C-glycosylphosphonyl analogs of 5-phosphorylribose 1-α-diphosphate: II. The "2-deoxy" analog

Article abstract of DOI:10.1006/bioo.1996.0004

D-ribo-2,5-anhydro-1,3-dideoxy-1-(((dihydroxyphosphinyl)oxy) hydroxyphosphinyl)-6-phosphorylhexitol, compound 4, along with 25% of its D-arabino epimer, was synthesized in 10 steps from 2-deoxyribonolactone. Compound 4 represents the 2-deoxy analog of the central metabolite 5-phosphorylribose 1-α-diphosphate (PRPP). The analog is a competitive inhibitor of yeast orotate phosphoribosyltransferase (PRTase) with K_i/K_m(PRPP) = 24 and is thus comparable to other C-glycosylphosphonyl PRPP analogs reported previously (R. W. McClard, and J. F. Witte (1990) Bioorg. Chem. 18, 165-178). The analog is an even better inhibitor of both human hypoxanthine-guanine PRTase and glutamine : amido PRTase from Escherichia coli with K_i/K_m(PRPP) values of 16 and 3, respectively. These results support the argument that the cis-diol grouping of nucleotides (or PRPP) plays a very minor role in binding of these substrates to PRTases.

Full text of DOI:10.1006/bioo.1996.0004

BIOORGANIC CHEMISTRY 24, 29–38 (1996)
ARTICLE NO. 0004
C-Glycosylphosphonyl Analogs of 5-Phosphorylribose
1-Ͱ-Diphosphate
II. The ‘‘2-Deoxy’’ Analog1
J
OHN F. WITTE AND
R
ONALD W. M
CCLARD
Arthur F. Scott Laboratory of Chemistry, Reed College, Portland, Oregon 97202
Received March 27, 1995
D-ribo-2,5-anhydro-1,3-dideoxy-1-(((dihydroxyphosphinyl)oxy)hydroxyphosphinyl)-6-phos-
phorylhexitol, compound 4, along with 25% of its
D-arabino epimer, was synthesized in 10
steps from 2-deoxyribonolactone. Compound 4 represents the 2-deoxy analog of the central
metabolite 5-phosphorylribose 1-Ͱ-diphosphate (PRPP). The analog is a competitive inhibitor
of yeast orotate phosphoribosyltransferase (PRTase) with Ki/Km(PRPP) ϭ 24 and is thus
comparable to other C-glycosylphosphonyl PRPP analogs reported previously (R. W. McClard,
and J. F. Witte (1990) Bioorg. Chem. 18, 165–178). The analog is an even better inhibitor of
both human hypoxanthine-guanine PRTase and glutamine : amido PRTase from Escherichia
coli with Ki/Km(PRPP) values of 16 and 3, respectively. These results support the argument
that the cis-diol grouping of nucleotides (or PRPP) plays a very minor role in binding of
these substrates to PRTases.
1996 Academic Press, Inc.
INTRODUCTION
Stable analogs of 5-phosphorylribose 1-Ͱ-diphosphate (PRPP, 1) are important
due to their applications in studying the mechanisms of phosphoribosyltransferases
(PRTases). In a previous paper (1) we described two C-glycosylphosphonyl analogs
that act as competent competitive inhibitors of orotate PRTase. We were unable
to attain the ‘‘exact’’ analog, 2, because upon acid hydrolysis of the intermediate
3 the generation of a free hydroxyl at C3 then allowed elimination of phosphate
with the formation of a phostone with the phosphonate P which then hydrolyzed
to the 1-phosphonate (1). Although the target (2) may still be attainable by another
protection/deprotection scheme, we elected to pursue the synthesis of compound
4, which we expected would evade the hydrolysis problem by virtue of its lack of
a 3-OH group.
We reasoned that the lack of this 3-OH (comparable to the 2-OH of PRPP)
should be a conservative change in the molecule’s behavior since the cis-hydroxyls
of PRPP (or conjugate nucleotide) seem to be of lower importance in the binding
of these molecules in the active sites of certain PRTases. For example, nucleotide
affinity columns constructed for the purposes of purification of bacterial orotate
PRTase (2) and mammalian hypoxanthine-guanine PRTase (3) are connected to
1 Part 1 of this series is M
CC
LARD, R. W., AND
W
ITTE, J. F. (1990) Bioorg. Chem. 18, 165–178.
29
0045-2068/96 $18.00
Copyright 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

30
WITTE AND MCCLARD
the resin through aldehydes generated from periodate-cleaved cis-diols of OMP
and GMP, respectively. Similarly, hypoxanthine-guanine (HG) PRTase is effectively
inhibited by acyclovir (4), which contains no ring hydroxyls at all. The most convinc-
ing rationale is provided by the recently published crystal structure of Salmonella
orotate PRTase (5), which binds OMP with the 2Ј,3Ј-cis hydroxyls apparently point-
ing out into the solvent with little or no apparent interaction with active site residues.
