Understanding of most neurodegenerative disorders suffer from the lack of the knowledge of their etiology. This creates a huge and significant challenge for designing and targeting effective therapies for the patients suffering from these untreatable, and terminal conditions. For example, although pathological features of Alzheimer’s disease (AD) are the deposition of Aβ peptides and tau aggregates, the therapies that have been directed toward these proteins have not had a remarkable success. Part of the problem could be that we are not targeting early etiological causes and missing opportunities to dramatically improve the patient outcomes by not targeting early causes of the disease. Recent studies have shown that substantial DNA damage occurs early in many neurodegenerative diseases (Rolig and McKinnon, 2000; McKinnon and Caldecott, 2007; McKinnon, 2009; Madabhushi et al., 2014), including AD (Shackelford, 2006; Shanbhag et al., 2019), Parkinson’s disease (PD; Milanese et al., 2018; Schaser et al., 2019; Martín-Jiménez et al., 2020; Gonzalez-Hunt and Sanders, 2021), Huntington’s disease (HD; Browne et al., 1997; Bogdanov et al., 2001; Giuliano et al., 2003; Illuzzi et al., 2009; Enokido et al., 2010; Bertoni et al., 2011; Tamura et al., 2011; Askeland et al., 2018; Massey and Jones, 2018), spinocerebellar ataxia type 3 (SCA3; Chatterjee et al., 2015; Gao et al., 2015; Chakraborty et al., 2020), and amyotrophic lateral sclerosis (ALS; Konopka and Atkin, 2018; Mitra et al., 2019; Konopka et al., 2020); while it is becoming increasingly clear that DNA damage may occur early and maybe an important early contributor to these diseases, it is unclear how DNA repair processes contribute to the dysfunction of neurons. Because our lab along with others, have made significant contributions toward understanding transcription-coupled DNA repair (TCR) processes in HD and other polyglutamine (polyQ) diseases, for the sake of this review, we will largely focus on DNA repair mechanisms and their relationship with TCR in HD.

Huntington’s disease is a rare genetic disorder with a prevalence of 5–10 individuals per 100,000 in the Caucasian population and many more people are at risk of developing the disease (Morrison, 2012; Baig et al., 2016; Rawlins et al., 2016). Some juvenile forms of the disease also exist but are rare, accounting for about 5% of the cases (Nance and Myers, 2001). The term “Juvenile HD” is applied to the cases of HD with disease onset before the age of 20. While HD symptoms that appear in adults primarily manifest with pure movement disorder or chorea (No authors listed, 1993; Vonsattel and DiFiglia, 1998; Rosenblatt et al., 2003; Ross and Tabrizi, 2011), patients with Juvenile HD develop symptoms of mental disturbance and rigidity rather than chorea (Nance and Myers, 2001; Geevasinga et al., 2006).

Huntington’s disease is a complex disease, and the disease symptoms vary between individuals but is typically manifested with three distinct sets of symptoms known as classical triad consisting of: (1) involuntary choreatic movements and motor coordination defects, (2) mild to moderate cognitive decline, and (3) psychiatric and behavioral abnormalities. At the onset, all the symptoms may not appear simultaneously, and symptoms often vary between individuals and the affected members of a family, however, the symptoms progress predictably with age (Ross et al., 1997; Walker, 2007). Typically, the phenotypes manifest in the middle-age (Ross et al., 1997; Walker, 2007) when they present with psychiatric and cognitive symptoms, although there are cases of juvenile onset at ages as young as 2 years (Nance and Myers, 2001). The motor coordination defects in adult-onset HD can be divided into the choreiform movements with gait disturbances that usually appear at the early stages of disease progression, and motor impairments such as bradykinesia and rigidity that are usually observed in the later stages of the disease (No authors listed, 1993; Vonsattel and DiFiglia, 1998; Rosenblatt et al., 2003; Ross and Tabrizi, 2011). Importantly, there is mounting evidence that cognitive deficits become apparent in mutant carriers several years prior to the onset of clinical symptoms (Jason et al., 1988; Hahn-Barma et al., 1998; Lawrence et al., 1998; Marder et al., 2000; Berrios et al., 2002). Cognitive impairment starts with subtle disturbances but progressively leads to detectable cognitive decline. In addition, HD patients cannot organize or plan simple tasks in daily life and lose flexibility of mind and often fail to make simple adjustments. Lastly, the HD patients can also be diagnosed with frequent depression, excessive mood swings and anxiety, and signs of apathy, irritability, impulsivity, and social disinhibition. Thus, HD is a complex disease, but subtle molecular changes likely occur well before the clinical manifestation of the disease, providing a window of opportunity for therapeutic interventions.

