Unraveling the links between neurodegeneration and Epstein-Barr virus-mediated cell cycle dysregulation

Abstract

The Epstein-Barr virus is a well-known cell cycle modulator. To establish successful infection in the host, EBV alters the cell cycle at multiple steps via antigens such as EBNAs, LMPs, and certain other EBV-encoded transcripts. Interestingly, several recent studies have indicated the possibility of EBV's neurotrophic potential. However, the effects and outcomes of EBV infection in the CNS are under-explored. Additionally, more and more epidemiological evidence implicates the cell-cycle dysregulation in neurodegeneration. Numerous hypotheses which describe the triggers that force post-mitotic neurons to re-enter the cell cycle are prevalent. Apart from the known genetic and epigenetic factors responsible, several reports have shown the association of microbial infections with neurodegenerative pathology. Although, studies implicating the herpesvirus family members in neurodegeneration exist, the involvement of Epstein-Barr virus (EBV), in particular, is under-evaluated. Interestingly, a few clinical studies have reported patients of AD or PD to be seropositive for EBV. Based on the findings mentioned above, in this review, we propose that EBV infection in neurons could drive it towards neurodegeneration through dysregulation of cell-cycle events and induction of apoptosis.

Graphical abstract

Highlights

• •Cell-cycle dysregulation lies at the interjection of oncogenesis and neurodegeneration.
• •Clinical and in-vitro evidences associates Epstein-Barr virus (EBV) with AD and PD.
• •As an oncogenic virus, EBV is well known to modulate the host cell-cycle.
• •Viral proteins EBNAs, LMPs and others like BARF-1 are involved in the process.
• •Thus, EBV mediated cell-cycle dysregulation could initiate neurodegenerative pathology in AD and PD.

1. Introduction

Human Herpesvirus-4 (HHV-4) is also called Epstein-Barr virus (EBV) after its discoverers Anthony Epstein, Yvonne Barr, and Burt Achong (Epstein et al., 1964). It is a double-stranded DNA virus, and upon infection, it can either integrate into the host cell genome or exist in an episomal form (Reisinger et al., 2006). As a group-I carcinogen, it is associated with various lymphatic or epithelial malignancies such as Hodgkin's and Non-Hodgkin's (Burkitt's) lymphoma or nasopharyngeal carcinoma. It is linked to non-neoplastic diseases like infectious mononucleosis (IM) and lymphoproliferative disorders (Niedobitek et al., 2001). EBV is also reported to aggravate gastric cancer (Sonkar et al., 2020). A study reported that EBV infection could get laterally transferred from its natural host cells of B-cell lineage to the cells of epithelial origin (Shannon-Lowe et al., 2006). Interestingly, recent reports from various groups have suggested EBV's involvement in neurological manifestations such as multiple sclerosis and other neurodegenerative disorders, including Alzheimer's and Parkinson's disease (Biström et al., 2021; Carbone et al., 2014; Bu et al., 2015). Not much to surprise, some recent reports have also highlighted the importance of molecular crosstalk between oncogenesis and neurodegeneration (Houck et al., 2019). Notably, most molecules shared among these phenomena are related to cell-cycle regulation (Seo and Park, 2020). Multiple studies have provided convincing evidence that cell-cycle dysregulation plays a critical role in the progression of neurodegenerative disorders (Yang and Gao, 2020; D. J. Bonda et al., 2010; Wang et al., 2009). Furthermore, EBV is well established to be capable of altering the host cell cycle. Various EBV antigens such as EBV nuclear antigen-1 (E1), E2, E3A, E3B, and E3C, EBV encoded small RNAs (EBERs), Bam-HI A rightward transcripts (BARTs), EBNA leader protein (E-LP) which is also known as E5, latent membrane proteins (LMP) −1, 2A, and 2B are involved in manipulating the host cell cycle (Yin et al., 2019) For example, the interaction of EBV latent genes with cellular oncogenes could promote G₁/S transition and halt apoptosis, leading to neoplastic transformation of the infected cells (Yin et al., 2019).

However, the neurovirulent and neuroinvasive capability of the virus is still debated; the genetic material of EBV and antiviral antibodies against the virus has been found in CSF samples of patients suffering from NDs (Carbone et al., 2014; Gate et al., 2020). A recent review beautifully shed light on the involvement of EBV in various NDs including AD, PD, MS, etc (Zhang et al., 2022). An earlier study done by Jha et al. has also established the neurotropic potential of EBV in primary neurons, Ntera2, and Sh-Sy5y cell lines (Jha et al., 2015). Additionally, our recent study has shown that EBV can infect and modulate glial cells (Tiwari et al., 2020; Jakhmola and Jha, 2021) in the brain, corroborating a previous finding by Menet et al., (1999). Also, we have observed that EBV can infect and alter the endothelial cells of the blood-brain barrier (Indari et al., 2021). These clues indicate the possibility of EBV being capable of establishing successful infection in neural cells: i.e., glial cells, and neurons. As stated earlier, EBV infection could cause deregulation of the cell cycle in the host cell. Therefore, in this review, we have tried to summarize the role of EBV infection in manipulating the host cell cycle and how dysregulation of these events in neurons could be involved in neurodegenerative disorders.

2. Understanding the role of EBV in the cell cycle and its regulation

The eukaryotic cell cycle is a highly regulated and coordinated process by which a cell gives rise to two daughter cells. A cell synthesizes various cellular components during this process, duplicates its genetic material, and eventually divides. A typical cell cycle consists of the interphase and the mitotic (M) phases. The interphase prepares for the upcoming division process. It can be further classified under three stages: gap-1 (G₁), synthesis (S), and gap-2 (G₂) phase [Fig. 1]. The gap phases function as a time interval allowing the cell to review its surroundings and prepare itself for upcoming cell cycle events. Although, under unfavorable conditions for division, a cell can enter a non-replicative state known as G₀, which occurs between the M-phase and the start of the next interphase. A cell in the G₀ phase neither divides nor prepares to divide. This G₀ state can be reversible (quiescent) or non-reversible (terminally differentiated). Certain cells of the body that are terminally differentiated, like neurons (Anda et al., 2016), and cardiac cells (Broughton and Sussman, 2019), remain in this inactive G₀ stage under physiological conditions.

Fig. 1.

Phases of the cell cycle. The cell cycle is mainly divided into the Interphase and Mitotic phase. Various heterodimers of cyclins and CDK complexes positively regulate the cell-cycle progression. On the other hand, the CKIs (CDK inhibitors) obstruct the cell-cycle progression if needed.

The regulation of this well-tuned process is orchestrated mainly with the help of certain intracellular molecules, namely: cyclins (Cyc), cyclin-dependent kinases (CDKs), and CDK inhibitors (CKIs). The Cyclin/CDK heterodimer drives a cell through various checkpoints by acting at specific points in the cycle, phosphorylating downstream proteins and modulating their activity (Satyanarayana and Kaldis, 2009) [Fig. 1].

