Purification, Characterization, and Molecular Cloning of A Novel Rat Liver Dopa/Tyrosine Sulfotransferase*

A novel sulfotransferase was purified from the rat liver cytosol to electrophoretic homogeneity via five column chromatography steps (hydroxylapatite I, DEAE Bio-Gel, ATP-agarose I, hydroxylapatite II, and ATP-agarose II). The minimum molecular weight of the puri- fied enzyme was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis to be (cid:59) 33,000. Gel filtration chromatography revealed a native molecular weight of (cid:59) 34,000, indicating the enzyme being present in the monomeric form. The purified sulfotransferase displayed enzymatic activities, with a pH optimum of 9.25, toward various tyrosine and 3,4-dihydroxyphenyl- alanine (Dopa) isomers, except DL - ortho -tyrosine. Thyroid hormones, as well as dopamine and p -nitrophenol, could also be used as substrates. The apparent K m value of the enzyme (designated the Dopa/tyrosine sulfotrans- ferase) for L -Dopa, determined at a constant 14 (cid:109) M of 3 (cid:42) -phosphoadenosine was per- formed on 10% polyacrylamide gels using the method of Laemmli (30). The native molecular weight ( M r ) of the purified Dopa/tyrosine sulfo- transferase was determined by gel filtration chromatography using a Sephacryl S-200 column (2.6 (cid:51) 90 cm). Molecular weight standards including bovine serum albumin ( M r 67,000), ovalbumin ( M r 43,000), carbonic anhydrase ( M r 29,000), chymotrypsinogen A ( M r 25,800), and cytochrome c ( M r 12,400) were used for calibration. Protein determination was based on the method of Bradford (35) with bovine serum albumin as the standard.

Sulfation represents an important mechanism in vivo for the biotransformation and excretion of a variety of compounds (1)(2)(3). Upon sulfation, some peptides or proteins undergo changes in their biological activities to fulfill particular biochemical/physiological needs (3). For low molecular weight xenobiotics or endogenous compounds such as steroid hormones, catecholamines, and bile acids, sulfation may increase the water solubility and facilitate their excretion from the body by endowing them with (additional) charged properties (1,2). In relation to this latter aspect, the biochemistry and functional relevance of the excretion of free tyrosine-O-sulfate (TyrS) 1 in mammalian urine have remained intriguing questions for the past 40 years.
Free TyrS was first reported to be present in human urine by Tallan et al. (4). Similar findings were made subsequently for other mammalian species including rat, rabbit, and mouse (5,6). Because none of the mammalian arylsulfatases could effectively catalyze its desulfation (7,8), the free TyrS produced by mammalian cells has been generally considered a modified amino acid destined to be excreted. Concerning the biochemical origin of the free TyrS excreted, two distinct mechanisms for its generation have been suggested: i) the enzymatic sulfation of L-tyrosine forming free TyrS and ii) the turnover (degradation) of tyrosine sulfated proteins, thereby releasing free TyrS. Aiming at demonstrating the first of these two mechanisms, a great number of studies using various cell homogenates or purified aryl sulfotransferases have, however, persistently failed to reveal the enzymatic activity that catalyzes the sulfation of free L-tyrosine (1, 9 -12). Among the different mammalian aryl sulfotransferases characterized, only the rat liver type IV aryl sulfotransferase, also named the "tyrosine-ester sulfotransferase," can use tyrosine esters or N-terminally located tyrosine residues as substrates. This enzyme, however, cannot catalyze the sulfation of unmodified L-tyrosine (13). In view of the widespread occurrence of the post-translational tyrosine sulfation among proteins of multicellular eukaryotic organisms (14), it has become increasingly accepted that the free TyrS excreted in mammalian urine is probably derived exclusively from the degradation of tyrosine sulfated proteins (1,6,(15)(16)(17). In support of this hypothesis, free Tyr[ 35 S] was shown to be generated when tyrosine 35 S-sulfated peptides or proteins were either injected into rabbits (15) or added to the medium of cultured cells (18). However, to quantitatively account for the excreted TyrS reported to be approximately 28 mg/day/normal adult human (4), the turnover of a large amount of tyrosine sulfated proteins would be needed. This has therefore continued to raise the question whether the turnover of tyrosine sulfated proteins * This work was supported in part by funds from the University of Texas Health Center at Tyler and a Joint Research Grant from the Monbusho International Scientific Research Program, Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM  is truly the sole source of the free TyrS produced and released by mammalian animals.
By employing radioactive 3Ј-phosphoadenosine 5Ј-phospho-[ 35 S]sulfate (PAP[ 35 S]) as the sulfate donor, we have recently obtained evidence showing the enzymatic sulfation of L-p-tyrosine in several mammalian cell lines (19,20). We have further demonstrated that 3,4-dihydroxyphenyl-alanine (Dopa) and mtyrosine can be sulfated more efficiently by the sulfotransferase(s) present in the cytosol of HepG2 human hepatoma cells. 2 An important question therefore is whether, in mammalian cells, there is a single enzyme catalyzing the sulfation of all Dopa and tyrosine isomers or if there are distinct sulfotransferases responsible for the sulfation of different Dopa and tyrosine isomers. Furthermore, is (are) the enzyme(s) different from the previously reported sulfotransferase(s) (21)(22)(23)(24)?
