Product release is rate-limiting in the activation of the prodrug 5-fluorocytosine by yeast cytosine deaminase

Yeast cytosine deaminase (yCD), a zinc metalloenzyme, catalyzes the hydrolytic deamination of cytosine to uracil. The enzyme is of great biomedical interest because it also catalyzes the deamination of the prodrug 5-fluorocytosine (5FC) to form the anticancer drug 5-fluorouracil (5FU). yCD/5FC is one of the most widely used enzyme/prodrug combinations for gene-directed enzyme prodrug therapy for the treatment of cancers. A pH indicator assay has been developed for the measurement of the steady-state kinetic parameters for the deamination reaction. Transient kinetic studies have shown that the product release is a rate-limiting step in the activation of the prodrug 5FC by yCD. The rate constant of the chemical step for the forward reaction (250 s(-)(1)) is approximately 8 times that of the product release (31 s(-)(1)) and approximately 15 times k(cat) (17 s(-)(1)). The transient kinetic results are consistent with those of the steady-state kinetic analysis in the sense that the k(cat) and K(m) values calculated from the rate constants determined by the transient kinetic analysis are in close agreement with those measured by the steady-state kinetic analysis. NMR experiments have demonstrated that free 5FU is in slow exchange with its complex with yCD but has a low affinity for yCD. The transient kinetic and NMR results together suggest that the release of 5FU is rate-limiting in the activation of the prodrug 5FC by yCD and may involve multiple steps.     

Yeast cytosine deaminase mutants with increased thermostability impart sensitivity to 5-fluorocytosine

Prodrug gene therapy (PGT) is a treatment strategy in which tumor cells are transfected with a ‘suicide’ gene that encodes a metabolic enzyme capable of converting a nontoxic prodrug into a potent cytotoxin. One of the most promising PGT enzymes is cytosine deaminase (CD), a microbial salvage enzyme that converts cytosine to uracil. CD also converts 5-fluorocytosine (5FC) to 5-fluorouracil, an inhibitor of DNA synthesis and RNA function. Over 150 studies of CD-mediated PGT applications have been reported since 2000, all using wild-type enzymes. However, various forms of CD are limited by inefficient turnover of 5FC and/or limited thermostability. In a previous study, we stabilized and extended the half-life of yeast CD (yCD) by repacking of its hydrophobic core at several positions distant from the active site. Here we report that random mutagenesis of residues selected based on alignment with similar enzymes, followed by selection for enhanced sensitization to 5FC, also produces an enzyme variant (yCD-D92E) with elevated T_m values and increased activity half-life. The new mutation is located at the enzyme's dimer interface, indicating that independent mutational pathways can lead to an increase in stability, as well as a more subtle effect on enzyme kinetics. Each independently derived set of mutations significantly improves the enzyme's performance in PGT assays both in cell culture and in animal models.     

Crystal structure of yeast cytosine deaminase. Insights into enzyme mechanism and evolution

Yeast cytosine deaminase is an attractive candidate for anticancer gene therapy because it catalyzes the deamination of the prodrug 5-fluorocytosine to form 5-fluorouracil. We report here the crystal structure of the enzyme in complex with the inhibitor 2-hydroxypyrimidine at 1.6-A resolution. The protein forms a tightly packed dimer with an extensive interface of 1450 A2 per monomer. The inhibitor was converted into a hydrated adduct as a transition-state analog. The essential zinc ion is ligated by the 4-hydroxyl group of the inhibitor together with His62, Cys91, and Cys94 from the protein. The enzyme shares similar active-site architecture to cytidine deaminases and an unusually high structural homology to 5-aminoimidazole-4-carboxamide-ribonucleotide transformylase and thereby may define a new superfamily. The unique C-terminal tail is involved in substrate specificity and also functions as a gate controlling access to the active site. The complex structure reveals a closed conformation, suggesting that substrate binding seals the active-site entrance so that the catalytic groups are sequestered from solvent. A comparison of the crystal structures of the bacterial and fungal cytosine deaminases provides an elegant example of convergent evolution, where starting from unrelated ancestral proteins, the same metal-assisted deamination is achieved through opposite chiral intermediates within distinctly different active sites.     

