Severe MTHFR Mutations

Examining the severe forms of MTHFR deficiency (residual enzyme activity of 0-20%), no one common mutation has been found. However, nine mutations have been identified in the patients diagnosed with this severe type. In order to identify the specific mutations, they began by designing primers from the cDNA that would generate 250-300 bp fragments that overlapped by 50-75 bp at each end. Then they performed RT-PCR on the patients’ RNA with these primer pairs. The PCR products were analyzed using the SSCP protocol (single strand conformation polymorphisms), which used direct silver staining to detect single strands,” (Dockhorn-Dworniczak, B et al, 1991). “The electrophoretic mobility of single-stranded nucleic acid depended not only on its size but also on its sequence. [It’s been] found that most single base changes in up to 200-base fragments [can] be detected as mobility shifts.” (Orita, M., Suzuki, T. & Hayashi, K., 1989). Then, “the PCR products that showed a shift on the SSCP gels were purified by NuSieve (FMC Bioproducts) and sequenced directly (Cycle sequencing kit, GIBCO) to identify the change. If the change affected a restriction site, then the PCR product was digested with the appropriate restriction endonuclease and analyzed on polyacrylamide gels,” (Goyette et al, 1994). To screen for the same mutation in other patients, their RT-PCR products were first digested by the same restriction enzyme and analyzed by southern blotting using radiolabelled cDNA by standard technique. (Goyette et al, 1994)

Generally, the mutations are a nonsense codon, a 5’ splice site mutation, and seven missense mutations. Since each of these has only been found in one or two families, it reaffirms that MTHFR deficiency is a genetic disease resulting from genetic alterations. Moreover, seven of the nine mutations occur in CpG dinucleotides, which are known to be prone to mutation (Rozen, 1996); it’s likely that these are also recurrent mutations.

In Rozen’s study, the five unrelated patients with the lowest activity levels of the MTHFR enzyme (0-3%) were homozygous for one of three mutations. They all had an onset of symptoms within the first year of life, a severe form of the deficiency, and  for some an early death. One of them is a nonsense mutation 559C®T (R®X). The two missense mutations are 764C®T in exon 4 (proline to leucine) and 692C®T also in exon 4, which converted an evolutionary conserved threonine to methionine (Goyette et al, 1994). The 692C®T mutation is associated with the early onset phenotype, indicating a critical role in enzyme activity, or it is part of an important domain of the protein. In addition, it is located proximally to the binding site for the folate substrate. Therefore, the mutation may alter the binding efficiency of the folate. Since the 764C®T mutation does not occur in a conserved region, its importance has yet to be discovered.

The other six mutations result in a higher residual enzyme activity level (6-20%) than the previous three mutations. Typically, onset is in the second decade of life for these individuals, and they are compound heterozygous for the gene (Goyette et al, 1994, 1995). It’s interesting to note that “there is no overlap between mutations seen in the early-onset patients and those with later onset of symptoms,” (Rozen, 1996). One of the mutations results in a deletion of 57 bp (19 amino acids) due to the activation of a cryptic splice site. The splice site is the result of another mutation (G®A) at 792 bp in a conserved GT dinucleotide region of the intron's 5' end (Goyette et al. 1995). The other missense mutations convert an arginine to a cysteine (C®T at 985, 1015, and 1081 bp) or arginine to glutamine (G®A at 167 and 482 bp). More specifically, the 1081C®T substitution occurs in a hydrophilic sequence of amino acids, which has been identified as the linker region between the catalytic and regulatory domains of MTHFR (Goyette et al, 1994). This linker domain is highly charged and thus, it has been further suggested that it is probably located on the outside surface of the protein. However, the original arginine, a charged hydrophilic residue, is converted to an uncharged polar cysteine, which most likely affects the stability of the enzyme (Goyette et al, 1995). The 985C®T substitution occurs in an evolutionarily conserved region (in a stretch of nine conserved amino acids, further implying the importance of this amino acid and its location) and may also be involved in binding the FAD cofactor (the absence of the FAD cofactor completely inactivates the MTHFR enzyme). Additionally, the 1015C®T substitution is likely to be located in this region as well, yielding similar results (but it has not been found to be in an evolutionarily conserved region). Overall, it is important to realize that “all the mutations identified thus far are located in the 5’ end of the coding sequence, the region thought to encode the catalytic domain of MTHFR,” (Goyette et al, 1995).

A summary of the mutations discovered by Goyette et al (1995) can be seen here.