Towards the mechanism of trimeric purine nucleoside phosphorylases: stopped-flow studies of binding of multisubstrate analogue inhibitor - 2-amino-9-[2-(phosphonomethoxy)ethyl]-6-sulfanylpurine.

The binding of multisubstrate analogue inhibitor - 2-amino-9-[2-(phosphonomethoxy)ethyl]-6-sulfanylpurine (PME-6-thio-Gua) to purine nucleoside phosphorylase from Cellulomonas sp. at 20 degrees C, in 20 mM Hepes buffer with ionic strength adjusted to 50 mM using KCl, at several pH values between 6.5 and 8.2, was investigated using a stopped-flow spectrofluorimeter. The kinetic transients registered after mixing a protein solution with ligand solutions of different concentrations were simultaneously fitted by several association reaction models using nonlinear least-squares procedure based on numerical integration of the chemical kinetic equations appropriate for given model. It is concluded that binding of a PME-6-thio-Gua molecule by each of the binding sites is sufficiently well described by one-step process, with a model assuming interacting binding sites being more probable than a model assuming independent sites. The association rate constants derived from experimental data, assuming one step binding and independent sites, are decreasing with an increase in pH, changing from 30 to 6 microM(-1)s(-1) per binding site. The dissociation rate constants are in the range of 1-3 s(-1), and they are rather insensitive of changes in pH. Interestingly, for each pH value, the one-step binding model with interacting sites results in the association rate constant per site 1.5-4 times smaller for the binding of the first ligand molecule than that for the binding of the second one. Decrease of association constants with pH indicate that the enzyme does not prefer binding of the naturally occurring anionic form of the 6-thioguanine ring (pK(a) 8.7) resulting from a dissociation of N(1)-H. This finding supports the mechanism in which hydrogen bond interaction of N(1)-H with Glu204 (Glu 201 in mammalian PNPs) is crucial in the catalytic process. Results obtained also indicate that, in contrast to transition-state analogues, for which binding is followed by a conformational change, binding of multisubstrate analogue inhibitors to trimeric PNPs is a one-step process.

Stopped-flow studies of guanine binding by calf spleen purine nucleoside phosphorylase.

The binding of guanine to calf spleen purine nucleoside phosphorylase at 20 degrees C, in 20 mM Hepes-NaOH buffer, pH 7.0, at several ionic strength between 5 and 150 mM was investigated using a stopped-flow spectrofluorimeter. The kinetic transients registered after mixing a protein solution with ligand solutions of different concentrations were simultaneously fitted by several association reaction models using nonlinear least-squares procedure based on numerical integration of the chemical kinetic equations appropriate for given model. It is concluded that binding of a guanine molecule by each of the binding sites is a two-step process and that symmetrical trimeric calf spleen purine nucleoside phosphorylase represents a system of (identical) interacting binding sites. The interaction is visible through relations between the rate constants and non-additivity of changes in "molar" fluorescence for different forms of PNP-guanine complexes. It is also probable that electrostatic effects in guanine binding are weak, which indicates that it is the neutral form of the ligand which is bound and dissociated by PNP molecule.

Calf spleen purine nucleoside phosphorylase: structure of its ternary complex with an N(7)-acycloguanosine inhibitor and a phosphate anion.

The calf spleen purine nucleoside phosphorylase (PNP) ternary complex with an N(7)-acycloguanosine inhibitor and a phosphate ion has been crystallized in the cubic space group P2(1)3, with unit-cell parameter a = 94.11 A and one monomer per asymmetric unit. X-ray diffraction data were collected using synchrotron radiation (Station X31, EMBL Outstation, DESY, Hamburg). The crystal structure was refined to a resolution of 2.2 A and R and R(free) values of 17.5 and 24.5%, respectively. The acyclonucleoside inhibitor is bound in the active site in an inverted ('upside-down') orientation of the purine base compared with natural substrates. The side chain of Asp243 forms two hydrogen bonds with the base ring: N(delta) donates a hydrogen to N(3) and O(delta) accepts a hydrogen from the guanine N(2)-amino group. N(1)--H of the base is hydrogen bonded to O(epsilon) of Glu201, while N(9) accepts a hydrogen bond from Thr242 O(gamma). In addition, a water molecule (W417) bridges the N(2)-amino group of the base and O(epsilon) of Glu201. In the phosphate-binding site, a phosphate ion is bound to Ser33, His64, Arg84, His86, Ala116 and Ser220. The acyclic chain of the N(7)-acycloguanosine inhibitor is in a folded conformation and together with a water molecule (W388) occupies the pentose-binding site, with possible hydrogen bonds to Tyr88 O(eta) and His257 N(delta 1). This new binding mode fully accounts for the previously observed substrate properties of 7-beta-D-ribofuranosides of hypoxanthine and guanine. It also provides a new starting point for the design of inhibitors of PNP for therapeutic and other applications.