Molecular imaging of gene expression and efficacy following adenoviral-mediated brain tumor gene therapy

Cancer gene therapy is an active area of research relying upon the transfer and subsequent expression of a therapeutic transgene into tumor cells in order to provide for therapeutic selectivity. Noninvasive assessment of therapeutic response and correlation of the location, magnitude, and duration of transgene expression in vivo would be particularly useful in the development of cancer gene therapy protocols by facilitating optimization of gene transfer protocols, vector development, and prodrug dosing schedules. In this study, we developed an adenoviral vector containing both the therapeutic transgene yeast cytosine deaminase (yCD) along with an optical reporter gene (luciferase). Following intratumoral injection of the vector into orthotopic 9 L gliomas, anatomical and diffusion-weighted MR images were obtained over time in order to provide for quantitative assessment of overall therapeutic efficacy and spatial heterogeneity of cell kill, respectively. In addition, bioluminescence images were acquired to assess the duration and magnitude of gene expression. MR images revealed significant reduction in tumor growth rates associated with yCD/5-fluorocytosine (5FC) gene therapy. Significant increases in mean tumor diffusion values were also observed during treatment with 5FC. Moreover, spatial heterogeneity in tumor diffusion changes were also observed revealing that diffusion magnetic resonance imaging could detect regional therapeutic effects due to the nonuniform delivery and/or expression of the therapeutic yCD transgene within the tumor mass. In addition, in vivo bioluminescence imaging detected luciferase gene expression, which was found to decrease over time during administration of the prodrug providing a noninvasive surrogate marker for monitoring gene expression. These results demonstrate the efficacy of the yCD/5FC strategy for the treatment of brain tumors and reveal the feasibility of using multimodality molecular and functional imaging for assessment of gene expression and therapeutic efficacy.     

Alanine-scanning mutagenesis reveals a cytosine deaminase mutant with altered substrate preference

Suicide gene therapy of cancer is a method whereby cancerous tumors can be selectively eradicated while sparing damage to normal tissue. This is accomplished by delivering a gene, encoding an enzyme capable of specifically converting a nontoxic prodrug into a cytotoxin, to cancer cells followed by prodrug administration. The Escherichia coli gene, codA, encodes cytosine deaminase and is introduced into cancer cells followed by administration of the prodrug 5-fluorocytosine (5-FC). Cytosine deaminase converts 5-FC into cytotoxic 5-fluorouracil, which leads to tumor-cell eradication. One limitation of this enzyme/prodrug combination is that 5-FC is a poor substrate for bacterial cytosine deaminase. The crystal structure of bacterial cytosine deaminase (bCD) reveals that a loop structure in the active site pocket of wild-type bCD comprising residues 310-320 undergoes a conformational change upon cytosine binding, making several contacts to the pyrimidine ring. Alanine-scanning mutagenesis was used to investigate the structure-function relationship of amino acid residues within this region, especially with regard to substrate specificity. Using an E. coli genetic complementation system, seven active mutants were identified (F310A, G311A, H312A, D314A, V315A, F316A, and P318A). Further characterization of these mutants reveals that mutant F316A is 14-fold more efficient than the wild-type at deaminating cytosine to uracil. The mutant D314A enzyme demonstrates a dramatic decrease in cytosine activity (17-fold) as well as a slight increase in activity toward 5-FC (2-fold), indicating that mutant D314A prefers the prodrug over cytosine by almost 20-fold, suggesting that it may be a superior suicide gene.     

AdoMet-dependent methyl-transfer: Glu119 is essential for DNA C5-cytosine methyltransferase M.HhaI