While the toxicity arising from the aggregates adopted by mHTT is largely hypothesized to be the cause and progression of the disease (Davies et al., 1997; DiFiglia et al., 1997; Benn et al., 2005; Hoffner et al., 2007; Sawada et al., 2007; Bauer and Nukina, 2009; Legleiter et al., 2010; Sontag et al., 2012), it is still a subject of intense debate whether the aggregates (Legleiter et al., 2010; Sontag et al., 2012) play any toxic role in HD pathogenic mechanisms or whether they are actually neuro protective. It also remains to be established whether the aggregates or oligomeric forms of the mutant protein are toxic for the HD neurons (Hoffner et al., 2007; Shirendeb et al., 2011). Arguments in favor of the aggregate hypothesis include interactions of HTT aggregates with various chaperone proteins, including heat shock proteins HSP40 and HSP70, and components of the neuronal proteasome system, resulting in sequestration of the proteins involved in protein clearance system. This might affect the protein folding machinery as well as clearance of the misfolded proteins in HD (Davies et al., 1997; DiFiglia et al., 1997; Muchowski et al., 2000). In fact, the HDJ-2 and HSP70 chaperones were found to co-localize with HTT aggregates in the brain of R6/2 HD transgenic mice expressing the N-terminal truncated fragment of mHTT (Mangiarini et al., 1996), and exogenous expression of chaperons HDJ-1 and HSP70 reduced the amount aggregates formed by mHTT and ameliorated the toxicity in cell models of HD (Jana et al., 2000). This unusual protein interactions might result in the inactivation of other key neuronal proteins adding to the complexity of the disease pathology. Importantly, increasing the levels of chaperons facilitates the clearance of the aggregates and reduces the neurotoxicity in HD models (Figure 1), suggesting that protein folding machinery and/or protein clearance systems are compromised in HD. However, the precise mechanisms of how mHTT disrupts these fundamental cellular processes in HD remains to be established.

Over the last two decades it has become clear that both HTT and mHTT are present in the nucleus, directly interact with various transcription factors, cofactors, and RNA polymerase, and regulate transcription of neuronal genes (Cha, 2007; Benn et al., 2008; Ross and Tabrizi, 2011; Valor, 2015; Jimenez-Sanchez et al., 2017). Both HTT and mHTT interact with several nuclear transcription factors and cofactors, including cyclic AMP-response element-binding protein (CBP; McCampbell et al., 2000; Steffan et al., 2000; Nucifora et al., 2001), TATA-binding protein (TBP; Huang et al., 1998), the tumor suppressor transcription factor p53 (Steffan et al., 2000; Bae et al., 2005), the general transcription factors TFIID and TFIIF (Zhai et al., 2005), and specificity protein 1 (SP1; Dunah et al., 2002). The RNA polymerase large subunit A (POLR2A) also interacts with HTT and is detected in nuclear inclusions in the HD brain (Huang et al., 1998; Suhr et al., 2001). Additionally, the transcription cofactor TAF_II130 (a cofactor for CREB-dependent transcription) has also been shown to bind to expanded polyQ sequences to interfere with CREB-dependent transcription in HD (Shimohata et al., 2000a). Moreover, peroxisome proliferator-activated receptor γ (PPARγ) coactivator 1α (PGC-1α), a transcriptional coactivator (Lin et al., 2002, 2005) that works in combination with other transcription factors like PPARγ may play an important role in HD pathogenesis (Cui et al., 2006; Weydt et al., 2006). mHTT has been shown to interfere with the binding of the CREB/TAF4 complex on PGC-1α promoter, repressing its expression, and evidence suggest that reduced PGC-1α level/activity might contribute to neurotoxicity in HD (Cui et al., 2006; Weydt et al., 2006). Moreover, the repressor element 1 (RE1)-silencing transcription (REST) is a master regulator of neuronal gene transcription, repressing their transcription (Huang et al., 1999). Several REST target genes e.g., BDNF (brain-derived neurotrophic factor) are known to be downregulated in HD, and reduced BDNF activity contribute to HD pathogenesis (Zuccato et al., 2001; Canals et al., 2004). Overexpression of BDNF in HD transgenic mouse brain can rescue HD-like phenotypes (Xie et al., 2010). Evidence also suggest that REST regulates transcription of microRNAs, many of which regulate neuronal gene expression and are dysregulated in HD (Johnson et al., 2008; Johnson and Buckley, 2009). Further, HTT has been shown to bind REST in the cytoplasm and thus prevents REST-mediated suppression of BDNF (Zuccato et al., 2003, 2007). The presence of mHTT results in reduced interaction between mHTT and REST, and consequently nuclear levels of REST increases, repressing BDNF expression (Zuccato et al., 2003, 2007). Based on these observations, it is hypothesized that HTT probably assists in the assembly of transcription factor and co-activator into multi-protein active transcription complexes, which regulate target gene expression in neurons, and that mHTT perturbs either the assembly or the function of these transcription complexes (Luthi-Carter and Cha, 2003; Ross and Tabrizi, 2011; Kumar et al., 2014; Figure 1).