Here we describe the 10-step synthesis of 4, as a mixture of its
D
-ribo and -arabino
D
epimers, from 2-deoxyribonolactone and demonstrate that this analog of PRPP is
as good or, in some cases, better than previously published (1) C-glycosylphos-
phonyl analogs.
RESULTS AND DISCUSSION
The synthesis of 4 began with deoxyribose which was oxidized catalytically with
Pd/O2/Mg(OH)2, by the method previously developed in this laboratory, to the
corresponding lactone in almost quantitative yield (6). The resulting lactone was
selectively protected at C5 with TBDMSi-Cl catalyzed by imidazole in DMF solution
(7) to give 5 in 65% yield, the major by-product being the bis-TBDMSi derivative
which was readily separated by chromatography. Introduction of the TBDPSi group
at C3 was readily accomplished with TBDMSi-Cl in the presence of DMAP and
CH2Cl2 to yield 6 in quantitative yield. Reduction of the protected lactone with
DIBAL in toluene at Ϫ78ЊC during 1 h furnished the corresponding protected
deoxyribose, 7, in quantitative yield (8). Reaction of this material with the sodium
salt of tetramethyl methylenebisphosphonate (from tetramethyl methylenebisphos-
phonate and NaH in glyme) during 24 h at room temperature under Horner–
Emmons conditions (9) yielded the desired methylene phosphonate compound 8,
isolated as a mixture of epimers at C2. The TBDMSi group at C6 was selectively
removed with 80% aqueous acetic acid at room temperature during 48 h to give
the corresponding C6 hydroxy compound, 9, in quantitative yield. Reaction of this
material with diphenyl chlorophosphate in CH2Cl2 and DMAP introduced the
desired diphenyl phosphoryl group to provide a quantitative yield of compound
10, by analogy to previous work in this lab (1). Removal of the TBDPSi protecting
group at C4 was accomplished with HFиPyr in methylene chloride at 0ЊC to furnish
compound 11 in 65% yield. Some substitution of -OPh by -F occurred at the C6

C-GLYCOSYLPHOSPHONYL ANALOGS OF PRPP
31
S
CHEME 1. (a) TBDMSi-Cl/imidazole/DMF, 16 h. (b) TBDPSI-Cl/DMAP/CH2Cl2 , 16 h. (c) DIBAL,
toluene, Ϫ78ЊC, 1 h. (d) [(MeO)2P(O)]2CHϪ, glyme. (e) aq. HOAc. (f) ClP(O)(Oph)2/DMAP. (g)
HF · pyridine/CH2Cl2 . (h) TMS-Br. (i) 1. 1 eq. Bu4Hϩ OHϪ/MeCN; 2. ImHϩ (PhO)P(O)(OϪ)(Im), 65ЊC.
(j) H2/Pt on C/PtO2 , 80% aq. HOAc.
phosphorus under these conditions. The formation of the crucial phosphonylphos-
phate anhydride 12 was accomplished in 55% yield in two steps following our
previous work (1). First the methyl ester groups were removed with TMSBr at 0ЊC
and the resulting phosphonic acid was allowed to react with the imidazolide of
monophenyl phosphate at 65ЊC in acetonitrile during 24 h to yield the product in
65% yield. Final deprotection was accomplished by hydrogenolysis (Pt/C/1 atm H2,
80% aqueous acetic acid) to furnish the desired analog, 4, in nearly quantitative
yield from 12.
From the point of the formation of compound 8, the products of each successive
reaction were carried through as a chromatographically inseparable mixture of
epimers at C2. The mixture of isomers of the product compound 4 was determined
by NMR spectral analysis (conducted by Michael Tears and John Grutzner at
Purdue University) to be composed of approximately 75% of the desired ‘‘Ͱ anomer’’
and 25% of the ‘‘ͱ anomer’’ (ratio based on integrations obtained from inverse-
gated 31P NMR spectrum).2
It is worth noting that the ready availability of aldonolactones, by a procedure
developed in this laboratory (6), provides a convenient means for the protection/
deprotection of the acetal/ketal functionality along the synthetic path involving a
sugar derivative; this study is a modest example. The classical protection of the
anomeric carbon via formation of an alkyl glycoside (10), a notable example being
2,3,5-tribenzylribose, proceeds with low yields and multiple byproducts. The synthe-
2 Initial peak assignments were readily determined by H/H and P/H correlation spectroscopy and
the respective orientation of the ring protons was determined through application of the Karplus
relationship. After these assignments had been made it was possible to assign multiplicities and coupling
constants to each of the protons of the major isomer. Thus the small coupling between H5 and H4
results from a trans relationship between these protons and establishes a trans configuration between
the phosphoryl group at C6 and the hydroxyl at C4. The 10.7 and 6.5 Hz (coupling constants dimished
by the presence of the hydroxyl at C4) couplings between H3ͱ–H2 and H3ͱ–H4, respectively, establishes
a single cis relationship between these protons. These data translate into a 2,4-cis and a 4,5-trans
relationship among the ring substituents and an overall configuration for the major isomer as depicted
as compound 4. In addition, the smaller trans couplings between H4–H3Ͱ(1.5 Hz) and H2–H3Ͱ(5.0
Hz) are also consistent with this assignment.

