Current advances in the role of classical non-homologous end joining in hematologic malignancies

1.1 Classical non-homologous end joining general introduction

The Ku70/80 complex serves as the primary sensor for DSBs, binding to the damaged DNA ends and initiating the recruitment of downstream NHEJ components. DNA-dependent protein kinase catalytic subunit (DNA-PKcs) is subsequently recruited, displacing Ku and positioning itself at the DNA termini to facilitate further repair. Key scaffolding proteins, including X-ray repair cross complementing protein (XRCC4), DNA ligase IV (LigIV) and XRCC4-like factor (XLF), assemble at the site, enabling end-bridging through intricate protein–protein interactions, while Artemis processes hairpin DNA ends to prepare them for repair. Opposing DNA-PKcs molecules undergo trans-autophosphorylation, leading to structural reorganisation of the complex and eventual dissociation of DNA-PKcs, thereby exposing the DNA ends for processing and ligation. PAXX interacts with XRCC4 and XLF, providing additional structural stability to the complex and functioning redundantly with XLF to reinforce the scaffolding network. PNKP, a human kinase and phosphatase, and TDP1, an enzyme, play a role in processing tricky DNA ends, paving the way for ligation. Besides, polymerases like Polμ, Polλ interacts with the complex by fitting into gaps in non-complementary ends, so as to lower the extent of deletions. Artemis is considered to be the major nuclease in cNHEJ, as other nucleases, namely aprataxin and APLF, also play their roles in the reparation.^{1-3, 5, 6} Furthermore, MRI/Cyren, a cell-cycle-specific inhibitor of cNHEJ, plays a dual role by enhancing the retention of cNHEJ factors and compensating for the absence of XLF in lymphocytes, ensuring repair efficiency.^{7, 8}

1.2 Ku70/80 heterodimer

Ku70 and Ku80 are synthesised from the XRCC6 and XRCC5 genes. During the initial step of DNA end recognition, the Ku complex detects DSBs, binds tightly to the DNA termini, protects them from further damage, and signals other NHEJ proteins to join the repair process.^{6, 9}

Ku70, widely studied for its role in DNA repair, also participates in diverse biological processes such as aging, cellular senescence, apoptosis, and the replication of certain viruses. Additionally, research has revealed its function in innate immunity, where it acts as a pattern recognition receptor. Upon detecting viral DNA in human cells, Ku70 initiates an antiviral immune response by inducing the release of interferons and pro-inflammatory cytokines.¹⁰

Studies on knockout mice demonstrate that deleting either the Ku70 or Ku80 gene in p53-mutated mice results in similar phenotypic consequences, including increased sensitivity to clastogens, accelerated aging, and a heightened risk of pro-B-cell lymphoma, although the impact of Ku70 deletion is less pronounced compared to Ku80.¹¹ Earlier investigations in mice further indicate that the deletion of one or both genes does not significantly elevate lymphoma incidence but is strongly associated with premature aging.¹² Additionally, Ku80/p53 double-deficient mice exhibit reduced resistance to experimentally induced medulloblastoma, while DAOY cells, a human medulloblastoma cell line, are naturally devoid of nuclear Ku70.^{13, 14} Despite these findings in murine models, there remains a lack of robust and conclusive evidence regarding the effects of Ku70 or Ku80 gene knockout in human cells, highlighting the need for further research in this area.

Through targeted gene sequencing, a study identified altered expression of XRCC5 and XRCC6 in B-cell lymphoma samples, particularly diffuse large B-cell lymphomas (DLBCLs). These changes led to defects in NHEJ, contributing to genomic instability in these tumours.¹⁵ Similar findings were observed in human pre-B cell samples associated with leukaemia, where heterozygous Ku70 mutant pre-B leukaemic cells exhibited increased sensitivity to DNA-damaging agents, a preference for microhomology-mediated end joining (MMEJ), and elevated markers of cellular aging following radiation exposure.¹⁶

Recent research has highlighted Ku70 as a critical factor linked to the frequency of chromosomal translocations in human lymphocytes after radiation exposure and in T-cell acute lymphocytic leukaemia (ALL). This connection suggests that Ku70 could serve as a valuable biomarker for assessing radiation-induced damage and may also emerge as a potential therapeutic target in the cancer treatment.¹⁷

Additionally, a cross-sectional study involving 95 acute myeloid leukaemia (AML) patients revealed that reduced expression of XRCC6, the gene encoding Ku70, correlated with improved prognosis.¹⁸

Single nucleotide polymorphisms (SNPs) are naturally occurring variations in the DNA sequence that are prevalent within populations and can significantly impact gene regulation and functional outcomes. This study further identified that SNP rs2267437, in the XRCC6 gene can suppress its expression, leading to reduced levels of its encoded protein, Ku70, as well as a worsened prognosis in AML.¹⁸ Besides, in acute lymphoblastic leukaemia, SNP rs5751129 of XRCC6 suppresses its trasnscription, resulting in an augmented odds; for XRCC5 rs828907, a higher odds was observed too.¹⁹ Another study highlighted that acetylation of Ku70 can inhibit the NHEJ repair pathway, thereby promoting apoptosis. Emerging evidence suggests that Ku70 is a potential target for histone deacetylase inhibitors and plays a pivotal role in the induction of programmed cell death.²⁰ Additionally, OTUD5, a specific deubiquitinase for Ku80, enhances its stability and functions as a positive regulator of NHEJ repair, ensuring efficient DNA damage response.²¹ In a separate investigation, it was discovered that TOX interferes with the binding of the Ku70/80 complex to DNA damage sites, impairing NHEJ and promoting genomic instability in T-cell ALL.²²