The role of Glu119 in S-adenosyl-L-methionine-dependent DNA methyltransferase M.HhaI-catalyzed DNA methylation was studied. Glu119 belongs to the highly conserved Glu/Asn/Val motif found in all DNA C5-cytosine methyltransferases, and its importance for M.HhaI function remains untested. We show that formation of the covalent intermediate between Cys81 and the target cytosine requires Glu119, since conversion to Ala, Asp or Gln lowers the rate of methyl transfer 10(2)-10(6) fold. Further, unlike the wild-type M.HhaI, these mutants are not trapped by the substrate in which the target cytosine is replaced with the mechanism-based inhibitor 5-fluorocytosine. The DNA binding affinity for the Glu119Asp mutant is decreased 10(3)-fold. Thus, the ability of the enzyme to stabilize the extrahelical cytosine is coupled directly to tight DNA binding. The structures of the ternary protein/DNA/AdoHcy complexes for both the Glu119Ala and Glu119Gln mutants (2.70 A and 2.75 A, respectively) show that the flipped base is positioned nearly identically with that observed in the wild-type M.HhaI complex. A single water molecule in the Glu119Ala structure between Ala119 and the extrahelical cytosine N3 is lacking in the Glu119Gln and wild-type M.HhaI structures, and most likely accounts for this mutant's partial activity. Glu119 has essential roles in activating the target cytosine for nucleophilic attack and contributes to tight DNA binding.     

A negative selection scheme based on the expression of cytosine deaminase in plastids

The enzyme cytosine deaminase (CD) encoded by codA catalyzes deamination of cytosine to uracil. CD is present in prokaryotes and in many eukaryotic micro-organisms, but is absent in higher plants. 5-fluorocytosine (5FC) is metabolized in CD-expressing cells, causing cellular death. A chimeric codA has been introduced into the tobacco plastid genome and 5FC was used to select against tissue culture cells and seedlings expressing CD. This negative selection scheme will be useful in identifying nuclear genes which control plastid gene expression in higher plants.     

Structure of the D142N mutant of the family 18 chitinase ChiB from Serratia marcescens and its complex with allosamidin

Catalysis by ChiB, a family 18 chitinase from Serratia marcescens, involves a conformational change of Asp142 which is part of a characteristic D(140)XD(142)XE(144) sequence motif. In the free enzyme Asp142 points towards Asp140, whereas it rotates towards the catalytic acid, Glu144, upon ligand binding. Mutation of Asp142 to Asn reduced k(cat) and affinity for allosamidin, a competitive inhibitor. The X-ray structure of the D142N mutant showed that Asn142 points towards Glu144 in the absence of a ligand. The active site also showed other structural adjustments (Tyr10, Ser93) that had previously been observed in the wild-type enzyme upon substrate binding. The X-ray structure of a complex of D142N with allosamidin, a pseudotrisaccharide competitive inhibitor, was essentially identical to that of the wild-type enzyme in complex with the same compound. Thus, the reduced allosamidin affinity in the mutant is not caused by structural changes but solely by the loss of electrostatic interactions with Asp142. The importance of electrostatics was further confirmed by the pH dependence of catalysis and allosamidin inhibition. The pH-dependent apparent affinities for allosamidin were not correlated with k(cat), indicating that it is probably better to view the inhibitor as a mimic of the oxazolinium ion reaction intermediate than as a transition state analogue.     

A sensitive genetic assay for the detection of cytosine deamination: determination of rate constants and the activation energy

Previously it has not been possible to determine the rate of deamination of cytosine in DNA at 37 degrees C because this reaction occurs so slowly. We describe here a sensitive genetic assay to measure the rate of cytosine deamination in DNA at a single cytosine residue. The assay is based on reversion of a mutant in the lacZ alpha gene coding sequence of bacteriophage M13mp2 and employs ung- bacterial strains lacking the enzyme uracil glycosylase. The assay is sufficiently sensitive to allow us to detect, at a given site, a single deamination event occurring with a background frequency as low as 1 in 200,000. With this assay, we determined cytosine deamination rate constants in single-stranded DNA at temperatures ranging from 30 to 90 degrees C and then calculated that the activation energy for cytosine deamination in single-stranded DNA is 28 +/- 1 kcal/mol. At 80 degrees C, deamination rate constants at six sites varied by less than a factor of 3. At 37 degrees C, the cytosine deamination rate constants for single- and double-stranded DNA at pH 7.4 are 1 x 10(-10) and about 7 x 10(-13) per second, respectively. (In other words, the measured half-life for cytosine in single-stranded DNA at 37 degrees C is ca. 200 years, while in double-stranded DNA it is on the order of 30,000 years.) Thus, cytosine is deaminated approximately 140-fold more slowly when present in the double helix. These and other data indicate that the rate of deamination is strongly dependent upon DNA structure and the degree of protonation of the cytosine. The data suggest that agents which perturb DNA structure or facilitate direct protonation of cytosine may induce deamination at biologically significant rates. The assay provides a means to directly test the hypothesis.     