32
WITTE AND MCCLARD
TABLE 1
Inhibition of PRTases by the PRPP Analog Compound 4
a
b
Enzyme
Observed Ki /Km(PRPP)
Corrected Ki /Km
yeast orotate PRTase
Human HG PRTase
32
20
4
24
16
3
E. coli Gln : amido PRTasec
a The Km(PRPP) for yeast orotate PRTase is equivalent to the Kdiss in this case (11).
b Corrected for the fact that compound 4 is a 75 : 25 mixture of the Ͱ : ͱ ‘‘Anomers,’’
i.e., the
D-ribo and D-arabino epimers.
c Performed by Dana Wolle and Howard Zalkin (Purdue University).
sis of 5-trityl-2,3-isopropylideneribose is also cumbersome. The scheme employed
in this paper (i.e., oxidation to the lactone followed by reduction back to the aldose
at the time the aldehyde oxidation state is required) involves a sequence of reactions
that give distinct products in high yields.
Compound 4 acts as a respectable competitive inhibitor of three different PRTases
from various sources as shown in Table 1.
By comparison, previously reported C-glycosylphosphonyl PRPP analogs 13 and
14 showed Ki/Km(PRPP) values of 16 and 54, respectively (1). Clearly, with regard
to yeast orotate PRTase, removal of the inverted 3-OH from 14 results in tighter
binding than for compound 4. Interestingly, the latter is slightly weaker as an
inhibitor than is 13 with its rather bulky acetonide functionality.
These results with compound 4, in addition to other related observations (1–4),
reveal the lack of relative importance of the sugar cis-diol in the nucleotide or
PRPP in binding to orotate PRTase. This conclusion accords well with the recently
published crystal structure of Salmonella orotate PRTase, which clearly reveals that
the cis-diol grouping of bound OMP seems to point directly out into solvent.
Compound 4 is an even better inhibitor (based on Ki/Km values) of HG PRTase
and glutamine : amido PRTase. It would appear that in these cases as well that the
2-OH of PRPP is not very crucial to binding.
EXPERIMENTAL
General
2-Deoxy-D-ribonolactone was synthesized from
D-deoxyribose as described by
Witte et al. (6). NMR spectra were obtained on a Bruker AC 300 instrument run
1
at the following frequencies: H, 300.13 MHz; 13C, 75.47 MHz; 31P, 121.50 MHz;

C-GLYCOSYLPHOSPHONYL ANALOGS OF PRPP
33
coupling constants are given in Hz. All spectra were performed in CDCl3 except
as noted. Routine assignments of structures were based on analyses of 13C-DEPT,
1H-1H-COSY, and 13C-1H heteronuclear correlation (HETCOR) experiments. In
the following ‘‘major’’ refers to the deduced major (
D
-ribo or ‘‘Ͱ’’) epimer, and
‘‘minor’’ refers to the minor ( -arabino or ‘‘ͱ’’) epimer. Organic extracts obtained
D
from aqueous mixtures were routinely dried with solid Na2SO4. Chromatography
was performed using EM Science silica gel 60 (230–400 mesh) with solvents as
indicated. TLC was performed on EM Science Kieselgel 60 F254 plates.
Syntheses
5-(t-Butyldimethylsilyl)-2-deoxy- -ribonolactone (5). To a solution of 2-deoxyri-
D
bonolactone (26.4 g, 0.2 mol), dried by evaporation with CH3CN (3 ϫ 30 ml) in
dry DMF (300 ml) was added TBDMSi-Cl (31.4 g, 0.2 mol) and imidazole (13.7 g,
0.2 mol) at 0Њ with stirring under N2. The resulting solution was allowed to stand
for 16 h, whereupon water (200 ml) was added and the mixture extracted with
CH2Cl2 (3 ϫ 100 ml). The extracts were dried and all volatiles removed at reduced
pressure. The resulting viscous semicrystalline material was stirred with MeOH
(300 ml) and the crystalline material filtered off, washed with MeOH (50 ml) to
give 7.26 g of the bis-TBDMSi derivative. The filtrates and washings were combined
and evaporated under reduced pressure. The residue was chromatographed with
hexane : EtOAc (2 : 3) to give the desired derivative (31.5 g, 65%). 1H NMR: ͳ 0.09
(6H, 2s, -Si-CH3), 0.91 (9H, s, -Si-C(CH3)3), 2.42 (1H, dd, J2,2 ϭ 18.0, J2,3 ϭ 3.5,
Ј
C2-H), 2.71 (1H, bs, -OH), 2.92 (1H, dd, J2 ϭ 18.0, J2 ϭ 9.0 Hz, C2-HЈ), 3.82
Ј
,2
Ј
,3
(2H, d, J4,5 ϭ 4.5, C5-H), 4.43 (1H, m, C4-H), 4.58 (1H, dt, J ϭ 10.1, 1.5, C3-H).