Purine nucleoside phosphorylases: properties, functions, and clinical aspects.

The ubiquitous purine nucleoside phosphorylases (PNPs) play a key role in the purine salvage pathway, and PNP deficiency in humans leads to an impairment of T-cell function, usually with no apparent effects on B-cell function. This review updates the properties of the enzymes from eukaryotes and a wide range of prokaryotes, including a tentative classification of the enzymes from various sources, based on three-dimensional structures in the solid state, subunit composition, amino acid sequences, and substrate specificities. Attention is drawn to the compelling need of quantitative experimental data on subunit composition in solution, binding constants, and stoichiometry of binding; order of ligand binding and release; and its possible relevance to the complex kinetics exhibited with some substrates. Mutations responsible for PNP deficiency are described, as well as clinical methods, including gene therapy, for corrections of this usually fatal disease. Substrate discrimination between enzymes from different sources is also being profited from for development of tumour-directed gene therapy. Detailed accounts are presented of design of potent inhibitors, largely nucleosides and acyclonucleosides, their phosphates and phosphonates, particularly of the human erythrocyte enzyme, some with Ki values in nanomolar and picomolar range, intended for induction of the immunodeficient state for clinical applications, such as prevention of host-versus-graft response in organ transplantations. Methods of assay of PNP activity are reviewed. Also described are applications of PNP from various sources as tools for the enzymatic synthesis of otherwise inaccessible therapeutic nucleoside analogues, as coupling enzymes for assays of orthophosphate in biological systems in the micromolar and submicromolar ranges, and for coupled assays of other enzyme systems.

Crystal structure of the purine nucleoside phosphorylase (PNP) from Cellulomonas sp. and its implication for the mechanism of trimeric PNPs.

The three-dimensional structure of the trimeric purine nucleoside phosphorylase (PNP) from Cellulomonas sp. has been determined by X-ray crystallography. The binary complex of the enzyme with orthophosphate was crystallized in the orthorhombic space group P212121 with unit cell dimensions a=64.1 A, b=108.9 A, c=119.3 A and an enzymatically active trimer in the asymmetric unit. X-ray data were collected at 4 degrees C using synchrotron radiation (EMBL/DESY, Hamburg). The structure was solved by molecular replacement, with the calf spleen PNP structure as a model, and refined at 2.2 A resolution. The ternary "dead-end" complex of the enzyme with orthophosphate and 8-iodoguanine was obtained by soaking crystals of the binary orthophosphate complex with the very weak substrate 8-iodoguanosine. Data were collected at 100 K with CuKalpha radiation, and the three-dimensional structure refined at 2.4 A resolution. Although the sequence of the Cellulomonas PNP shares only 33 % identity with the calf spleen enzyme, and almost no identity with the hexameric Escherichia coli PNP, all three enzymes have many common structural features, viz. the nine-stranded central beta-sheet, the positions of the active centres, and the geometrical arrangement of the ligands in the active centres. Some similarities of the surrounding helices also prevail. In Cellulomonas PNP, each of the three active centres per trimer is occupied by orthophosphate, and by orthophosphate and base, respectively, and small structural differences between monomers A, B and C are observed. This supports cooperativity between subunits (non-identity of binding sites) rather than existence of more than one binding site per monomer, as previously suggested for binding of phosphate by mammalian PNPs. The phosphate binding site is located between two conserved beta- and gamma-turns and consists of Ser46, Arg103, His105, Gly135 and Ser223, and one or two water molecules. The guanine base is recognized by a zig-zag pattern of possible hydrogen bonds, as follows: guanine N-1...Glu204 O(epsilon1)...guanine NH2...Glu204 O(epsilon2). The exocyclic O6 of the base is bridged via a water molecule to Asn246 N(delta), which accounts for the inhibitory, but lack of substrate, activity of adenosine. An alternative molecular mechanism for catalysis by trimeric PNPs is proposed, in which the key catalytic role is played by Glu204 (Glu201 in the calf and human enzymes), while Asn246 (Asn243 in the mammalian enzymes) supports binding of 6-oxopurines rather than catalysis. This mechanism, in contrast to that previously suggested, is consistent with the excellent substrate properties of N-7 substituted nucleosides, the specificity of trimeric PNPs versus 6-oxopurine nucleosides and the reported kinetic properties of Glu201/Ala and Asn243/Ala point variants of human PNP.

