Identification and purification of sulfotransferases for 20-hydroxysteroid from the larval fat body of a fleshfly, Sarcophaga peregrina.

Sulfotransferase (ST) activity for 20-hydroxyecdysone (20E) was identified in a larval fat body lysate of the fleshfly, Sarcophaga peregrina, but not in the hemolymph. The activity was highly sensitive to 2,6-dichloro-4-nitrophenol (DCNP) (IC50=0.61 microM), a specific inhibitor of phenol ST (P-ST), but insensitive to triethylamine, a hydroxysteroid ST inhibitor. These results suggest that 20E-specific ST enzymes belong to the P-ST family, despite the fact that 20E is a hydroxysteroid. In addition to 20E ST activity, a relatively high level of 2-naphthol ST activity was detected in the fat body lysate. The ST activity for both substrates transiently decreased to the 50% of maximal levels, 6 hrs after induction of pupation. The ST enzymes were separated on a DEAE-cellulose column. The 20E-ST enzymes were eluted around 50 mM KCl as two separate peaks of close proximity and the P-ST was eluted at 0.1 M KCl. The 20E ST enzymes were further purified using 3'-phosphoadenosine 5'-phosphate (PAP)-agarose affinity column chromatography. Both of the eluted active fractions demonstrated 43-kDa proteins on SDS-polyacrylamide gel. Photoaffinity labeling with [35S]-3'-phosphoadenosine 5'-phosphosulfate (PAPS) showed 43-kDa bands in the fat body lysate, as well as in the purified fractions. These results suggest that the 43-kDa proteins catalyze 20E sulfation within the fat body of S. peregrina.

Purification and characterization of a lymph node sulfotransferase responsible for 6-O-sulfation of the galactose residues in 2'-fucosyllactose and other sialyl LewisX-related sugars.

A microsomal galactose-6-O-sulfotransferase (Gal-6-O-Stase) from porcine lymph nodes, able to transfer the sulfate group from adenosine 3'-phosphate 5'-phosphosulphate (PAPS) onto 2'-fucosyllactose (2'-FL) and other sialyl LewisX (sLex)-related sugars, has been purified and characterized. The enzyme was purified to about 35,000-fold by a combination of conventional and affinity chromatographic steps. The purified enzyme preparation exhibited two protein bands at around 80-90 and 170 kDa on 7.5% SDS-PAGE under reducing conditions. Both of these protein bands always comigrated in the gel when peak fractions containing Gal-6-O-Stase activity from the 3',5'-ADP-agarose column were subjected to 6% SDS-PAGE under reducing conditions. These protein bands also showed similar binding patterns to WGA (wheat germ agglutinin), Con A (concanvalin A), and EBA (elderberry agglutinin). Similarly, when the enzyme preparation after the hydroxylapatite step was photolabeled with 8-azido-[32P]-PAPS, both 80-90 and 170 kDa protein bands were labeled in a specific manner. These results suggest a possible association of these two protein bands with the enzyme activity. The carbohydrate substrate specificity of this enzyme suggests that it is well suited to catalyze the sulphonation at the C-6 position of the galactose residues of oligosaccharides that are structurally similar to sLex. Furthermore, a survey of several porcine organs revealed that this enzyme was selectively expressed in lymphoid tissues such as lymph nodes (peripheral and mesenteric) and spleen. These findings suggest that this enzyme may be involved in the assembly of 3'-sialyl-6'-sulfo Lewisx, the major capping group of HEV-ligands for L-selectin.

Kinetics of PAPS translocase: evidence for an antiport mechanism.

In order to gain an understanding of the mechanisms involved in the transfer of 3'-phosphoadenosine 5'-phosphosulfate (PAPS) from the cytosol where it is synthesized to the Golgi lumen where it serves as the universal sulfate donor for sulfate ester formation in higher organisms, we have undertaken a kinetic characterization of the PAPS translocase from rat liver Golgi. Analyzing the PAS translocase activity in both intact Golgi vesicles and in a reconstituted liposome system, we have determined a number of physical and kinetic parameters. Strong competitive inhibition in zero-trans uptake experiments only with beta-methylene PAPS and adenosine 3',5'-biphosphate (PAP) suggest the transporter is highly specific for the 3'-phosphate. The demonstration of trans acceleration as observed by stimulation of transport activity under exchange conditions suggests that the translocase is a carrier with distinct binding sites accessible from both faces of the membrane. The behavior of the PAPS translocase in the presence of equilibrium concentrations of PAP supports the function of an antiport mechanism. Thus the translocase is characterized by its kinetic properties as a specific transporter of PAPS which acts through an antiport mechanism with PAP as the returning ligand. This characterization of the transport activity has proved instrumental in the identification of an approximate 230 kDa Golgi membrane protein as the PAPS translocase protein [Ozeran, J.D., Westley, J., & Schwartz, N.B. (1996) Biochemistry 35, 3695-3703 (accompanying paper)].

Identification and partial purification of PAPS translocase.