13C NMR: ͳ -5.81 and -5.75 (2C, -Si-CH3), 18.04 (-Si-C(CH3)3), 25.63 (-Si-C(CH3)3)
38.80 (C2), 62.94 (C5), 69.37 (C3), 88.14 (C4), 177.10 (CuO).
5-(t-Butyldimethylsilyl)-3-(t-butyldiphenylsilyl)-2-deoxy- -ribonolactone (6). To a
D
solution of 5 (23.5 g, 0.096 mol) in dry CH2Cl2 (250 ml) was added with stirring
under N2 TBDPSi-Cl (28.9 g, 27.3 ml, 0.11 mol) and DMAP (13.42 g, 0.11 mol) at
0ЊC. The resulting solution was allowed to stand at room temperature for 16 h.
Water (200 ml) was added and the layers separated. The organic layer was dried,
filtered, and volatiles removed at reduced pressure. The syrupy residue (53.4 g)
chromatographed with hexane : EtOAc (5 : 1) gave 44.1 g (97%) of 6, whose NMR
spectra indicate the presence of two apparently rotationally constrained (at Si)
1
conformers. H NMR: ͳ -0.06 to 0.12 (6H, 4s, -Si-CH3), 0.77 to 1.10 (18H, 4s, -Si-
C(CH3)3), 2.37 to 2.84 (2H, m, C2-H and C2-HЈ), 3.13 (1H, dd, C5-H of major
conformer), 3.58 (1H, dd, C5-HЈ of major conformer), 3.72 to 3.83 (2H, m, C5-H
and C5-HЈ of minor conformer), 4.31 to 4.37 (1H, br m, C4-H), 4.45 (1H, dt, C3-
H of major conformer), 4.51 (1H, dt, C3-H of minor conformer), 7.33 to 7.68
(10H, 2m, arom). 13C NMR: ͳ -5.86, -5.68, -5.43, -0.01 (-Si-CH3); 18.11, 19.00 (-Si-
C(CH3)3); 25.65, 25.70, 25.78, 26.78 (-Si-C(CH3)3); 39.05 (C2); 62.45, 62.48 (C5);
69.60, 70.91 (C3); 88.10, 88.16 (C4); 127.96, 130.15, 132.85, 133.07, 135.62, 135.66
(arom); 176.12 (CuO).
5-(t-butyldimethylsilyl)-3-(t-butyldiphenylsilyl)-2-deoxy-
D-ribose (7). To a solu-
tion of 6 (4.8 g, 0.01 mol) in dry toluene (25 ml) was added with stirring under N2

34
WITTE AND MCCLARD
DIBAL (20.2 ml, 1
M
in cyclohexane) at Ϫ78ЊC for 60 min. The reaction was
quenched with MeOH (5 ml) followed by saturated Na2SO4 (50 ml) and ether (100
ml) and allowed to warm to room temperature with vigorous stirring. The inorganic
precipitate was filtered off. The filtrate was extracted with ether (2 ϫ 100 ml) and
the combined extracts were dried, filtered, and evaporated under reduced pressure
to give 5.0 g of a colorless oil. The oil was chromatographed with hexane : EtOAc
(5 : 1) to give 4.32 g of 7 (95%) as a mixture of anomers each with magnetically
1
nonequivalent t-butyl groups. H NMR: ͳ 0.05 to 0.18 (6H, 4s, -Si-CH3), 0.88 to
1.20 (18H, 4s, -Si-C(CH3)3), 1.91 to 2.24 (4H, several m, -C2-H), 3.17 to 3.75 (2H,
m, -C5-H), 3.91 to 4.12 (1H, m, -C4-H), 4.23 to 4.52 (1H, m, C3-H), 5.35 to 5.58
(1H, m, C1-H), 7.22 to 7.70 (10H, 2m, arom). 13C NMR: ͳ -5.82, -5.76, -5.72,
-5.56 (-Si-CH3); 18.15, 18.18, 18.91 (-Si-C(CH3)3); 25.72, 25.77, 26.82, 26.91 (-Si-
C(CH3)3); 41.98, 45.53 (C2); 63.32, 63.87 (C5); 73.82, 75.10 (C3); 87.30, 87.46 (C4);
99.32, 99.90 (C1); 127.71, 127.83, 129.99, 130.01, 132.82, 133.47, 133.63, 133.62, 135.68,
135.75 (arom).
