Cyclin-dependent kinase 5 (CDK5), a proline-directed serine/threonine-protein kinase, was originally purified from bovine brain and defined as a neuronal CDC2 (CDK1)-like kinase (NCLK) (Roder and Ingram 1991; Hellmich et al. 1992). Later on, CDK5 demonstrated capability to induce the Alzheimer-like characteristics by phosphorylation of tau protein (Baumann et al. 1993). p35 (CDK5R1) was then characterized as a regulatory subunit of CDK5 to activate its kinase activity, with subsequent identification of its isoform p39 (CDK5R2) (Tsai et al. 1994; Tang et al. 1995). CDK5/p35 is the first example of a CDC2-like kinase with neuronal function. A truncated form of p35, p25, was purified with CDK5 as a hetero-dimer exhibiting activity in vitro and regarded as a novel regulatory subunit of CDK5 (Lew et al. 1994). Accumulation of p25 is found in brains of AD patients and conversion of p35 to p25 plays a major role in triggering pathological events, leading to neurodegeneration (Patrick et al. 1999; Cruz et al. 2003). Inhibition of CDK5 prevents β-amyloid (Aβ)-induced neuronal death (Alvarez et al. 1999). CDK5/p35 kinase is found essential for neurite outgrowth, neuronal migration, laminar configuration of the cerebral cortex, as well as dynamics of actin skeleton reorganization (Nikolic et al. 1996, 1998). The Phosphorylation of DARPP-32 indicates the involvement of CDK5 in dopamine signaling pathway, neurotransmission, and cocaine addiction (Bibb et al. 1999, 2001). Nowadays, apart from neuronal functions, CDK5 signaling is also involved in cardiometabolic disorders and cancer events. CDK5 can regulate insulin expression and secretion in β-cell (Ubeda et al. 2004; Lee et al. 2008) and affect insulin sensitivity by phosphorylation of PPARγ (peroxisome proliferator-activated receptor γ) in obesity (Choi et al. 2010, 2011). CDK5-mediated hyperphosphorylation of SIRT1 contributes to endothelial senescence and atherosclerosis (Bai et al. 2012). Tumorgenesis and metastasis are also affected by CDK5 (Goodyear and Sharma 2007; Bisht et al. 2015).
Expression and Regulation
Human CDK5 gene is located on chromosome 7q36 and encodes a 31 kD polypeptide containing 292 amino acids. CDK5 belongs to the cyclin-dependent kinase (CDK) family and can bind to cyclin D1, D2, D3, and E (Miyajima et al. 1995; Malumbres et al. 2009; Nagano et al. 2013). Unlike other family members, the enzymatic activity of mammalian CDK5 is mainly dependent on noncyclin proteins, including p35 and p39 (Baumann et al. 1993). Cyclin I is also indicated to activate CDK5 in kidney podocytes (Nagano et al. 2013). In contrast to cell cycle CDKs, CDK5 does not require phosphorylation of its activation loop (T-loop). Both p35 and p39, containing 307 and 369 amino acids respectively, have CDK5 activation domains in their C-terminal region (Tsai et al. 1994; Tang et al. 1995). The crystal structure has explained the ability of p35 to activate CDK5 and that the activation domain has a tertiary structure which resembles cyclin A in the CDK2-cyclin A complex (Tarricone et al. 2001). Binding to regulatory subunits is sufficient for CDK5 to get fully stretched and activated. The available amount of p35 and p39 determines CDK5 kinase activity. p35 and p39 are short-lived proteins, with half-life of 30 and 120 min respectively, and are degraded by ubiquitin-protease system (Patrick et al. 1998; Patzke and Tsai 2002; Minegishi et al. 2010). This degradation is stimulated by CDK5-mediated phosphorylation of p35 at threonine138 (Kamei et al. 2007). When encountering oxidative stress or under pathological stimuli, increased influxes of Ca2+ induce protease calpain activity, facilitating