Research

Deciphering the Folding Pathway of the p53 Tumor Suppressor Protein

Matthew J. Gage
AZCC/NAU

Jesse Martinez
AZCC/UA

Abstract:

p53 is a key protein involved in protection against tumor formation and has become one of the most intensely studied tumor suppressors. However, there is a very poor understanding of the effects that various mutations have on the folding of the protein. It is our hypothesis that many of the mutations that effect p53 function affect the protein’s ability to fold properly. We intend to use a series of cell lines developed by Dr. Martinez that show defects in folding of p53 to begin an initial characterization of the folding pathway of p53. Specifically, we intend to 1) Determine in which mutant cell lines the p53 protein shows an increased propensity for aggregation, a sign of defects in the folding pathway, 2) identify chaperones that are essential for proper folding of the p53 molecule and 3) examine specific interactions between various domains of the protein. These studies will provide the necessary insight to begin to develop a model of the folding pathway of p53.
 

Introduction:

The p53 tumor suppressor protein is a key protein in the cellular response to DNA damage. p53 is central to many cellular anti-cancer mechanisms including arrest, apoptosis, and senescence (1). In part, due to this critical function, over 50% of all cancers involve a mutation in the p53 gene. p53 is a tetramer with each monomer having 393 amino acids. There are numerous post-translational modifications that are involved in the regulation of the protein. Each monomer consists of a number of domains, as shown in Figure 1.

Figure 1: Domain structure of the p53 monomer. Each p53 domain has four primary domains. The N-terminus consists of the activation domain, which upregulates the level of p53 in response to DNA damage. The main body of the protein is the DNA Binding Domain (DBD). At the C-terminus are the tetramerization domain (which is responsible for the oligomerization of the protein) and the non-specific DNA binding domains.

Figure 2: Binding of isolated p53 DNA Binding Domain to DNA. The complete structure of the p53 protein has not determined due to large regions of flexibility in the protein. The individual DNA Binding Domain was first determined. The p53 molecule binds through the major helices in the DNA binding domain in a classical helix-turn-helix motif. However, the association between the four DBD domains is not clear. Image from the EBI website (http://industry.ebi.ac.uk/~pow/Work/The_Campus/).
Figure 3. A. Proposed sandwich model of the p53 tetramer. Structures of the DNA-binding domain (DBD) and tetramerization domain (TD) are shown in ribbon diagram. The N-terminal transactivation domain (TAD) is shown in space filling representation. (Figure from Klein et al (2)).

Figure 3. B. Crystal structure of the tetramerization domain of p53. Shown are the four -helices and the associated strands in what is believed to be the orientation in the full-length protein. The helices pack together in a dimer of dimers packing arrangement.
     

Specific Aim #1 - Determination of the Oligomeric State of the p53 Protein in the Mutant Cell Lines (Eric Fanucci)

Figure 4: p53 DBD domain hot spot mutations. Shown is a ribbon diagram of the DNA-Binding domain of the p53 protein with four of the most prevalent p53 mutations (“hot-spot” mutations) shown in ball and stick representation.

Figure 5: Expression of p53 hot spot mutations. Shown is a western blot of the soluble proteins from cells transfected with either the native p53 gene or a “hot-spot” mutant. Lane 1 are control Hct116 cells. Lane 2 are p53-null Hct116 cells. Lane 3 is blank. Lanes 4-9 are p53-null Hct116 cells transfected with R282Q, wtp53, R248Q, G245S, R273H and R249S respectively. Notice that expression levels of the mutants are different than the wt-p53 and that the R273H mutant actually forms a higher molecular weight species than any of the other proteins.

Similar experiments to those shown in figure 5 are being performed on the mutant cell lines developed in Dr. Martinez’s lab to determine the oligomeric state of the p53 in those cells.
     

Specific Aim #2 - Identification of Chaperone and Folding-Associated Proteins that Regulate p53 Function
Previously, 20 cell lines were generated that had non-functional p53, although not all of the cell lines had mutations in the p53 gene. These cells are sensitive to heat stress (Figure 6) and have p53 sequestered in the cytoplasm. One cell line does not induce hsp70 in response to heat stress (Figure 7). This suggests a role for heat shock/chaparonin proteins in p53 activity and folding. The goal of this aim is to use siRNA to silence various heat shock and chaparonin proteins to determine which proteins are associated with p53 folding.
   
ALTR cells are highly sensitive to heat stress
Hsc70 is required for p53 nuclear importation in vivo

Figure 7: Induction of heat shock protein 70 is aberrant in some of the p53 resistant cell lines. Shown are hsp70 levels in various cell lines following heat shock. Notice that no hsp70 is seen in the ALTR 17 cells and many of the ALTR cell lines have differing levels of hsp70 relative to the control A1-5 cells.

Figure 8: Suppression of hsc70 expression also inhibits p53 nuclear localization. Expression of hsc70 in Saos2 cells was suppressed using siRNA. The cells were transiently transfected with a plasmid which expressed a GFP-p53 chimeric protein. The fraction of cells with nuclear (N) verses nuclear/cytoplasmic (N+C) p53 was determined visually using fluorescent microscopy. The extent of knockdown was determined by Western blotting for hsc70 (panel A). The fraction of cells with GFP-p53 in the nucleus is graphed in panel B. This suggests a role for hsc70 in nuclear translocation and regulation of p53 in addition to folding of the protein.
     

Specific Aim #3 - Determination of the Association Between Various Domains in p53 (Casey Goodyear, Justin Saul)

Interactions between the C-terminus and the DNA Binding Domain: There are currently two models for how the C-terminus of p53 regulates the transition between specific and non-specific DNA binding. In one model, the C-terminus binds to the DNA Binding Domain (DBD) when p53 binds DNA non-specifically and that post-translational modifications disrupt this interaction, leading to specific DNA binding. In the other model, there is no interaction between the C-terminus and the DBD. We are using FRET to determine if there is an interaction between the C-terminus and the DBD. (Casey)

Figure 9: Two models of p53 regulation by the C-terminus.

A. C-terminus binds DBD domain of p53 and inhibits binding of p53 to specific DNA sequence (4).

Figure 9. B. C-terminus binds to DNA non-specifically. Once the protein is activated, p53 binds DNA through the C-terminus through specific DNA-protein interactions (5).
       

Figure 10: Diagram detailing the basics of FRET. The fluorescent emission from one molecule overlaps with excitation of a second molecule. If the two are close enough in space, transfer occurs and a distance can be measured.

Figure 11: DBD of p53 with the one tryptophan in the domain highlighted. We will be using the one tryptophan in the C-terminus as a donor in FRET experiments to determine if the C-terminus can bind the DBD in the full-length protein.
 

Association of the DNA Binding Domains of p53:

One of the major questions about the structure of p53 has been how the DNA binding domains are associated in the functional tetramer. Several recent NMR studies have produced a model of the interface between DNA binding domains in both dimeric and tetrameric forms of the protein (2,6-8). These studies have also demonstrated that there is a fair amount of flexibility along this interface. The binding site for the fluorescent probe Lumio is being inserted along this interface to develop a system for measuring formation of this interface in folding studies. It is believed that this will also provide a system to measure the dynamic motions within the protein. (Justin)

Figure 12. The dimeric p53 DBD-DNA complex in comparison to the crystal structure of the p53 DBD in complex with 53BP2 (2). The location of the split Lumio binding domain is indicated by the arrow.
Figure 13A: Structure of the Lumio molecule. Lumio is a derivative of modified fluorescene. The EDT moeities quench the natural fluorescence of the fluorescence.


Figure 13. B. The EDT moeities are displaced in the presence of a Cys-Cys-Pro-Gly-Cys-Cys motif. (Figure from Invitrogen website)

       

References
  1. Supanska, A.; Dziembowska, M.; Ellert-Miklaszewska, A.; Gaweda-Walerych, K.; Kaminska, B. Cyclosprine A Induces Growth Arrest or Programmed Cell Death of Human Glioma Cells. Neurochemistry International 2005, 47, 430-441.
  2. Klein, C., Planker, E., Diercks, T., Kessler, H., Kunkele, K. P., Lang, K., Hansen, S., and Schwaiger, M. (2001) J Biol Chem276(52), 49020-490273.
  3. Jeffrey, P. D., Gorina, S., and Pavletich, N. P. (1995) Science267(5203), 1498-1502
  4. Ayed, A., Mulder, F. A. A., Yi, G.-S., Lu, Y., Kay, L. E., and Arrowsmith, C. H. (2001) Nature Structural Biology8(9), 756-760
  5. Hupp, T. R., and Lane, D. P. (1994) Curr Biol4(10), 865-875
  6. Kitayner, M., Rozenberg, H., Kessler, N., Rabinovich, D., Shaulov, L., Haran, T. E., and Shakked, Z. (2006) Mol Cell22(6), 741-753
  7. Ma, B., Pan, Y., Gunasekaran, K., Venkataraghavan, R. B., Levine, A. J., and Nussinov, R. (2005) Proc Natl Acad Sci U S A102(11), 3988-3993

Questions? Comments? Contact us