Artemis:DNA-PKcs: A new drug target for cancer treatment?

Cells encounter DNA double-strand breaks (DSBs) every day. Because they can lead to loss of genetic information, DSBs must be repaired. Nonhomologous DNA end joining (NHEJ) is the major pathway for repairing DSBs, and the key enzymes include Artemis nuclease and DNA-dependent protein kinase catalytic subunit (DNA-PKcs). For the first time, Dr Go Watanabe and his team from the University of Southern California, USA, describe the basal state of Artemis:DNA-PKcs complex and reveal binding sites for Artemis on DNA-PKcs. These sites can be pharmacological targets for cancer. Blocking the assembly of the Artemis:DNA-PKcs complex would affect DNA repair mechanisms in cancer cells, halting their growth and spread.

DNA, the double-stranded molecule of life, is supported by remarkable molecular machinery that regulates and maintains its structural integrity. When cells divide and multiply, or encounter ionizing radiation or oxidative free radicals, double-strand breaks (DSBs) occur. DSBs are the most dangerous type of DNA lesions because failure to resolve DSBs leads to extensive loss of genetic information. Therefore, cells must promptly fix the damage by joining the two correct DNA ends together. In vertebrate cells, nonhomologous DNA end joining (NHEJ) is the major pathway for repairing DSBs and involves several enzymes which facilitate specific chemical reactions.

The key enzymes regulating NHEJ are Artemis nuclease and DNA-dependent protein kinase catalytic subunit (DNA-PKcs). A nuclease is a type of enzyme that cuts DNA; DNA-PKcs is a kinase, which adds phosphate groups to the surface of proteins to change their properties and is only active when it encounters a broken DNA end. These two enzymes form a tight complex in cells, called the Artemis:DNA-PKcs complex. Other core enzymes involved in the NHEJ process are Ku70/80 proteins, the DNA polymerases (µ, λ and TdT), and the DNA ligase complex (XRCC4:DNA ligase IV, XLF and PAXX).

When DSBs occur, the NHEJ process starts with Ku70/80 proteins binding to the DNA ends at the break site, like first responders securing the damage site. Then, DNA ligase IV complex can come to the site and join the DNA ends together. However, parts of DNA ends are often damaged and require ‘end-processing’ by nucleases and polymerases before the joining step. The key nuclease responsible for removing damaged DNA in NHEJ is Artemis:DNA-PKcs complex. When Artemis:DNA-PKcs complex binds the broken DNA ends, DNA-PKcs becomes its self-activated phosphorylated form, which activates Artemis as an endonuclease for specific ‘DNA-cutting’ activity. In vertebrates, this endonuclease activity of Artemis is not only important for removing a variety of damaged DNA but also essential to produce antigen receptors, antibodies and T cell receptors, in order to develop the immune system.

Understanding the structure of Artemis:DNA-PKcs complex prior to its activation is important to increase the chances of successful treatment of some types of cancers. This is because chemotherapy kills cancer cells through a variety of mechanisms, which include DSBs. Malignant cells can, however, acquire resistance to the treatment due to their robust DNA repair mechanisms, continuing to grow and spread uncontrollably.

Go Watanabe and his team from the University of Southern California, USA, made an exciting experimental discovery in 2022. Using cryogenic electron microscopy (cryo-EM), a technique that employs extremely low temperatures (around -196°C), the researchers obtained snapshots of the basal state of Artemis:DNA-PKcs complex, which the researchers believe exists in cells, prior to the activation of DNA-PKcs and Artemis. This is the first study to document this state using cryo-EM.

New structural information

By analysing the cryo-EM data, Watanabe’s team discovered that the region responsible for the enzymatic activity of Artemis is positioned externally to DNA-PKcs. When studying structures of dynamic multiprotein complexes, many scientists deliberately introduce covalent bonds, or ‘crosslinks’ between amino acids of those flexible protein regions that yield a poorly defined signal. This helps to make the complex more rigid, augmenting the resolution of the technique. Watanabe emphasises: ‘Crosslinking was avoided in our samples because it can lead to a false degree of stabilisation of features within the complexes that are intrinsically flexible.’ That is why the researchers were able to observe that the catalytic domain of Artemis is highly dynamic and flexible and positions multiple places relative to DNA-PKcs.

