Legend to figure on purine and thymidine synthesis.

Top: Purine nucleotide biosynthesis. Purines are synthesized at the level of the nucleotide; that is, with the ribose and phosphate already in place. The purine ring is put together from 2-carbon (glycine) and one-carbon units attached to tetrahydrofolate, and nitrogen from glutamine. Ribose and phosphate are put together to form phosphoribosyl pyrophosphate (PRPP). The amino group of glutamine is added to the ribose to form phosphoribosylamine, and then additional pieces are added to form the two rings of IMP. IMP is hypoxanthine (a purine base) + ribose + phosphate. AMP and GMP are synthesized from IMP by further modification to the hypoxanthine ring. See Purine Chapter for exact structures if you are interested.

There is an enzyme that catalyzes the conversion of AMP back to IMP, so AMP can serve a source of AMP and GMP. The same cannot be said for GMP, which cannot be converted back to IMP in mammals.

At 2 steps in the pathway, a one-carbon unit is added to the forming ring, donated by formyl-tetrahydrofolate (H4-folate, or THF, or FH4). The formyl carbon comes from the side group of serine (-CH2-OH). These one-carbon groups must be attached to FH4 in order to be incorporated into the ring. FH4 is formed from folate supplied in the medium via the enzyme dihydrofolate reductase (DHFR) . Thus if there is no folate there can be no purine biosynthesis. And if there is no DHFR activity then there can be no purine biosynthesis. DHFR activity can be missing because of mutation in the gene for the enzyme or by adding a powerful inhibitor of the enzyme, methotrexate (MTX or amethopterin; aminopterin is a very similar drug.).

Other drugs that will inhibit the purine biosynthetic pathway include azaserine, a glutamine analog, that inhibits the addition of an amino group at one step. Alanosine inhibits the path from IMP to AMP (but not to GMP). Mycophenolic acid (MPA) inhibits the path to GMP but not to AMP.

Another route to purine nucleotides is via salvage pathways that use pre-formed purines or purine nucleosides if they are available in the medium (or in the diet). Adenosine can be salvages using adenosine kinase that adds a phosphate to form AMP. The AMP can then be used to form GMP.

Three purine bases can be salvaged by adding ribose-phosphate from PRPP: adenine via adenine phosphoribosyltransferase (APRT) and hypoxanthine and guanine via the mammalian enzyme hypoxanthine-guanine phosphoribosyltransferase (HGPRT, or just HPRT) that accepts both substrates. The E. coli enzyme accepts xanthine as a substrate as well. Thus in the presence of hypoxanthine or adenine the de novo biosynthetic pathway is not needed. However, guanine will not suffice since it a cannot be back-converted to AMP.

If a drug is added that inhibits an enzyme in the de novo pathway to IMP, then the cell becomes dependent on the a functional salvage pathway. Thus in the presence of azaserine and hypoxanthine or adenine, a wild type cell will grow, using its salvage enzymes APRT or HPRT, respectively.

Analogs of the salvage pathway substrates that are non-toxic as bases, can become toxic if they serve as substrates for the salvage pathway enzyme. Thus 6-thioguanine itself in not toxic, but 6-TGMP is toxic. In a wild type cell with an active HPRT, 6TG becomes toxic. However, mutant cell lacking HPRT activity will be resistant to 6TG. Such mutants are easy to isolate because HPRT is X-linked, so there is only one active copy of the gene in mammals, and so only one copy of the gene need be mutated. Even spontaneous mutant that may appear at a frequency of 1 in a million are easily selected in 6TG-containg medium.

Similarly, APRT- mutants (more difficult to isolate as they are autosomally linked) are resistant to adenine analogs such as 8-azaadenine and diaminopurine.

Bottom:

Thymidylate biosynthesis. Once again synthesis of the base takes place at the nucleotide level. The de novo path leads first to UMP, a precursor of RNA. After a temporary phosphorylation to UDP, the sugar is reduced to the deoxy form, yielding dUDP, which is promptly dephosphorylated to dUMP. dUMP is the substrate for thymidylate synthetase, which takes a methylene group from methylene-FH4 and transfers it to the uracil ring to form 5-methyl-uracil (i.e., thymine), reducing the methylene to methyl in the process. The reduction is at the expense of FH4, which ends up as FH2. This FH2 must then be re-reduced to FH4, carried out by DHFR. Unique among the one-carbon transfers mediated by FH4, this transfer uses FH4 stoichiometrically ; in all other transfers, the FH4 acts catalytically (picks up a one-carbon unit and then transfers it without being changed itself).

Here a salvage pathway exists for the nucleoside (thymidine = thymine + deoxyribose), not the base (thymine). Thymidine kinase (TK) phosphorylates thymidine to yield TMP which then goes on to be incorporated into DNA as the triphosphate. Thymidine could be produced from the breakdown of DNA or it could be supplied in the medium (or the diet). In the presence of thymidine the steps from UMP to TMP are not needed. If de novo TMP synthesis is blocked by an inhibitor of thymidylate synthetase, then wild type cells can grow if thymidine is provided. 5-Fluorouracil powerfully inhibits thymidylate synthetase directly, actually forming a covalent bond to the enzyme. MTX inhibits the thymidylate synthetase reaction by preventing the formation of one of its substrates, methylene-FH4, by inhibiting the FH2 to FH4 reaction. Thus in the presence of MTX all the FH4 should end up as FH2, which is not active as a one-carbon transfer agent. 5-Bromodeoxyuridine is an analog of thymidine that can be phosphorylated by TK to form BrdUMP, which gets incorporated into DNA. Large amounts of BrdU in DNA can be toxic, especially if cells are exposed to long wavelength UV light (~ 310 nm) which is otherwise benign. Another analog of thymidine that becomes toxic when phosphorylated is trifluorothymidine. Mutant cells selected for resistant to BrdU or trifluorothymidine are usually TK-, as these drugs are harmless if not phosphorylated.

HAT medium contains the purine base hypoxanthine, amethopterin (a synonym for MTX) or aminopterin (similar), and thymidine. Wild type cells can grow in this medium using 2 salvage pathways to get around the block to de novo pathways to purine nucleotides and to TMP caused by the aminopterin or amethopterin (MTX). Mutant cells lacking either HPRT or TK cannot grow in HAT medium.