Thymidylate Synthase

Thymidylate synthase (EC 2.1.1.45 for canonical ThyA and EC 2.1.1.148 for flavin-dependent ThyX) is the conserved methyltransferase enzyme that catalyzes the reductive methylation of 2'-deoxyuridine-5'-monophosphate (dUMP) at carbon-5 (C5) of the pyrimidine ring to produce 2'-deoxythymidine-5'-monophosphate (dTMP), providing the sole de novo intracellular pathway for synthesizing thymidylate required for DNA replication and repair. Because ribonucleotide reductase reduces ribonucleotides (ADP, GDP, CDP, and UDP) directly into 2'-deoxyribonucleotides without acting on thymine ribonucleotides, cells initially generate 2'-deoxyuridine nucleotides (dUDP → dUTP → dUMP via dUTPase) rather than dTTP. Thymidylate synthase bridges this gap by installing the C5-methyl group that distinguishes the DNA base thymine (5-methyluracil) from the RNA base uracil.

In the evolutionary transition from the primordial RNA World to modern cellular genomes, thymidylate synthase executed the second of two sequential biochemical innovations proposed by Patrick Forterre and Peter Reichard. First, ribonucleotide reductase eliminated the reactive 2'-hydroxyl group of ribose to create uracil-containing DNA (U-DNA), stabilizing the sugar-phosphate backbone against alkaline transesterification cleavage by >10⁶-fold; second, thymidylate synthase replaced uracil with thymine to create canonical thymine-containing DNA (T-DNA), enabling DNA repair enzymes to recognize and excise mutagenic cytosine deamination lesions.

The Spontaneous Hydrolytic Deamination of Cytosine

While replacing ribose with 2'-deoxyribose solved the physical instability of the RNA backbone, early U-DNA genomes remained vulnerable to a chemical lesion of the nucleobases: the spontaneous hydrolytic deamination of cytosine (C) into uracil (U). In physiological aqueous solution (37 °C, pH 7.4), water attacks carbon-4 (C4) of cytosine to release ammonia (NH₃) with a first-order rate constant of k ≈ 7 × 10^-13 s^-1 in double-stranded DNA and k ≈ 1 × 10^-10 s^-1 in single-stranded DNA or RNA, generating 100 to 500 uracil lesions per human diploid genome every day.

Because uracil base-pairs with adenine (A) just as thymine does, any unexcised C → U deamination event in a template strand directs the incorporation of adenine instead of guanine (G) during the next round of replication, fixing a permanent C:G → T:A (or U:A) transition mutation. In an RNA genome or a primordial U-DNA genome (preserved today in select bacteriophages such as Bacillus subtilis phages PBS1 and PBS2), repair enzymes cannot distinguish a newly formed mutagenic uracil (arising from cytosine deamination in a G•U mismatch) from a legitimate ancestral uracil in an A•U base pair.

By evolving thymidylate synthase to tag all legitimate nascent pyrimidine-adenine pairs with a C5-methyl group (5-methyluracil = thymine)—paired with deoxyuridine triphosphatase (dUTPase), which hydrolyzes cytoplasmic dUTP to dUMP + PP_i to prevent DNA polymerase from incorporating uracil—cells rendered uracil an illegal base in DNA. Whenever uracil-DNA glycosylase (UDG) scans the minor groove of T-DNA and encounters an unmethylated uracil residue, it cleaves the N-glycosidic bond to initiate base-excision repair (BER), restoring the original C:G base pair and lowering genomic mutation rates by orders of magnitude.

Canonical ThyA: Covalent Catalysis and Dual Folate Utilization

Classical thymidylate synthase (ThyA in bacteria, TYMS in humans) is an obligate homodimer studied by Daniel V. Santi, Robert M. Stroud, and Frank Maley. Among folate-dependent enzymes, ThyA uses its cofactor—N⁵,*N*¹⁰-methylenetetrahydrofolate (CH₂H₄folate)—both as the one-carbon methylene donor and as the two-electron hydride reductant:

1. Nucleophilic Michael Addition at C6: Once dUMP and CH₂H₄folate bind in an ordered sequential mechanism, a conserved active-site cysteine thiolate (Cys146 in Lactobacillus casei, Cys195 in human TYMS) executes a nucleophilic 1,4-conjugate (Michael) addition onto carbon-6 (C6) of the dUMP pyrimidine ring. This covalent attachment converts C5 into a nucleophilic C5-enol / carbanion intermediate stabilized by hydrogen bonding to conserved arginine and asparagine residues.
2. Ternary Covalent Adduct Formation: The nucleophilic C5 carbanion of dUMP attacks the electrophilic iminium methylene carbon (N5=CH₂⁺) of ring-opened CH₂H₄folate, forming a transient covalent ternary intermediate unifying the enzyme (via Cys-S–C6), the dUMP substrate, and the tetrahydrofolate cofactor (via C5–CH₂–N5).
3. Proton Abstraction and β-Elimination: A general base abstracts the acidic C5-proton (H5) of the pyrimidine ring, triggering β-elimination of the tetrahydrofolate cofactor to leave an exocyclic C5=CH₂ methylene intermediate still covalently tethered to the catalytic cysteine at C6.
4. 1,3-Hydride Transfer and Product Release: 7,8-Tetrahydrofolate (THF) transfers a hydride ion (H^-) directly from its C6 position to the exocyclic methylene group of the pyrimidine intermediate, reducing C5=CH₂ into the C5–CH₃ methyl group of thymine while simultaneous β-elimination expels the Cys195 thiolate from C6, releasing dTMP and oxidized 7,8-dihydrofolate (DHF).