There is ample evidence to suggest that subtle alterations of synapses occur before overt neuronal death, and that these changes are predictive of the onset of behavioral problems associated with HD (Morton et al., 2001; Li H. et al., 2003; Joshi et al., 2009; Raymond et al., 2011). Huntingtin is highly expressed in the presynaptic terminals of nerve cells (DiFiglia et al., 1995), and various proteins that regulate both the presynaptic and postsynaptic function have been shown to interact with HTT (Li et al., 2002; Li J. Y. et al., 2003; Barron et al., 2021). These proteins affect a range of pre and postsynaptic functions, including endocytosis, synaptic homeostasis, and axonal transport on the presynaptic side, and postsynaptic receptor localization and dendritic protein transport at the post-synapse.

At the presynaptic terminal, HTT can regulate the normal exocytosis, endocytosis, and axonal transport. For example, Complexin II – a protein involved in neurotransmitter release (Rizo and Südhof, 2002), is decreased in HD patients (Morton and Edwardson, 2001), R6/2 transgenic mice (Morton et al., 2001) and neuronal cell model of HD (Edwardson et al., 2003). Another protein involved in docking of synaptic vesicles to the membrane is rabphilin 3A, the level of which is progressively decreased in R6/1 mouse model of HD (Smith et al., 2005). Also, reduction of HTT-interacting proteins involved in endocytosis might impair endocytic and intracellular trafficking in HD (Kalchman et al., 1997; Li et al., 1998; Sittler et al., 1998; Modregger et al., 2002; Singaraja et al., 2002; Trushina et al., 2004). HTT-interacting protein 1 (HIP1) is closely related HIP12, and these proteins are orthologs of the yeast Sla2p, which is involved in endocytosis (Engqvist-Goldstein et al., 1999). HIP1 interacts with clathrin and AP-2, whereas HIP12 interacts with F-actin and clathrin light chain (Legendre-Guillemin et al., 2002). The most studied of these proteins is HTT-associated protein 1 (HAP1) (Li et al., 1995) that interacts with the p150 Glued subunit of dynactin (Engelender et al., 1997; Li et al., 1998), which in turn, interacts with dynein, the motor protein involved in retrograde transport. These two proteins interact to facilitate efficient vesicular transport along the microtubules (Gill et al., 1991). Studies have shown that the binding of HAP1 with HTT is increased in HD (Li et al., 1995), and increased binding of mHTT with HAP1 depletes HAP1 from its normal functional site at the dynein/dynactin complex, negatively impacting axonal transport in HD (Gauthier et al., 2004).