cleavage of p35 and p39 to N-terminal truncated forms, p25 and p29 (Patzke and Tsai 2002; Smith et al. 2006). Because p25 has about five fold longer half-life than p35, CDK5/p25 complex is more stable with prolonged activity (Patrick et al. 1999). The nuclear protein SET can enhance CDK5 activity by its physical interaction with the N-terminal regions of p35 and p39 (Qu et al. 2002). By contrast, C42 can form a complex with p35 and specifically inhibits CDK5 activation (Ching et al. 2002). A C-terminal 172 amino acid domain of the DNA-binding protein, dbpA, also binds to CDK5 but inhibits its activity (Moorthamer et al. 1999). Ribosomal protein L34 is discovered as a novel inhibitor of CDK5, suggesting role of CDK5 in translational regulation (Moorthamer and Chaudhuri 1999). Despite wide expression in different cell and tissue types, CDK5 is mainly active in postmitotic neurons due to predominant expression of p35 and p39 in nervous system (Tsai et al. 1994; Tang et al. 1995). In neuronal cells, the subcellular distribtuion of CDK5 is determined by binding with different subunit. p39 localizes the active CDK5 complex in the perinuclear region and at the plasma membrane as does p35 (Asada et al. 2008, 2012). In addition, CDK5 also shows nuclear/cytoplasm translocation, which is critically involved in cell cycle control. CDK5 has no intrinsic nuclear localization signal (NLS) but contains two weak nuclear export signals (NES). Its nuclear localization relies on its binding to the cyclin-dependent kinase inhibitor p27, whereas its cytoplasmic localization is achieved through the NES-CRM-1 nuclear export mechanism (Zhang et al. 2010a). Structurally, CDK5/p27 interaction requires two protein faces involving threonine 17 on CDK5 (Zhang et al. 2010a) and tryptophan 60 on p27 (Kawauchi et al. 2006).
Cell Cycle Control
Ectopic expressed CDK5 does not promote cell cycle progression or change cell cycle distribution in several mammalian cell lines, so roles of CDK5 in cell cycle regulation are questioned (van den Heuvel and Harlow 1993). However, CDK5 plays crucial roles for neuronal cell cycle arrest. Postmitotic neurons cannot reenter a new cell cycle (Herrup and Yang 2007). Deficiency of CDK5 in neurons leads to loss of cell cycle arrest and cell death (Cicero and Herrup 2005). Latter studies reveal that dynamic nuclear/cytoplasmic translocation of CDK5 determines cell cycle arrest. β-amyloid peptide-treated primary neurons reenter a new cell cycle. They show localization of CDK5 changes from nuclear to cytoplasmic compartment. Loss of nuclear CDK5 is also observed in cycling cells (NIH 3T3), when cells reenter a cell cycle after serum starvation (Zhang et al. 2008). Blocking nuclear export abolishes cell cycle reentry (Zhang et al. 2008). The interaction between CDK5 and p27 contributes to nuclear localization. Moreover, p27 tightly regulates mitogenic signals to the cell cycle at the restriction point and is essential for G0 arrest (Coats et al. 1996). Transcription factor E2F1, a major factor regulating the G0/G1- to S-phase transition (Trimarchi and Lees 2002), also interacts with CDK5. The assembling of E2F1-CDK5-p35 complex in neuronal nucleus excludes the E2F1 cofactor, DP1, thus inhibiting E2F1 binding to cell cycle gene promoter and driving the cycle (Zhang et al. 2010b). The p27 can affect assembling of E2F1-CDK5-p35 trimer because the N-terminal truncation of CDK5 which lacks the binding domain with p27 loses its interaction with E2F1 (Zhang and Herrup 2011). In turn, CDK5/p35 directly phosphorylates and stabilizes p27, maintaining the amount of p27 in postmitotic neurons (Kawauchi et al. 2006).