Wild-type enzyme as a reporter of inhibitor binding by catalytically impaired mutant enzymes.

A method for the determination of inhibition constants for catalytically-debilitated mutant enzymes is described. The inhibitor is partitioned between the mutant and wild-type enzymes. Catalytic rates of the wild-type enzyme are used as the signal of inhibitor binding to the mutant enzyme. The method is validated with scytalone dehydratase, the Y50F mutant, and a potent inhibitor. The K(i) value for Y50F determined by this method is 0.49 +/- 0.10 nM. The K(i) value determined using the Y50F catalytic report for inhibitor binding in the absence of wild-type enzyme is 0.20 +/- 0.030 nM. The wild-type enzyme binds the inhibitor ten-fold less tightly, thus indicating that the hydrogen-bonding interaction between the Y50 hydroxyl group and the inhibitor (suggested by X-ray crystallography) is weak. The method is most useful when the catalytic activity of the wild-type enzyme is the most sensitive report of inhibitor binding and the mutant enzyme is greatly crippled in catalytic activity.     

Sequence-dependent enhancement of hydrolytic deamination of cytosines in DNA by the restriction enzyme PspGI

Hydrolytic deamination of cytosines in DNA creates uracil and, if unrepaired, these lesions result in C to T mutations. We have suggested previously that a possible way in which cells may prevent or reduce this chemical reaction is through the binding of proteins to DNA. We use a genetic reversion assay to show that a restriction enzyme, PspGI, protects cytosines within its cognate site, 5′-CCWGG (W is A or T), against deamination under conditions where no DNA cleavage can occur. It decreases the rate of cytosine deamination to uracil by 7-fold. However, the same protein dramatically increases the rate of deaminations within the site 5′-CCSGG (S is C or G) by ~15-fold. Furthermore, a similar increase in cytosine deaminations is also seen with a catalytically inactive mutant of the enzyme showing that endonucleolytic ability of the protein is dispensable for its mutagenic action. The sequences of the mutants generated in the presence of PspGI show that only one of the cytosines in CCSGG is predominantly converted to thymine. Our results are consistent with PspGI ‘sensitizing’ the cytosine in the central base pair in CCSGG for deamination. Remarkably, PspGI sensitizes this base for damage despite its inability to form stable complexes at CCSGG sites. These results can be explained if the enzyme has a transient interaction with this sequence during which it flips the central cytosine out of the helix. This prediction was validated by modeling the structure of PspGI–DNA complex based on the structure of the related enzyme Ecl18kI which is known to cause base-flipping.     

Site-directed mutagenesis identifies catalytic residues in the active site of Escherichia coli phosphofructokinase.

Six active site mutants of Escherichia coli phosphofructokinase have been constructed and characterized using steady-state kinetics. All but one of the mutants (ES222) have significantly lower maximal activity, implicating these residues in the catalytic process. Replacement of Asp127, the key catalytic residue in the forward reaction with Glu, results in an enzyme with wild-type cooperative and allosteric behavior but severely decreased Fru6P binding. Replacement of the same residue with Tyr abolishes cooperativity while retaining sensitivity to allosteric inhibition and activation. Thus, this mutant has uncoupled homotropic from heterotropic allostery. Mutation of Asp103 to Ala results in an enzyme which retains wild-type Fru6P-binding characteristics with reduced activity. GDP, which allosterically activates the wild-type enzyme, acts as a mixed inhibitor for this mutant. Mutation of Thr125 to Ala and Asp129 to Ser produces mutants with impaired Fru6P binding and decreased cooperativity. In the presence of the activator GDP, both these mutants display apparent negative cooperativity. In addition, ATP binding is now allosterically altered by GDP. These results extend the number of active site residues known to participate in the catalytic process and help to define the mechanisms behind catalysis and homotropic and heterotropic allostery.     

The rate of hydrolytic deamination of 5-methylcytosine in double-stranded DNA.