In this paper, we report the purification, characterization, and molecular cloning of a single species of the enzyme, designated the "Dopa/tyrosine sulfotransferase," from the rat liver. Comparison of the nucleotide sequence and the deduced amino acid sequence of the cloned cDNA with sequences of known aryl (phenol) sulfotransferases, as well as the data from the biochemical characterization, demonstrated the Dopa/tyrosine sulfotransferase to be a novel enzyme.

EXPERIMENTAL PROCEDURES
Materials-L-Dopa, D-Dopa, L-p-tyrosine, D-p-tyrosine, DL-m-tyrosine, DL-o-tyrosine, ninhydrin, aprotinin, antipain, benzamidine, soybean trypsin inhibitor, phenylmethylsulfonyl fluoride, 2,6-dichloro-4nitrophenol (DCNP), ATP, adipic acid dihydrazide-agarose, 5Ј-AMP, Hepes, Ches, Taps, 3-[dimethyl(hydroxymethyl)methylamino]-2-hydroxypropanesulfonic acid (Ampso), and Caps were products of Sigma. PAPS was a generous gift from Unitika, Ltd. (Japan). L-Tyrosine (disodium salt) was purchased from Research Organics, Inc. Bio-Gel HTP hydroxylapatite and DEAE Bio-Gel A were from Bio-Rad Laboratories. ATP-agarose was prepared by coupling sodium periodate-oxidized ATP to adipic acid dihydrazide-agarose using the procedure of Lamed et al. (25). TyrS, dopamine-O-sulfate, and a mixture of L-Dopa 3-O-sulfate and L-Dopa 4-O-sulfate (collectively referred to as DopaS) were synthesized according to the procedure developed by Jevons (26). Rat liver Lambda ZAP II cDNA library and XL1-Blue MRFЈ Escherichia coli host strain were purchased from Stratagene. SuperScript Preamplification System and LipofectAMINE were from Life Technologies, Inc. Taq polymerase was purchased from Perkin-Elmer. Cycle sequencing kits were products of Applied Biosystems, Inc. The mammalian expression vector, pMSG⅐CMV, was kindly provided by Dr. Nakayama at Miyazaki Medical College. BcaBEST labeling kit, Exonuclease III, mung bean nuclease, and DNA ligation kit were products of Takara Shuzo. All restriction endonucleases were from New England Biolabs. Carrier-free sodium [ 35 S]sulfate and [␣-32 P]dCTP (3,000 Ci/mM) were from ICN Biomedicals. Chromatogram cellulose TLC plates were from Eastman Kodak Company. COS-7 SV40 transformed African green monkey kidney cells (ATCC CRL 1651) were obtained from the American Type Culture Collection. Rabbit antiserum against the rat liver Dopa/tyrosine sulfotransferase was prepared according to the procedure previously described (27). All other chemicals were of the highest grades commercially available.
Preparation of the Rat Liver Cytosol-Rat liver (190 g) rinsed thoroughly with ice-cold phosphate-buffered saline was ground through a USA standard testing sieve (35 mesh) and made into a 1:2 (w/v) suspension in a buffer A containing 10 mM Tris-HCl (pH 7.4), 250 mM sucrose, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride. The preparation was homogenized by 20 strokes in a tight fitting Teflon glass homogenizer placed on ice. The crude homogenate was centrifuged at 10,000 ϫ g for 20 min at 4°C, and the supernatant collected was further subjected to ultracentrifugation at 140,000 ϫ g for 2 h at 4°C. The supernatant containing cytosolic proteins was used for the purification as described below.
Purification of the Rat Liver Dopa/Tyrosine Sulfotransferase-Unless otherwise indicated, all buffer solutions used in the purification were at pH 7.4 and supplemented with 1 mM dithiothreitol. All operations described below were carried out at 4°C.
First Bio-Gel HTP Hydroxylapatite Column Chromatography-The rat liver cytosol (400 ml) was applied onto a Bio-Gel HTP column (4.5 ϫ 25 cm) pre-equilibrated with 10 mM Tris-HCl. After loading, 200 ml of 10 mM Tris-HCl was passed through the column to remove unbound proteins. The proteins bound on the column were eluted using a linear potassium phosphate buffer gradient composed of 350 ml each of 10 mM and 400 mM potassium phosphate buffer. The eluted fractions (spanning from 150 to 300 mM) containing the Dopa/tyrosine sulfotransferase activity were combined and dialyzed overnight against 10 mM Tris-HCl.
DEAE Bio-Gel A Anion Exchange Chromatography-The dialyzed fraction was applied onto a DEAE Bio-Gel A column (4.5 ϫ 25 cm) pre-equilibrated with 10 mM Tris-HCl. The bound proteins were eluted with an NaCl gradient composed of 350 ml each of 0 mM and 350 mM NaCl solution containing 10 mM Tris-HCl. The eluted fractions (spanning from 130 to 250 mM) containing the Dopa/tyrosine sulfotransferase activity were combined and dialyzed overnight against 10 mM Tris-HCl.