Neuronal Survival, Migration, and Synaptic Plasticity
CDK5 activity is essential for a variety of neuronal functions such as neuronal survival, neuronal migration, and synaptic plasticity (Hisanaga and Endo 2010). CDK5-null mice exhibit perinatal lethality associated with unique lesions in brain development, including neuronal loss, neuronal migration defects, disrupted forebrain layering, as well as loss of cortical laminar structure (Ohshima et al. 1996) (Takahashi et al. 2010). CDK5 phosphorylates neuregulin receptors (ErbB), supporting neuregulin-mediated neuronal survival via PI3K/Akt activation (Li et al. 2003). CDK5 associates with and phosphorylates Bcl-2 to maintain its survival function (Cheung et al. 2008). By comparison, phosphorylation of JNK3 by CDK5 inhibits JNK3 activity and subsequent c-Jun phosphorylation, reducing neuronal apoptosis (Li et al. 2002). CDK5 phosphorylates and inhibits MEK1, repressing ERK induced apoptosis in cortical neurons (Zheng et al. 2007). CDK5 can protect neurons against oxidative damage via phosphorylating nuclear factor-erythroid 2-related factor-2 (Nrf2). Phosphorylated Nrf2 induces antioxidant gene expression and boosts glutathione metabolism (Jimenez-Blasco et al. 2015). Besides that, CDK5 regulates locomotion mode which covers most of neuronal migration routes in the developing cerebral cortex (Nishimura et al. 2010). CDK5 also contributes to proper position of the cortical neurons (Ohshima et al. 2001), facial branchiomotor, and inferior olive neurons (Ohshima et al. 2002) in the developing mouse brain. These biological functions mainly rely on CDK5-mediated phosphorylation of a number of substrates, including PAK1 (Rashid et al. 2001), focal adhesion kinase (Fak) (Xie et al. 2003), doublecortin (DCX) (Tanaka et al. 2004), microtubule-associated protein (MAP1B) (Kawauchi et al. 2005), G protein regulated inducer of neurite outgrowth 1 (Grin1) (Contreras-Vallejos et al. 2014), and drebrin (Tanabe et al. 2014). Ras-related protein/guanine nucleotide-exchange factor (RapGEF) signaling is critically involved in neuronal migration (Bos et al. 2001). By phosphorylation of RapGEF1(Utreras et al. 2013) or RapGEF2 (Ye et al. 2014), CDK5 facilitates proper neuronal migration in the cerebral cortex.
The induction of synaptic plasticity as well as spatial learning is affected in CDK5 or p35-null mice (Ohshima et al. 2005; Takahashi et al. 2010; Mishiba et al. 2014). By phosphorylation of numerous substrates, CDK5 is involves in regulating every aspect of synaptic function, from neurotransmitter release, vesicle cycling, and ion channel modulation to protein clustering and synaptic structural change along with spine formation (Lai and Ip 2009). CDK5 functions as an integral element in synaptic homeostatic scaling by influencing the number (Kim and Ryan 2010) and size (Marra et al. 2012) of synaptic vesicles when driving neurotransmitter release. CDK5-dependent phosphorylation of dephosphin protein contributes to maintaining synaptic vesicle endocytosis in nerve terminals (Tan et al. 2003). CDK5 can regulate polarity of neuropeptide-containing dense-core vesicles (DCVs) by promoting DCV trafficking in axons (Goodwin et al. 2012). NR2B, the subunit of N-methyl-D-aspartate receptor (NMDAR), improves synaptic plasticity and memory. CDK5 helps NR2B-containing NMDAR to localize in synaptic membrane, thereby modulating NMDAR-mediated synaptic currents (Plattner et al. 2014). CDK5 dynamically regulates Ca2+ channel. Phosphorylation of N-type Ca2+ channel by CDK5 enhances (Su et al. 2012), whereas phosphorylation of P/Q-type voltage-dependent Ca2+ channel reduces Ca2+ influx and neurotransmitter release (Tomizawa et al. 2002). Postsynaptic density 95 (PSD-95), a postsynaptic scaffolding protein, is implicated in synaptic maturation, strength, and plasticity. CDK5 phosphorylates PSD-95 and regulates the clustering of PSD-95/NMDA receptors at synapse (Morabito et al. 2004). CDK5 inhibits protein interaction between PSD-95 and Mdm2 (ubiquitin E3 ligase) and reduces PSD-95 ubiquitination, thereby repressing its subsequent interaction with clathrin adaptor protein complex AP-2 (Bianchetta et al. 2011). Through phosphorylation of TrkB, a receptor of neurotrophin brain-derived neurotrophic factor (BDNF), CDK5 activates Rac1 during dendritic spine remodeling, which is crucial for spatial memory formation (Lai et al. 2012). Both Cyclic AMP and CDK5 participate in phosphorylation of Wiskott–Aldrich syndrome protein family verprolin homologous protein 1(WAVE1) in neurons, which plays an essential role in the formation of the filamentous actin cytoskeleton, and thus in the regulation of dendritic spine morphology (Kim et al. 2006).