The modified base, 5-methylcytosine, constitutes approximately 1% of human DNA, but sites containing 5-methylcytosine account for at least 30% of all germline and somatic point mutations. A genetic assay with a sensitivity of 1 in 10(7), based on reversion to neomycin resistance of a mutant pSV2-neo plasmid, was utilized to determine and compare the deamination rates of 5-methylcytosine and cytosine in double-stranded DNA for the first time. The rate constants for spontaneous hydrolytic deamination of 5-methylcytosine and cytosine in double-stranded DNA at 37 degrees C were 5.8 x 10(-13) s-1 and 2.6 x 10(-13) s-1, respectively. These rates are more than sufficient to explain the observed frequency of mutation at sites containing 5-methylcytosine and emphasize the importance of hydrolytic deamination as a major source of human mutations.     

Analysis of the dominance of mutations in cAMP-binding sites of murine type I cAMP-dependent protein kinase in activation of kinase from heterozygous mutant lymphoma cells.

Structural lesions in cAMP-binding sites of regulatory (R) subunit of cAMP-dependent protein kinase caused identical increases in apparent constants for cyclic nucleotide-dependent kinase activation in preparations from cells that were hemizygous or heterozygous for mutant R1 subunit expression. No wild-type kinase activation was observed in extracts from heterozygous mutant cells. This "dominance" was investigated by characterizing expression of wild-type and mutant R1 subunits and properties of protein kinase from S49 mouse lymphoma cell mutants heterozygous for expression of wild-type R1 subunits and R1 subunits with a lesion (Glu200) that inactivates cAMP-binding site A. By both studies of cAMP dissociation and two-dimensional gel analysis, wild-type R subunits comprised about 35% of total R1 subunits in heterozygous mutants. Synthesis of wild-type and mutant R1 subunits was equivalent, but wild-type subunits were degraded preferentially. Hydroxylapatite chromatography revealed a novel R1 subunit-containing species from heterozygous mutant preparations whose elution behavior suggested a trimeric kinase consisting of an R1 subunit dimer and one catalytic (C) subunit. Wild-type R1 subunit was found only in dimer and "trimer" peaks; the tetrameric kinase peak contained only mutant R1 subunit. It is concluded that C subunit binds preferentially to mutant R1 subunit in heterozygous cells forming either tetrameric kinase with mutant R1 subunit homodimers or trimeric kinase with R1 subunit heterodimers. This preferential binding results both in suppression of wild-type kinase activation and differential stabilization of mutant R1 subunits.     

The specific binding of nuclear protein(s) to the cAMP responsive element (CRE) sequence (TGACGTCA) is reduced by the misincorporation of U and increased by the deamination of C.

Point mutations in the cAMP-responsive element (CRE) of the rat somatostatin gene promoter/enhancer sequence (TGACGTCA) were used as a model for assessing the effect of uracil, deriving either from misincorporation during DNA synthesis (T----U) or cytosine deamination (C----U), on the binding of sequence/specific regulatory proteins. The results show that the T----U conversion in both strands of the CRE palindromic sequence reduces its affinity for the CRE binding factor(s), suggesting the crucial role of the methyl group contributed by T for the correct recognition of the sequence. On the other hand, deamination of C in the CpG central dinucleotide (CpG----UpG) causes an increase of binding affinity which is further enhanced by the contemporary deamination in both strands. Then, both uracil misincorporation and cytosine deamination alter the binding to CRE sequence in vitro, suggesting that uracil, if not removed by uracil DNA-glycosylase, could be dangerous for cellular functions.     

Escherichia coli maltodextrin phosphorylase: contribution of active site residues glutamate-637 and tyrosine-538 to the phosphorolytic cleavage of alpha-glucans

The role of Escherichia coli maltodextrin phosphorylase (EC 2.4.1.1) active site residues Glu637 and Tyr538 which line the sugar-phosphate contact region of the enzyme was investigated by site-directed mutagenesis. Substitution of Glu637 by an Asp or Gln residue reduced kcat to approximately 0.2% of wild-type activity, while the Km values were affected to a minor extent. This indicated participation of Glu637 in transition-state binding rather than in ground-state binding. 31P NMR analysis of the ionization state of enzyme-bound pyridoxal phosphate suggested that Glu637 is also involved in modulation of the protonation state of the coenzyme phosphate observed during catalysis. Despite loss of proposed hydrogen-bonded substrate contacts, the Tyr538Phe mutant enzyme retained more than 10% activity; the apparent affinity of all substrates was slightly decreased. Mutations at either site affected the error rate of the enzyme (ratio of hydrolysis/phosphorolysis). Besides a role in substrate binding, the hydrogen-bond network of Tyr538 supports the exclusion of water from the active site.     