First ATP-Agarose Column Chromatography-The dialyzed fraction was loaded onto an ATP-agarose column (2.5 ϫ 4 cm). After loading, the bound proteins were eluted with an NaCl gradient composed of 120 ml each of 0 and 350 mM NaCl solution containing 10 mM Tris-HCl. The eluted fractions (spanning from 125 to 200 mM) containing the Dopa/ tyrosine sulfotransferase activity were pooled.
Second Bio-Gel HTP Column Chromatography-The Dopa/tyrosine sulfotransferase eluate from the first ATP agarose column was directly applied through a Bio-Gel HTP hydroxylapatite column (2.5 ϫ 10 cm) pre-equilibrated with 10 mM Tris-HCl. The bound proteins were eluted with a linear gradient composed of 150 ml each of 10 and 350 mM of potassium phosphate buffer. The fractions (spanning from 150 to 210 mM) containing the Dopa/tyrosine sulfotransferase activity were combined and dialyzed overnight against 10 mM Tris-HCl.
Second ATP-Agarose Column Chromatography-The eluate from the second Bio-Gel HTP column was applied onto a second ATP-agarose column (1.5 ϫ 5 cm) pre-equilibrated with 10 mM Tris-HCl. The bound proteins were eluted with a linear gradient composed of 100 ml each of 0 M and 75 M PAPS in 10 mM Tris-HCl. The Dopa/tyrosine sulfotransferase eluted throughout the entire PAPS concentration range was found to be electrophoretically homogeneous.
Enzymatic Assay-The activities of the Dopa/tyrosine sulfotransferase were assayed using PAP[ 35 S] as the sulfate donor. The standard assay mixture, with a final volume of 50 l, contained 50 mM Ampso-NaOH (pH 9.25), 250 mM sucrose, 25 mM NaF, 1 mM 5Ј-AMP, protease inhibitors (30 g/ml aprotinin, 30 g/ml antipain, 300 g/ml benzamidine, and 30 g/ml soybean trypsin inhibitor), 14 M PAP[ 35 S] (4.4 Ci/mmol), and 1 mM substrate (Dopa, tyrosine, etc.). The reaction was started by the addition of the enzyme preparation, allowed to proceed for 60 min at 37°C, and terminated by heating at 100°C for 3 min. The precipitates formed were cleared by centrifugation. The clear supernatant was subjected to the analysis of 35  , spotted onto a 20 ϫ 20-cm cellulose TLC plate, and analyzed according to a two-dimensional thinlayer separation procedure previously developed (28). Briefly, the plate was first subjected to high voltage electrophoresis (1,000 V for 70 min) in 7.8% (v/v) acetic acid/2.5% (v/v) 88% formic acid (pH 1.9). After electrophoresis, the plate was air-dried and subjected in the second dimension to ascending chromatography in n-butanol/88% formic acid/ isopropanol/H 2 O (3:1:1:1, v/v/v/v). Upon completion of the chromatography, the plate was sprayed with ninhydrin solution (0.5% in acetone). The ninhydrin-stained spot of the sulfated product was scraped off, suspended in 0.5-ml aliquots of H 2 O, and mixed with 4 ml of scintillation mixture (Ecolume, ICN Radiochemicals). The radioactivity associated with Tyr[ 35  N-terminal and Internal Partial Amino Acid Sequence Analysis-Nterminal amino acid sequence determination was performed according to the method of Matsudaira (29). Briefly, the purified Dopa/tyrosine sulfotransferase was subjected to SDS-PAGE (30) and electrotrans-ferred onto a Millipore Immobilon-P SQ membrane. The blotting was performed at a constant 200 mA for 6 h in 10 mM Caps-NaOH (pH 11). The blotted membrane was briefly stained with 0.1% Ponceau S in 5% acetic acid to reveal the protein band. After extensive washing with water, the membrane piece containing the bound Dopa/tyrosine sulfotransferase was used for the analysis of the N-terminal sequence. To determine the internal amino acid sequences, the Dopa/tyrosine sulfotransferase bound on the Immobilon-P membrane was digested with endoproteinase Lys-C, and the peptide fragments were purified using high performance liquid chromatography (HPLC). Purified peptide fragments were subjected to the N-terminal sequence determination according to the method of Matsudaira (29). The amino acid sequencing described above was performed using the sequencing facilities at The Rockefeller University.
Cloning of the Rat Liver Dopa/Tyrosine Sulfotransferase cDNAs-The reverse transcriptase-polymerase chain reaction (PCR) technique was employed to prepare the DNA probe used for cDNA library screening. Total RNA was isolated from the liver specimen of a 3-month-old male Sprague-Dawley rat using the method of Chomczynski and Sacchi (31). The first strand cDNA was synthesized using the SuperScript Preamplification system (Life Technologies, Inc.) with oligo(dT) as the primer. With the first strand cDNA as the template, a PCR reaction performed in a 50-l reaction mixture using the Taq  cycles of 30 s at 94°C, 30 s at 50°C, and 1 min at 72°C. The reaction mixture was applied onto a 2% agarose gel, separated by electrophoresis, and visualized by ethidium bromide staining. A major 452-nucleotide PCR product was detected and excised from the gel, and the DNA was isolated by spin filtration. Upon verification of the sequence by cycle sequencing, the purified PCR product was subcloned into the EcoRV site of pBluescript SK(Ϫ) and transformed into E. coli XL1-Blue MRFЈ. The recombinant plasmid was purified using alkali lysis method, and the PCR product insert was cut out by digestion with EcoRI and HindIII. The insert DNA was purified by agarose gel electrophoresis followed by spin filtration. The purified insert DNA was labeled with [␣-32 P]dCTP using random primers, and the labeled DNA was used as the probe for screening the cDNA encoding the Dopa/tyrosine sulfotransferase in a rat liver Lambda ZAP II cDNA library. Approximately 3 ϫ 10 6 plaques from the library were screened with 32 P-labeled DNA probe by hybridization on nylon membrane filters. Nylon membrane filter replicas of plaques/XL1-Blue MRFЈ grown on 150-mm Petri dishes upon prehybridization for 2 h at 65°C were hybridized with 32 P-labeled DNA probe overnight at 65°C. Hybridized membranes were washed once with 2 ϫ SSC (8.765 g of NaCl and 4.41 g of sodium citrate) plus 0.1% SDS and twice with 0.1 ϫ SSC plus 0.1% SDS at 65°C, followed by autoradiography to reveal the positive cDNA clones.
DNA Sequence Determination and Analysis-The positive cDNA clones were subjected to double-stranded sequencing according to the cycle sequencing method using Taq dye primer cycle sequencing kits (Applied Biosystems, Inc.) with Ϫ21M13 or M13 reverse primer. Serial deletional mutants were prepared by using Exonuclease III and mung bean nuclease according to the method of Henikoff (32). The nucleotide sequence, as well as the deduced amino acid sequence, of the full-length cDNA was analyzed using the E-mail servers at NCBI and EMBL for sequence homology to other known aryl sulfotransferases.
Transient Expression of the Dopa/Tyrosine Sulfotransferase in COS-7 Cells-COS-7 cells, normally maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, were used as the host cells for the expression of the enzyme. Dishes (60 mm) of COS-7 cells were individually transfected with 2 g of pMSG⅐CMV vector only or a pMSG⅐CMV derivative containing the sequence of D/TST-11 using the LipofectAMINE-mediated procedure. Incubation was for 18 h at 37°C, according to the manufacturer's instructions. The transfected cells were incubated at 37°C in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. At the end of a 48-h incubation, the cells were washed twice with phosphate-buffered saline and homogenized in buffer A containing 10 mM Tris-HCl (pH 7.4), 250 mM sucrose, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride. Aliquots of the homogenates prepared were assayed for the Dopa/tyrosine sulfotransferase activities or subjected to the Western blot analysis for the presence of the immunoreactive enzyme using rabbit anti-rat liver Dopa/ tyrosine sulfotransferase antiserum. The conditions for the Western blot analysis were as described previously (33).
Miscellaneous Methods-PAP[ 35 S] (15 Ci/mmol) was synthesized from ATP and [ 35 S]sulfate using the sulfate-activating enzymes ATP sulfurylase and adenosine 5Ј-phosphosulfate kinase from Bacillus stearothermophilus as described previously (34). SDS-PAGE was performed on 10% polyacrylamide gels using the method of Laemmli (30). The native molecular weight (M r ) of the purified Dopa/tyrosine sulfotransferase was determined by gel filtration chromatography using a Sephacryl S-200 column (2.6 ϫ 90 cm). Molecular weight standards including bovine serum albumin (M r 67,000), ovalbumin (M r 43,000), carbonic anhydrase (M r 29,000), chymotrypsinogen A (M r 25,800), and cytochrome c (M r 12,400) were used for calibration. Protein determination was based on the method of Bradford (35) with bovine serum albumin as the standard.

RESULTS
Purification of the Rat Liver Dopa/Tyrosine Sulfotransferase-Preliminary experiments showed that similar to several mammalian cell lines previously studied (19,20), the Dopa/tyrosine sulfotransferase was present predominantly in the cytosolic fraction of the rat liver. The specific activity of the enzyme purified from the rat liver cytosol, with L-Dopa as the substrate, was determined to be 2,153 pmol/min/mg protein, indicating a 760-fold purification over its specific activity in the rat liver cytosol (Table I). It was noted that, during the elution from the DEAE Bio-Gel A anion exchange chromatography step, the Dopa/tyrosine sulfotransferase activity was well separated from the major phenol sulfotransferase activity (data not shown). As shown in Fig. 1, the purified Dopa/tyrosine sulfotransferase migrated as a single protein band upon SDS-PAGE under reducing conditions. A key step during the purification of the Dopa/tyrosine sulfotransferase was the first ATP-agarose affinity chromatography, which furnished a considerably greater efficiency of purification than the other two types of column chromatography (cf. Fig. 1, lanes 2-4). A second ATP-agarose affinity chromatography was used in the final step of purification to remove small amounts of contaminating proteins and, at the same time, improve the specific activity of the purified Dopa/tyrosine sulfotransferase by removing denatured enzyme molecules that had lost the PAPS (nucleotide) binding activity. In this last purification step, the Dopa/tyrosine sulfotransferase was eluted not with an NaCl gradient but with a PAPS gradient. Because PAPS is a co-substrate needed during the sulfation reaction, its presence in the purified Dopa/ tyrosine sulfotransferase fraction may also help stabilize the Dopa/tyrosine sulfotransferase. No significant loss of the enzyme activity was observed during several weeks of storage at 4°C. However, freezing and thawing readily caused the puri- Rat Liver Dopa/Tyrosine Sulfotransferase fied Dopa/tyrosine sulfotransferase to lose its enzymatic activity.