Throughout the cell cycle, the activity of Cyclin/CDK complexes is controlled by timely production of the two proteins, phosphorylation/dephosphorylation of CDKs, controlled degradation of Cyclins, and binding of CKI proteins to specific complexes (Tarn and Lai, 2011; Suryadinata et al., 2010). The stage-specific expressions of cyclin/CDK complexes and respective CKIs are shown in [Table 1]

Table 1.

The majorly involved Cyclins and Cyclin-dependent kinases (CDKs) at different cell cycle phases and respective checkpoints.

Cell cycle stage	Active Cyclins	Active CDKs	Active CKI	Checkpoint
G1- phase (early)	Cyclin D1/2/3	CDK-4	Ink family (p15, p16, p18, and p19)
G1- phase (late)	Cyclin D1/2/3	CDK-6		G1 checkpoint
G1/S- phase transition	Cyclin E	CDK-2	pRb
S- phase	Cyclin A	CDK-2	Cip/Kip family (p21, p27, and p57)
G2/M- phase transition	Cyclin A	CDK-1		G2 checkpoint
M- phase	Cyclin B	CDK-1		Mitotic checkpoint

As an oncogenic virus, EBV has been reported to alter the host cell cycle at multiple steps. It can modulate protein-protein interactions, redistribute proteins, or encode homologs of cellular proteins (Fan et al., 2018). Previous studies have reported direct interaction between several EBV proteins and the host cell cycle proteins. The specific proteins of EBV acting at various stages of the cell cycle are discussed in the following section:

2.1. The G₁ phase

A cell begins the division process with the initiation of the G₁-phase, which calls for an external stimulus in the form of growth factors. Continuous growth factor supply is required until the cell passes through the restriction point in G₁ (Pardee, 1974); after this, the process becomes growth factor-independent, and the cell is now committed to complete DNA replication and division. Unsurprisingly, this step becomes the first target for virus-mediated cell-cycle manipulation. Multiple studies involving EBV have shown that the virus infection could promote the expression of neurotropic growth factors such as brain-derived neurotropic factor (BDNF) and glial cell-derived neurotropic factor (GDNF) (Ng et al., 2012; Jakhmola and Jha, 2021) in the infected cells. Clinical reports evaluating EBV meningoencephalitis/meningitis have corroborated the claim, as an increase in neurotropic factors was observed in the patient's CSF samples (Chiaretti et al., 2014; Kozko et al., 2018).

Once a cell enters the G₁-phase, it grows to its usual size and synthesizes mRNA and protein that will be used in the forthcoming stages of the cell cycle. As depicted in Table 1, progression through the G₁-phase is under the regulation of CDK-4 and D type Cyclins (Cyclin D1/2/3). Various viruses, including EBV, modulate the expression and activity of cyclin/CDK complexes to gain control of the cellular machinery (Tavakolian et al., 2020). These intracellular pathogens have evolved impressive strategies to modulate the kinase activity of the Cyclin/CDK complex by interacting with the individual subunits of the heterodimer or altering the activity of CKIs. For instance, the EBV nuclear antigen-3C (E3C) interacts with the D-type cyclins D1 and D2 (Saha et al., 2011; Pei et al., 2018). E3C can form stable complexes with G₁ phase cyclins- D1 and D2/CDK-6 heterodimer [Fig. 2a] (Saha et al., 2011; Pei et al., 2018). By upregulating the expression of D-type cyclin, EBV can deregulate the activities of the G₁ phase.

Fig. 2.

Epstein-Barr virus utilizes different mechanisms to manipulate the cell cycle. Modulating protein-protein interactions of the host cell cycle proteins is one of the most favored strategies.

a) EBV can interact with positive cell cycle regulators to promote cell cycle progression. EBV antigen, E3C, interacts with various cyclins such as Cyc D1, E, and A to accelerate the cell cycle progression at respective steps. Further, another antigen, E3B, is also reported to interact with Cyc A and result in the expedition of S-phase transition. Another EBV antigen, LMP1, reportedly stimulates the expression of EGFR, that in turn acts as a mitogenic signal to promote G₁/S transition.

b) Alternatively, EBV can interact with negative cell cycle regulators to either hinder their activity and release cells from the arrest stage or halt cell cycle progression. (i). EBV antigen E1 interacts with HAUSP to obstruct the degradation of Mdm2, thereby allowing blockage of p53 from acting as a stimulator of p21. Another EBV antigen, E3A, also sequesters p21. Together, these molecules inhibit the cell cycle via p53 and p21; and aid in progression through G₁ to the S phase. Further, progression through the G₁ phase is reported to be expedited by E3C via interaction with various Cyclin/CDK complexes. E3C also regulates the pRb cascade by governing its degradation through the SCF complex or by interacting with and thereby stabilizing AURKB to promote pRb phosphorylation. The E3C is also reported to promote the degradation of p27 via the SCF complex, thereby releasing the Cyclin A/CDK-2 complex from inhibition, leading to the advancement of the cell cycle through the S phase. (ii). EBV antigen, Rta, is reported to upregulate the expression of p21 and thus facilitate inhibition of the Cyclin E/CDK-2 complex. Rta may also mediate sequestration of CDK1 and CDK2 by upregulating 14-3-3 σ, thereby obstructing G₁/S transition. Further, another EBV antigen, E5, is also known to modulate the cell cycle at the G₂ phase by inhibiting Mdm2-mediated poly-ubiquitination of p53, thus facilitating p53-mediated p21 activation resulting in blocking the activity of the respective cyclin/CDK complexes.

Further, the latent membrane protein 1 (LMP-1) of EBV can also influence the activity of the Cyclin D/CDK-4 complex by interacting with and altering the activity of CKI, p16^INK14a. This interaction facilitates the progression of the cell cycle through the G₁ phase by releasing the Cyclin D/CDK-4 complexes from p16^INK14a-mediated inhibition of its kinase activity [Fig. 2b(i)] (Yang et al., 2000; Ohtani et al., 2003; Dawson et al., 2012). LMP-1 is reported to reduce the expression levels of p16^INK14a by blocking the transcriptional activity of various factors regulating its expressions, such as Ets2, bmi-1, JunB, 14–3-3σ, and SNF5.

2.2. The G₁/S phase transition

Moving ahead in the cell cycle, the G₁/S phase transition is governed by the activity of the Cyclin E/ CDK-2 complex. Few studies have reported that EBV alters the kinase activity of the Cyclin E/ CDK-2 complex through the interaction of E3C with cyclin E [Fig. 2a]. However, the downstream effect of the association between E3C and cyclin E is not yet described (Knight et al., 2004). Several studies proclaimed that EBV employs LMP-1 to modulate the kinase activity of various cyclin/CDK complexes at the G₁/S phase transition. LMP-1 is reported to enhance cyclins' promoter activity in the late G₁ and G₁/S transition phase via the EGFR and STAT3 signaling pathways. Under the influence of LMP-1, the EGFR directly binds to cyclins- D1 and E, thereby accelerating the G₁/S phase transition (Tao et al., 2005; Xu et al., 2013) [Fig. 2a].