D-Ribo- and
D-arabino-2,5-anhydro-6-(t-butyldimethylsilyl)-4-(t-butyldiphenyl-
silyl)-1,3-dideoxy-1-dimethoxyphosphinylhexitol (8). To a suspension of NaH (0.33
g, 0.014 mol) in dry glyme (40 ml) was added dropwise, with stirring under N2, a
solution of tetramethyl methylenebisphosphonate (3.48 g, 0.015 mol in glyme, 10
ml) at 20Њ. After the evolution of H2 had ceased, a solution of 7 (4.16 g, 8.7 mmol)
in glyme (10 ml) was added rapidly. The resulting mixture was allowed to stand at
room temperature for 16 h. During this time, a gummy material precipitated and
TLC (hexane/EtOAc, 1 : 4) indicated that the reaction was complete. The reaction
was quenched with 10% KH2PO4 (100 ml) and glyme was removed under reduced
pressure. The aqueous mixture was extracted with CH2Cl2 (3 ϫ 30 ml). The com-
bined extracts were dried, filtered, and concentrated under reduced pressure to
give a pale yellow oil. This material was chromatographed with EtOAc : hexane
(1 : 1) to furnish 4.00 g (70% of theory) of 8 (mixture of the Ͱ and ͱ ‘‘anomers’’)
as a colorless viscous oil. 1H NMR: ͳ -0.89 to 1.02 (6H, 3s, -Si-CH3), 0.79 and 1.11
(18H, 2s, -Si-C(CH3)3), 1.48 to 2.43 (4H, several m, C1-H and C3-H), 3.15 to 3.42
(2H, 2m, C6-H), 3.68 to 3.79 (6H, 12s, -P(O)CH3), 3.93 and 4.01 (1H, 2m, C5-H),
4.30 to 4.60 (2H, several m, C2-H and C4-H), 7.32 to 7.70 (10H, 2 m, arom). 13C
NMR: ͳ -5.70, -5.67, -5.52 (-Si-CH3); 18.16, 18.97, 19.02 (-Si-C(CH3)3); 25.77, 26.98
(-Si-C(CH3)3); 31.37 (d, C1 of major, JCP ϭ 138), 32.3 (d, C1 of minor, JCP ȁ 140),
41.25 (d, C3 of minor, JCP ϭ 7.5); 42.56 (d, C3 of major, JCP ϭ 8.2); 63.55 (C6 of
major); 63.64 (C6 of minor); 73.27 (C2 of major); 73.94 (C2 of minor); 74.96 (C4
of minor); 75.39 (C4 of major); 86.96 (C5 of minor); 88.02 (C5 of major); 127.60,
129.64, 129.72, 133.52, 133.67, 133.82, 135.67, 135.72 (arom). 31P NMR (1H-dec,
relative to ext. 85% H3PO4): ͳ 31.22 (major), 31.43 (minor).
D
-Ribo- and
D
-arabino-2,5-anhydro-4-(t-butyldiphenylsilyl)-1,3-dideoxy-1-
dimethoxyphosphinylhexitol (9). A solution of 8 (4.00 g, 6.76 mmol) in 80% aqueous
acetic acid was allowed to stand at 20ЊC for 2 days, at which time TLC indicated
that starting material was no longer present. All volatiles were removed under
reduced pressure, followed by high vacuum to give 3.20 g (94%) of pure 9 as a
1
colorless viscous oil. H NMR: ͳ 1.05 and 1.09 (-Si-C(CH3)3), 1.84 to 2.32 (4H, m,
C1-H and C3-H), 3.04 to 3.52 (2H, m, C6-H), 3.66 to 3.78 (6H, 8s, -P(O)CH3), 3.90

C-GLYCOSYLPHOSPHONYL ANALOGS OF PRPP
35
to 4.02 (1H, m, C5-H), 4.21 to 4.67 (2H, m, C2-H and C4-H), 7.32 to 7.70 (10H,
2m, arom). 13C NMR: ͳ -0.12, -0.10 (-Si-CH3); 18.88, 18.95 (-Si-C(CH3)3); 26.79,
26.82 (-Si-C(CH3)3); 29.68 (d, C1 of major, JCP ϭ 140); 32.04 (d, C1 of minor,
JCP ϭ 136); 41.48 (d, C3 of minor, JCP ϭ 8.2); 42.33 (d, C3 of major, JCP ϭ 5.2);
51.73, 51.82, 51.90, 51.99, 52.50, 52.30 52.39, 52.59 (2C, split by P, -P(O)CH3); 62.31
(C6 of minor); 62.61 (C6 of major); 73.12 (d, C2 of major, JCCP ϭ 6.0); 73.61 (C2
of minor); 74.47 (C4 of minor); 75.09 (C4 of major); 86.54 (C5 of minor); 88.82 (C5
of major); 127.64, 127.65, 127.69, 129.72, 129.78, 129.79, 129.87, 133.25, 133.55, 133.60,
135.61, 135.63, 135.65 (arom). 31P NMR (1H-dec, relative to ext. 85% H3PO4): ͳ
31.27 (major), 32.25 (minor).