Characterization of the Purified Rat Liver Dopa/Tyrosine Sulfotransferase-The purified rat liver Dopa/tyrosine sulfotransferase was subjected to characterization with respect to its physicochemical and enzymatic properties as described below.
Molecular Weight-Based on its electrophoretic mobility relative to the molecular weight standards co-electrophoresed during the SDS-PAGE under reducing conditions (Fig. 1), the minimum molecular weight of the purified Dopa/tyrosine sulfotransferase was determined to be approximately 33,000. Gel filtration chromatography revealed a molecular weight of approximately 34,000 for the native Dopa/tyrosine sulfotransferase (figure not shown). These results combined indicate that the purified Dopa/tyrosine sulfotransferase was present in monomeric form.
Partial Amino Acid Sequences-Repeated attempts to determine the N-terminal amino acid sequence of the purified Dopa/ tyrosine sulfotransferase were unsuccessful, indicating that the enzyme, similar to some other cytosolic sulfotransferases (36,37), is N-blocked. Upon digestion with endoproteinase Lys-C, three proteolytic fragments were purified by HPLC and sequenced. Fig. 2 shows the alignment of the three partial amino acid sequences with the homologous sequences of other sulfotransferases reported previously. Although the Dopa/tyrosine sulfotransferase displayed higher degrees of sequence homology to phenol (aryl) sulfotransferases from mammalian animals, some degrees of homology to the sequences of even the two plant flavonol sulfotransferases (fcFST3 and fcFST4Ј) were observed. Among the 18 sulfotransferases listed for comparison, one unknown rat liver sulfotransferase (cDNA clone ST1B1; previously identified by Yamazoe et al. (38)) contains sequences identical to the three partial amino acid sequences determined for the purified Dopa/tyrosine sulfotransferase. However, the deduced amino acid sequence of the full-length cDNA encoding the Dopa/tyrosine sulfotransferase (see below) was found to differ from that of clone ST1B1 at position 68 (with a glycine residue instead of a glutamic acid residue).
Enzymatic Sulfation of Tyrosine and Dopa Isomers-During the purification process, the activity of the Dopa/tyrosine sulfotransferase was determined using DL-m-tyrosine or L-Dopa as substrates. The purified enzyme was found to display enzymatic activities toward all tyrosine and Dopa isomers tested except DL-o-tyrosine.
pH Optimum-As shown in Fig. 3, with L-Dopa as the substrate, the pH optimum of the purified Dopa/tyrosine sulfotransferase was determined to be 9.25. Although considerably lower enzymatic activities were observed in Ches buffer when compared with other buffers (Taps, Ampso, and Caps) at cor-responding pH values, the pH optimum remained unchanged at 9.25.
Kinetic Parameters-To determine the apparent K m values of the Dopa/tyrosine sulfotransferase for L-Dopa and D-Dopa, the enzymatic assays were performed in the presence of varying concentrations of L-or D-Dopa with a constant 14 M of PAPS. Results obtained indicated that in the low concentration range, L-Dopa was sulfated at higher rates than was D-Dopa. When the concentration was above 0.25 mM, however, higher rates of sulfation were observed for D-Dopa. Based on the Lineweaver-Burk plots plotted using the data obtained, the apparent K m values of the Dopa/tyrosine sulfotransferase for L-Dopa and D-Dopa were determined to be 0.76 and 3.44 mM, respectively (Table III). Their corresponding V max values were calculated to be, respectively, 3,521 and 13,699 pmol/min/mg protein. The kinetic constants for p-nitrophenol and dopamine, two commonly used substrates for phenol sulfotransferases, were also determined. As shown in Table III, the apparent K m values of the Dopa/tyrosine sulfotransferase for p-nitrophenol and dopamine, calculated from the Lineweaver-Burk plots, were 30.9 and 0.24 mM, respectively. Their corresponding V max values were calculated to be, respectively, 125,000 and 5,435 pmol/min/mg protein (Table III). It is worthwhile pointing out that although the apparent K m values for L-Dopa, D-Dopa, and p-nitrophenol are different, their calculated V max /K m values are similar. Because the same concentration of the purified enzyme was used in these experiments, the K cat /K m or the catalytic efficiency of the Dopa/tyrosine sulfotransferase for these substrates is almost the same. With triiodothyronine as the substrate, a strong substrate inhibition effect was observed when the concentration of triiodothyronine was greater than 0.5 mM.
Effects of Cationic Salts on the Enzymatic Activity-The effects of cationic salts on the Dopa/tyrosine sulfotransferase activity were measured. The addition of different divalent cationic salts, such as MgCl 2 , MnCl 2 , and CoCl 2 , exerted virtually no effects on the activities of the purified enzyme. With L-Dopa as the substrate, increasing concentrations of NaCl or KCl resulted in the inhibition of the Dopa/tyrosine sulfotransferase activity. A 50% decrease in the enzymatic activity was found when the concentration of NaCl or KCl in the reaction mixture was increased to 200 mM.