These studies highlight the importance of HTT in normal synaptic function. It is no surprise then that various genetic therapies directed toward lowering mutant copy of HTT, which also reduced wild-type HTT level (Tabrizi et al., 2019), led to devastating consequences manifesting with motor and memory deficits (Kwon, 2021). It is interesting to note that both DNA damage and synaptic dysfunction are early signs of HD. However, there is no evidence that these two phenomena are linked or whether these processes are disrupted independent of one another in HD. One possibility is that synapses are disrupted because their energy demands are not met in HD. Most of the energy needed for synaptic function is supplied by mitochondria that localize to synapses and to the nodes of Ranvier in the motor neurons. There is plenty of evidence to suggest that mitochondrial function is disrupted in HD (Orr et al., 2008; Shirendeb et al., 2011; Siddiqui et al., 2012), and this could directly affect various synaptic processes. Second possibility is that in HD, there might a decrease in the interactions of HTT with its synaptic partners, much like we observed in DNA damage repair where mHTT can bind to the native DNA repair complex partners and decrease the efficacy of the complex to repair DNA damages (Gao et al., 2019). Finally, we know that inefficiency of DNA repair can disrupt transcription (Gao et al., 2019), which might affect the synaptic protein turnover leading to decreased synaptic function. However, to our knowledge, no study has yet demonstrated a direct link between DNA repair deficiency and synaptic dysfunction. Intriguingly, mutations that affect DNA repair have implications for many neurological diseases including ALS, AD, PD, and HD (Browne et al., 1997; Rolig and McKinnon, 2000; Bogdanov et al., 2001; Giuliano et al., 2003; Illuzzi et al., 2009; Bertoni et al., 2011; Madabhushi et al., 2014; Maynard et al., 2015; Askeland et al., 2018; Massey and Jones, 2018; Mitra et al., 2019; Shanbhag et al., 2019; Konopka et al., 2020). Thus, it would be interesting to investigate whether there is a direct link between synaptic demise and DNA repair and whether DNA repair is upstream of synaptic dysfunction in HD and related neurodegenerative diseases (Figure 1).

It is often debated whether persistent accumulation of DNA damage that is consistently observed in HD brain is the cause or a consequence of HD pathophysiology. Emerging evidence indicates that mHTT interferes with the DNA repair process in the post-mitotic neurons, leading to the accumulation of DNA damage (Qi et al., 2007; Enokido et al., 2010; Tamura et al., 2011). The wide-spread occurrence of unresolved DNA damages/lesions in postmortem HD patient brains, and especially in the pre-symptomatic HD transgenic mouse brain early during the disease progression suggest that accumulation of DNA damage/lesions in HD could be a cause rather than a symptom of HD (Enokido et al., 2010; Tamura et al., 2011; Gao et al., 2019). However, these observations do not rule out the possibility that DNA damages accumulate independently due to other possible causes of neurodegeneration in HD. Indeed, both could contribute to the HD disease pathology but recent evidence from genome-wide association (GWAS) studies of HD have found many genes that are involved in DNA damage repair in neurons (e.g., FAN1, LIG1, MLH1, MSH3, PMS1, and PMS2) as strong modifiers of age at onset and disease severity (Genetic Modifiers of Huntington’s Disease [GeM-HD] Consortium, 2015; Maiuri et al., 2019; Tabrizi et al., 2020), suggesting that impaired or defective DNA repair mechanism is mechanistically linked to HD pathomechanism. Second, many recent studies support the idea that presence of DNA damage/lesions is one of the early pathological hallmarks in HD brain (Browne et al., 1997; Bogdanov et al., 2001; Giuliano et al., 2003; Illuzzi et al., 2009; Bertoni et al., 2011; Askeland et al., 2018; Massey and Jones, 2018), indicating that DNA damage occur early in the disease progression and persistence of these lesions might trigger a cascade of pro-degenerative pathways in HD. Third, the presence of substantially higher oxidized bases in genomic DNA, in cell and animal models of HD, and in human postmortem brains (Browne et al., 1997; Bogdanov et al., 2001; Giuliano et al., 2003; Illuzzi et al., 2009; Bertoni et al., 2011; Askeland et al., 2018; Massey and Jones, 2018) also indicate problems with DNA repair system(s) in HD. Importantly, accumulation of oxidized bases in DNA, with frequent DNA deletions in HD postmortem brains strongly support the idea (Horton et al., 1995; Browne et al., 1997; Polidori et al., 1999; Siddiqui et al., 2012). Fourth, transgenic expression of mHTT in mouse brain also induces DNA damage early in the disease progression (Bogdanov et al., 2001; Gao et al., 2019), suggesting that DNA repair system(s) that maintains neuronal genome integrity might be affected early in the disease progression. Finally, many studies have implicated HTT in double strand break (DSB) repair mechanism, and mHTT has been shown to interact with Ku70, an essential DSB repair protein and a regulatory component of the DNA-dependent protein kinase (DNA-PK) (Enokido et al., 2010). Interaction and sequestration of Ku70 by mHTT has been suggested to impair non-homologous end-joining (NHEJ)-mediated DNA DSB repair in HD neurons (Enokido et al., 2010). DSBs are the most lethal form of DNA lesions in postmitotic cells like neurons and unlike other cells, neurons do not use homologous recombination. Instead, NHEJ pathway is used in neurons to repair these potentially lethal lesions (Rass et al., 2007; McKinnon, 2009; Lieber, 2010). Importantly, stimulating DSB repair activity by overexpressing Ku70 in mouse or Drosophila model of HD can rescue neurodegeneration (Enokido et al., 2010; Tamura et al., 2011), suggesting that mHTT impairs NHEJ-mediated DSB repair in HD. These findings strongly support the idea that failure of the neuronal DNA repair mechanisms due to polyQ expansion in HTT could be an important early contributing factor to the downstream degenerative events commonly observed in HD. Why would the DNA repair process be so important and effect the neurons disproportionately? We know that most of the mature neurons in adult brains are postmitotic and these repair processes maybe more important in neurons than other cells that can undergo apoptosis in case of heavy damage to their DNA. Also, we know that neurons are metabolically highly active, even at resting state, generating excessive levels of potentially damaging reactive oxygen species (ROS), and these damaging free radicals have been demonstrated to induce various types of lesions in neuronal DNA, including DNA single-strand breaks, oxidative damages, base deletions, DSBs (Lindahl and Barnes, 2000). Therefore, if not efficiently resolved, these potentially lethal and damaging lesions can accumulate during routine neuronal activities and become highly toxic if left unresolved. Thus, while DNA damage might contribute to HD pathology, and that HTT may play a critical role in DNA repair processes in neurons, until recently, it was not clear how the native HTT protein participates in the DNA repair processes in normal neurons, and how polyQ expansions in mHTT disrupts this vital genome maintenance process in neurons.