Because every molecule of dTMP synthesized by ThyA consumes one molecule of reduced tetrahydrofolate and generates oxidized DHF, ongoing DNA synthesis requires the coupled "thymidylate cycle": dihydrofolate reductase (DHFR) uses NADPH to reduce DHF back to THF, and serine hydroxymethyltransferase (SHMT) transfers a methylene group from L-serine to THF (forming glycine and CH₂H₄folate).

Non-Homologous Flavin-Dependent ThyX and Oncology Pharmacology

In 2002, Hannu Myllykallio and colleagues discovered that roughly 30% of sequenced bacterial and archaeal genomes—including human pathogens such as Mycobacterium tuberculosis, Helicobacter pylori, Borrelia burgdorferi, and Treponema pallidum—lack the thyA and folA (DHFR) genes yet synthesize dTMP de novo. These organisms encode ThyX, a non-homologous homotetrameric flavoprotein that evolved independently of ThyA. In ThyX, CH₂H₄folate serves only as the one-carbon methylene donor (releasing reduced THF directly without generating DHF), while a tightly bound flavin adenine dinucleotide (FAD) prosthetic group—a classic nucleotide coenzyme reduced to FADH₂ by NADPH—provides the hydride equivalent to reduce the exocyclic methylene intermediate into the C5-methyl group of dTMP.

In clinical oncology, Charles Heidelberger's fluorinated pyrimidine antimetabolite 5-fluorouracil (5-FU) (and its oral prodrug capecitabine) remains a standard therapy for colorectal, pancreatic, breast, and head-and-neck carcinomas. Inside cells, salvage enzymes convert 5-FU into 5-fluoro-2'-deoxyuridine-5'-monophosphate (FdUMP). FdUMP enters the TYMS active site and undergoes Steps 1 and 2 of the catalytic cycle: Cys195 attacks C6 of FdUMP, and C5 attacks CH₂H₄folate to form the covalent ternary adduct. At Step 3, however, the C5-fluorine atom cannot be abstracted as a fluoronium cation (F⁺), trapping the enzyme in an irreversible Enzyme–FdUMP–CH₂H₄folate suicide adduct (potentiated clinically by leucovorin). Deprived of dTMP, tumor cells undergo uracil misincorporation and UDG-mediated double-strand DNA breaks known as thymineless death. Complementary antifolate drugs inhibit TYMS directly (raltitrexed, pemetrexed) or block upstream THF regeneration at DHFR (methotrexate, trimethoprim, pyrimethamine).

Why does DNA contain thymine instead of uracil?

In water at physiological temperature, the DNA base cytosine (C) spontaneously loses its amino group (hydrolytic deamination) to become uracil (U) roughly 100 to 500 times per human cell per day. Because uracil base-pairs with adenine (A), an unrepaired C → U conversion causes a permanent C:G → T:A mutation during DNA replication. If uracil were a standard component of DNA (as it is in RNA), repair enzymes could not tell which uracils were original and which were mutated cytosines. By using thymidylate synthase to add a C5-methyl tag to all legitimate uracils—converting them into thymine (5-methyluracil)—the cell allows uracil-DNA glycosylase (UDG) to recognize and remove any unmethylated uracil in DNA.

How does the chemotherapy drug 5-fluorouracil (5-FU) kill cancer cells?

Once inside a dividing cancer cell, 5-fluorouracil (5-FU) is metabolically converted into FdUMP (5-fluoro-2'-deoxyuridine-5'-monophosphate), a molecular mimic of the natural substrate dUMP. When thymidylate synthase attempts to methylate FdUMP, its catalytic cysteine (Cys195) forms a covalent bond to C6 and links C5 to the folate cofactor (N⁵,*N*¹⁰-methylenetetrahydrofolate). To finish the reaction, the enzyme must remove a proton (H⁺) from C5, but FdUMP has a tightly bound fluorine atom at C5 that cannot be removed as F⁺. This traps the enzyme in an irreversible covalent adduct, starving the tumor cell of dTTP and triggering fatal DNA fragmentation (thymineless death).

Why do some pathogenic bacteria survive without dihydrofolate reductase (DHFR)?

In humans and model bacteria like Escherichia coli, canonical thymidylate synthase (ThyA) oxidizes its folate cofactor into dihydrofolate (DHF) during every catalytic turnover, requiring dihydrofolate reductase (DHFR) to recycle DHF back into tetrahydrofolate (THF). However, ~30% of bacteria—including Mycobacterium tuberculosis and Helicobacter pylori—use an alternative flavin-dependent enzyme called ThyX. Because ThyX uses FADH₂ and NADPH for the reduction step instead of oxidizing the folate ring, it releases reduced THF directly and allows those microbes to replicate DNA with little or no DHFR activity.

1. Carreras, C. W., & Santi, D. V. (1995). "The catalytic mechanism and structure of thymidylate synthase." Annual Review of Biochemistry, 64, 721–762. https://doi.org/10.1146/annurev.bi.64.070195.003445
2. Myllykallio, H., Lipowski, G., Leduc, D., Filee, J., Forterre, P., & Liebl, U. (2002). "An alternative flavin-dependent mechanism for thymidylate synthesis." Science, 297(5578), 105–107. https://doi.org/10.1126/science.1072113
3. Longley, D. B., Harkin, D. P., & Johnston, P. G. (2003). "5-Fluorouracil: Mechanisms of action and clinical strategies." Nature Reviews Cancer, 3(5), 330–338. https://doi.org/10.1038/nrc1074
4. Finer-Moore, J. S., Santi, D. V., & Stroud, R. M. (2003). "Lessons and questions from the structure of thymidylate synthase generating a methyl group." Biochemistry, 42(2), 248–256. https://doi.org/10.1021/bi020599a