Effect of DCNP on the Enzymatic Activity-DCNP, a commonly used inhibitor for aryl (phenol) sulfotransferases (1,2,21), was tested for its inhibitory effect on the Dopa/tyrosine sulfotransferase activity. It was found that the purified Dopa/ tyrosine sulfotransferase was inhibited by submillimolar levels of DCNP, with an IC 50 of approximately 7 ϫ 10 Ϫ4 M. A similar IC 50 value for DCNP has previously been determined for the thermolabile M form but not the thermostable P form of human phenol sulfotransferase (1,22).
Molecular Cloning of the Rat Liver Dopa/Tyrosine Sulfotransferase-To determine unequivocally the identity of the rat liver Dopa/tyrosine sulfotransferase as a novel enzyme, we have cloned and sequenced its cDNA. Repeated screening of the rat liver Lambda ZAP II cDNA library yielded 14 positive cDNA clones. The three largest cDNA inserts, ranging from 1,500 to 2,400 base pairs, were subjected to preliminary nucleotide sequencing. The analysis revealed that one of them, designated clone D/TST-11, contained an initiation codon and a 3Ј region encoding the poly(A) tail and thus appeared to contain the full-length sequence. The nucleotide and deduced amino acid sequences are presented in Fig. 4. Because the purified Dopa/tyrosine sulfotransferase was found to be N-blocked, the ATG codon encoding the N-terminal methionine residue was assigned based on i) the predicted molecular weight that matches the data from SDS-PAGE and gel filtration chromatography and ii) the sequence alignment in comparison with known aryl sulfotransferases (see below). The open reading frame, beginning at base residue 91, encompasses 897 nucleotides and encodes a 299-amino acid polypeptide. The predicted molecular weight, 34,762, is in agreement with the results (33,000 and 34,000, respectively) obtained through SDS-PAGE and gel filtration chromatography using the purified Dopa/ tyrosine sulfotransferase. The termination codon, located at nucleotide residues 988 -990, was followed by a 292-nucleotide 3Ј-untranslated sequence that includes a poly(A) tract. Two polyadenylation signals (ATTAAA) (39) located 180 and 19 nucleotides, respectively, upstream from the poly(A) tract were found. The authenticity of the cDNA was indicated by the inclusion of the three partial amino acid sequences obtained through direct amino acid sequencing of the purified Dopa/ tyrosine sulfotransferase, and by the expression of the functionally active recombinant enzyme that cross-reacted with the  antiserum against the purified Dopa/tyrosine sulfotransferase (see below). As shown in Fig. 5, the deduced amino acid sequence of the rat liver Dopa/tyrosine sulfotransferase (rD/TST) cDNA displays 72.2/52.6, 72.6/52.4, 72.5/51.9, and 71.9/53.6% similarity/identity to the amino acid sequences of rat liver phenol sulfotransferase, human thermolabile phenol sulfotransferase, human thermostable phenol sulfotransferase, and mouse phenol sulfotransferase (based on analysis using the Program Manual for the Wisconsin package, version 8). The deduced amino acid sequence of the Dopa/tyrosine sulfotransferase differed from that of clone ST1B1 (38) by a glycine residue instead of a glutamic acid residue at position 68.
Expression of the Cloned Rat Liver Dopa/Tyrosine Sulfotransferase in COS-7 Cells-The recombinant protein was expressed in COS-7 cells and subjected to functional characterization and examination of the immunoreactivity toward the rabbit antiserum against the purified rat liver Dopa/tyrosine sulfotransferase. As shown in Fig. 6, a 33-kDa protein crossreactive toward the antiserum against the purified Dopa/tyrosine sulfotransferase was expressed specifically when the COS-7 cells were transfected with an expression vector (pSMG⅐CMV) that contained the full-length cDNA encoding the Dopa/tyrosine sulfotransferase. When the cell homogenates were assayed for the Dopa/tyrosine sulfotransferase activity, it was found that the sample prepared from the cells transfected with the expression vector inserted with the full-length cDNA indeed exhibited a highly elevated Dopa/tyrosine sulfotransferase activity (Table IV). DISCUSSION Since the discovery of the excretion of free TyrS in human urine (4), the questions concerning the functional relevance and the formation of TyrS by the enzymatic sulfation of tyrosine have remained unresolved for nearly forty years. A consensus formed (following Huttner's discovery of the widespread occurrence of the post-translational tyrosine sulfation of eukaryotic proteins (14)) is that free TyrS is generated primarily through the turnover of tyrosine sulfated proteins in vivo. We have indeed demonstrated earlier (16) that exogenous tyrosine 35 S-sulfated proteins added to the medium could be endocytosed by cultured cells and degraded intracellularly to generate free Tyr[ 35 S]. A metabolic labeling experiment using the same cells, however, showed a considerable discrepancy between the amount of the free Tyr[ 35 S] generated and the amount of tyrosine 35 S-sulfated proteins turned over during a 48-h time course monitored. This finding had prompted our interest in investigating further the possibility of the sulfation of free tyrosine.