Interactions of calf spleen purine nucleoside phosphorylase with antiviral acyclic nucleoside phosphonate inhibitors: kinetics and emission studies.

Association between calf spleen purine nucleoside phosphorylase and a series of phosphonylalkoxyalkyl derivatives of purine bases was studied by inhibition kinetics and fluorimetric titrations. Dissociation constants, determined by fluorimetric titration in phosphate-free conditions, were lower than inhibition constants in 1 mM phosphate, and inhibition was still weaker in 50 mM phosphate, in accord with the postulated bisubstrate analogue character of this class of inhibitors.

Synthesis of 6-aryloxy- and 6-arylalkoxy-2-chloropurines and their interactions with purine nucleoside phosphorylase from Escherichia coli.

The phase transfer method was applied to perform the nucleophilic substitution of 2,6-dichloropurines by modified arylalkyl alcohol or phenols. Since under these conditions only the 6-halogen is exchanged, this method gives 2-chloro-6-aryloxy- and 2-chloro-6-arylalkoxy-purines. 2-Chloro-6-benzylthiopurine was synthesized by alkylation of 2-chloro-6-thiopurine with benzyl bromide. The stereoisomers of 2-chloro-6-(1-phenyl-1-ethoxy)purine were obtained from R- and S-enantiomers of sec.-phenylethylalcohol and 2,6-dichloropurine. All derivatives were tested for inhibition with purified hexameric E. coli purine nucleoside phosphorylase (PNP). For analogues showing IC50 < 10 microM, the type of inhibition and inhibition constants were determined. In all cases the experimental data were best described by the mixed-type inhibition model and the uncompetitive inhibition constant, Kiu, was found to be several-fold lower than the competitive inhibition constant, Kic. This effect seems to be due to the 6-aryloxy- or 6-arylalkoxy substituent, because a natural PNP substrate adenine, as well as 2-chloroadenine, show mixed type inhibition with almost the same inhibition constants Kiu and Kic. The most potent inhibition was observed for 6-benzylthio-2-chloro-, 6-benzyloxy-2-chloro-, 2-chloro-6-(2-phenyl-1-ethoxy), 2-chloro-6-(3-phenyl-1-propoxy)- and 2-chloro-6-ethoxypurines (Kiu = 0.4, 0.6, 1.4, 1.4 and 2.2 microM, respectively). The R-stereoisomer of 2-chloro-6-(1-pheny-1-ethoxy)purine has Kiu = 2.0 microM, whereas inhibition of its S counterpart is rather weak (IC50 > 12 microM). More rigid (e.g. phenoxy-), non-planar (cyclohexyloxy-), or more bulky (2,4,6-trimethylphenoxy-) substituents at position 6 of the purine base gave less potent inhibitors (IC50 = 26, 56 and > 100 microM, respectively). The derivatives are selective inhibitors of hexameric "high-molecular mass" PNPs because no inhibitory activity vs. trimeric Cellulomonas sp. PNP was detected. By establishing the ligand-dependent stabilization pattern of the E. coli PNP it was shown that the new derivatives, similarly as the natural purine bases, are able to form a dead-end ternary complex with the enzyme and orthophosphate. It was also shown that the derivatives are substrates in the reverse synthetic direction catalyzed by E. coli PNP.