A recent study highlighting the importance of DNA damage in neurodegenerative disorders has shown that either genetic or pharmacological ablation of DDR kinase ATM activity significantly ameliorates neurotoxicity in HD animal models (Lu et al., 2014). This report strongly supports the emerging view that persistent accumulation of unresolved DNA damage in neuronal genome and the subsequent chronic activation of the DDR ATM signaling pathway probably is one of the major contributing factors in HD pathogenic mechanism(s). Our recent publication also supports this idea because the presence of mHTT or mutant ataxin-3 in the transcription-linked DNA repair complex dramatically impairs its DNA repair activities, resulting in DNA damage/strand break accumulation and chronic activation of the DDR pathway in HD (Gao et al., 2019) as well as in SCA3 (Gao et al., 2015) respectively. Furthermore, our study suggests that HTT may help stimulate TCR complex activity and DNA repair, while mHTT reduces DNA repair activity thereby, impairing DNA damage repair and enhancing the DNA strand breaks in genome.

Persistent and cumulative accumulation of DNA damage/strand breaks can result in chronic activation of the DDR serine-threonine kinase ATM as well as DNA-PK, which in turn phosphorylate the tumor suppressor protein p53 (Kurz and Lees-Miller, 2004; Guo et al., 2010). The transcription factor p53 is the primary target of DDR pathway, and many of the functions of ATM/DNA-PK are p53-dependent and activated p53 regulates a wide variety of cellular processes such as transcription, cell-cycle regulation, DDR and DNA repair, and cell death (Miller et al., 2000; Kurz and Lees-Miller, 2004; Culmsee and Mattson, 2005; Chang et al., 2012). Aberrant activation of p53 pathway has also been reported in several other polyQ-expansion-associated hereditary ataxias including HD (Bae et al., 2005; Wang et al., 2010; Chou et al., 2011). The activated p53 (phosphorylated p53) can transactivate expression of many proapoptotic genes such as BAX, Bcl2L2 (encoding BIM), PMA1P1 (encoding NOXA), BBC3 (encoding PUMA), triggering apoptotic pathways (Oda et al., 2000; Nakano and Vousden, 2001; Shibue et al., 2003; Chipuk et al., 2004, 2005; Cregan et al., 2004; Kuwana et al., 2005; Gao et al., 2015). Persistence of unrepaired DNA damage in genome thus can facilitate p53-depndent activation of pro-degenerative pathways in HD as described earlier (Bae et al., 2005). In response to DNA strand breaks, ATM directly phosphorylates p53, which in turn activates transcription of many pro-apoptotic genes (Oda et al., 2000; Chipuk et al., 2004). Expression of mHTT has been shown to increase p53 target gene expression, whereas deleting p53 in the HD transgenic mouse brain rescues the behavioral abnormalities (Bae et al., 2005) highlighting the fact that mHTT-mediated chronic activation of DDR-p53 pathway contributes to neurotoxicity in HD. These data indicate that HTT is important in DNA strand break repair, whereas mHTT impairs DNA repair activity, resulting in DNA strand break/damage accumulation, and cumulative accumulation of DNA damage results in chronic activation of ATM-dependent p53 signaling pathway to triggers neurotoxicity in HD.