D
-Ribo- and
D
-arabino-2,5-anhydro-4-(t-butyldiphenylsilyl)-1,3-dideoxy-1-di-
methoxyphosphinyl-6-diphenoxyphosphinylhexitol (10). To a solution of 9 (3.25 g,
6.80 mmol) in dry CH2Cl2 (100 ml) was added with stirring under N2 diphenyl
chlorophosphate (2.19 g, 1.92 ml, 8.16 mmol) and DMAP (1.00 g, 8.16 mmol). The
resulting solution was allowed to stand at room temperature for 36 h. Water (100
ml) was added and the layers were separated. The organic layer was washed succes-
sively with 1
M
HCl (25 ml), 10% NaHCO3 (100 ml), and water (100 ml) and dried.
CH2Cl2 was removed under reduced pressure to give 5.50 g of a pale yellow viscous
oil which was chromatographed (EtOAc) to furnish 4.35 g (92%) of 10 as a colorless
1
viscous oil. H NMR: ͳ 1.05 and 1.09 (9H, 2s, -Si-C(CH3)3), 1.41 to 2.39 (4H, m,
C1-H and C3-H), 3.58 to 3.67 and 3.90 to 3.96 (2H, 2m, C6-H), 3.67 to 3.73 (6H,
6s, -P(O)CH3), 3.97 to 4.02 (1H, m, C5-H), 4.24 to 4.37 (1H, m, C4-H), 4.46 to 4.58
(1H, m, C2-H), 7.08 to 7.64 (20H, 4m, arom). 13C NMR (major): ͳ 18.89, (-Si-
C(CH3)3); 26.78, (-Si-C(CH3)3); 30.85 (d, C1, JCP ϭ 139); 42.58 (d, C3, JCP ϭ 8.2);
51.96, 52.04, 52.36, 52.45, (2C, split by P, -P(O)CH3); 68.24 (C6, JCOP ϭ 6.8); 73.69
(C2); 74.76 (C4); 85.61 (C5, JCCOP ϭ 8.2); 119.82, 119.89, 125.25, 127.74, 127.80,
127.84, 129.63, 129.66, 129.84, 129.94, 130.02, 132.91, 133.00, 133.16, 133.25, 135.60,
150.24, 150.33 (arom). 31P NMR (1H-dec, relative to ext. 85% H3PO4): ͳ -11.76
(phosphate, major), -11.72 (phosphate, minor), 30.66 (phosphonate, major), 30.72
(phosphonate, minor).
D-Ribo- and
D-arabino-2,5-anhydro-1,3-dideoxy-1-(dimethoxyphosphinyl)-6-(di-
phenoxyphosphinyl)hexitol (11). To a solution of 10 (0.86 g, 1.23 mmol) in CH2Cl2
(10 ml) in a Teflon tube was added with stirring 70% HF/pyridine (1 ml) at 0ЊC.
The reaction was allowed to proceed for 1 h at 0ЊC and then quenched with water
(5 ml). The mixture was extracted with CH2Cl2 (3 ϫ 10 ml). The extracts were
dried, filtered, and evaporated under reduced pressure to a viscous oil. Chromatog-
1
raphy with EtOAc : IPA (4 : 1) yielded 0.36 g (62%) of pure 11. H NMR: ͳ 1.62 to
2.39 (4H, m, C1-H and C3-H), 3.65 to 3.72 (6H, 6s, -P(O)CH3), 3.58 to 3.67 and
3.90 to 3.96 (2H, 2m, C6-H), 4.00 (1H, br m, C5-H of major), 4.10 (1H, m, presumed
C5-H of minor), 4.20 to 4.30 (3H, m, C4-H and C6-H’s), 4.38 to 4.48 (1H, m, C2-
H), 7.12 to 7.40 (10H, 2m, arom). 13C NMR: ͳ 30.77 (d, C1, major, JCP ϭ 139),
31.51 (d, C1, minor, JCP ϭ 134); 40.50 (d, C3, minor, JCCCP ϭ 6.8); 41.78 (d, C3,
major, JCCCP ϭ 6.8); 51.97, 52.06, 52.14, 52.21, 52.36, 52.45 (2C, split by P, -P(O)CH3);
68.58, 68.67, 68.76 (C6, JCOP unclear); 72.34 (C4, minor); 72.52 (C4, major), 73.18
(C2, major); 73.34 (C2, minor); 83.72 (C5, d, minor, JCCOP ϭ 7.5); 84.70 (C5, d,
major, JCCOP ϭ 7.5); 119.78, 119.80, 119.86, 125.32, 129.64, 150.13, 150.23 (arom).