A checkpoint guards the G1/S phase transition often referred to as “a point of no return.” After passing this restriction point, the process becomes growth factor independent, and the cell is now committed to complete DNA replication and division. A member of the pocket protein family, pRb is one of the critical regulators of the G₁/S transition checkpoint (Y. J. Wang, et al., 2001). pRb acts as a guardian and lets the cell pass through the restriction point only under suitable growth conditions and if there is no DNA damage or metabolic disturbances. The regulation of pRb activity is mainly controlled by its timely phosphorylation (inactivation) and dephosphorylation (activation) at T373 and S608 by the G₁ phase Cyclin/ CDK complex (Cyclin D and CDK-4/6 complex) (Dowdy et al., 1993; Beijersbergen et al., 1995). This leads to its dissociation from a complex with transcription factor E2F (Bartek et al., 1996), and the subsequent release of E2F activates downstream genes required for the progression of the cell cycle.

Previous reports have listed interactions of pRb with various nuclear antigens of EBV (Szekely et al., 1993; Saha and Robertson, 2013). One such study suggested the role of E3C in influencing cell cycle regulation via controlling the pRb activity (J. S. Knight et al., 2005). The study conducted by Knight et al. indicates that E3C mediates the phosphorylation of pRb by controlling the kinase activity of Cyclin D1/ CDK-4 complex [Fig. 2b(i)]. It was also revealed that besides regulating the phosphorylation status of pRb, E3C mediates its ubiquitination. The study further elaborated that E3C recruits the SCF^Skp2-E3 ubiquitin ligase complex to affect the stability of pRb in EBV-transformed cells. The conserved domain region of E3C from amino acids 140–149, which was responsible for regulating the SCF^Skp2 complex, is also crucial for regulating pRb. It explicitly shows that E3C usurps SCF^Skp2 in EBV-transformed cells to target and regulate the levels of pRb [Fig. 2b(i)] (Maruo et al., 2006). Additionally, E3C is also reported to influence the regulation of pRb activity via aurora kinase B (AURKB) (Jha et al., 2013). The study done by Jha et al. demonstrates that the direct interaction of E3C with AURKB stabilizes the protein. This interaction reduces ubiquitination of AURKB, thus maintaining its phosphorylating activity towards pRb leading to cell proliferation [Fig. 2b(i)]. The following hyper-phosphorylated state of pRb would ultimately accelerate the G₁/S transition and cause the cell cycle to proceed unchecked.

Additionally, a study in virus-infected Akata cells has reported an interaction between EBV-immediate early lytic gene product BRLF-1 and the pRb (Zacny et al., 1998). BRLF-1 is speculated to interact with two regions of pRb, from amino acids 39–89 and 249–309. The interaction possibly occurs first at one site, followed by secondary interaction at the other. Although the BRLF1 does not directly interact with E2F bound to pRb, its binding region on pRb was observed to lie outside the pocket region. A correlation between the BRLF-1 binding and E2F displacement from the pRb was still observed. This indicates a potential role of BRLF-1 in regulating the cell cycle beyond relief of pRb-mediated E2F repression and thus cell cycle proliferation.

Furthermore, EBV is also known to target pRb via the binding of EBV nuclear antigen 5 (E5/EBNA-LP) during B-cell transformation (Szekely et al., 1993). E5 localizes with and binds to pRb without possessing an LXCXE motif. However, the pRb binding region of E5 was found to be on the N-terminal of the protein as a 66-amino acid long peptide, which is also the site for p53 binding. Therefore, the binding of E5 to pRb is competitively affected by p53. An in-vitro study revealed that the binding of p53 in a dose-dependent fashion could inhibit the formation of the E5-pRb complex, indicating that p53 competes with pRb for E5 binding. At the same time, the vice-versa cannot be feasible. Additionally, a study reported that E5 affects the pRb/p53 cascade by binding to its negative regulator, Mdm2 [Fig. 2b(ii)]. This binding of E5 with Mdm2 prevents Mdm2-mediated polyubiquitination of p53 and further degradation [Fig. 2b(ii)]. The formation of the E5- Mdm2-p53 trimolecular complex causes the cell to bypass G₁ arrest and proceed toward the S-phase of the cell cycle (Kashuba et al., 2011). The nuclear antigen E3C of EBV also directly interacts with p53 and attenuates the p53-mediated transcription of downstream genes and apoptosis. The interacting domain of p53 on E3C is mapped near the N-terminal at 130–190 amino acid residues (Yi et al., 2009).

An EBV protein Rta is known to transcriptionally upregulate the expression of 14-3-3σ, which negatively regulates the cell cycle progression [Fig. 2b(ii)] (Dar et al., 2014; Gupta et al., 2020). Though 14-3-3σ is under the control of p53, EBV-Rta could induce its expression in a p53-independent manner. Induction of 14-3-3σ would cause the sequestration of CDK-1 and 2 in the cytoplasm. Coupled with the diminished activity of cyclin E/CDK-2, as mentioned in the earlier section, this is reported to finally lead the cell cycle to arrest at the G₁/S transition state in EBV infected cells [Fig. 2b(ii)] (Huang et al., 2012). Furthermore, EBV-Rta induces p21^WAF1/CIP1 (CDKN1A) expression. By up-regulating the p21^WAF1/CIP1 expression levels, EBV diminishes the activity of the Cyclin E/CDK-2 complex (Huang et al., 2012). The reduced kinase activity of the cyclin E/CDK-2 complex, along with certain other conditions, results in cell cycle arrest at the G₁/S transition phase and aids in viral reactivation. The binding of LMP-1 to p16^INK14a relieves the growth arrest at the G1 phase, thus initiating G₁/S transition by targeting the downstream mediators of the pRb-p16^INK14a pathway.

Another effective strategy of viruses to hack the cellular machinery is protein redistribution. In order to control of the cell cycle, EBV mainly disturbs the localization and distribution of cyclins, p53 and survivin protein. During the G₁/S phase transition of the cell cycle, Cyclin D1 is primarily localized in the nuclear region [Fig. 3a] (Baldin et al., 1993). However, it is transported to the cytoplasm for ubiquitin-proteasome-mediated degradation upon GSK3β-mediated phosphorylation at T286 (Diehl et al., 1998). Remarkably, in a study by Saha et al., E3C was shown to play a dual role in blocking the poly-ubiquitination and GSK-3β mediated phosphorylation of Cyclin-D1, leading to its increased nuclear localization [Fig. 3a] (Pei et al., 2018).

Fig. 3.

EBV modulates the cellular protein distribution. EBV antigens could also redistribute the cellular proteins to facilitate the survival of the virus inside the host cell. (a) Cyclin D1- E3C mediates the nuclear localization of cyclin D1 by blocking GSK-3β. (b) Survivin and p53- E3C aids in nuclear localization of survivin and p53. (c) MRPS18-2- E3C also enhances the nuclear localization of MRPS18-2 to facilitate the E2F-dependent gene transcription.