Crystallization and preliminary X-ray studies of purine nucleoside phosphorylase from Cellulomonas sp.

The commercially available enzyme purine nucleoside phosphorylase (PNP) from Cellulomonas sp. was purified by ion--exchange chromatography, partially sequenced and crystallized in two different crystal forms using the hanging-drop vapour-diffusion technique. Crystal form A grows as polyeders and/or cubes in the cubic space group P4232 with unit-cell dimension a = 162.5 A. Crystal form B appears as thick plates in the space group P212121 with unit-cell dimensions a = 63.2, b = 108.3 and c = 117.4 A. Both crystal forms contain three monomers (one trimer) in the asymmetric unit.

Crystal structure of the ternary complex of E. coli purine nucleoside phosphorylase with formycin B, a structural analogue of the substrate inosine, and phosphate (Sulphate) at 2.1 A resolution.

The ternary complex of purine nucleoside phosphorylase from E. coli with formycin B and a sulphate or phosphate ion crystallized in the hexagonal space group P6122 with unit cell dimensions a=123.11, c=241.22 A and three monomers per asymmetric unit. The biologically active hexamer is formed through 2-fold crystallographic symmetry, constituting a trimer of dimers. High-resolution X-ray diffraction data were collected using synchrotron radiation (Daresbury, England). The crystal structure was determined by molecular replacement and refined at 2.1 A resolution to an R-value of 0.196. There is one active centre per monomer, composed of residues belonging to two subunits of one dimer. The phosphate binding site is strongly positively charged and consists of three arginine residues (Arg24, Arg87 and Arg43 from a neighbouring subunit), Ser90 and Gly20. It is occupied by a sulphate or phosphate anion, each oxygen atom of which accepts at least two hydrogen bonds or salt-bridges. The sulphate or phosphate anion is also in direct contact with the ribose moiety of formycin B. The ribose binding site is composed of Ser90, Met180, Glu181 and His4, the latter belonging to the neighbouring subunit. The base binding site is exposed to solvent, and the base is unspecifically bound through a chain of water molecules and aromatic-aromatic interactions. In all monomers the nucleosides are in the high syn conformation about the glycosidic bonds with chi in the range 100 to 130 degrees. The architecture of the active centre is in line with the known broad specificity and the kinetic properties of E. coli PNP.

7-Deazapurine 2'-deoxyribofuranosides are noncleavable competitive inhibitors of Escherichia coli purine nucleoside phosphorylase (PNP).

A series of 7-deazapurine 2'-deoxyribofuranosides were synthesized according to already known procedures and their substrate and inhibitor properties with purified E. coli purine nucleoside phosphorylase were examined. In agreement with previous findings, substrate activity was not detected for any of the compounds tested. Most of the nucleosides showed weak inhibition in the preliminary screening, i.e. at a concentration of about 100 microM. However some combinations of 6-chloro, 6-amino or 6-methoxy substituents with bulky hydrophobic groups at position 7 of the base and/or chloro, amino, methoxy or methylthio group at position 2 markedly enhanced affinity of such modified nucleosides for the E. coli enzyme. The most potent inhibition was observed for two nucleosides: 6-chloro- and 2-amino-6-chloro-7-deazapurine 2'-deoxyribofuranosides that show inhibition constants Ki = 2.4 and 2.3 microM, respectively. Several other compounds were also found to be good inhibitors, with inhibition constants in the range 5-50 microM. In all instances the inhibition was competitive vs. the nucleoside substrate 7-methylguanosine. Inhibition constants for 7-deazapurine nucleosides are in general several-fold lower than those observed for their purine counterparts. Therefore 7-deaza modification together with substitutions at positions 2, 6 and 7 of the base is a very promising approach to obtain competitive noncleavable inhibitors of E. coli PNP that may bind to the enzyme with inhibition constants in the microM range.

Crystal structure of calf spleen purine nucleoside phosphorylase in a complex with hypoxanthine at 2.15 A resolution.