1.3 DNA-dependent protein kinase catalytic subunit

The DNA-PK complex, a heterotetramer with a molecular weight of around 650 kDa, is formed by the association of DNA-PKcs, Ku70/80, and a DNA duplex. DNA-PKcs, encoded by the PRKDC gene and a member of the phosphatidylinositol 3-kinase-related kinase family, is the second factor recruited in the NHEJ pathway. This complex is essential for activating and phosphorylating a wide array of proteins involved in the repair process.^{5, 6}

Studies in mice reveal that DNA-PKcs-null mutations result in severe combined immunodeficiency (SCID), marked by the absence of mature B and T cells due to impaired V(D)J recombination.²³ The absence of DNA-PKcs also promotes MMEJ and increases chromosomal translocations, which drive the development of cancers such as leukaemias and lymphomas.²⁴ Mice lacking both PARP1 and DNA-PK exhibit a higher incidence of T-cell lymphoma, driven by p53 mutations and telomere fusion.^{25, 26} Additionally, a human case with a DNA-PKcs missense mutation showed defective Artemis activation and impaired NHEJ.²⁷

Recent studies have delved deeper into the functional roles of kinases, particularly their catalytically inactive forms. For instance, DNA-PKcs, when rendered inactive, has been shown to induce higher levels of genomic instability compared to its complete absence. This observation highlights the intricate challenges and potential risks associated with developing and deploying targeted kinase inhibitors, such as those aimed at DNA-PKcs, in the cancer treatment. It underscores the critical importance of precision in therapeutic approaches to avoid unintended consequences.²⁸

Complementing this perspective, a bioinformatics analysis revealed that DNA-PK activation is prevalent in the majority of AML blast samples, positioning it as a viable therapeutic target in AML. The study suggests that inhibiting DNA-PK could disrupt cancer cell survival mechanisms and enhance the effectiveness of existing treatments. In clinical practice, combining multikinase inhibitors with DNA-PK-specific inhibitors may provide substantial therapeutic advantages, potentially improving patient outcomes and addressing resistance to conventional therapies.²⁹

In the context of melanomas, recent research indicates that suppressing the DNA-PK activity can enhance tumour immunogenicity by upregulating neoantigen expression and presentation while expanding the population of neoantigen-reactive T cells. However, whether this effect translates to hematologic cancers remains to be validated.³⁰

Additionally, DNA-PK has been implicated as a transcriptional regulator, driving the expression of target enzymes that contribute to anthracycline chemoresistance in AML. This finding further emphasises the multifaceted role of DNA-PK in cancer biology and its potential as a therapeutic target.³¹

DNA-PK inhibitors have emerged as promising therapeutic agents in hematologic malignancies, with distinct effects across different leukaemia subtypes. AZD-7648, a potent DNA-PK inhibitor, has been shown to exert multiple anti-tumour effects in vitro, including the inhibition of cell multiplation, induction of DNA lesion, activation of apoptotic pathways, and stasis of the cell cycle in AML and chronic myeloid leukaemia (CML) cells. However, its therapeutic impact appears to be less pronounced in ALL, as evidenced by findings from a separate study.^{32, 33} Meanwhile, CC-115, a dual inhibitor targeting both TORK and DNA-PK, has shown clinical potential in chronic lymphocytic leukaemia (CLL) by disrupting critical signalling pathways essential for cell proliferation.³⁴ Peposertib (M3814), another DNA-PK inhibitor, effectively blocks the NHEJ pathway, enhancing p53-mediated cytotoxicity in acute leukaemia when combined with therapies that induce DSBs, such as ionising radiation or topoisomerase II inhibitors.³⁵ Additionally, in pre-B ALL cell experiments, the combination of the DNA-PK inhibitor NU7441 and the telomerase inhibitor MST-312 significantly increased apoptosis rates, suggesting a potential therapeutic strategy for this subtype.³⁶

In MM, inhibition of DNA-PKcs could enhance the efficacy of a DNA-damaging-inducing agent doxorubicin, but caused by immunity-related pathway rather than defective cNHEJ.³⁷

In DLBCL, somatic mutations in DNA-PKcs have been observed.³⁸

1.4 Artemis

The activation of Artemis, a 5′ to 3′ endonuclease crucial for resolving hairpin DNA structures, is mediated by DNA-PKcs.^{5, 31} Together, these proteins are integral to V(D)J recombination, a process vital for the development of lymphoid cells and the establishment of a functional immune system.^38-41