Another cellular protein whose localization is affected by EBV infection is p53. A study done by Gou et al. demonstrated that LMP-1 induces the nuclear localization of p53 and survivin [Fig. 3b] (Guo et al., 2012). It is reported to upregulate the expression and phosphorylation of these two proteins. Survivin also possesses a p53-binding element in its promoter region. Besides, it facilitates the cell cycle to progress from G₁ to S-phase by interacting with CDK-4 in the nucleus. Although p53 is predominantly reported as a tumor suppressor protein increasing evidence suggests its overexpression and accumulation be linked with nasopharyngeal carcinoma (NPC) (Saha et al., 2009; Banerjee et al., 2013; Pei et al., 2016). Under physiological and DNA damage conditions, the expression of LMP-1 is found to be associated with p53 expression in EBV transformed cells. p53 is essentially required to stimulate the expression of LMP-1 in response to DNA damage. Ectopic p53 stimulates endogenous LMP-1 expression, subsequently blocking DNA damage-mediated apoptosis. Interferon regulatory factor 5 (IRF-5), a direct target of p53, is implicated in this process. In response to DNA damage-induced p53, IRF-5 increases the expression of LMP-1 by binding to and activating an LMP-1 promoter-reporter construct (Wang et al., 2017). Notably, IRF-5 is a well-known tumor suppressor gene that heightens apoptotic signaling in response to DNA damage. However, LMP-1 is predominantly considered a viral oncogene that blocks apoptosis. Therefore, it was evident that LMP-1 blocks apoptosis via modulating IRF-5 in EBV-infected cells to maintain latency (Hu and Barnes, 2009).

Apart from the proteins mentioned above, another cellular protein implicated in the cell cycle, MRPS18-2 (mitochondrial ribosomal protein S18-2), is also dislocated to the nucleus by E3C [Fig. 3c] (Kashuba et al., 2011). MRPS18-2 binds explicitly to both the hypo- and hyperphosphorylated forms of Rb protein. This binding occurs through a site in the small pocket of pRb, which is also involved in its interaction with E2F1. MRPS18-2 and E2F1 competitively interact with pRb, further elevating the free E2F1 level. The levels of nuclear MRPS18-2 protein are also affected by E3C. The EBV protein E3C binds to and raises the level of MRPS18-2, which may help in the progression of EBV-infected B cells into the S phase. These strategies would competitively inhibit pRb binding to E2F1 and lift the block to S-phase entry.

2.3. The S phase

The passage of a cell into the S phase initiates the synthesis of DNA, which generates two replicas of the genetic material. The progression of this cell cycle stage is under the governance of the Cyclin A/CDK-2 complex. The EBV nuclear antigen E3C is well known to interact with Cyclin/CDK complexes active at various cell cycle stages, including the one in the S-phase. As described in the previous section, E3C forms a stable complex with Cyclin A and promotes the Cyclin/CDK complex's kinase activity. Additionally, a study conducted by Knight and Robertson in 2004 demonstrated weak interactions between E3B and Cyclin A, owing to the conserved region in E3B that is similar to the Cyclin A binding site of E3C. However, the precise role of this binding is still unclear [Fig. 2a] (Knight et al., 2004).

The binding of E3C to Cyclin A decreases the association of p21^WAF1/CIP1 with the Cyclin A/ CDK-2 complex, thus rescuing it from p21^WAF1/CIP1-mediated inhibition, thereby aiding in cell cycle progression through the S phase (Knight and Robertson, 2004). Furthermore, the inhibitory activity of p21^WAF1/CIP1 against the S phase Cyclin/CDK complex has been modulated by E3A of EBV (Tursiella et al., 2014). It contributes to apoptotic resistance in EBV-infected Burkitt's lymphoma cell lines through fine-tuning the expression of p21^WAF1/CIP1. The E3A-mediated repression of p21^WAF1/CIP1 prevents the cell from succumbing to p53 and pRb-mediated cell cycle arrest and ensures continuous proliferation of the cell through the S phase [Fig. 2b(i)] (Tursiella et al., 2014).

Furthermore, an in-vitro study revealed that the N-terminal amino acids (130–159) of E3C are responsible for binding with Cyclin A and restricting the p27^KIP1-mediated inhibition of Cyclin A/CDK-2 kinase activity (Tursiella et al., 2014). At the same time, C-terminal domain amino acids (957–990) might be playing a role in stabilizing the elements of this complex [Fig. 2a] (Knight and Robertson, 2004; Saha et al., 2011). E3C might serve as a bridge between SCF^Skp2 and p27^KIP1, resulting in the degradation of p27^KIP1, thereby relieving its inhibitory effect on the Cyclin A/CDK-2 complex (Jason S. Knight et al., 2005; Knight Jason et al., 2005). Thus, the enhanced activity of the Cyclin A/CDK-2 complex would cause the release of the cell from arrest at the S phase [Fig. 2b(i)] (Tursiella et al., 2014).

2.4. The G₂ phase and G₂/M phase transition

After the completion of DNA duplication in the S phase comes the G₂-phase, in which the cell growth continues along with protein synthesis as the cell prepares for M-phase. The regulation of progress through the G₂ phase is controlled by Cyclin B and CDK-1 (cdc2). A study done by Mauser et al. has demonstrated that EBV can induce G2/M block in HeLa cell lines and normal human fibroblasts by reducing the levels of Cyclin B1/CDK-1 [Fig. 2a] (Mauser et al., 2002). The EBV lytic protein Zta, also known as BZLF-1 (BamHI Z fragment leftward open reading frame 1), was demonstrated to decrease the transcript and protein expression levels of Cyclin B1, thus inducing cell cycle arrest at the G₂/M phase transition.

Before moving on to the M-phase, the cell must go through checkpoints at the G₂/M phase transition to ensure error-free and successful completion of DNA replication (Hartwell and Weinert, 1989). If DNA damage is identified, the cell undergoes growth arrest instead of repair without proceeding to the M-phase. Interestingly, E3C of EBV is demonstrated to disrupt this restriction by binding and inactivating the Chk2 (checkpoint kinase 2), the effector molecule of the ATM/ATR signaling pathway that regulates the G₂/M checkpoint. The binding of E3C to Chk2 induces phosphorylation of Cdc25c (cellular phosphatase) at Ser216, which leads to its sequestration in the cytoplasm by 14-3-3σ thus permitting activation of Cyclin B/CDK-1 complex and bypassing of the G₂/M checkpoint [Fig. 2b(ii)] (Choudhuri et al., 2007).

Such contradictory reports suggest that blocking the cell cycle progression at certain stages would be advantageous under some circumstances, while promoting the progression would be more beneficial in some others. As described in the previous sections, a careful balance between the two processes is essential for establishing EBV infection, which is fine-tuned by various EBV proteins.

3. Evidence and implications of EBV-mediated cell cycle dysregulation in neurodegenerative disorders

The association between neurodegenerative pathology and different members of the Herpesviridae family has been consistently reported over the past two decades (Costa Sa et al. 2019a; Phuna and Madhavan, 2022 Mar 14). Although the role of EBV has been understated and frequently debated in this context (Carbone et al., 2014), the evidence does confirm the connection. The post-mortem human brain tissue and cerebrospinal fluid (CSF) samples of patients suffering from neurodegenerative disorders like multiple sclerosis (MS), Alzheimer's disease (AD), Parkinson's disease (PD), etc. have shown seropositivity for EBV and the presence of antibodies against EBV (Kleines et al., 2011). Interestingly, demographic studies done on elderly population that progressed to cognitive impairment from a healthy state showed elevated levels of anti-EBV antibodies (Shim et al., 2016). As described earlier, EBV is highly efficient in manipulating the cell cycle; and various pathophysiological conditions of the central nervous system (CNS), including chronic neurodegenerative disorders and acute damage, have been linked with abnormal cell cycle activation and progression (Wang et al., 2009; Woulfe et al., 2016). Therefore, in this review, we have focused on NDDs implicating virus-mediated cell cycle dysregulation as a possible mode of pathogenesis.