Trimeric calf spleen purine nucleoside phosphorylase has been complexed with hypoxanthine via phosphorolysis of inosine in the presence of phosphate. The resulting, "Michaelis" complex (three hypoxanthine molecules per trimer), presumed to be formed under these conditions, crystallized in the cubic space group P2(1)3, with unit cell dimension a = 94.11 A and one monomer in the asymmetric crystal unit; the biologically active trimer is located on the crystallographic 3-fold axis. High-resolution X-ray diffraction data were collected using synchrotron radiation (EMBL outstation, Hamburg, c/o DESY). The crystal structure has been determined by molecular replacement and refined at 2.15 A resolution to an R-value of 0.18. In the hypoxanthine binding site, a cis-peptide bond between Asn243 and Lys244 is observed. Side-chains of GIu201 and Asn243, as well as one integral water molecule located in the base binding site, form hydrogen bonds with the hypoxanthine N-1 H, N-7 H and O-6. A second water molecule links the base positions N-3 and N-9 with an adjacent pocket, which presumably is the phosphate-binding site. This pocket is filled completely by a cluster of six water molecules. Hence all possible donor/acceptor-positions of hypoxanthine are saturated by hydrogen-bonding to protein side-chains or integral water molecules. Purine nucleoside phosphorylase isolated form human tissues is a primary target for chemotherapeutic intervention, and the more stable calf enzyme has similar physico-chemical and kinetic properties, as well as response to inhibitors. Hence the high-resolution structure presented here may serve for design of inhibitors with potential pharmacological applications.

Nicotinamide riboside, an unusual, non-typical, substrate of purified purine-nucleoside phosphorylases.

Nicotinamide 1-beta-D-riboside (Nir), the cationic, reducible moiety of the coenzyme NAD+, has been confirmed as an unusual substrate for purified purine-nucleoside phosphorylase (PNP) from a mammalian source (calf spleen). It is also a substrate of the enzyme from Escherichia coli. The Km values at pH 7, 1.48 mM and 0.62 mM, respectively, were 1-2 orders of magnitude higher than for the natural substrate inosine, but the Vmax values were comparable, 96% and 35% that for Ino. The pseudo first-order rate constants, Vmax/Km, were 1.1% and 2.5% for the calf spleen and E. coli enzymes. The aglycon, nicotinamide, was neither a substrate nor an inhibitor of PNP. Nir was a weak inhibitor of inosine phosphorolysis catalyzed by both enzymes, with Ki values close to the Km for its phosphorolysis, consistent with simple competitive inhibition; this was further confirmed by Dixon plots. Phosphorolysis of the fluorescent positively charged substrate 7-methylguanosine was also inhibited in a competitive manner by both Ino and Nir. Phosphorolysis of Nir by both enzymes was inhibited competitively by several specific inhibitors of calf spleen and E. coli PNP, with Ki values similar to those for inhibition of other natural substrates. The pH dependence of the kinetic constants for the phosphorolysis of Nir and of a variety of other substrates, was extensively investigated, particularly in the alkaline pH range, where Nir exhibited abnormally high substrate activity relative to the reduced reaction rates of both enzymes towards other anionic or neutral substrates. The overall results are discussed in relation to present concepts regarding binding and phosphorolysis of substrates by PNP based on crystallographic data of enzyme-inhibitor complexes, and current studies on enzymatic and nonenzymatic mechanisms of the cleavage of the Nir glycosidic bond.

Kinetics of phosphorolysis of 3-(beta-D-ribofuranosyl)adenine and 3-(beta-D-ribofuranosyl)hypoxanthine, non-conventional substrates of purine-nucleoside phosphorylase.