Sulfation of all macromolecules in higher organisms requires the high-energy donor 3'-phosphoadenosine 5'-phosphosulfate (PAPS). PAPS is synthesized via the sequential actions of two cytoplasmic enzymes, ATP sulfurylase and APS kinase, and then must be transferred across the Golgi membrane for utilization by lumenal sulfotransferases. Following the kinetic characterization of the PAPS translocase as a specific transporter that act through an antiport mechanism with PAP as the returning ligand [Ozeran, J.D., Westley, J., Schwartz, N.B. (1996) Biochemistry 35, 3685-3694 (accompanying paper)], the present study describes the identification and physical characterization of the PAPS translocase from rat liver Golgi membranes. The following evidence suggests the PAPS translocase is a membrane spanning protein of approximately 230 kDa: isolation by affinity chromatography on beta-methylene PAPS matrices of a 230 kDa Golgi membrane protein concomitant with PAPS translocase activity; demonstration that the 230 kDa protein possesses the only PAPS binding site accessible to the cytoplasmic face of intact Golgi membranes, while several other PAPS binding proteins are labeled in solubilized membrane preparations; reduction in size of the 230 kDa membrane protein and loss of PAPS translocase activity following protease treatment; estimation via hydrodynamic analysis of a molecular size of the membrane protein associated with PAPS translocase activity; and correlation of beta-methylene PAPS binding and labeling of the 230 kDa Golgi protein with PAPS translocase activity in artificial liposomes. These and the accompanying data have permitted the identification of the first of a potentially large class of Golgi membrane nucleotide-metabolite transporters.

Synthesis and utilization of a nonhydrolyzable phosphoadenosine phosphosulfate analog.

3'-Phosphoadenosine 5'-phosphosulfate (PAPS) functions as the high-energy sulfate donor for sulfate ester synthesis in all higher organisms. This activated sulfate, like its adenosine 5'-phosphosulfate precursor, is both chemically labile and vulnerable to sulfohydrolase degradation. These obstacles have limited the utility of the native PAPS in the purification and mechanistic description of the numerous PAPS-utilizing enzymes. This paper describes the synthesis of the 2'- and 3'-isomers of a nonhydrolysable, and thus stable, PAPS analog, beta-methylene-PAPS, from the previously described beta-methylene-APS (L. Callahan et al., Anal. Biochem. 177, 67-71, 1989). The method involves phosphorylation of beta-methylene-APS with trimetaphosphate and separation of the resulting mixed 2'(3')-isomers by ion-pair reverse-phase HPLC. The utilization of this analog as an inhibitor of APS kinase and PAPS translocase, two of the numerous PAPS-utilizing activities, as well as an affinity ligand for purification of APS kinase, is described.

Dibasic amines as competitive ions improve the resolution between polyanionic nucleotides.

Aliphatic diamines when used as single ion pairing reagents were capable of resolving 3'-,5'- and 2'-,5'- nucleotidyl diphosphates from one another while conventional ion pairing reagents did not separate these positional isomers. The use of 1,2-diamines resulted in the greatest resolution while increasing spacing between the amino groups progressively reduced the resolution while increasing the retention volume. A competitive ion pairing system was also developed using triethylamine as an additional ion pairing reagent. Using this system ethylenediamine, 1,2- and 1,3-diaminopropane were nearly equivalent in their ability to resolve adenosine 3'-phosphate 5'-phosphate, from adenosine 2'-phosphate 5'-phosphate, and adenosine 3'-phosphate 5'-beta-methylenephosphosulfate (3'-mePAPS) from adenosine 2'-phosphate 5'-beta-methylenephosphosulfate (2'-mePAPS), respectively. The ability to easily resolve these positional isomers allows the use of a more simplified synthetic procedure that does not involve the use selective protecting groups to specifically phosphorylate the 2' or 3' hydroxyl group. We have used this procedure on a semipreparative scale to obtain small quantities of both mePAPS and 2'-mePAPS for use in enzymatic studies.

Inhibition of M and P phenol sulfotransferase by analogues of 3'-phosphoadenosine-5'-phosphosulfate.

Structural analogues of the sulfate donor 3'-phosphoadenosine-5'-phosphosulfate (3',5'-PAPS) were examined for their ability to inhibit dopamine and phenol sulfation by the M and P forms of phenol sulfotransferase (PST), respectively. The Ki values for each of the adenosine derivatives were calculated from the rate equation for PST. For both M and P PST, the naturally occurring product 3'-phosphoadenosine-5'-phosphate, (3',5'-PAP), was shown to be the most effective inhibitor. The weakest inhibitors of the two sulfotransferases were 5'-adenosine phosphosulfate and the three AMP derivatives, which were less than 1,000 times as effective as 3',5'-PAP. 5'-ATP, 2',5'-PAPS, 2',5'-PAP, and 5'-ADP were similar in their inhibition of M and P PST and were all approximately 100 times less effective than the natural end product. These data reveal that there is a rigid structural requirement for binding of the ribose portion of adenosine to both M and P PST that involves the groups on both the 3' and 5' positions. The effectiveness of binding to the two enzymes may depend on both steric factors as well as the distribution of negative charges on the ribose ring.