β-Cell and Insulin Secretion
CDK5 promotes pancreatic β-cell survival and proliferation via phosphorylating Fak (Daval et al. 2011) and retinoblastoma (Rb) (Draney et al. 2016). CDK5 is also involved in regulating gene transcription and protein secretion of insulin, a key glucose-lowering hormone. But this regulation is complex, depending on culture conditions (high vs. low glucose and duration). Glucose (20 mM, 24 h) upregulates CDK5/p35 activity through increasing p35 mRNA and protein expression in INS-1 cell. The active CDK5/p35 further activates insulin promoter (Ubeda et al. 2004). Munc18-1, an essential regulator in membrane fusion, is a substrate of CDK5. CDK5-dependent phosphorylation promotes Ca2+-dependent insulin exocytosis in primary β-cell treated with 10 mM glucose (Lilja et al. 2004). Phosphorylation of phospholipase D2 (PLD2) by CDK5 is an important process in the generation of phosphatidic acid and subsequent fusion event, which controls insulin secretion (Lee et al. 2008). Moreover, CDK5-mediated phosphorylation of β2-syntrophin can enhance insulin secretion by promoting the mobilization of cortical granules in β-cell (Schubert et al. 2010).
Endothelial Cell and Adipocyte
CDK5 expression is low in quiescent whereas high in proliferating endothelial cells. The highest CDK5 expression is detected when cells prepare for division (Sharma et al. 2004). CDK5 is required for endothelial cell motility and angiogenesis via regulating the activity of small GTPase Rac1 (Liebl et al. 2010). By phosphorylation of the forkhead transcription factor 2 (Foxc2), CDK5 is essential for lymphatic vessel development and lymphatic vascular remodeling (Liebl et al. 2015). CDK5-dependent phosphorylation of endothelial nitric oxide synthase (eNOS) at serine 113 (Chang et al. 2010) or serine 116 (Cho et al. 2010) reduces nitric oxide production. In 3T3-L1 adipocyte, CDK5/p35 interacts with glucose transporter type 4(GLUT4). The activity of CDK5 is stimulated by insulin which is dependent on PI3K pathway. The active CDK5 phosphorylates synaptotagmin homolog E-Syt1 and contributes to protein interaction between E-Syt1 and GLUT4, which leads to enhanced glucose uptake in adipocytes (Lalioti et al. 2009). However, a contradictory statement is proposed that CDK5-dependent phosphorylation of Rho family GTP-binding protein TC10α is responsible for its inhibitory function on insulin-stimulated GLUT4 translocation in 3T3-L1 adipocyte (Okada et al. 2008).