36
WITTE AND MCCLARD
31P NMR (1H-dec, relative to ext. 85% H3PO4): ͳ -11.51 (phosphate, major), -11.43
(phosphate, minor), 30.88 (phosphonate, major), 31.30 (phosphonate, minor).
D
-ribo- and
D
-arabino-2,5-anhydro-1,3-dideoxy-6-(diphenoxyphosphinyl)-1-
((((phenoxy)hydroxyphosphinyl)oxy)hydroxyphosphinyl)hexitol, ammonium salt
(12). To a solution of 11 (1.34 g, 2.87 mmol) in CH2Cl2 (40 ml) was added TMS-
Br (1.58 g, 1.34 ml, 10.3 mmol) with stirring under N2 at 0ЊC. The reaction was
allowed to warm to room temperature and after 3 h all volatiles were removed
under high vacuum. The resulting viscous mass was dissolved in anh. MeOH (10
ml) and evaporated in vacuo (Ͻ1 mm). This material was dissolved in 50% aqueous
MeOH and taken to pH 4.5 with aqueous 0.4
M
tetrabutylammonium hydroxide.
Volatiles were removed (Ͻ1 mm) and the residue was dried by azeotropic distillation
with anh. MeCN (3 ϫ 10 ml). An aqueous solution of disodium phenyl phosphate
dihydrate (0.82 g, 3.16 mmol) was passed over a column of Dowex-50 (Hϩ form).
Effluent was collected until the pH reached 4.5. The effluent was evaporated in
vacuo and dried by evaporation with MeCN (3 ϫ 20 ml). The resulting solid was
dissolved in anh. MeCN (10 ml) and carbonyl diimidazole (0.52 g, 3.16 mmol) was
added with stirring under N2. CO2 was evolved as stirring was continued at 20ЊC
for 2 h, whereupon a solution of the mono(tetrabutylammonium) phosphonate
(described above) in anh. MeCN (10 ml) was added. The mixture was stirred at
50ЊC for 16 h. All volatiles were removed at 1 mm. The resulting gum was dissolved
in water (20 ml) and passed over a Dowex-50 column (NHϩ4 form). The column
was eluted with water until the pH of the effluent reached 5.0. The effluent was
evaporated under reduced pressure to a gummy residue. The residue was chromato-
graphed with MeCN : H2O : NH3 (90 : 10 : 1), followed by the same mixture at
80 : 10 : 1. The fractions containing the desired phosphonylphosphate were combined
and concentrated under reduced pressure to yield 1.23 g (65%) of a gum. This
material was taken directly to the next step.
Deprotection of 12 to
D
-ribo- and -arabino-2,5-anhydro-1,3-dideoxy-1-(((dihy-
D
droxyphosphinyl)oxy)hydroxyphosphinyl)-6-phosphorylhexitol, (4). The gum from
above was dissolved (in 50% aqueous acetic acid (40 ml). PtO2 (1.00 g) and 10%
Pt/C (1.00 g) were added and the resulting suspension was stirred rapidly under
H2 at 1 atm until the uptake of H2 ceased at 105% of theory. The reaction mixture
was filtered and the filtrate was evaporated at 0.1 mm to yield a colorless gum
1
which was found to be the desired 4 as the free acid. H NMR (D2O; referenced
to D4-TMS-propionate at ͳ ϭ 0): ͳ 1.89 (m, C3-HA, minor), 1.93 (m, C3-HA, major),
1.98 (m, C1-HA, major ϩ minor), 2.1 (m, C1HB, minor), 2.20 (ddd, C3-HB, major),
2.31 (m, C1-HB, major), 2.58 (m, C3-HB, minor), 3.75 to 3.87 (m, C6-HA and C6-
HB, major ϩ minor), 3.97 (dt, C5-H, major), 4.02 (m, C5-H, minor), 4.36 (m, C4-
1
H, major ϩ minor), 4.44 (apparent nonet, C2-H, major ϩ minor). H coupling
constants: Major isomer, H1A to (H1B, -14.5; H2, 9.3; P, 19.2), H1B to (H2, 5.0; P,
19.2), H2 to (H3A, 10.7; H3B, 5.0; P, 5.2), H3A to (H3B, -13.5; H4, 6.5), H3B to (H4,
1.5), H4 to (H5, 2.6), H5 to (H6A, 6.0; H6B, 6.0), H6A to (H6B, -14.5; P, 5.5), H6B
to (P, 5.5). Minor isomer, H2 to (H3A, 8.5; H3B, 7.0), H3A to (H3B, -13.5; H4, 6.5),
H3B to (H4, 7.0), H4 to (H5, 10.6), H5 to (H6A, 6.3; H6B, 4.6), H6 to (P, 5.5). 13C
NMR (D2O; referenced to D4-TMS-propionate at ͳ ϭ 0): ͳ 36.67 (d, C1, major,
JCP ϭ 134), 37.34 (d, C1, minor, JCP ϭ 134); 42.45 (d, C3, minor, JCCCP ȁ 4); 42.98

C-GLYCOSYLPHOSPHONYL ANALOGS OF PRPP
37
(d, C3, major, JCCCP ϭ 4.1); 66.53 (d, C6, minor, JCOP ȁ 5); 67.17 (d, C6, major,
JCOP ϭ 4.9); 74.74 (C4, minor); 75.42 (C4, major), 77.13 (C2, minor); 77.72 (C2,
major); 85.78 (d, C5, minor, JCCOP ȁ 8); 87.60 (C5, d, major, JCCOP ϭ 8.1). 31P NMR
1
(D2O; H-dec; relative to ext. 85% H3PO4): ͳ -7.58 (phosphate anhydride, major
and minor epimers superimposed, JPOP ϭ 24.3), 2.71 (6-phosphate, major), 2.78
(6-phosphate, minor), 11.79 (phosphonate anhydride, major, JPOP ϭ 24.4), 11.84
(phosphonate, minor, JPOP ϭ 24.4).