In our recent studies (19,20), we have obtained conclusive evidence that sulfation of L-p-tyrosine does occur in several mammalian cell lines. It is, however, unclear why mammalian cells should carry out the sulfation of an amino acid needed for protein synthesis. To convert L-p-tyrosine to L-p-TyrS, a compound destined for excretion (4), would seem to be counterproductive in terms of cellular economy. The question that should be raised then is whether L-p-tyrosine truly represents the physiological substrate of the enzyme, initially designated the "tyrosine sulfotransferase" (19). Using HepG2 human hepatoma cells as a model, we have demonstrated in a more recent study 2 that other tyrosine derivatives, e.g. Dopa and m-tyrosine isomers, are in fact better substrates for sulfation than is L-p-tyrosine. These results are summarized in the schematic diagram shown in Fig. 7. In view of the large number of aryl sulfotransferases that have been identified, it is tempting to ask whether, in mammalian cells, there is one single enzyme catalyzing the sulfation of all Dopa and tyrosine isomers or instead that there are multiple sulfotransferases responsible for the sulfation of individual Dopa and tyrosine isomers. To find an answer to this question, we have decided to isolate the enzyme(s) from rat liver for further characterization.
The rat liver has been more exhaustively studied with regard to aryl sulfotransferases (1, 2) than other mammalian tissues. At least six different types of aryl sulfotransferases have been identified and characterized (13,23,24). The rat liver is therefore the best model for investigating whether the Dopa/tyrosine sulfotransferase activities are associated with a new enzyme(s) or instead a known aryl sulfotransferase(s). In the present study, a single Dopa/tyrosine sulfotransferase was purified from the rat liver. It was noted that for all five chromatography steps during the purification, single symmetric peaks of elution of the Dopa/tyrosine sulfotransferase, as monitored by standard assays using either L-Dopa or DL-m-tyrosine as substrate, were observed. The existence of multiple enzymes catalyzing the sulfation of individual Dopa and tyrosine isomers, therefore, seemed unlikely. SDS-PAGE and gel filtration chromatography revealed the enzyme to be present in the monomeric form. In contrast to these data obtained with the purified FIG. 4. Nucleotide and deduced amino acid sequences of the rat liver Dopa/tyrosine sulfotransferase cDNA. Nucleotides are numbered in the 5Ј to 3Ј direction with the adenosine of the translation initiation codon designated as ϩ1. The translation stop codon is indicated by an asterisk. The polyadenylation signals and the two primer sequences used in the reverse transcriptase-PCR are underlined. The three partial amino acid sequences determined using the purified enzyme are double underlined.
Dopa/tyrosine sulfotransferase, most, if not all, of the known aryl sulfotransferases were shown to be present in the dimeric form (21,24). The purified rat liver Dopa/tyrosine sulfotransferase was found to be capable of catalyzing the sulfation of all Dopa and tyrosine isomers, except DL-O-tyrosine. The specific activity of the purified enzyme with L-Dopa as the substrate was 2,153.4 pmol/min/mg protein. This value is considerably lower than those previously reported for phenol sulfotransferases from rat liver with either simple phenols or monoamines as substrates (1,2,21,22). However, some sulfotransferases that utilize endogenous compound as substrates, e.g. the tyrosylprotein sulfotransferase (5,700 pmol/min/mg) (40) and the dopamine-sulfating sulfotransferase (7,735 pmol/min/ mg) (41), also displayed specific activities in the same order of magnitude as that determined for the Dopa/tyrosine sulfotransferase. Furthermore, because tyrosine and Dopa are important precursors for the synthesis of proteins and/or catecholamines, there may be regulatory mechanisms in vivo for the Dopa/tyrosine sulfotransferase activity. Thyroid hormones  6. Expression of the recombinant rat liver Dopa/tyrosine sulfotransferase in COS-7 cells. The figure shows the autoradiograph taken from the Immobilon-P membrane used in the Western blot analysis for the presence of the recombinant rat liver Dopa/tyrosine sulfotransferase. Samples analyzed were homogenates prepared from untransfected COS-7 cells (lanes 1 and 2), COS-7 cells transfected with pSMG⅐CMV only (lanes 3 and 4), and COS-7 cells transfected with pSMG⅐CMV harboring the full-length cDNA encoding the Dopa/tyrosine sulfotransferase (lanes 5-8). triiodothyronine and thyroxine, as well as dopamine and pnitrophenol, could also be used as substrates by the purified Dopa/tyrosine sulfotransferase. Although the broad substrate specificity seems to be a rule rather than exception for aryl sulfotransferases that have been studied, it should be pointed out that these latter compounds (thyroid hormones, dopamine, and p-nitrophenol) are more effectively used by other aryl (phenol) sulfotransferases (42,43) with K m values 2-3 orders of magnitude lower than those determined for the purified rat liver Dopa/tyrosine sulfotransferase. Furthermore, the Dopa/ tyrosine sulfotransferase represents the only known enzyme that is capable of catalyzing the sulfation of Dopa and tyrosine isomers. In contrast to the HepG2 Dopa/tyrosine sulfotransferase, which showed higher activities toward D-form Dopa and tyrosine isomers and a remarkable divalent cation dependence, 2 the rat liver enzyme displayed higher activities toward the L-form substrates and showed no significant changes in activity in the presence of a variety of divalent cations. Initial attempts to determine the N-terminal amino acid sequence of the purified Dopa/tyrosine sulfotransferase showed it to be N-blocked. Three internal partial amino acid sequences were obtained by sequencing the HPLC-purified fragments derived from the digestion of the purified enzyme with endoproteinase Lys-C. The alignment of the partial amino acid sequences of the rat liver Dopa/tyrosine sulfotransferase with the homologous sequences from other sulfotransferases provided the first clue that the Dopa/tyrosine sulfotransferase is a novel enzyme. The three partial amino acid sequences, however, completely matched those found in the deduced amino acid sequence of an unidentified sulfotransferase cDNA clone (ST1B1) reported by Yamazoe et al. (38). Because the ST1B1 cDNA clone encodes the only rat liver sulfotransferase that remains unidentified to date, we decided to clone and express it in COS-7 cells for functional characterization with respect to its identity as the Dopa/tyrosine sulfotransferase. Two regions, WDNKCKM and WKNYFTM, of the deduced amino acid sequence of the ST1B1 clone were chosen for designing degenerate oligonucleotide primers for reverse transcriptase-PCR. Using the 452-nucleotide PCR product as the probe for screening, a full-length cDNA clone was isolated and sequenced. The deduced amino acid sequence of the isolated cDNA contained the three partial amino acid sequences derived from the protein sequencing of the purified Dopa/tyrosine sulfotransferase and was found to be identical to that of clone ST1B1 except for a glycine residue instead of a glutamic acid residue at position 68. Whether this difference reflects the presence of isozymes in the rat liver remains to be clarified. The identity of the cDNA isolated was further verified by the expression in transfected COS-7 cells of a functional 33-kDa Dopa/tyrosine sulfotransferase that displayed immunologic cross-reactivity toward the antiserum against the purified rat liver Dopa/tyrosine sulfotransferase. These results have thus unequivocally confirmed the identity of the rat liver Dopa/tyrosine sulfotransferase as a novel enzyme, being distinct from all known sulfotransferases previously characterized.
The important question remains of whether the sulfation of Dopa and tyrosine isomers is functionally relevant. Although the precise physiological involvement of the Dopa/tyrosine sulfotransferase still awaits further clarification, some possibilities could be put forth by taking into account the metabolic roles of its substrates, in particular L-Dopa and L-m-tyrosine. L-Dopa is generally known as the biosynthetic precursor of catecholamines including dopamine, norepinephrine, and epinephrine (44). L-meta-Tyrosine has been shown to be present in vivo (45,46) and is capable of crossing the blood-brain barrier (47). Quantitative analysis showed that although L-p-tyrosine represents the predominant species, L-m-tyrosine constitutes a significant amount (2.8%) of the total tyrosine circulating in blood (46). Using bovine adrenal medulla extract or rat brain homogenate, it has been demonstrated that L-m-tyrosine was produced through the meta-hydroxylation of L-phenylalanine (48,49). Furthermore, in vivo studies have shown that L-mtyrosine could be converted to L-Dopa (50 -52) or m-tyramine (47,53), a decarboxylated product of L-m-tyrosine with neurotransmitter activity. Considering that L-Dopa and L-m-tyrosine are both involved in the biosynthesis of neurotransmitters, it would be important to regulate the concentrations of these compounds in vivo. A hypothetical role for the sulfation of L-Dopa and L-m-tyrosine therefore is that under normal circumstances, sulfation may be employed as a safeguard against the overproduction of L-Dopa and L-meta-tyrosine that if not prevented, might lead to the overproduction of catecholamines and consequently some neurological problems. When L-Dopa or L-m-tyrosine exceeds the normal concentration range, sulfation reaction may provide a mechanism by which they can be readily excreted. In this regard, the Dopa/tyrosine sulfotransferase may occupy a unique position related to the neurotransmitter metabolism. For other Dopa and tyrosine isomers, as well as thyroid hormones (triiodothyronine and thyroxine), a similar role for sulfation in facilitating their excretion can also be proposed. It is to be pointed out that the sulfation of dopamine and other catecholamines has been reported (2,54). The enzyme responsible for the sulfation of catecholamines, the M form phenol sulfotransferase, has been shown to be predominantly present in neuronal cells (41). Whereas the M form phenol sulfotransferase serves to catalyze the sulfation of catecholamines that may have already exerted their neurotransmitter function, the Dopa/tyrosine sulfotransferase discovered in our studies functions to catalyze the sulfation of their biosynthetic precursors (L-Dopa, L-p-tyrosine, and L-m-tyrosine), thereby preventing the overproduction of catecholamines.
Finally, it is to be noted that the co-elution with synthetic L-p-TyrS standard upon ion-exchange column chromatography was the major, if not exclusive, criteria used for the identification of free TyrS excreted in mammalian urine (4 -6). This procedure, however, is unlikely to provide enough resolution needed to distinguish between different sulfated tyrosine and/or Dopa isomers. It will be important to investigate, using more precise methods, the true identity (or identities) of the TyrS excreted in mammalian urine. Such information will be valuable in delineating the real substrate(s) for the Dopa/tyrosine sulfotransferase in vivo.