Unlike other cell types, neurons, once terminally differentiated, are supposed to have lost their proliferation capability. Most of the CNS neurons enter this post-mitotic quiescent state while going through embryonic development and remain in the “prolonged G₀ phase” in the adult nervous system (Frade and Ovejero-Benito, 2015a). These cells are unable to re-enter the cell cycle. However, surprisingly, various structures in a typical adult human brain express several genes that encode regulators of G₁/S transition (Kruman et al., 2004; Koeller et al., 2008; Lopes et al., 2009). These genes include cyclin D1, CDK-4, Rb proteins, E2Fs, and CKIs [Fig. 4]. In fact, most of these gene transcripts are translated to proteins in normal adult neurons. Traditionally, the presence of core cell cycle regulators in adult neurons is attributed to their functions in neuronal migration, maturation, and synaptic plasticity (Herrup and Yang, 2007). Nevertheless, these proteins could potentially force the cell to re-enter the cell cycle upon induction by various factors under certain conditions. Certain mitogenic stimuli including viral infections could induce the aberrant re-entry of neurons into the cell cycle, ultimately leading to cell death. Though, the association of cell cycle reentry with synaptic failure and neuronal death is well established by various studies, the functional connection between the two phenomena remains elusive (David J. Bonda et al., 2010; Barrio-Alonso et al., 2018a). When subjected to acute stress conditions such as lack of neurotrophic factors, DNA damage, oxidative stress, excitotoxicity, and sudden mitogenic stimuli, these cells would reactivate the cell cycle by aberrant expression of cell cycle proteins (Hernández-Ortega et al., 2011). Under such circumstances, neurons re-enter the cell cycle essentially to facilitate the recovery from damage caused by the insult. However, the regenerative capacity is limited to specialized neuronal progenitor cells located in the subventricular (SVZ) and subgranular zone of the dentate gyrus in the hippocampal region of the mammalian brain [Fig. 4a].

Fig. 4.

Neuronal cell-cycle. Mitogenic signals received by neural progenitor cells in the subventricular zone (SVZ) and dentate gyrus (DG) give new cells. However, mitogenic stimulation could result in ectopic re-entry of the cells into the cell cycle, ultimately leading to neurodegeneration in the rest of the brain.

Nonetheless, if post-mitotic neurons are forced to re-enter the cell cycle, they die even before any sign of DNA synthesis appears (Herrup and Busser, 1995; Herrup, 2004; Marlier et al., 2020 May 31) [Fig. 4b]. This process, traditionally referred to as “abortive cell cycle re-entry,” is characterized by the increased activity of cyclin D-CDK4/6 and deregulation of E2F transcription factors, ultimately followed by cell death (Marlier et al., 2020 May 31). In this regard, E2F1 may act as a trigger of neuronal apoptosis. It has been reported to activate two pro-apoptotic signaling pathways in cerebellar granule cells and cortical neurons, namely, p53-independent activation of Bax/caspase-3 and the induction of the CDK1/FOXO1/Bad pathway (Schmidt-Kastner et al., 2000; Zhang et al., 2020). In addition, p130/E2F4, a repressive complex that maintains neurons' post-mitotic state, is also deregulated. This dysregulation results in the induction of neuronal apoptosis through the upregulation of B-myb and C-myb (Liu and Greene, 2001). Overall, these observations indicate that different environmental conditions trigger various signaling pathways that can elicit cell cycle reactivation and cell death in specific neuronal phenotypes.

Aberrant induction of the cell cycle in glial cells causes them to activate and proliferate, leading to glial scar formation and the production of inflammatory factors. This microenvironment plays a crucial role in developing neurodegenerative pathology (Woulfe et al., 2016; Caggiu et al., 2019). However, the forced re-entry of terminally differentiated neurons into the cell cycle leads them to death instead of proliferation (Hernández-Ortega et al., 2011; Frade and Ovejero-Benito, 2015a). Although in specific pathologies, various types of machinery are involved in distinct ways of neuronal demise. The abnormal cell cycle re-entry (CCR) of neurons leading to death may be a common pathway among different neurodegenerative conditions [Fig. 5]. Acute injury to CNS, such as stroke or trauma, is often accompanied by neuronal apoptosis and is usually associated with blockade of G₁/S transition (Wu et al., 2011). Whereas, in chronic circumstances such as neurodegenerative diseases like multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), Alzheimer's disease (AD), and Parkinson's disease (PD), however, some of the neurons show S-phase markers (Bonda et al., 2009). In such neurons, successful completion of DNA synthesis is depicted by S-phase proteins such as phosphorylated Mcm2, and aneuploidy supports the notion (D. J. Bonda et al., 2010).

Fig. 5.

Schematic diagram of cell-cycle deregulation-mediated neurodegeneration. The interdependent cascade of cellular signaling activated by various factors such as environmental, genetic, or external mitogenic stimuli; results in cell-cycle re-entry of post-mitotic neurons. However, mitotic catastrophe and an overstimulated immune system lead the cell towards neurodegeneration instead of cell cycle completion.

3.1. EBV-induced CCR in Alzheimer's disease

AD is characterized by progressive and irreparable damage inflicted on the neurons, leading to cognitive impairment. According to WHO reports, it is the predominant cause of senile dementia globally, contributing 60–70% of the total cases (Carbone et al., 2014). Between 2020 and 2040, incidents of dementia associated with AD are speculated to increase by more than 300% in South-East Asian countries, including India (Rizzi et al., 2014; Ferri et al., 2005). Though a few key elements in the disease pathology are well-acknowledged, explicit molecular mechanisms leading to the disease pathogenesis are yet to be understood. While not the sole cause of neurodegeneration, amyloid-β (Aβ) plaques, and neurofibrillary tangles are hallmarks of AD and critical in the disease development [Fig. 5] (Braak et al., 1998; LaFerla and Oddo, 2005). Damage induced by oxidative stress is another crucial facet of neurodegeneration related to AD (Zhang et al., 2014; Kamat et al., 2016; Kang et al., 2017). Additionally, viral infections could be a risk factor for AD susceptibility (Carbone et al., 2014).