The researchers also described the Artemis tail bound on DNA-PKcs in such detail that it is now possible to understand why a specific mutation on DNA-PKcs, in which the amino acid leucine is changed to arginine, interferes with the Artemis interaction, leading to a condition of the immune system known as radiosensitive T- B- severe combined immunodeficiency (RS-SCID). While the lack of Artemis activity in RS-SCID causes a heightened sensitivity to radiotherapy among its sufferers, the team discovered that blocking the conversion from the basal state to the activated state will be a promising strategy for developing treatments for leukaemia and other types of blood cancers. Watanabe explains: ‘However, that’s just a small portion of the Artemis tail. Artemis has a very long unstructured tail consisting of >300 amino acids. Where exactly the rest of the Artemis tail is located is still not clear.’

A pioneering research effort

Structural and biochemical studies on Artemis:DNA-PKcs are notoriously difficult. The purification of high-quality DNA-PKcs is particularly challenging because it tends to yield a mixture of different states of the protein, with varying levels of activity and degrees of phosphorylation. Watanabe points out that ‘if the protein is not properly purified and handled, results can easily be misleading.’ Interestingly, Watanabe observed a mysterious disulfide bond, a bridge between two sulfur atoms holding two arms of DNA-PKcs. He was puzzled by this finding since the conditions used in his protein purification protocol are aimed, as a standard, at disrupting these disulfide bond formations. ‘It is not clear why this covalent bond still exists even if the reducing condition was used during purification,’ says Watanabe. ’I am still skeptical about this finding, but I was always curious to know what’s holding these two flexible arms of such dynamic protein, so this is interesting.’ He believes such a bizarre finding warrants further investigation in future studies.

Another complication is posed by the very bulky nature of DNA-PKcs, which comprises 4128 amino acids. This DNA-PKcs forms huge multiprotein complexes with other NHEJ proteins, and these complexes (eg, Artemis:DNA-PKcs:Ku70/80:DNA, ~6200 amino acids, ~730 kDa) are also very dynamic and flexible. Thus, analysing the native state of complexes with a conventional polyacrylamide gel electrophoresis (PAGE) is challenging. In general, the overall size of the complexes is too large to be detected in conventional 4% polyacrylamide gel. Making a gel with a lower concentration was not practical because the efficient polymerisation reaction does not occur at such low concentration and the solution remains fluid-like. But it was obvious that in a 0% gel, protein complexes can travel without a problem. To get around this problem, Watanabe had to slightly alter the typical composition of the gel, adopting an agarose-acrylamide composite native gel electrophoresis technique, to produce a 2% polyacrylamide gel with the help of a little bit of agarose, which gives vital mechanical support. However, this composite gel is still very soft and fragile like ‘silken tofu’ so the researchers needed to be extremely careful when handling it.

Blocking the activation of Artemis

Thanks to the optimised composite gel system, the team demonstrated that a peptide of XRCC4, a part of the DNA ligase IV complex involved in NHEJ, which supposedly binds to a cleft region of DNA-PKcs, can compete with Artemis – disrupting the Artemis:DNA-PKcs complex formation. What also sets apart the researchers’ study is that they established a system which allows any researcher to design more specific peptide sequences as potential inhibitors of the Artemis:DNA-PKcs interaction and test them in any laboratory. One can use this system to analyse other large macromolecular complexes. The researchers explain that the mutually exclusive binding site, coined an Art-X4 cleft, within DNA-PKcs for the Artemis tail and the XRCC4 tail could be one of the therapeutic target sites to design disruptors of Artemis conversion to an active form as well as disruptors of the interaction with XRCC4.

Blocking the assembly of the Artemis:DNA-PKcs complex and inhibiting the activation of Artemis in acute lymphoblastic leukaemia via specific chemotherapeutics would affect DNA repair mechanisms in the transformed cells, halting their growth and spread. Watanabe’s team explains that the advantage of this approach is that Artemis inhibition has no clinical consequences for cells other than lymphocytes unless patients are given radiation therapy. Watanabe adds that ‘Artemis inhibitors would not affect memory B and T cells (types of lymphocytes that are part of the immune system) in the patients, thus largely preserving their existing immunity.’

A new drug target?

A significant finding in the study identifies a mutually exclusive binding site for Artemis and XRCC4 on DNA-PKcs – the Art-X4 cleft. The team also shows that an XRCC4 peptide disrupts the Artemis:DNA-PKcs complex. Further studies aimed at developing inhibitors of the Artemis:DNA-PKcs complex could prove a useful pharmacological strategy in the selective treatment of conditions where introducing DNA breaks could arrest the undesired growth of harmful cells.