Aberrant activation of CDK5 triggers pathological events in the central nervous system, leading to neurodegenerative disorders such as Alzheimer’s (AD) and Parkinson’s diseases (PD). AD is the most common form of dementia in elderly population, featured by progressive impairment of cognitive functions. The predominant hallmarks of AD pathology include extracellular deposition of β-amyloid peptide in senile plaques, intracellular accumulation of hyperphosphorylated tau protein in neurofibrillary tangles (NFTs), neuronal loss, and synaptic dysfunction (Liu et al. 2016). In human AD postmortem brain, a significant elevation of CDK5 activity is detected, accompanied by accumulation of p25 and active calpain (Grynspan et al. 1997; Lee et al. 1999; Tseng et al. 2002). Animal studies confirm the contribution of CDK5 deregulation to neuronal loss in AD (Cruz and Tsai 2004). PD is a common movement disorder, which is characterized by the early selective dopaminergic neuron loss in the substantia nigra and formation of Lewy bodies. Elevated calpain levels as well as increased CDK5/p35 are also observed in postmortem PD brains (Mouatt-Prigent et al. 1996; Nakamura et al. 1997). Calpain-mediated cleavage of p35 into p25 has been regarded as the predominant culprit of hyperactivation of CDK5 activities, leading to this pathological event (Cruz et al. 2003).
In AD pathology, CDK5/p25 forms a vicious cycle with Aβ. Aβ is derived from β-secretase (BACE1)-mediated cleavage of glycoprotein APP. CDK5/p25 can promote Aβ generation and accumulation by direct and indirect regulations of APP. Direct phosphorylation of APP by CDK5/p25 at threonine688 is found to not only increase Aβ formation but also transform GSK-3β activity for APP to modulate Aβ generation (Wilkaniec et al. 2016). Enhanced BACE1 transcription by CDK5-mediated phosphorylation of STAT3 (Wen et al. 2008) as well as elevated presenilin level via phosphorylation of PS1 by CDK5 (Lau et al. 2002) can lead to increased production of Aβ from APP. CDK5-mediated phosphorylation of Foxo3 initially rescues cells from oxidative stress by upregulating Mn-superoxide dismutase (MnSOD) but eventually promotes neuronal death and aberrant Aβ processing via activating genes of Bim and FasL (Shi et al. 2016). In turn, Aβ generation triggers Ca2+ influx, thereby further activating calpain-dependent cleavage of p35 to p25 (Kawahara 2010). Aβ-stimulated CDK5 activation can induce p38 activation via generation of reactive oxygen species (ROS) in neuronal cells, thus increasing expression of c-Jun, which is overexpressed in AD and significantly contributes to neurodegenerative diseases (Chang et al. 2010). Aβ can increase S-nitrosylation of CDK5 and activates CDK5 activity by evoking calcium imbalance (Qu et al. 2012). The increased S-nitrosylation induces nitrosylation of dynamin-related protein 1 (Drp1), contributing to excessive mitochondrial fragmentation, with subsequent bioenergetics compromise and dendritic spine loss (Qu et al. 2011). This positive feedback loop results in accumulation of Aβ in the postmitotic neuron, which induces neurotoxicity associated with amyloid cascade and finally leads to neuronal loss. CDK5/p25 has been shown potent capability to phosphorylate tau at various sites directly or indirectly. The direct phosphorylation of tau leads to NFT formation in AD. This modification of tau impairs its capability to assemble tubulin into microtubules, thereby inducing dysfunctions of cytoskeleton and axonal transport. Phosphorylation by CDK5 also causes tau oligomerization and then aggregation into intraneuronal tangles of paired helical filaments (PHF), the main components of NFTs. GSK-3β is another important kinase involved in tau phosphorylation and proved to be activated by binding of p25 (Chow et al. 2014). CDK5-mediated phosphorylation of protein phosphates 1 contributes to increased tau phosphorylation. PI3K/Akt pathway can be activated by CDK5, leading to increased tau accumulation (Dickey et al. 2008). In AD transgenic mice, c-Abl can further promote CDK5 activity by phosphorylation of tyrosine15 and promote tau hyperphosphorylation (Zukerberg et al. 2000).