Over the past few years, an increasing number of reports have been implicating aberrant cell cycle re-entry of neurons in AD pathogenesis (Bonda David et al., 2010). Interestingly, aberrant cell cycle markers in AD neurons are present way before obvious disease indicators; therefore, mitotic dysfunction is indicated as a predisposition to disease onset and early development (Lombardi and Lasagni, 2016). Markers of ectopic cell cycle activation in AD-affected neurons include altered expression of cell cycle-mediating mitotic proteins like Ras (Wright and Harding, 2010) and G-proteins (Thathiah and De Strooper, 2011), expression of S-phase proteins, aneuploidy, etc., are found to be concordant with disease development. Recently published studies provide more evidence of the importance of cell cycle dysregulation in AD pathogenesis. Firstly, compared to healthy controls, the Ras protein and its downstream mediators such as Raf, MAPK, and MEK1/2 have been activated in AD neurons (Wright and Harding, 2010; Thathiah and De Strooper, 2011; Tan et al., 2012). The Ras pathway is exclusively implicated in the phase transition from G₀ to the G₁ stage of the cell cycle via its interaction with Cyclin D1. Interestingly, activation of the Ras-MAPK-dependent pathway by LMP-1 of EBV is essential for malignant transformation in B-cells (Roberts and Cooper, 1998). EBV could probably influence the Ras-MAPK pathway in neurons and initiate the development of neurodegenerative pathology via modulating the cell cycle.

Additionally, compared to the age-matched controls AD neurons display elevated levels of other cell-cycle markers, indicating an exit of neurons from the quiescence stage (G₀) and re-entry into the cell cycle (Yang et al., 2003; Bonda et al., 2009; Tan et al., 2012). In particular, the cytoplasm of diseased neurons express Cyclin D, CDK-4, and Ki67, and elevation of the Cyclin D/CDK-4 complex in the cells corroborates the fact that these cells have indeed crossed the G₁ phase and are no longer in G₀ [Fig. 4] (McShea et al., 1997; Nagy et al., 1997; Ueberham, 2003; Heine et al., 2004; Ueberham and Arendt, 2005; Aulia and Tang, 2006; Currais et al., 2009a; van Leeuwen and Hoozemans, 2015). As described in previous sections, EBV has long been known to influence various cyclin/CDK complexes. There are high chances that EBV might modulate the activity of the Cyclin D/CDK-4 complex during the initiation of the cell cycle at the G₁ phase and force them towards abortive cell-cycle reentry.

Moreover, even transition through S-phase is indicated by successful DNA replication in AD neurons, along with the expression of S-phase proteins such as phosphorylated pRb and Mcm2 (Thakur et al., 2008; Bonda et al., 2009). The presence of aneuploidy corroborates this possibility (Arendt, 2012; Barrio-Alonso et al., 2018b). The premature chromosome separation (PCS), a phenomenon associated with successful DNA replication and an aberrant mitotic phenotype, provides persuasive proof for CCR in AD neurons (Spremo-Potparević et al., 2008). Studies particularly illustrate PCS happening in the G₂ of the cell cycle immediately after DNA duplication (Zhu et al., 1999). It is noteworthy that EBV, as an oncogenic virus, has been known to induce chromosomal rearrangements and instability (Lacoste et al., 2010). The viral protein BNRF1 induces centrosome amplification (Shumilov et al., 2017), while the BGLF4 kinase of the virus induces premature chromosome condensation through activation of condensin and topoisomerase II (Lee et al., 2007). The above-mentioned studies indicate the possibility of EBV-mediated chromosomal modulation in neurons, as a probable step towards developing neurodegenerative pathology.

Further, oxidative stress in neurons, a significant contributor to AD pathogenesis, has been consistently associated with cell cycle aberration markers (Zhu et al., 2001; Klein and Ackerman, 2003). Although it is one of the key players in disease onset and development, the precise source of this oxidative stress and its role in AD pathophysiology have been ambiguous (Smith et al., 2000; Scheff et al., 2016; Uddin and Kabir, 2019). For instance, in immunohistochemically stained neurons, oxidative stress markers like 8-hydroxyguanosine appear decades before the general signs of AD (Abe et al., 2002). Interestingly, Aβ in its soluble, non-aggregated form is an effective antioxidant and a high-valence metal chelator. Initially, it may exist as a respondent mechanism to relieve oxidative stress (Lee et al., 2006; Cheignon et al., 2018). However, oxidization renders Aβ insoluble, forming aggregates and incurring the cellular burden. It is noteworthy that members of the Herpesviridae family, like EBV, can stimulate Aβ fibrillation, a protective measure against brain infection (Eimer et al., 2018).

Recent studies have revealed a direct and indirect causal link between oxidative stress and cell cycle aberrations (Bresgen et al., 2003; Limoli and Giedzinski, 2003; Taniai et al., 2014). Specifically, a ‘two-hit hypothesis’ has been suggested to implicate oxidative stress and cell cycle malfunctioning conjointly, resulting in AD neurodegeneration (Zhu et al., 2007; Bonda et al., 2009). The disease pathogenesis, provoked by either of the two elements, advances as a combined process. Oxidative stress, as one factor, destructs neurons to evoke other factors involved in cell cycle dysfunction eventually or vice-versa. For instance, in sporadic AD (or late-onset), which is more prevalent, a steady-state of oxidative imbalance has been observed that predisposes the affected neurons to develop cell cycle malfunctions as a ‘second hit.’ Although neurons can counter the effect of acute oxidative stress, chronic large-scale accumulation of reactive oxygen species will require the cell to adapt and attain a state of oxidative imbalance. While the cell might survive with these adaptations for decades, eventual secondary abnormalities in the cell cycle ultimately result in its death. These two 'hits' finally launch the cell on a deteriorative course of oxidative stress, inflammation, Aβ aggregation, and mitotic dysfunction, followed by cell death, giving rise to the conditions responsible for disease pathology (Zhu et al., 2007). Based on the above-mentioned studies, it would not be farfetched to claim the association of EBV or HHV infection-mediated oxidative stress as a possible contributor to neurodegeneration in AD.

Besides, some crucial genes implicated in AD pathogenesis may be involved in abnormal CCR of neurons and subsequent neurodegeneration. In particular, three genes, the Aβ precursor protein (AβPP) gene, Presenilin 1 and 2 (PS-1/2) gene homologs, are linked with early-onset AD and have been reported to play a vital role in the cell cycle and its regulation (Porquet et al., 2015). AβPP, an integral transmembrane protein, undergoes proteolytic cleavage to produce Aβ peptide. The PS-1/2 genes are responsible for the proteolytic cleavage of AβPP and are hence involved in the regulation of the cell cycle as well as AD pathogenesis (De Strooper et al. 1998, 1999). In-vitro, both the protein and the resultant peptide have been shown to possess mitogenic activity (Greenberg et al., 1994).

