natural purines. However, the salvage pathways share the
common approach of catalyzing the addition of nucleobases
to anomerically activated ribose. Two enzyme classes, purine
nucleoside phosphorylase (PNP) and phosphoribosyltrans-
ferases (PRTs), catalyze the addition of structurally diverse
nucleobases to ribose activated as ribose-1-phophate or
5-phosphoribosyl 1-pyrophosphate, respectively.10-12 The
mechanistic commonality inherent in the salvage solutions
to a wide variety of nucleotide biosyntheses has prompted
us to investigate the PRT systems for their utility in
biocatalytic generation of synthetic nucleotide analogues.
Laboratory-scale biocatalytic generation of nucleotide
analogues suffers from potential practical drawbacks includ-
ing the specificity of the enzymatic systems involved, diffi-
culty in obtaining purified enzymes, and the separation of
polar reaction products from polar starting materials and
byproducts. To address these concerns, we have developed
a biocatalytic system for the synthesis of a range of nucleo-
tide analogues. Herein, we present a one-step biocatalytic
process for the synthesis of several nucleotide analogues from
commercially available starting materials. E. coli whole cell
crude extract is used as the biocatalyst, and nucleotides are
purified in a single step using anion-exchange chromatog-
raphy.
Table 1. Base Analogues Assayed for Percent Conversion by
PRT Enzymes in This Study
We have recently generated a mutant variant of E. coli
hypoxanthine phosphoribosyltransferase (HPRT) as a product
of a directed evolution study in which an error-prone PCR
library of the hpt gene in a protein expression host was
selected for improved in vivo transformation of triazole
carboxamide 13 to ribavirin monophosphate (the nucleotide
and peptide sequence is reported in the Supporting Informa-
tion). To further investigate the synthetic utility of this
mutant, designated 8B3PRT (V157A, Y173H), we examined
the ability of this enzyme to utilize a structurally diverse
range of commercially available purine and purine base
analogues (Table 1).
To facilitate the use of 8B3PRT as a practical catalyst,
we desired to obviate the need for enzyme isolation and
developed our system using a simple cell-free preparation.
E. coli BL21(DE3) cells harboring plasmid pRAS1001, a
pET28a-based plasmid-containing the mutant hpt gene, were
induced at OD ) 0.6 with IPTG and incubated for 3 h. Two
milliliter aliquots of cells were pelleted by centrifugation and
treated with a commercial detergent (BugBuster, Novagen,
Inc.), each providing enough catalyst for 30 mg scale
reactions. The cell preparation was added to buffered solu-
tions of phosphoribosyl pyrophosphate (PRPP) and purine
base analogues to catalyze the formation of nucleotides in
less than 2 h at 37 °C.
a PRT extracts (6 µL) were added to an aqueous solution of 1 mM PRPP,
30 µL of base analogue (3-25 mM), 12 mM Tris pH 7.3, and 12 mM
MgCl2 to a final reaction volume of 600 µL. Percent conversion measured
by NMR. N.D. ) not determined.
development of a general in vitro assay for PRPP turnover.
For this purpose, a molybdate-based assay for the detection
of phosphate was modified to detect for the formation of
pyrophosphate by adding an excess of inorganic pyrophos-
phatase to the enzymatic reactions (Figure 1). In addition to
observing efficient turnover for triazole carboxamide 13, we
observed enhanced turnover for a wide variety of purine base
analogues. As pyrophosphate turnover is an indirect indica-
tion of product formation, percent conversion was addition-
ally determined via 1H NMR, assaying for the disappearance
of PRPP and formation of product (Table 1). For this
purpose, 10% D2O was added to reactions, and a W5-
WATERGATE water suppression pulse sequence was em-
ployed to eliminate the water resonance13 and accurately
measure the ratio of the anomeric protons (H1′). Often,
conversion of PRPP was nearly quantitative, and since PRPP
is a highly unstable technical grade compound containing
contaminating inorganic phosphates and degradation prod-
ucts, we used percent conversion to determine reaction
efficiency in lieu of percent yield.
Previously reported assays for PRT activity are dependent
upon coupled enzyme assays detecting the turnover of natural
nucleobases such as hypoxanthine or adenosine. Thus, the
assay for nonnatural base analogue turnover required the
Purine phosphoribosyltransferase mutant 8B3PRT dem-
onstrated enhanced activity and relaxed specificity in pro-
cessing a wide variety of nucleoside base analogues. In the
case of ribavirin monophosphate formation from triazole
(10) Lee, C. C.; Craig, S. P.; Eakin, A. E. Biochemistry 1998, 37, 3491-
3498.
(11) Munagala, N.; Sarver, A. E.; Wang, C. C. J. Biol. Chem. 2000,
275, 37072-37077.
(12) Guddat, L. W.; Vos, S.; Martin, J. L.; Keough, D. T.; De Jersey, J.
Protein Sci. 2002, 11, 1626-1638.
(13) Liu, M. L.; Mao, X. A.; Ye, C. H.; Huang, H.; Nicholson, J. K.;
Lindon, J. C. J. Magn. Reson. 1998, 132, 125-129.
4180
Org. Lett., Vol. 9, No. 21, 2007