Furthermore, persistence of DNA damage can also trigger pro-apoptotic signaling in another ATM and/or DNA-PK-dependent but p53-independent pathway in HD. There is ample evidence suggesting that persistent accumulation of DNA damages and subsequent phosphorylation-dependent chronic activation of the DDR kinase ATM and/or DNA-PK can also result in the phosphorylation of another downstream tyrosine kinase c-Abl (encoded by the mammalian homolog of the v-Abl oncogene from the Abelson murine leukemia virus) (Baskaran et al., 1997; Kharbanda et al., 1997; Shafman et al., 1997). Experimental evidence suggests that the activated c-Abl kinase (phosphorylated form of c-Abl) can constitutively associates with the protein kinase C delta (PKCδ), resulting in phosphorylation and subsequent nuclear translocation of PKCδ (Yuan et al., 1998; Yoshida, 2008; Adwan et al., 2011). Several studies have shown that cytosolic retention of PKCδ is required for cell survival and function while its phosphorylation and subsequent nuclear translocation can activate the apoptotic pathways (Basu et al., 2001; Yoshida, 2008; Adwan et al., 2011). However, it remains to be tested whether chronic activation of the DDR pathways and aberrant activation of the c-Abl→PKCδ pathway contributes to degeneration of neurons in HD. The presence of unrepaired DNA damages in neuronal genome and subsequent chronic activation of the DNA damage sensor ATM and/or DNA-PK kinases thus can concomitantly activate several pro-degenerative signaling pathways in neurons, and presence of chronic but low-grade DDR pathways in neurons can lead to neuronal dysfunction, and finally their premature demise in HD (Figure 4).

In response to DNA damage, several DDR kinases e.g., ATM, ATR, DNA-PK, p53, chk1 and 2 (checkpoint kinase 1 and 2), etc. are phosphorylated, and the phosphorylated form of these effector DDR kinases coordinately regulate DNA repair, neuronal survival, and their fate (Kurz and Lees-Miller, 2004; Guo et al., 2010). The important question that needs vigorous investigation is whether blocking ATM and/or DNA-PK with small molecule inhibitors can block neurodegeneration in HD. Therefore, significant efforts are being made to develop therapeutic approaches to block ATM or DNA-PK kinases (Lu et al., 2014). It is important to note that the DDR kinase also play important role in mediating DNA repair processes in the neurons by triggering phosphorylation-dependent activation of DNA damage repair proteins such as BRCA1, p53, PNKP, Ku70, ligase IV, etc. Therefore, one must consider the possibility that therapeutic approaches for HD aiming to block the DDR kinases with small molecule inhibitors may lead to impaired DNA repair in neurons, resulting in adverse consequences. Therefore, understanding the detailed molecular mechanism of DNA repair in neurons, and how polyQ expansion in HTT causes neurotoxicity affects these processes will provide more rational and targeted therapeutic strategies.