Apart from their earlier mentioned interaction with AβPP, these homologous genes (PS-1/2) have also been associated with centrosomal assembly in the dividing cell, thus linking mitogenic alterations with neurodegeneration (Judge et al., 2011). For instance, deficiency of PS-1/2 in HeLa cells caused their accelerated transition from G₁ to S phase, while their overexpression resulted in G₁ phase arrest of the cell cycle (Janicki, 2000). Several studies with transgenic mice with AβPP, PS-1, and PS-2 mutations have also demonstrated that neurons exhibit CCR several months before amyloid deposition and full AD pathogenesis (D. J. Bonda et al., 2010). Also, the report suggests that these events occur in an anatomical pattern resembling the progression sequence of neuronal damage susceptibility observed in AD. Interestingly, though these mice show significant CCR as early as 6 months of age, their cognition and cellular functioning are maintained at a near-normal level for extended periods (Zhu et al., 2007). These mutant mice showed a phenotype of ‘mitotic steady-state,’ which ultimately evoked further abuse in the form of a ‘second hit’ of oxidative stress. Since the importance of PS- 1/2 is well established in the onset of familial AD cases, the possibility of its role as a link between cell cycle aberration and neurodegeneration cannot be overlooked. Additionally, AβPP-binding protein 1 (AβPP-BP1), which binds to the carboxy-terminal domain of the AβPP, is a multifunctional adapter protein that is also implicated in the regulation of mitotic transition from S to M-phase (Chen et al., 2000). Therefore, a neuron overexpressing AβPP-BP1 would be pushed in S-phase, exhibiting DNA replication and expressing the corresponding cell cycle markers Cyclin B/cdc2, ultimately propelling the cell towards apoptosis. As these phenotypes are apparent in AD neurodegeneration, AβPP-BP1 could be partially responsible for CCR.

Further, multiple factors could be responsible for inducing vulnerable neurons into improper re-entry into the cell cycle. According to the reports, mitogenic signaling induced by viral infections plays a pivotal role. Interestingly, EBV seropositivity has been demonstrated consistently over the decades in patients suffering from AD. However, the role of EBV in the disease remained debatable and underexplored. Several recent studies have provided convincing proof implicating EBV in the etiology of AD. A study done by Gate et al. showed that adaptive immune changes mediated by EBV were involved in AD pathogenesis (Kang and Liu, 2020). They reported the presence of CD8⁺ T effector memory CD45RA+ (TEMRA) cells specific for EBV as a part of adaptive immunity in AD patients.

Regardless of the cause, the outcomes of cell cycle dysfunction evident in AD are the same. The aberrant re-entry of already differentiated adult neurons into the cell cycle results in cellular malfunction and premature cell death, ultimately leading to neurodegeneration (Lee et al., 2009). As the cell cycle involves complex interactions of a wide variety of Cyclins and CDKs, which are crucial for cell proliferation and survival, any modulation of their expression, function, or control is likely to affect the cell negatively. Evidence suggests that the inability of neurons to complete the M-phase after the initial mitotic induction ultimately leads the cell to its death (Lombardi and Lasagni, 2016). Although many studies have reported neurons to complete or at least enter into the S-phase (indicated by DNA replication, chromosome maintenance protein expression, and binucleation events), some cells to be in the G₂ phase, and an entrance to M-phase, no study depicted completion of M-phase in the neurons (Currais et al., 2009b; Hardwick and Philpott, 2014; Frade and Ovejero-Benito, 2015b; Sharma et al., 2017; Walton et al., 2019). These neurons appear to experience a “mitotic catastrophe,” a phenomenon indicating the inability of the cell to complete the cell cycle due to failure in its regulation. Unfortunately, as these neurons have already passed the mitotic point of no return (i.e., G₁-phase/S-phase), they ultimately succumb to death instead of proliferating.

Therefore, modulations in the cell cycle and its control system may play a crucial role in developing neurodegenerative pathology. Although the definite mechanisms resulting in such mitotic dysfunctions are not entirely understood, various factors seem to play, including oxidative damage, viral infections, etc. A complex and reciprocal relationship between these factors and CCR seems likely in AD. Altogether, these facts indicate that the neurodegenerative pathology in AD might result from cell-cycle modulation mediated by EBV.

3.2. EBV induced CCR in Parkinson's disease

PD is categorized as a progressive neurodegenerative disorder occurring either sporadically or due to hereditary mutations in genes such as parkin, a-synuclein, and ubiquitin C-terminal hydrolase L1 gene (Braak et al., 1998). The pathophysiology of PD is marked by the presence of proteinaceous aggregates known as Lewy bodies (LBs) and pigmentation in dopaminergic neurons of substantia nigra pars compacta (SNpc) [Fig. 4] (Antony et al., 2013). However, the mechanism of pathogenesis leading to neuronal death in PD is still unclear. Previous studies have advocated the role of environmental factors, such as viral infections, that conspire with a permissive genetic background to initiate the neurodegenerative pathology of PD (Jang et al., 2009; Wang et al., 2020). However, the precise identity of these viral infections remains elusive, and the definite mechanism underlying this association remains unclear to date. For instance, in many cases, a direct viral infection of nigral neurons has been demonstrated or implicated in disease pathology, whereas in others, virally induced autoimmune mechanisms are held responsible (Olsen et al., 2019; Pajares et al., 2020).

Specifically, the molecular mimicry exhibited by a repeat region in latent membrane protein 1 (LMP-1) encoded by EBV and the C-terminal region of alpha-synuclein corroborates the statement (Caggiu et al., 2019). In genetically predisposed individuals, oligomerization of alpha-synuclein forming aggregates is believed to occur due to its cross-reactivity with EBV LMP-1 antibodies. These antibodies target a critical repeat region of alpha-synuclein. However, it is believed that EBV infection is essential but not sufficient alone for the development of PD within the average human lifespan. Considerable studies indicate that environmental factors collaborate with the susceptible genetic background to initiate the PD pathology. Together, both factors contribute to disease onset and progression and critical modulation of its temporal profile. In the face of EBV-induced α-syn autoimmunity, the host genetics may govern the immune response (e.g., HLA-DR, LRRK2) against EBV proteins and α-syn aggregation kinetics. The host genes SNCA, PARKIN, DJ-1, and PINK-1, along with factors like oxidative stress respectively, dictate the aggregation propensity, oligomer or aggregate clearance ability of the cell, and its capacity to deal with the consequences of aggregated α-syn, etc. (Caggiu et al., 2019).

Interestingly, cell-cycle dysregulation has also been linked to the development of PD pathophysiology. Several studies with post-mortem tissues of PD patients have reported the presence of p35 and CDK-5 in LBs in the locus coeruleus, neocortex, and substantia nigra (He et al., 2020; Allnutt et al., 2020). It suggests that CDK-5 is probably involved in LB formation. There is unquestionable evidence suggesting that proteinaceous aggregates of α-syn are key constituents of LBs that play a central role in PD pathogenesis (Takeda et al., 1998; Olanow and Brundin, 2013). Many studies demonstrated the co-localization of EBV with these protein aggregates (Woulfe et al., 2016). EBV signatures in α-syn aggregates have led scientists to hypothesize that PD is an EBV-induced autoimmune phenomenon. CDK-5 is known to play a crucial role in the cytoarchitecture of the CNS and is thereby indirectly linked to cell-cycle progression. Also, many studies have corroborated the implication of phase-related CDKs in PD neurodegeneration. Markedly, EBV has been repeatedly associated with modulation of Cyclin-CDK complexes, as described in previous sections. For instance, (the association of Cyclin E with EBV) Cyclin E, a CDK-2 activator, acts as a substrate of the parkin ubiquitin ligase complex, whose association is reported with familial forms of PD [Fig. 4] (Staropoli et al., 2003). Overexpression of parkin has been shown to rescue primary midbrain dopaminergic neurons from kainic acid (a neuro-excitotoxin)-induced death by reducing Cyclin E buildup in the cell. In the dopaminergic neurons of the substantia nigra of post-mortem PD brain tissue, ample cytoplasmic pRb, E2F-1, and PCNA immunostaining along with DNA duplication were also observed [Fig. 4] (Braak et al., 1998; Frade and Ovejero-Benito, 2015a).

