Isolation of Temperature-sensitive p53 Mutations from a Comprehensive Missense Mutation Library

TL;DR

Screening 2,314 p53 missense mutants in yeast identified 142 temperature-sensitive (ts) p53 mutants, including 131 previously unreported. The ts mutants clustered in β-strands of the DNA-binding domain, and most tested also showed ts transactivation in mammalian cells.

Key findings

• From 2,314 p53 missense mutations, 142 ts mutants were selected (6.1%); 131 were previously unreported. Of these, 131 (92%) mapped within the core DNA-binding domain (residues 97–286), and 70 (53.4%) mapped on β-strands; 15 residues were ts hot spots: Thr155, Arg158, Met160, Ala161, Val172, His214, Ser215, Pro223, Thr231, Thr253, Ile254, Thr256, Ser269, Glu271, and Glu285.
• In p53-deficient Saos-2 cells, 54 mutants were tested and 48 (89%) were ts in at least one of six promoters; 16 were ts in all six promoters, while 32 were ts in a limited number.
• Among six mutants examined with conformation-sensitive antibodies, P219S and T253I showed an obvious ts increase in the PAb1620/PAb240 ratio; R175H and R273H were not ts by this assay.
• In the IARC database, 10.3% (110 of 1,066) of distinct single-nucleotide missense mutants and 10.4% (1,254 of 12,032) of total mutation reports were thought to be ts.

Methods

• A library of 2,314 p53 missense mutations was stably expressed in haploid or diploid *Saccharomyces cerevisiae* strains with p53-responsive fluorescent reporters, including p21WAF1, MDM2, BAX, 14-3-3, p53AIP1, GADD45, Noxa, and p53R2; fluorescence was measured at 30 and 37 °C. Ts mutants met the criteria M30/W30 ≥ 0.7, M37/W37 ≤ 0.5, and M30/M37 ≥ 2.
• Mammalian validation used p53-deficient human osteosarcoma Saos-2 cells: 54 mutants were tested by luciferase assay with six promoters at 32 and 37 °C, with ts defined by t test p < 0.001; immunoprecipitation used PAb1620 and PAb240, and mapping used the 1TUP structure visualized with Cn3D.

Limitations

• The paper states that ts criteria were strict and differed between yeast and mammalian systems, so some ts mutants may be defined in one system but not the other; previously reported ts mutants may have been missed because of different assays, weak ts phenotypes, or mechanisms absent in yeast, such as V143A requiring human ASPP2.
• Evident in the text: only 54 of 142 yeast-isolated ts mutants were tested in mammalian cells, and only six were examined by conformation-sensitive antibodies, so mammalian ts frequency and epitope findings rest on subsets.
• The paper notes that ts transactivity spectra differed in part between human and yeast cells and speculates this may be due to different p53 binding elements, temperatures, and post-translational or protein-interaction contexts.