The properties of two non-conventional substrates of the calf-spleen and Escherichia coli purine nucleoside phosphorylases (PNP), 3-(beta-D-ribofuranosyl)adenine (RibfAde) and 3-(beta-D-ribofuranosyl)hypoxanthine (RibfHyp), are described. In contrast to Ado, RibfAde is a substrate for the mammalian enzyme. With the calf enzyme, the pseudo-first-order rate constants (Vmax/K(m)) for phosphorolysis of RibfAde and RibfHyp are 3% and 13%, respectively, that for phosphorolysis of Ino, while for E. coli PNP the corresponding values are 22% and 30%, respectively. The Michaelis constants (K(m)) for RibfAde were 800 microM (calf PNP) and 150 microM (E. coli PNP). For RibfHyp, the corresponding K(m) values were 220 microM and 260 microM. Two well-characterized inhibitors of calf spleen PNP [9-(2-fluoro-3,4-dihydroxybutyl)guanine] and E. coli PNP (formycin A) were found to inhibit phosphorolysis of RibfAde and RibfHyp with the same inhibition constants as for Ino. Moreover, the inhibition was competitive, which indicates that phosphorolysis of 3-beta-nucleosides occurs at the same active site as for the natural substrate Ino. In particular, the substrate properties of both 3-beta-nucleosides are consistent with their binding to the enzyme in the conformation anti to the imidazole ring about the glycosidic bond, which is superimposable on the structure of natural 9-beta-nucleosides in the conformation anti to the pyrimidine ring. The results are examined in relation to present concepts regarding the binding of substrates and inhibitors at the active site(s) of these enzymes.

2-Chloro-2'-deoxyadenosine (cladribine) and its analogues are good substrates and potent selective inhibitors of Escherichia coli purine-nucleoside phosphorylase.

2-Chloro-2'-deoxyadenosine (CldAdo), a nucleoside that has proven useful in the treatment of several chronic lymphoid malignancies, and its analogue, 2-bromo-2'-deoxyadenosine, are both effective inhibitors of the bacterial (Escherichia coli) purine-nucleoside phosphorylase (PNP), with Ki values of 4.5 microM and 6.3 microM, respectively. The examination of a series of base-modified analogues of CldAdo has shown that several other compounds have similar inhibitor properties, and has indicated that 6-benzyloxy-2-chloro-9-(2'-deoxy-beta-D-ribofuranosyl)purine is the most potent inhibitor with a Ki value of 0.5 microM, competitive with respect to inosine (Ino). CldAdo itself and its base-modified analogues, discounting those substituted at C(8), are also substrates for the E. coli PNP and undergo rapid glycosidic bond cleavage. CldAdo is degraded with substrate efficiency, i.e. Vmax/Km similar to that observed for Ino (130%), although the individual kinetic constants, Km and Vmax, are both approximately an order of magnitude lower than for Ino. All compounds tested are totally inactive as substrates and inhibitors for mammalian (calf spleen) PNP and therefore constitute a new class of potent selective, although cleavable, inhibitors of bacterial phosphorylases. 8-Bromo-2-chloro-2'-deoxyadenosine and 8-thio-2-chloro-2'-deoxyadenosine are the only base-modified CldAdo derivatives showing inhibitory activity against MOLT-3 (acute T-cell leukemia) and U-937 (histiocytic lymphoma) cells and, as shown in this study, are resistant to degradation by E. coli PNP. The above-mentioned results suggest that both analogues could be effective as oral cytotoxic agents that are noncleavable by enteric bacteria.

Calf spleen purine nucleoside phosphorylase: purification, sequence and crystal structure of its complex with an N(7)-acycloguanosine inhibitor.

Calf spleen purine nucleoside phosphorylase was purified to homogeneity and its amino acid sequence was determined. The complex of the enzyme with an N(7)-acycloguanosine inhibitor crystallized in the cubic space group P2(1)3, with unit cell dimension a = 94.02 A and one monomer in the asymmetric crystal unit. The biologically active trimer is formed by the crystallographic three-fold axis. The structure was solved by molecular replacement methods, using the model of the human erythrocyte enzyme, and refined at a resolution of 2.9 A to an R-factor of 0.21. The orientation of the inhibitor at the active site is examined in relation to the catalytic activity of the enzyme in the phosphorolysis of N(7)-beta-D-purine nucleosides.

Fluorescence of tyrosine and tryptophan in proteins using one- and two-photon excitation.

We examined the emission spectra of tyrosine- and tryptophan-containing proteins using one-photon (270-310 nm) and two-photon (565-610 nm) excitation. Emission spectra for two-photon excitation of native and denatured human serum albumin and of three purine nucleoside phosphorylases indicated an absence of the tyrosine emission normally seen for one-photon excitation below 290 nm. We examined the one-photon and two-photon excitation spectra of tyrosine-tryptophan mixtures to determine the origin of selective excitation of the tryptophan residues. These results confirmed a short-wavelength shift of the tyrosine two-photon excitation spectrum relative to that of tryptophan, as recently reported by Rehms and Callis (1993) Chem. Phys. Lett. 208, 276-282.

Purine nucleoside phosphorylase: inhibition by purine N(7)- and N(9)-acyclonucleosides; and substrate properties of 7-beta-D-ribofuranosylguanine and 7-beta-D-ribofuranosylhypoxanthine.

A series of 10 N(7)- and N(9)-acyclonucleosides of guanine and 8-substituted guanines (8-Br, 8-SH and 8-NH2), and two N(7)-acyclonucleosides of hypoxanthine, were tested for their ability to inhibit purine nucleoside phosphorylase (PNP) (E.C. 2.4.2.1) from human erythrocytes and rabbit kidney. The acyclic chains contained a nitrogen in place of a carbon at the 3', 4' or 5' position and, in one case, an ether oxygen at the 2' position. Most striking was the finding that one of the N(7)-acyclonucleoside analogues, 7-[(1,3-dihydroxypropyl-2)amino]ethylguanine, proved to be a 3-fold more effective inhibitor than its corresponding N(9) counterpart, with Ki = 5 vs 14 microM for the human enzyme and 0.7 vs 2.3 microM for the rabbit enzyme. Both analogues, as well as the others examined, inhibited phosphorolysis competitively with respect to nucleoside substrates (inosine with the human enzyme and guanosine with the rabbit enzyme). The foregoing logically led to the finding that the 7-beta-D-ribosides of guanine (N7Guo) and hypoxanthine (N7Ino) were weak substrates of PNP from human erythrocytes, calf spleen and E. coli. With the human enzyme the pseudo-first-order rate constants (Vmax/Km) for phosphorolysis of N7Guo and N7Ino were 0.08 and 0.02% that for Ino. The Michaelis constants (Km) for N7Guo were 27 (calf PNP), 108 (human PNP) and 450 microM (E. coli PNP). For N7Ino the corresponding Km values were 1.52, 1.26 and 0.64 mM. Four previously well-characterized N(9)-acyclonucleoside inhibitors of calf spleen PNP were found to inhibit phosphorolysis of N7Ino by the same enzyme 2-10-fold more effectively than the parent Ino. The overall results, along with the known excellent substrate properties of N(7)-alkyl- Guo and Ino (Bzowska et al. J Biol Chem 263, 9212-9217, 1988), were examined in relation to present concepts regarding binding of substrates and inhibitors at the active site(s) of these enzymes.

Linear free energy relationships for N(7)-substituted guanosines as substrates of calf spleen purine nucleoside phosphorylase. Possible role of N(7)-protonation as an intermediary in phosphorolysis.

Quantitative structure-activity relationships (QSAR) for a series of N(7)-substituted guanosines as substrates for calf spleen purine nucleoside phosphorylase (PNP) were developed, and compared with those for acid hydrolysis of these analogues. There is no correlation between the rates for enzymatic phosphorolysis and acid hydrolysis, indicating that for the enzymatic reaction labilization of the glycosidic bond is not the only, nor the predominant, effect of N(7)-substitution. Multiple regression analysis of the enzymatic process revealed that optimal substrate properties (minimal Michaelis constant) are associated with the Taft electronic constant equal zero and a substituent size, parametrized by the Taft steric constant, smaller than that for a methyl group. These results support the hypothesis of protonation of the N(7)-position of the base by the enzyme as a catalytic mechanism for calf spleen PNP. Attention is drawn to the postulated similar mechanism of action of other purine N-glycosidases, including plant antiviral proteins which function as RNA N-glycosidases, and possibly some DNA N-glycosidases which function as repair enzymes.