Enzymes and Assays
Orotate PRTase was purified to homogeneity from baker’s yeast by a combination
of literature methods. Initial purification was carried out according to the procedure
of Umezu et al. (12) through the DEAE cellulose step with the omission of the
ethanol fractionation. The ammonium sulfate step was changed to a fractionation
of 0–55% of saturation at 4ЊC. Enzyme that eluted in the second peak of activity
from the DEAE column was then applied to OMP-Sepharose and eluted with OMP
as described by Dodin (2). The enzyme routinely had specific activities Ͼ50 unit/
mg at 25ЊC and was homogeneous as determined by SDS gel electrophoresis. It
was assayed by observing the decrease in absorbance at 295 nm due to the conversion
of orotic acid to OMP (molar absorptivity ⌬␧ ϭ 3.35 mMϪ1). Human HG PRTase
was obtained from Professor C. C. Wang (University of California at San Francisco).
It was assayed by the spectrophotometric method of Giacomello and Salerno (13).
The activity of glutamine : amido PRTase from Eschenichia coli was determined by
measuring the PRPP-dependent formation of glutamate from glutamine according
to Messenger and Zalkin (14); these assays were conducted by Dana Wolle and
Howard Zalkin at Purdue University.
ACKNOWLEDGMENTS
This work was funded, in part, by grants from the National Science Foundation (DMB 90-20503) and
the American Cancer Society (CH 426). The authors are also very grateful for the following assistance
in this study: Howard Zalkin and Dana Wolle (Purdue University) for the Gln : Amido PRTase enzyme
assays, Pamela Focia and C. C. Wang (University of California, San Francisco) for the gift human HG
PRTase, and John Grutzner and Michael Tears (Purdue University) for help in the interpretation of
NMR spectra of compound 4.
REFERENCES
1. MCCLARD, R. W., AND WITTE, J. F. (1990) Bioorg. Chem. 18, 165–178.
2. DODIN, G. (1981) FEBS Lett. 134, 20–24.
3. GUTENSOHN, W. (1977) Adv. Exp. Med. Biol. 76A, 586–590.
4. TUTTLE, J. V., KRENITSKY, T. A., AND
5. SCAPIN, G., GRUBMEYER, C., AND ACCHETTINI, J. C. (1994) Biochemistry 33, 1287–1294.
6. WITTE, J. F., FIRTH, R., AND LARD, R. W. (1995) Carbohydr. Lett. 1, 123–124.
7. ATWOOD, S. V., AND ARRETT, A. G. M. (1984) J. Chem. Soc. Perkin Trans. I, 1315–1322.
ELION, G. B. (1983) Biochem. Pharmacol. 32, 3011–3015.
S
MCC
B
8. OKABE, M., SUN, R.-C., TAM, S. K.-Y., TODARO, L. J., AND
COFFEN, D. L. (1988) J. Org. Chem.
53, 4780–4786.

38
WITTE AND MCCLARD
9. M
CC
LARD, R. W., AND
WITTE, J. F. (1992) in Handbook of Organophosphorus Chemistry (Engel,
R., Ed.), pp. 655–681. Marcel Dekker, New York.
10. BARKER, R., AND LETCHER, H. G. (1961) J. Org. Chem. 26, 4605–4611.
11. VICTOR, J., GREENBERG, L. B., AND LOAN, D. L. (1979) J. Biol. Chem. 254, 2647–2655.
12. UMEZU, K., AMAYA, T., YOSHIMOTO, A., AND OMITA, K. (1971) J. Biochem. 70, 249–267.
ALERNO, C. (1978) J. Biol. Chem. 253, 6038–6044.
ALKIN, H. (1979) J. Biol. Chem. 254, 3382–3392.
F
S
T
13. GIACOMELLO, A., AND
S
14. MESSENGER, L. J., AND
Z