Cytosine deamination plays a primary role in the evolution of mammalian isochores

DNA melting is rate-limiting for cytosine deamination, from which we infer that the rate of cytosine deamination should decline twofold for each 10% increase in GC content. Analysis of human DNA sequence data confirms that this is the case for 5-methylcytosine. Several lines of evidence further confirm that it is also the case for unmethylated cytosine and that cytosine deamination causes the majority of all C-->T and G-->A transitions in mammals. Thus, cytosine deamination and DNA base composition each affect the other, forming a positive feedback loop that facilitates divergent genetic drift to high or low GC content. Because a 10 degrees C increase in temperature in vitro increases the rate of cytosine deamination 5. 7-fold, cytosine deamination must be highly dependent on body temperature, which is consistent with the dramatic differences between the isochores of warm-blooded versus cold-blooded vertebrates. Because this process involves both DNA melting and positive feedback, it would be expected to spread progressively (in evolutionary time) down the length of the chromosome, which is consistent with the large size of isochores in modern mammals.     

Three-dimensional structure and catalytic mechanism of cytosine deaminase

Cytosine deaminase (CDA) from E. coli is a member of the amidohydrolase superfamily. The structure of the zinc-activated enzyme was determined in the presence of phosphonocytosine, a mimic of the tetrahedral reaction intermediate. This compound inhibits the deamination of cytosine with a K(i) of 52 nM. The zinc- and iron-containing enzymes were characterized to determine the effect of the divalent cations on activation of the hydrolytic water. Fe-CDA loses activity at low pH with a kinetic pK(a) of 6.0, and Zn-CDA has a kinetic pK(a) of 7.3. Mutation of Gln-156 decreased the catalytic activity by more than 5 orders of magnitude, supporting its role in substrate binding. Mutation of Glu-217, Asp-313, and His-246 significantly decreased catalytic activity supporting the role of these three residues in activation of the hydrolytic water molecule and facilitation of proton transfer reactions. A library of potential substrates was used to probe the structural determinants responsible for catalytic activity. CDA was able to catalyze the deamination of isocytosine and the hydrolysis of 3-oxauracil. Large inverse solvent isotope effects were obtained on k(cat) and k(cat)/K(m), consistent with the formation of a low-barrier hydrogen bond during the conversion of cytosine to uracil. A chemical mechanism for substrate deamination by CDA was proposed.     

The Impact of the HydroxyMethylCytosine epigenetic signature on DNA structure and function

We present a comprehensive, experimental and theoretical study of the impact of 5-hydroxymethylation of DNA cytosine. Using molecular dynamics, biophysical experiments and NMR spectroscopy, we found that Ten-Eleven translocation (TET) dioxygenases generate an epigenetic variant with structural and physical properties similar to those of 5-methylcytosine. Experiments and simulations demonstrate that 5-methylcytosine (mC) and 5-hydroxymethylcytosine (hmC) generally lead to stiffer DNA than normal cytosine, with poorer circularization efficiencies and lower ability to form nucleosomes. In particular, we can rule out the hypothesis that hydroxymethylation reverts to unmodified cytosine physical properties, as hmC is even more rigid than mC. Thus, we do not expect dramatic changes in the chromatin structure induced by differences in physical properties between d(mCpG) and d(hmCpG). Conversely, our simulations suggest that methylated-DNA binding domains (MBDs), associated with repression activities, are sensitive to the substitution d(mCpG) ➔ d(hmCpG), while MBD3 which has a dual activation/repression activity is not sensitive to the d(mCpG) d(hmCpG) change. Overall, while gene activity changes due to cytosine methylation are the result of the combination of stiffness-related chromatin reorganization and MBD binding, those associated to 5-hydroxylation of methylcytosine could be explained by a change in the balance of repression/activation pathways related to differential MBD binding.