Apart from human brain tissue samples, in-vivo and in-vitro models have been developed using neurotoxins like 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and 6-hydroxydopamine (6-OHDA) and are used to study the disease pathophysiology (Schober, 2004). These models reproduce a subset of PD features and are majorly used to study the disease. In one such study done by Neystat et al. on 6-OHDA-treated rats, activation of CDK-5 and its concomitant increase along with p35 expression levels were detected in the nigral neurons. These nigral neurons were observed to be in the late morphological state of apoptosis (Neystat et al., 2001). Furthermore, another report suggested an increase in the expression of CDK-5 after MPTP administration. Dominant-negative CDK-5 or inhibition of CDK-5 activity by general CDK inhibitors have shown beneficial effects in curtailing disease progression (Binukumar et al., 2015). The data thus far available indicate an imperative role of CDK-5 as a mediator in the pathogenesis of PD.

Additionally, immunohistochemical examination of primary rat midbrain neuron cultures treated with MPTP also displayed unusual cell cycle protein expression in the G₁-M phase transition. The observed phenotype also corroborated with the data obtained from human studies. The dopaminergic neurons of rat SNpc displayed markers signifying entry of neurons in the S and G₂ phase after 6-OHDA administration (Schober, 2004). Strategies targeting the G₁ phase, such as flavopiridol-mediated inhibition of CDK-4 and E2F knockout, have shown neuroprotective effects against PD pathology in in-vivo and in-vitro studies (Verdaguer et al., 2005).

Altogether, these observations suggest that dopaminergic neurons post mitosis subvert the G₁/S checkpoint and re-enter the cell cycle in PD but get arrested at the M-phase. Perhaps, the miscarried cell cycle is the process that leads the neurons towards apoptosis and eventual neurodegeneration in PD.

4. Outlook: how is EBV supposedly driving neurons' cell cycle re-entry (CCR) and leading to degeneration?

Various studies have linked the neurodegenerative pathology of AD, PD, and other disorders with Herpesviral infections (De Chiara et al., 2012; Hogestyn et al., 2018; Costa Sa et al. 2019b). However, the precise mechanisms by which these viruses exert their deteriorative effect on the nervous system are still elusive. Although these viruses are well known to modulate cell cycle events [Fig. 6] for their benefit, as discussed above, and dysregulation of the cell cycle is reported to be an early event in neurodegeneration; no such study exists linking them to the two. This review has discussed some poorly understood, intensely debated, and underexplored avenues of virus-mediated neurodegeneration. Additionally, we propose a possible mechanism of EBV-mediated cell cycle dysregulation in causing neurodegeneration.

Fig. 6.

Epstein-Barr virus antigens deregulate the cell cycle. Epstein-Barr virus (EBV) associated nuclear antigens (EBNA) – 1, 2, 3A, 3B, 3C, 5 (LP), LMPs, and Rta modulate the cell cycle at various stages. In brief, at the early G1 stage of the cell cycle, E3C and LMP-1 hinder the activity of p27 and p16, respectively, to block their inhibitory effect on the Cyclin D/CDK-4/6 complex. E3C also inhibits the cytoplasmic translocation of GSK-3b to prevent Cyclin D phosphorylation. In addition, E3C modulates the E2F-mediated transcription by interacting with pRb and AURKB; E3C also mediates the nuclear translocation of mitochondrial protein MRPS18-2 to facilitate its binding with pRb and thereby regulating the level of free E2F. During the late G1 phase, the formation of the Cyclin E/CDK-2 complex is affected by the binding of LMP-1 and E3C. Further, the p21^WAF1/CIP1 mediated inhibition of the Cyclin E/CDK-2 complex is obstructed by the binding of E3A and Rta antigens of EBV. Interaction of E3C with other host molecules such as p53 or Chk1/2 also affects cell cycle progression from G1 to the S phase. Along with E3, other EBV antigens such as E5 and E1 also modulate the Mdm-2-mediated apoptosis pathway in the late G1 phase of the cell cycle. E5 is reported to directly interact with Mdm-2, while E1 binds with HAUSP to promote ubiquitination of the Mdm-2 molecule, leading to its degradation and thereby modulating the host cell cycle. Progressing to the S phase of the cell cycle, EBV antigen E3B is reported to directly bind Cyclin A, while E3C and E3A curb the inhibitory effect of p27 and p21^WAF1/CIP1 respectively towards Cyclin A/CDK-1/2 complex. Rta binding to 14-3-3σ enhances its expression and thereby sequesters the activity of the Cyclin A/CDK-1/2 complex. During the G2 phase of the cell cycle, E3A binding to p21^WAF1/CIP1 obstructs its blocking activity towards the Cyclin /-CDK-1 complex. Furthermore, the binding of E3C to the p53-gemin complex leads to IRF-5-mediated upregulation of LMP-1 expression that causes p53-mediated apoptosis to stop.

Various factors that might act as a trigger are reported to initiate the cell cycle in terminally differentiated neurons. A trending hypothesis states the implication of viral infections in the etiology of neurodegeneration via modulating the cell cycle. First, EBV has been shown to associate with neurodegenerative diseases. Second, EBV, a well-known oncogenic virus, is reported to manipulate the cell cycle at various stages of protein-protein interaction, protein redistribution, or molecular mimicry, as mentioned above in detail. Interestingly, these steps are also crucial in developing neurodegenerative pathology. For instance, pRb, a protein responsible for guarding the G₁ to S transition, is known to be modulated by EBV and is also associated with PD pathology. Moreover, in addition to the direct interaction of EBV with the effector molecules, it might be acting by creating favorable conditions for cell cycle progression. EBV could also instigate cell cycle progression via ectopic and untimely accumulation of various CDKs and other modulators. These facts suggest a strong association between neurodegeneration and EBV-mediated cell-cycle dysregulation. Further studies in this direction could provide a tangible causative link between the two and help us establish the precise role of EBV in aggravating neurodegenerative pathology.

Funding

The authors thank the Council of Scientific and Industrial Research Grant No. 37(1693)/17/EMR-II. They also thank MHRD for providing fellowship to one of the authors (D.T.). The funding bodies have not played any role in the conceptualization and writing of the manuscript.

Data availability

Data sharing does not apply to this article as no data set was generated during the current study.

CRediT authorship contribution statement

Deeksha Tiwari: conceived the manuscript under the guidance of, Writing – original draft, review & editing, All authors have read and approved the final manuscript. Nitish Mittal: The final draft was prepared with the help of, Writing – review & editing, All authors have read and approved the final manuscript. Hem Chandra Jha: conceived the manuscript under the guidance of, Writing – review & editing, All authors have read and approved the final manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Peer Review Overview and Supplementary data

A Peer Review Overview and (sometimes) Supplementary data associated with this article: