p53 complexed with DNA (PDB ID: 1TUP) acts as a tumor suppressor that binds to DNA and inhibits the growth of tumor cells in humans (1). The protein receives significant attention in the scientific community from researchers who seek to understand the underlying chemical mechanisms that govern the biological effects of cancer, a disease that affects millions of Americans. It consists of three subunits with a molecular weight of 86840.62 Da and and an isoelectric point of 6.83 (6). As a tumor suppressor protein, p53 controls a cell cycle checkpoint necessary for maintaining genomic integrity by inducing the expression of the Cip1 cell cycle inhibitor, resulting in cell cycle arrest in the G1 phase and a subsequent suspension to cell division. Given its role in human cells as a protein that assesses DNA damage prior to cell division, a significant portion of its structure consists of two core domains involved in DNA binding, which provide insights into how p53’s tertiary structure inform its function in binding DNA. A 1993 study on the DNA-binding domain of p53 revealed that the core portion of p53 (residues 102-292) folds into a compact structure domain containing the sequence-specific DNA binding activity of the protein (2). A third core domain in p53 does not bind DNA; rather, it establishes protein-protein contacts that stabilize crystal packing (1). An analysis of the DNA binding region lends may lend insight into the understanding of the function of p53 in its role as a tumor suppressor.

The DNA-binding secondary structure consists of an antiparallel β-sheet sandwich containing four or five β-strands that act as a scaffold for two large loops, L2 and L3, and a loop-sheet-helix motif containing a smaller L1 loop (1,2). A tetrahedrally coordinated zinc atom, the only metal ligand in the structure of p53, holds the two large loops together in the ligand binding pocket. Zinc interacts with His-179 of the L2 loop and Cys-173 of the L1 loop through electrostatic interactions. From the loop-sheet-helix motif, the L1 loop and an α-helix fit in the major groove of DNA such that they lie on the edges of the base pairs. An arginine residue belonging to one of the two large loops fits in the DNA’s adjacent minor groove. Regarding secondary structure, a short, three-stranded β-sheet consisting of an S2- S2’ hairpin and four carboxyl-terminal residues of the extended S10 β-strand compose the loopsheet-helix motif. Its interactions with the L1 loop may be characterized by the hydrogen bonds between the backbone atoms in the L1 loop and the NH2-terminus of the α-helix in the loopsheet-helix motif. The L2 and L3 loops both occur between β strands belonging to the β sandwich, the former between S45 and S5, and the latter between S8 and S9. All loops exhibit a lack of backbone hydrogen bonding for which the zinc ligand compensates, maintaining the thermodynamic stability of the binding site between p53 and DNA (1,2).

There exist four sites at which p53 binds to DNA. All exhibit a variation of the pentamer consensus sequence PuPuPuC(A/T), where Pu resembles a purine nucleotide (1,3). The core domain of p53 binds to one of the four pentamers, which orient themselves in alternating directions; occasionally, base pairs outside the pentamer consensus participate in binding. One of the two core domains binds a consensus site while the other binds a nonconsensus site, thus achieving fewer DNA contacts. Two of the three purines in the consensus sequence make side chain contacts with one of the core domains in the major groove, including Lys-120 from the L1 loop which donates hydrogen bonds to a guanine base, and Cys-277 which occurs before the α-helix and receives a hydrogen bond from a cytosine base. Minor groove interactions exhibit more extensive hydrogen bonding; as an example,the guanidinium group of Arg-248 hydrogen bonds with a thymine base and an adenine (3).

While the loop-sheet-helix motif and the two large loops comprise the DNA-binding surface of the protein, the most critical aspect of p53’s structure, they also present a higher frequency of mutations than any other region within p53 (1). Deletion mutations to p53 result in the inactivation of tumor suppression and the proliferation of cancerous cells; hence, understanding the structure of the p53 core domain and the mechanism by which it binds to DNA are areas of academic interest (3). Frequently mutated residues include Arg-248 with 9.6 percent of the p53 mutations, Arg-273 with 8.8 percent, Arg-175 with 6.1 percent, and Gly-245 with 6.0 percent. These residues in p53’s primary structure also make more direct contact with DNA than most of the p53 residues. Arg-248 contacts the minor groove of DNA, while Arg-273 from the loopsheet-helix motif makes contact with a DNA backbone phosphate. Arginine’s role in binding to DNA within the structure of p53 can be explained by its positive charge at cellular conditions, which enables it to bind to negatively-charged DNA. Its side chain plays a crucial role in stabilizing the structure of the DNA binding surface of p53 through van der Waals, electrostatic, and hydrogen bonding interactions. If arginines were not present in the structure of p53, then interactions between p53 and the DNA to which it binds would be weaker. The guanidinium groups enable the arginines in p53 to act as hydrogen bond donors. Extensive weak interactions are enabled by the high frequency of polar residues in the structure of p53 (1,2).

Additional information about the structure of p53 may be gleaned by comparing p53 to a related protein. One such protein is p53 DNA-binding domain in the zinc-free oxidized state (PDB ID: 2P52) found in mice (4). The comparison protein in mice functions similarly to that in humans in that it acts as a tumor suppressor; however, it differs from the p53 protein found in humans in that it does not contain a zinc ligand. This structure, determined to be an appropriate comparison structure through use of the Dali server and PSI-BLAST searches, contains just one primary domain, unlike the p53 protein found in humans which contains multiple domains. The Dali server determined that the p53 DNA-binding domain in the zinc-free oxidized state closely matches the structure of p53 complexed with DNA in humans through a sum of pairs method to compare the three-dimensional structures of the two proteins. PSI-BLAST, which arrived at the same result, uses a slightly different method for comparing the structures of the two proteins in that it examines a protein database in search of proteins that contain amino acid sequences similar to that of p53 complexed with DNA in humans (4). Its E-value serves as a measure of sequence similarity by describing the number of search results one can “expect” to see by chance when searching a database of a particular size, where a lower E-value is associated with more significant sequence similarity. With an E-value of 9e-129 resulting from the PSI-BLAST, the structure of p53 DNA-binding domain in the zinc-free oxidized state in mice was determined to be similar to that of p53 complexed with DNA in humans. Additionally, with a Z-score of 15.8 resulting from the Dali server search, a value greater than 2, the structure of p53 DNA-binding domain in the zinc-free oxidized state was determined to be significantly similar to that of p53 complexed with DNA in humans (5). ExPASy analysis of p53 revealed an estimated molecular weight of 33,000 Da and an isoelectric point of 6.98 (6). Structurally, the primary difference between the two proteins is that the p53 DNA-binding domain in mice does not contain a zinc ligand, whereas the p53 complex with DNA in humans contains a zinc ligand instrumental to stabilization of the DNA binding site through ionic interactions. The L1 loop, L2 loop, and loopsheet helix motif from the core domains, in addition to the third structural domain, are all conserved in the structure of p53 found in mice.

Like p53 complexed with DNA in humans, the p53 DNA-binding domain in mice has tumor suppressor properties. It also initiates cell cycle arrest by binding mouse DNA and subsequently inducing apoptosis in cancerous cells. For this reason, researchers interested in tumor suppression in humans often select the p53 DNA-binding domain in mice as a target of study, employing mouse models of cancer in order to investigate cancer at a biochemical level. In both in-vivo and in-vitro contexts, p53 knockout mice can be compared to p53 mice in order to gain an understanding of tumor suppression in mice and, more broadly, humans. A 1996 study on tumor development in mice found that null (p53 -/-) knockout mice have an average tumor development time of 4.5 months, whereas heterozygous (p53 +/-) mice have an average tumor development of 18 months, illustrating the significance of p53 as a tumor suppressor protein (7).

Because of p53’s role as a tumor suppressor, it is a common target of drugs for pharmacologists. Efforts to develop drugs that can activate or potentially restore p53 pathways have focused on p53 inhibitors such as MDM2, a negative regulator of p53 (8). However, ontarget toxicities associated with p53 modulator inhibition have stalled their development. Other areas of pharmacological research regarding p53 include the development of gene therapy technologies that allow for the introduction of p53 through viruses that can deliver the protein to human cells.

Cancer remains one of the greatest problems facing the scientific community and the population at large; as a result, the study of p53 and its complex with DNA provide significant insights into cell growth, cell death, and the treatment of cancer. In approximately half of all cancer cases, mutations to p53 prevent tumor suppression and result in the uncontrolled growth of human cells (9). Additional research into p53 using mouse models containing the p53 DNA- binding domain may reveal information that could ultimately result in the elusive cure for cancer in humans.