sources.
S. mansoni-A Puerto Rican strain of S. mansoni was maintained as described previously (40). Crude adult worm tissue extracts were prepared and used directly for TPH activity measurements as described earlier (40). Total RNA was extracted from S. mansoni using the TRIzol reagent (Life Technologies, Inc.). Poly(A ϩ ) RNA from adult S. mansoni was purified from total RNA using oligo(dT)-cellulose columns (Amersham Pharmacia Biotech).
Cloning of Full-length SmTPH-A partial S. mansoni cDNA sequence (576 bp) homologous to other TPH sequences was isolated by homology RT-PCR. Oligonucleotide primers were synthesized based on a predicted genomic TPH sequence from the free-living nematode Caenorhabditis elegans (cosmid ZK1290; GenBank TM accession no. U21308) and used for PCR amplification of adult S. mansoni oligo(dT) reverse transcribed cDNA. The sense and antisense primers targeted a region of the predicted catalytic domain that is highly conserved among all aromatic amino acid hydroxylases (see Fig. 3). Primer sequences and RT-PCR conditions are described elsewhere (40).
The 5Ј-end of SmTPH was obtained using a RT-PCR method that targets the conserved 5Ј-end spliced leader (SL) sequence of S. mansoni transcripts (41,42 Fig. 1), which corresponded to nucleotides 9 -32 of the S. mansoni 36-nucleotide SL sequence (41) was used as a sense primer, while oligonucleotide A (see Fig. 1) was used as an antisense primer. An aliquot (2 l of 1:10 dilution) of the PCR product was similarly subjected to a second PCR reaction (25 cycles; same cycling parameters as above) using the same sense primer (SmSL) and a nested reverse primer (primer B; see Fig. 1). The resulting PCR product was gel-purified, cloned into the vector PCR 2.1 (Invitrogen), and sequenced by the dideoxy chain termination method.
The 3Ј-end of SmTPH cDNA was amplified from an adult S. mansoni cDNA library in pcDNA3.1(ϩ) (Invitrogen). An aliquot of the plasmid library (0.4 g) was subjected to PCR (30 cycles) using a SmTPHspecific sense primer (primer C; see Fig. 1) and an antisense pcDNA3.1(ϩ) vector-specific primer flanking the multiple cloning site (VSP1, 5Ј-GGAGGGGCAAACAACAGATGG-3Ј). The PCR cycling parameters were as above except for an annealing temperature of 55°C. An aliquot of the PCR product was subjected to a second round of PCR amplification (25 cycles) using a nested SmTPH-specific sense primer (primer D; see Fig. 1) and pcDNA3.1(ϩ) vector-specific-primer (VSP2, 5Ј-TAGAAGGCACAGTCGAGGC-3Ј). Amplified products were cloned into pCR2.1, and DNA was sequenced as before.
For expression studies, the complete coding sequence of SmTPH was amplified by RT-PCR and subcloned into a T7 polymerase-based pET prokaryotic expression vector (43). Oligo(dT) reverse transcribed cDNA was subjected to 35 cycles of PCR with primers that targeted the entire coding sequence of SmTPH (primers S and E; see Fig. 1) and a proofreading DNA polymerase (Pwo; Roche Molecular Biochemicals) according to the manufacturer's procedure. To facilitate further subcloning into the expression vector, enzyme restriction sites NdeI and BamHI were incorporated at the 5Ј-end of the sense and antisense primers, respectively. The resulting PCR product was gel-purified, digested with NdeI and BamHI (Life Technologies, Inc.), and ligated into pET15b vector (Novagen), which was linearized by the same two restriction enzymes. Cloning into pET15b introduces an N-terminal oligohistidine fusion tag, which adds 20 amino acid residues of vector-derived sequence to the expressed SmTPH product. The final construct was confirmed by DNA sequencing of three independent clones and then used to transform E. coli host strain BL21(DE3)pLysS (Novagen).
For purification of recombinant SmTPH, cell pellets from 100 ml of induced bacterial cultures were thawed in 4 ml of 20 mM phosphate buffer (pH 7.4) containing 0.5 M NaCl, 0.2% Tween 20, 5% glycerol, 10 mM imidazole, and a mixture of protease inhibitors (1 mM phenylmethylsulfonyl fluoride and 50 g/ml each leupeptin and aprotinin). To promote cell lysis by the T7 resident lysogen, the cells were subjected to two cycles of rapid freeze-thawing followed by sonication on ice (seven pulses of 15 s separated by intervals of 30 s) using a vibra cell sonicator (Sonics and Material, Danbury, CT) set at 20% maximal power. Cell lysates were centrifuged at 12,000 ϫ g for 15 min and 4°C. The resulting pellet was resuspended in 2.5 ml of the same buffer and similarly sonicated and centrifuged as above. The two supernatants were pooled and used for direct enzymatic assays and for subsequent purification of the expressed enzyme.
Recombinant SmTPH was purified by immobilized metal (nickel) affinity chromatography (44) using the HisTrap kit (Amersham Pharmacia Biotech) for purification of histidine-tagged proteins, as described previously (40). Excess imidazole was removed by gel filtration through a Sephadex G-25 column (PD-10; Amersham Pharmacia Biotech), and the purified enzyme was stored in 50 mM HEPES (pH 7.5) containing 0.2 M NaCl, 10% glycerol, 0.05% Tween 20, and 1 mM dithiothreitol. Purified enzyme preparations (0.2 mg/ml) were stable for at least 4 days at 4°C and could be stored at Ϫ80°C for at least 1 month with no significant loss in activity.
TPH Enzymatic Assays-TPH activity was measured using the tritiated water release method (45,46) with few modifications. The standard assay was performed in a 100-l reaction of 50 mM HEPES (pH 7.5) containing 400,000 cpm of L-[5-3 H]Tryptophan and enough unlabeled L-tryptophan to make a final concentration of 100 M, 0.2 mg/ml catalase, 0.4 mM NADH, 10 milliunits of dihydropteridine reductase, 10 M dithiothreitol, and either purified SmTPH (0.6 g) or crude adult S. mansoni tissue extract (50 -100 g of protein). The reaction was started with the addition of 200 M BH 4 , unless indicated otherwise, and the samples were incubated for 10 min at 37°C. Preliminary experiments revealed that TPH activity increased linearly up to 12 min of incubation under these conditions. The reaction was terminated by the addition of 1 ml of activated charcoal (7.5% (w/v) in 1 M HCl) to each sample. After centrifugation (2000 ϫ g for 10 min), aliquots of the supernatants were radioassayed in 10 ml of scintillation mixture (ICN). Enzyme activity data were analyzed using Lineweaver-Burk plots or by computer-assisted, nonlinear curve fitting to the Michaelis-Menten model. All kinetic parameters (K m and apparent K m (S 0.5 )) were determined using the program Enzyme Kinetics (version 1.C; DogStar Software) and were obtained from two to three independent experiments, each performed in duplicates or triplicates.
TPH Stability Assay-The stability of SmTPH was assessed in comparison with that of recombinant rabbit brain TPH similarly expressed in E. coli. In preparation for these experiments, the complete coding sequence of rabbit brain TPH cDNA (13) was subcloned into pET15b and expressed as a histidine-tagged protein in BL21(DE3)pLysS E. coli. Bacterial cells expressing rabbit TPH or SmTPH were lysed, as described above, and the corresponding soluble fractions containing expressed enzyme were passed through a Sephadex G25 PD10 (Amersham Pharmacia Biotech) column equilibrated with the same HEPES buffer described earlier. Stability was measured as a function of time according to the procedure of Mockus et al. (25). Briefly, aliquots (100 g of protein) of the crude SmTPH or rabbit TPH extracts were preincubated at 37°C for varying lengths of time (0, 10, 20, 40, and 80 min) and then assayed for TPH activity. The data were calculated as a percentage of the initial level of activity (t ϭ 0) for each enzyme.
Developmental Expression of SmTPH mRNA in S. mansoni-Semiquantitative RT-PCR was employed for the determination of SmTPH mRNA levels. Total RNA (ϳ2 g) from two developmental stages of S. mansoni (cercaria and adults) were subjected to DNase I treatment (amplification grade; Life Technologies, Inc.) followed by a standard RT-PCR reaction (1 min at 94°C, 14 -36 cycles of 30 s at 94°C, 30 s at 53.5°C, 60 s at 72°C, and 7 min at 72°C) using primers (sense, primer D; antisense, primer F; see Fig. 1) that amplify an 873-bp cDNA product from SmTPH. PCR was standardized by simultaneous amplification of a constitutively expressed control housekeeping gene, S. mansoni ␣-tubulin (47,48), as described previously (49). The PCR primers used for the amplification of the 558-bp S. mansoni ␣-tubulin cDNA fragment were as follows: sense, 5Ј-CTTATCGTCAACTTTTCCATCC-3Ј; antisense, 5Ј-GGAAGTGGATACGAGGATAAGG-3Ј (modified from Ref. 48). Standard curves were generated to ensure that the PCR assay was in the exponential phase of synthesis after 34 cycles for SmTPH or 24 cycles for ␣-tubulin. The resulting PCR products were cloned in PCR2.1 and confirmed by DNA sequencing. Densitometric analysis of the ethidium bromide-stained RT-PCR products were performed with the NIH Image program version 1.61 (Bethesda, MA).
Other Methods-Size exclusion chromatography of purified SmTPH was performed on a Sephacryl-200HR gel filtration column (10-mm inner diameter ϫ 50 cm; Bio-Rad), as described previously (27). Protein concentrations were measured by the method of Bradford (50), using the Bio-Rad protein assay kit and bovine serum albumin as a standard. Reducing SDS-polyacrylamide gel electrophoresis was performed according to the method of Laemmli (51) using precast 10% acrylamide gels from Novex, Inc. For Western blot analysis of SmTPH, aliquots of purified enzyme (0.5-1 g) were electrophoresed as above, transferred onto nitrocellulose (52), and reacted with a sheep polyclonal antibody (1:500 dilution) raised against rabbit TPH (Chemicon International) followed by a peroxidase-conjugated rabbit anti-sheep IgG (Pierce) as the secondary antibody (1:1000 dilution).

RESULTS
Cloning of the Full-length SmTPH cDNA and Protein Sequence Analyses-A partial TPH sequence was first obtained by RT-PCR using oligo(dT) reverse-transcribed S. mansoni cDNA and C. elegans primers (40) that targeted a region conserved among all aromatic amino acid hydroxylases. A 576-bp product was sequenced and found to have high homology with TPH sequences from other species. The missing 5Ј-and 3Ј-ends were subsequently obtained by an anchored PCR-based strategy. The 5Ј-end was amplified in a RT-PCR reaction that targeted the conserved SL sequence of S. mansoni (40,41). A 741-bp product corresponding to the 5Ј-end of TPH was sequenced and found to carry a complete S. mansoni SL (nucleotides 9 -36 of the SL sequence; see Fig. 1) including the last four nucleotides, which were not part of the SmSL primer used in the anchored PCR reaction. This finding suggests that SmTPH is trans-spliced at the 5Ј-end to the S. mansoni SL, just as described previously for several other S. mansoni cDNAs (40 -42). The 3Ј-end of SmTPH was PCR-amplified directly from a S. mansoni cDNA plasmid library, using TPH-specific and vector-derived primers (see Fig. 1). The resulting product (1000 bp) contains a potential polyadenylation sequence (AATATA) (53) upstream of a poly(A) tail and thus is presumed to represent the 3Ј-end of the full-length transcript. Fig. 1 shows the nucleotide and predicted amino acid sequence of SmTPH. The composite cDNA reveals a single open reading frame of 1494 bp encoding a predicted protein of 497 amino acids with a calculated molecular mass of 58 kDa. Protein sequence analysis revealed the presence of two consensus sites (Ser 151 and Thr 130 ) for phosphorylation by the Ca 2ϩ /calmodulin-dependent protein kinase type II and a consensus leucine zipper motif (SmTPH amino acid positions 358 -377). BLAST analysis (54) of the predicted protein sequence indicated that SmTPH is highly homologous to tryptophan hydroxylase from other species. The dendrogram in Fig. 2 shows that SmTPH is more related to TPH sequences than to those of the other two aromatic amino acid hydroxylases, TH and PAH. Based on pairwise CLUSTAL protein alignments (55), SmTPH shares high amino acid sequence homology (65-67%) with vertebrate TPH sequences (13, 18 -23). In contrast, there is considerably less homology (57%) with the only other invertebrate sequence available, a Drosophila enzyme that has been described as a TPH/PAH hybrid (56). Structurally, the Drosophila enzyme appears to be more closely related to PAH than any of the TPH sequences (Fig. 2).
An amino acid sequence alignment of SmTPH and other tryptophan hydroxylase sequences is shown in Fig. 3. A high degree of sequence conservation (up to 82% homology) is apparent in the carboxyl-terminal two-thirds of the sequence (amino acid positions 143-455 of SmTPH), a region that comprises the conserved catalytic domain of TPH (15,16,21,24). Several structural motifs that are characteristic of TPH and other aromatic amino acid hydroxylases are also present in this conserved core region, including a potential iron binding site (SmTPH His 311 , His 315 , and Glu 330 ) (2, 57, 58) and the signature peptide PEPD-CHELLGHVP. This latter is part of a conserved 27-amino acid sequence (SmTPH Tyr 289 -His 315 ; see Fig. 3) that forms a cofactor (BH 4 ) binding site in PAH (59) and possibly TH and TPH as well. Amino acid sequence homology decreases significantly at the amino-terminal end (amino acid positions 1-142 of SmTPH) and within a C-terminal sequence of approximately 40 amino acids. The latter region contains a potential intersubunit binding domain (SmTPH Leu 462 -Ile 480 ), consisting of a leucine heptad repeat interspersed by a 4,3hydrophobic repeat (26,27) (Fig. 1).
Expression of SmTPH-The complete coding sequence of SmTPH (1494 bp) was amplified directly from S. mansoni mRNA by RT-PCR, cloned into the prokaryotic pET15b vector, and expressed in E. coli BL21(DE3)pLysS. To ensure maximal TPH enzyme activity, bacterial cultures were supplemented with iron, which is limiting in a bacterium (13)(14)(15). Cloning of SmTPH into pET15b introduces an N-terminal oligohistidine fusion tag, which permitted the subsequent purification of the enzyme by nickel affinity chromatography. This tag, consisting Induction of SmTPH-transformed BL21(DE3)pLysS cells produced large amounts of active TPH. Western blot analysis of soluble bacterial protein extracts identified a predominant band of the expected size that reacted with a rabbit anti-TPH antibody (Fig. 4). No cross-reactivity was detected between the antibody and lysates prepared from E. coli transformed with pET15b vector containing no insert (Fig. 4). The average specific activity of SmTPH in crude bacterial extracts was 3.5 Ϯ 0.5 nmol/min/mg of protein with about 80% of total enzyme activity being recovered in the soluble fraction. The remaining activity was retained in the pellet in the form of inclusion bodies.
Purification and Macromolecular Structure of SmTPH-SmTPH was purified by nickel affinity chromatography as an N-terminal histidine fusion tag. Based on specific activity measurements, SmTPH was purified about 45-fold to a final specific activity of 170 Ϯ 5.0 nmol/min/mg of protein. The yield from 100 ml of induced bacterial culture was 0.66 -0.8 mg of purified SmTPH. Coomassie Blue staining and densitometric analysis of the purified protein showed a predominant Western positive band of ϳ60 kDa, which was consistent with the expected size of the SmTPH monomer (Fig. 4). The purified enzyme was subjected to size exclusion chromatography to determine the oligomeric organization of the protein. Calculations of the molecular weight from a standard V e /V o plot suggest that SmTPH exists as a tetramer with approximate molecular mass of ϳ240 kDa (Fig. 5).
Kinetic Properties of Purified SmTPH-All three aromatic amino acid hydroxylases display an absolute requirement for tetrahydrobiopterin as a cofactor (2). Fig. 6A is a representative BH 4 saturation curve showing that the activity of the purified SmTPH is also dependent on its cofactor BH 4 . Removing the cofactor abolished all enzymatic activity. The K m for BH 4 was measured by varying the concentration of BH 4 (2.5-200 M) and keeping the concentration of the substrate, tryptophan, constant at 100 M. Purified SmTPH displayed a K m for BH 4 of 6.7 Ϯ 0.7 M (mean Ϯ S.E.) and a V max value of 163 Ϯ 13 nmol/min/mg of protein (mean Ϯ S.E.).
The K m for the substrate was investigated by varying tryptophan concentration (1-250 M) while fixing the BH 4 concentration at 200 M (Fig. 6B). SmTPH activity was inhibited at high concentrations of tryptophan (Ͼ100 M), in a similar fashion to what was previously reported for mammalian TPH (6,15,17,60). Since an accurate K m determination for tryptophan was not possible due to substrate inhibition, an apparent K m (S 0.5 ) was estimated at (mean Ϯ S.E.) 22 Ϯ 2.0 M.
Several known inhibitors of mammalian TPH, including pchlorophenylalanine, dopamine, and the product of tryptophan hydroxylation, 5-HTP (11,13,(61)(62)(63)(64)(65)(66), all caused inhibition of purified SmTPH (Table I). The parasite enzyme was not sensitive to feedback inhibition by 5-HT or by the products of 5-HT metabolism, N-acetyl-5-HT, and melatonin, even at high concentrations of 0.1 mM ( Table I). The inhibition by dopamine was found to be predominantly competitive with respect to the cofactor, BH 4 (Fig. 7A). In the case of 5-HTP, the inhibition was predominantly noncompetitive with respect to tryptophan. The addition of 50 M 5-HTP caused a marked 2-3-fold decrease in V max with very little change in the apparent K m for the substrate (Fig. 7B).
Low levels of catechols, in particular dopamine, have been shown to cause a sustained time-dependent inhibition of mammalian forms of tyrosine hydroxylase (67)(68)(69). To determine if SmTPH is similarly sensitive to this form of dopamine inhibition, aliquots of the purified enzyme (2.8 M) were preincubated with a stoichiometric amount of dopamine for 2, 4, 6, 10, and 15 min at 30°C and then assayed for TPH activity. No change in SmTPH activity was detected compared with a control sample incubated for the same length of time in the absence of dopamine (data not shown).
Stability of SmTPH-The stability of SmTPH was compared with that of rabbit brain TPH (13) that was similarly subcloned into pET15b and expressed in E. coli BL21(DE3)pLysS strain as a histidine-tagged fusion protein. Initial attempts to purify the rabbit enzyme by nickel chelation chromatography yielded a very unstable enzyme (t1 ⁄2 Ͻ 10 min at 37°C) with low specific activity. Therefore, the comparison between SmTPH and the rabbit enzyme was carried out with crude bacterial lysates, which had similar specific activities of 3.3 nmol/min/mg of protein for SmTPH and 1.49 nmol/min/mg of protein for rabbit TPH, respectively. The latter value is essentially identical to what was previously reported for similar preparations of this enzyme (13). Aliquots of crude SmTPH or rabbit TPH containing the same amount of protein (ϳ100 g) were preincubated at 37°C for up to 80 min and then assayed for TPH activity. As can be seen in Fig. 8, SmTPH activity remained virtually unchanged over the incubation period, with an estimated halflife (t1 ⁄2 ) of ϳ21 h. In contrast, the rabbit brain TPH displayed a significantly shorter t1 ⁄2 of 54 min. This latter value is comparable with the previously reported t1 ⁄2 for recombinant rabbit TPH (25). The same experiment was repeated with aliquots (1.75 g) of purified SmTPH. The results (data not shown) produced an estimated t1 ⁄2 value for the pure enzyme of 99 min at 37°C.
TPH Activity in Crude S. mansoni Extracts-Crude tissue extracts of adult S. mansoni were prepared and tested directly for TPH enzymatic activity as described above. When compared with a boiled enzyme control, the specific activity level of the native S. mansoni TPH was ϳ0.02-0.04 nmol/min/mg of protein (data not shown). This level of activity is similar to that reported earlier for native TPH measured in rabbit brain extracts (13). When BH 4 was omitted from the assay mixture, no detectable levels of activity were obtained from the worm extracts. This illustrates that the native S. mansoni TPH has the same absolute requirement for the biopterin cofactor as the recombinant SmTPH.
SmTPH Developmental Expression in S. mansoni-The expression of SmTPH was examined by semiquantitative RT-PCR in two different developmental stages of S. mansoni, cercaria and adults. Expression levels were standardized by comparison with a constitutively expressed control gene from S. mansoni, ␣-tubulin (47,48). Fig. 9 indicates that the SmTPH expression level is approximately 2.5-fold higher in the cercarial stage than in the adult stage of S. mansoni. No ␣-tubulin PCR products were detected in a control reaction that lacked reverse transcriptase, thus ruling out the possibility of genomic DNA contamination. DISCUSSION This study describes the cloning and functional characterization of tryptophan hydroxylase from a lower invertebrate, the parasitic platyhelminth S. mansoni. This is the second member  Fig. 2 for relevant GenBank TM accession numbers) were aligned using the program MacVector (version 6.5; Oxford Molecular), according to the CLUSTAL method. Predicted conserved residues of the iron binding site are indicated by an asterisk. The overlined sequence represents a conserved signature motif present in all aromatic amino acid hydroxylases. Arrows indicate the conserved regions that were targeted for initial PCR cloning of a partial SmTPH sequence, as described under "Experimental Procedures." of the aromatic amino acid hydroxylase family identified in S. mansoni; we recently cloned a functional form of TH from this same parasite (40). The finding of these enzymes in such a primitive invertebrate raises interesting questions about the evolution of the three hydroxylases. There is general agreement that the three enzymes are derived from a common ancestral gene through a series of two gene duplications, the first of which gave rise to TH, whereas the second separated TPH from PAH (18,70,71). It has been suggested that the divergence of TPH and PAH occurred late in evolution, possibly after the emergence of arthropods (71). However, as pointed out by Boularand et al. (72), recent genome sequencing data have identified distinct predicted genomic sequences for all three enzymes in the nematode, C. elegans, suggesting that the two gene duplications occurred earlier than was previously thought. The finding of TPH and TH in S. mansoni strengthens this point and further suggests that the divergence of the aromatic amino acid hydroxylases may have occurred before the emergence of platyhelminths.
An alignment of SmTPH with other TPH sequences identified a core of high amino acid sequence identity in the middle to C-terminal region of the enzyme. This stretch of conserved sequence corresponds roughly to the previously defined catalytic domain of TPH (1) and includes several distinctive motifs, including the predicted iron binding site (57,58) and a putative biopterin binding domain (59). The N-terminal region, on the other hand, shows little sequence conservation across the different species of the enzyme and is particularly divergent in SmTPH, with identity scores of 14 -17% when compared with cognate mammalian sequences (see Fig. 3). The divergence at the N-terminal end is consistent with the notion that this region lies outside the catalytic domain and may serve a regulatory function (15, 21, 24, 25, 28, 65), just as shown for other aromatic amino acid hydroxylases (2, 3). The present identification of a putative phosphorylation site for Ca 2ϩ /calmodulin-   (26,27), which is present in all TPH sequences including SmTPH. The conservation of this structural motif in an otherwise divergent Cterminal end supports previous suggestions that the carboxyl region of TPH is not directly involved in catalysis, despite conflicting mutagenesis data (15,27) but rather constitutes a distinctive oligomerization domain (27). SmTPH was expressed in E. coli as a histidine-tagged protein, which was purified and partially characterized. The analysis showed that SmTPH shares many of the characteristics of mammalian TPH, both recombinant and native. Similar to other forms of the enzyme, SmTPH was found to form tetramers of about 240 kDa. In addition, SmTPH showed a characteristic absolute requirement for the reduced pterin cofactor, BH 4 . The K m for BH 4 was 4 -7-fold less than what was previously reported for purified mammalian brain TPH tagged to glutathione S-transferase (16) or maltose-binding protein (15). It is unknown if this discrepancy is due to the presence of large fusion tags on the two mammalian enzymes, which may have influenced the kinetic determination, or whether the parasite hydroxylase has a significantly higher affinity for the cofactor. With respect to the substrate, tryptophan, SmTPH exhibited a typical kinetic profile, with characteristic substrate inhibition at tryptophan concentrations above 100 M and an apparent K m (S 0.5 ) of about 22 M, similar to what was described previously for mammalian forms of the enzyme (6,15). Additional characterization of the parasite hydroxylase showed that the enzyme is sensitive to inhibition by the classic TPH inhibitor, p-chlorophenylalanine, as well as the immediate product of the reaction, 5-HTP, but not serotonin or its metabolites, N-acetylserotonin and melatonin. The lack of inhibition by serotonin was reported previously in crude brain extracts of native TPH (11). It is noteworthy that the inhibition of SmTPH by 5-HTP did not follow classical competitive kinetics, as might have been expected from standard product inhibition. Instead, 5-HTP inhibition (as a function of tryptophan) showed mixed, predominantly noncompetitive characteristics, with V max decreasing nearly 3-fold and the apparent K m increasing by about 50%. The significance of this inhibition profile is unclear, nor is it known if it is unique to the parasite. All available data on product inhibition of mammalian TPH stem from preparations of crude native enzyme (5,11), which is not directly comparable with the purified enzyme preparations used in this study. Additional research on the role of 5-HTP in TPH regulation is needed.
Previous studies have shown that TPH is susceptible to inhibition by catechol products of the TH reaction, in particular dopamine (62)(63)(64)(65)(66). Inhibition by dopamine is thought to be biologically relevant in regions of the nervous system where serotonergic and catecholaminergic neurons may interact. A large body of evidence on dopamine inhibition of TH shows the existence of two major mechanisms of hydroxylase inhibition, a time-dependent sustained inhibition seen at low (stoichiomet-  (25). The data are expressed as the percentage of initial activity for each enzyme measured at t ϭ 0. The initial specific activities (nmol/min/mg of protein) for SmTPH and rabbit TPH extracts were 3.3 and 1.46, respectively. All data points represent an n ϭ 3. t1 ⁄2 values were calculated from linear regression analysis of the data points obtained for each sample. ric) dopamine concentrations and competitive inhibition (with respect to the cofactor), which occurs at higher (M) concentrations of the neurotransmitter (4,73). In the present study, we found that SmTPH is similarly sensitive to inhibition by dopamine and that the inhibition at micromolar concentrations is essentially competitive with respect to BH 4 . We were unable to detect, however, any time-dependent enzyme inhibition at lower concentrations of dopamine. This points to important differences in the responses of TPH and TH to catechol inhibition.
Biochemical studies of TPH have been hindered by the difficulty in obtaining large amounts of purified active enzyme that is suitable for characterization. Even with the advent of molecular biology techniques, researchers have found that recombinant mammalian TPH overexpressed in E. coli tends to form inactive inclusion bodies (13,14), unless it is expressed with large fusion partners (15,16) and also becomes unstable upon purification. In contrast to mammalian TPH, however, the parasite enzyme was expressed as a soluble protein, which could be purified and was both active and stable. The specific activity of SmTPH was 2-13-fold higher than values reported for purified forms of recombinant mammalian TPH (15)(16)(17). In addition, results presented here showed that the half-life of a crude SmTPH extract was about 23 times longer than that of a similar preparation of rabbit TPH also expressed in E. coli. After purification, the rabbit enzyme lost activity very rapidly, whereas SmTPH remained relatively stable, with a half-life at 37°C of 99 min, and could be stored frozen with virtually no loss of activity. The reason for this dramatic difference in enzyme stability is unknown. Recent evidence has suggested that the notorious instability of mammalian TPH (10,11,14,25,63) may be associated, at least in part, with the enzyme's regulatory domain (25,60), roughly the same region that is least conserved in the parasite. This raises the interesting possibility that the stability of SmTPH is related to its distinctive N-terminal domain, which may stabilize activity more effectively than the cognate region of mammalian TPH. A stabilizing effect of the N-terminal regulatory domain on enzyme activity has been reported for the related hydroxylase, TH (25).
The finding of TPH in S. mansoni has clarified a long standing question of how this parasite obtains its serotonin. Just as other parasitic worms, S. mansoni has high levels of serotonin within its nervous system and is well known to rely heavily upon serotonin for a wide range of essential activities, among them the regulation of motility and carbohydrate metabolism. Earlier difficulties in identifying TPH activity in tissue extracts of S. mansoni led to the generalized belief that the parasite lacked the enzymatic capacity to synthesize serotonin endogenously and thus relied on the human host for a supply of the neurotransmitter. By cloning an active form of TPH from the parasite and also demonstrating TPH activity in crude worm extracts, the present study has shown clearly that the enzyme is present and active in S. mansoni. The earlier negative results were probably due to the paucity of the enzyme in the worm combined with the low sensitivity of the assay (74), both of which would limit the ability to detect TPH activity in the crude worm extracts. It should be noted, however, that the present results do not rule out the possibility that the parasite may recruit some exogenous serotonin from the host, possibly through a tegumental carrier (32,33,36), in addition to synthesizing the neurotransmitter endogenously. In this respect, it is interesting that the levels of SmTPH mRNA in the adult worm, which is strictly parasitic, are nearly 2.5 times lower that in a free living larval stage (cercaria). Although the difference is small, it is nonetheless surprising in light of the greater development of the serotonergic nervous system in the adult compared with the larva. The possibility exists that the adults rely both on endogenous synthesis and exogenous intake of serotonin, whereas the free living stage, which must rely entirely on biosynthetic activity, has proportionally higher levels of TPH. The significance of these results for the development of the parasite and its survival in the host is currently under investigation. FIG. 9. Developmental expression of SmTPH in S. mansoni. Semiquantitative RT-PCR was performed on total RNA extracted from two different developmental stages of S. mansoni (cercaria and adult), as described under "Experimental Procedures." The RT-PCR reactions in both developmental stages were standardized by simultaneous amplification of an internal control house keeping gene from S. mansoni (␣-tubulin). The SmTPH PCR product and that of the ␣-tubulin control are shown in the top panel. A negative PCR control was done using S. mansoni ␣-tubulin primers on total RNA samples subjected to a mock RT reaction (-RT). The PCR products were analyzed on a 1.2% agarose gel containing ethidium bromide followed by densitometric analysis. The lower panel shows a bar graph displaying the relative optical density (ROD ϭ optical density of SmTPH PCR product/optical density of ␣-tubulin control) obtained from adult and cercaria S. mansoni. Results are the mean Ϯ S.E. of three independent RT-PCR experiments, each done in duplicate. Unpaired t test showed that the ROD differences between the adults and cercaria are statistically significant (p ϭ 0.015).