An important unanswered question is how mHTT impacts specific neuronal populations in HD despite its ubiquitous expression in the brain cells. Why do different neurons and/or brain regions show variable sensitivity to genotoxic insults? Neither the transcriptional signatures nor the activity patterns of any two classes of neurons are likely to be identical; therefore, the regions of robust and high-fidelity DNA repair, as well as those that are poorly maintained, likely vary from neuron to neuron. Although our findings suggest that the efficiency of TCR predominantly impacts the cortex and striatum with little effect on the cerebellar DNA, in HD, they do not explain why mHTT predominantly impairs TCR in specific brain regions in HD. How might inactivation of a potentially key DNA repair mechanism like TCR differentially affect various brain regions in HD? One possibility is that several factors specifically regulate the activity of this complex in various brain regions, particularly under genotoxic stress conditions. Inactivation of these region-specific regulators could potentially be impacted differentially causing the difference in the levels of DNA damage in different brain regions. Alternatively, specific brain regions might be differentially vulnerable to impaired TCR complex function. Therefore, complete, and detailed elucidation of the composition of this unique transcription-linked DNA repair complex, particularly in different brain regions and understanding how its activity is regulated by these factors may clarify the selective vulnerabilities of neurons in the HD brain.

While it is often argued that ineffective DNA repair and/or cumulative accumulation of unresolved DNA lesion/damages in neuronal genome are the consequences, and not the causes of neuronal dysfunction, toxicity and degeneration, we diverge and hypothesize that in fact, impaired or defective DNA repair might be an early pathology in not only in HD but also in several other terminal neurodegenerative diseases like ALS (Konopka and Atkin, 2018; Mitra et al., 2019; Konopka et al., 2020), AD (Shackelford, 2006; Shanbhag et al., 2019), PD (Milanese et al., 2018; Schaser et al., 2019; Martín-Jiménez et al., 2020; Gonzalez-Hunt and Sanders, 2021), SCA7 (Niss et al., 2021), and SCA3 (Chatterjee et al., 2015; Gao et al., 2015; Chakraborty et al., 2020). It is possible that once DNA damage occurs due to faulty or impaired repair system, the presence of highly damaging and potentially lethal lesions may become irreversible in adult post-mitotic neurons, and no interventions will rescue the damages or phenotype after that point. However, our data argue against this point. Our goal must therefore be to prevent neurons from reaching that critical tipping point and identify therapeutic targets within this window of opportunity to stop or reverse the degeneration for diseases like HD.

Studies showing that brain-specific knockout of HTT in postnatal mouse brain triggers neurodegeneration (O’Kusky et al., 1999; Dragatsis et al., 2000), suggesting that HTT plays important roles in neuronal development and survival. On the other hand, immortalized cell lines or primary neurons overexpressing HTT are resistant to degeneration following exposure to degenerative stimuli including expression of mHTT, and transgenic mice overexpressing HTT show tremendous resistance to degeneration when exposed to excitotoxic and ischemic injury (Rigamonti et al., 2000; Ho et al., 2001; Zhang et al., 2003, 2006; Leavitt et al., 2006). Although these reports suggest that increased HTT levels dramatically improve neuronal resistance to degeneration, the precise mechanism by which this protein provides neuroprotection remains poorly understood. Emerging evidence, including our studies, suggest that HTT plays a critical role in DNA strand break repair possibly by assembling the macromolecular TCR complex. Based on our recent data we hypothesize that this complex “senses” DNA damage/lesions in the template DNA and orchestrates their repair during transcription to maintain sequence integrity of neuronal genome. On the other hand, mHTT-mediated loss of DNA repair and deubiquitinating activities might be critical proximal events that impair DNA repair as well as transcription in HD. This mechanism would provide a mechanistic link between transcriptional dysregulation, cumulative DNA damage accumulations and inappropriate and chronic activation of the ATM/DNA-PK-dependent pro-apoptotic pathways, which trigger early neurotoxicity in HD. While the final biological output triggered by impaired TCR and persistence of unrepaired DNA damages in HD remains to be understood, our study points to a possible TCR mechanism that may be disrupted by mHTT in HD (Gao et al., 2019). The molecular strategies that interfere with the interaction of mHTT with the TCR complex might help slow neurotoxicity and the functional decline in HD. Alternatively, molecular approaches to stimulate PNKP activity for efficacious DNA repair may be another way to combat transcriptional dysregulation in HD. Collectively, the various studies on DNA repair deficiency in HD (Qi et al., 2007; Enokido et al., 2010; Tamura et al., 2011) and our recent findings (Gao et al., 2019) may help clarify how mHTT compromises genome integrity and neuronal function in HD.

SP, RG, NZ, and KB contributed to the experiments described in this review article. PS and YW wrote and edited the manuscript. All authors read and contributed to the editing of the manuscript and approved the submission of all versions of the review article.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.