In addition, Artemis has been identified as a tumour suppressor, with its deficiency associated with immunodeficiency and an increased risk of developing lymphoma and non-lymphoid malignancies.⁴² In the context of ALL, targeting Artemis activity has emerged as a potential therapeutic strategy, as its inhibition can effectively suppress the proliferation of B- or T-ALL cells. However, this approach may come at the cost of compromised immune function due to its broader role in the lymphocyte development.³³ Experimental studies have demonstrated that pharmacological inhibition of Artemis significantly reduces the growth of B-ALL cell lines, while mature B-cell lines remain largely unaffected, highlighting its selective impact on immature lymphoid cells.³³ Structural investigations into the molecular architecture of Artemis have provided valuable insights, paving the way for the design of specific inhibitors that could enhance therapeutic precision in ALL treatment.⁴³ (See Table 1

In DLBCL, the expression of Artemis has been observed to be altered selectively.¹⁵

1.5 LigIV–XRCC4–XLF complex

The final step is implemented by the LigIV–XRCC4–XLF complex, which ligates the DNA ends.⁵

1.6 DNA ligase IV

Deficiency in LigIV has been shown to trigger extensive neuronal apoptosis and halt lymphocyte development, ultimately leading to late embryonic lethality in mice through the p53 pathway. However, mice with mutations in both LigIV and p53 are viable, exhibiting impaired lymphogenesis, a predisposition to pro-B-cell lymphomas, and an increased risk of medulloblastoma.^44-47

In a clinical case, a patient with LigIV deficiency presented with SCID. The absolute count of B cells was drastically reduced by 100-fold, while alphabeta T cells decreased by 10-fold, primarily due to defects in mechanisms responsible for generating immune diversity. These defects contribute to an elevated risk of tumourigenesis, particularly lymphoma and leukaemia.⁴⁸ Furthermore, recent evidence suggests that monoallelic LigIV missense mutations, through haploinsufficiency, can lead to immune dysregulation in humans, manifesting as autoimmunity or immunodeficiency.⁴⁹

Though LigIV SNP rs1805388 suppresses the expression of itself, no significant association with prognosis was observed in AML.¹⁸

1.7 X-ray repair cross complementing protein 4

Similar to LigIV deficiency, the absence of XRCC4 in mice leads to severe consequences, including embryonic lethality, widespread neuronal apoptosis, and impaired cellular proliferation. However, mice with double deficiencies in both XRCC4 and p53 can circumvent these effects, though they exhibit defective V(D)J recombination and impaired the lymphocyte development. These mice are predisposed to developing pro-B-cell lymphomas, likely due to heightened genomic instability.⁵⁰ Additionally, the combined deficiency of XRCC4 and p53 accelerates the onset of medulloblastoma, further underscoring the role of XRCC4 in maintaining genomic stability.⁵¹

XRCC4 has also been implicated in modulating apoptosis induced by chemotherapy agents such as paclitaxel and vincristine. This suggests that the regulation of XRCC4 could serve as a predictive biomarker for evaluating the efficacy of apoptosis-inducing therapies in the cancer treatment.⁵²

Findings indicate that genomic variants of XRCC4 genotypes may influence susceptibility to CML, and MM.^{53, 54} Global studies have explored the association between SNPs in XRCC4 and hematologic cancer risk. In childhood ALL, XRCC4 rs6869366 or XRCC4 rs28360071 are tied with increased odds, though the mRNA levels of XRCC4 are not altered; while in all childhood leukaemia, rs6869366 is also considered as a risk factor.^{19, 55}

1.8 XRCC4-like factor

Unlike other NHEJ factors, XLF deficiency in mice leads to a gradual decline in lymphocyte levels, which is generally less severe than SCID. Interestingly, XLF-deficient lymphocytes in mice do not exhibit significant defects in V(D)J recombination.⁵⁶ This phenomenon may be attributed to functional redundancy between XLF and other NHEJ components or DNA damage repair factors, such as PAXX and 53BP1.^{1, 56-59} However, mature B cells lacking XLF display moderate impairments in immunoglobulin heavy-chain class switch recombination. Mice deficient in both XLF and p53 do not show a pronounced predisposition to pro-B-cell lymphomas but are prone to developing medulloblastomas.⁶⁰

In humans, XLF deficiency can cause microcephaly, delayed neural development, and varying degrees of immune deficiency, characterised by reduced lymphocyte counts and increased radiosensitivity. Despite sharing the same genetic defect, clinical phenotypes and immunological profiles can vary significantly among individuals.^61-63

XLF- or PAXX-deficient mice are viable and exhibit only mild immune system impairments. However, mice lacking both genes experience embryonic lethality due to neuronal apoptosis, genomic instability, and nearly complete arrest of lymphogenesis. Studies have shown that XLF and PAXX collaborate in V(D)J recombination, while PAXX plays a critical and independent role in class switch recombination.^64-67 Remarkably, triple-deficient mice lacking XLF, PAXX, and p53 are viable but display a dramatic reduction in mature lymphocyte populations and a significant increase in pro-B-cell numbers.^{68, 69} Additionally, combined deficiency of XLF and MRI results in embryonic lethality,^{69, 8} whereas selective depletion of MRI in mice reduces class switch recombination activity in B lymphocytes and slows the proliferation of neuronal progenitor cells.⁷⁰