In AD progression, hyperactivation of CDK5/p25 is involved in neuron apoptosis as well as neuron death via multiple mechanisms. For example, CDK5/p25 is proved to induce the expression and activity of the p53 tumor suppressor by phosphorylation, thus promoting expression of its downstream proapoptotic target Bax (Lee et al. 2007). Phosphorylation of lamin A and B1 in neurons by CDK5 results in robust nuclear fragmentation and high neurotoxicity, which is an early and irreversible trigger for apoptosis (Chang et al. 2011). CDK5 is also responsible for neurotoxicity-induced apoptosis through phosphorylation of transcription factor MEF2 (Ke et al. 2015). Reduced activity of apurinic/apyrimidinic endonuclease 1 via phosphorylation by CDK5 leads to dysfunction of base excision repair following DNA damage and further neuronal death (Huang et al. 2010). Cell cycle reentry indicates the stress-induced conversion of mature neurons from the steady G0 state to reenter the cell cycle, which has been revealed in human AD brains (Lopes et al. 2009). Neuron cells encountering cell cycle reentry cannot go through G2/M checkpoint and then initiates apoptotic cell death pathways. CDK5 deregulation triggers reentry of cell cycle, which acts as a novel mechanism related with neuronal apoptosis and death (Folch et al. 2012). CDK5 deregulation triggered by Aβ or other stimuli can translocate CDK5 from nucleus to cytoplasm and impair assembling of E2F1-CDK5-p35-p27 in neuronal nucleus. Thus, CDK5 loses capability to control cell cycle arrest (Trimarchi and Lees 2002) (Zhang and Herrup 2011). CDK5 improves the CDK1/2/4 kinase activities via phosphorylation of cell division cycle proteins, Cdc 25A/B/C phosphatases, thus leading to cell cycle-driven death (Chang et al. 2012).
CDK5 has been implicated in the development of various types of cancers. Gene copy number of CDK5 is increased in human gastric cancer genome (Yang 2007). CDK5 promoter polymorphisms contribute to the genetic susceptibility to human lung cancer (Choi et al. 2009). Moreover, increased CDK5/p35 expression is detected in human hepatocellular carcinoma (HCC) (Ehrlich et al. 2015), non–small cell lung cancer (Liu et al. 2011), and nasopharyngeal carcinoma (NPC) (Zhang et al. 2015). Elevated p35 alone is found in invasive prolactin pituitary adenomas (Xie et al. 2016). The positive correlation between high level of CDK5 and lymph node metastasis is observed in NPC patients (Zhang et al. 2015) or patients with lung cancer (Liu et al. 2011). CDK5 regulates STAT3 activation by phosphorylation at serine 727, which promotes androgen receptor activity and supports prostate cancer growth (Hsu et al. 2013). CDK5 also modulates phosphorylation of Rb and activates downstream effectors, CDK2/cyclin A, thereby inducing medullary thyroid carcinoma (Pozo et al. 2013). Isoform A of phosphatidylinositol 3-kinase enhancer (PIKE-A), a prooncogenic factor, is phosphorylated by CDK5. This phosphorylation stimulates activity of PIKE-A GTPase and downstream Akt, thereby benefitting migration and invasion of human glioblastoma cells (Liu et al. 2008). CDK5-dependent phosphorylation can downregulate protein stability of caldesmon, in turn rescues motility and invasion/migration of melanoma cell (Bisht et al. 2015). Talin is a β-integrin tail-binding protein required for integrin activation. CDK5-dependent phosphorylation at head domain inhibits its ubiquitination and degradation. Thus, CDK5 controls talin head turnover, adhesion stability, and ultimately pheochromocytoma cell migration (Huang et al. 2009). Phosphorylation of FAK by CDK5 is found to be involved in TGFβ1 induced epithelial-mesenchymal transition in breast cancer (Liang et al. 2013). Despite information above, the impacts of CDK5 on tumor cell survival and migration have not been addressed in details yet.
Nowadays, inhibition of CDK5 is pharmacologically achievable. Acitvity of CDK5 can be blocked upon occupation of its ATP-binding pockets by specific compounds. Roscovitine ([2-(1-ethyl-2- hydroxyethylamino)-6-benzylamino-9-isopropylpurine, also CYC202, (R)-Roscovitine, Seliciclib]) is a chemically synthesized purine analogue displaying high efficiency and selectivity towards CDKs. Due to the lowest IC50 value (0.16 μM) of CDK5 compared with other CDKs (Meijer et al. 1997), roscovitine is widely used as CDK5 inhibitor in basic research or clinical trials. With rat model of traumatic brain injury, roscovitine reduces neuronal loss, glial activation, and neurologic defects (Hilton et al. 2008). Roscovitine downregulates p38-induced c-Jun expression upon Aβ stimulation, thereby decreasing neuron death in primary neuron cells (Chang et al. 2010). Roscovitine can exert protective effect for β-cell against glucotoxicity (Ubeda et al. 2006). It also restores antisenescence functions of SIRT1 and exhibits therapeutic potential for combating vascular senescence and atherosclerosis in vivo (Bai et al. 2012, 2014). Roscovitine has been tested in several clinical trials in patients with non–small cell lung cancer, hepatocellular carcinoma, and undifferentiated nasopharyngeal carcinoma. But only partial tumor response is observed in some patients (Benson et al. 2007; Le Tourneau et al. 2010; Hsieh et al. 2009). A truncated form of p35 consisting of residues 154–279 demonstrated high affinity to CDK5 and is termed as CDK5 inhibitory peptide (CIP) (Kesavapany et al. 2004, 2007). CIP is found to selectively inhibit CDK5/p25 complex activity without affecting CDK5/p35 or mitotic CDK activities both in vitro and in vivo. Tau hyperphosphorylation and apoptosis is reduced in primary neurons by CIP (Zheng et al. 2005; Kesavapany et al. 2007). Overexpression of CIP in mice can reduce neuroinflammation by antagonizing deleterious effects of tau and amyloid pathologies (Sundaram et al. 2013). CIP can protect β-cell against glucotoxicity induced by high glucose and recover insulin secretion (Zheng et al. 2013). A 24-residue peptide, p5 fragment, spanning CIP residues Lys245-Ala277 is determined as a more effective peptide that selectively inhibit CDK5/p25 activities, demonstrating reduced Aβ-neuronal loss in cortical neurons (Binukumar et al. 2015). Calpain inhibitors also have potential to inhibit aberrant CDK5/p25 hyperactivities. Calpain activation by abnormal Ca2+ influx is closely related NMDAR activities. An NMDAR antagonist, D-2-amino-5-phosphonovalerate, and a non-NMDAR antagonist, 6-cyano-7-nitroquinoxaline-2, are utilized to reduce activation of calpain, however fail in clinical trials due to unacceptable side effects. A transactivating regulatory protein-metabotropic glutmate receptor 1 is later developed as an alternative to general receptor antagonist and shows neuroprotective effects (Xu et al. 2009). Inhibition of calpain activity or its downstream targets also exhibits neuroprotection. Apart from the intrinsic inhibitor of calpain, calpastatin, a number of reversible or irreversible exogenous caplain inhibitors are developed including AJK275, MDL-28170, PD150606, SJA6017, A-705253, SNJ-1945, and Calpeptin (Yildiz-Unal et al. 2015).
In summary, it is certain that CDK5 is of great importance in the development of CNS. It is also functionally important in various types of cells and physiological contexts. However, aberrant expression or activation of CDK5 turns it into an active participant of multiple pathogenic events. Inhibition of CDK5 is pharmacologically accessible. Although preclinical studies have shown therapeutic potentials of CDK5 inhibitors against neurodegeneration, high glucose induced glucotoxicity, endothelial senescence as well as tumorigenesis and metastasis, its translational values have not been supported by sufficient clinical evidence. The limited clinical trials only show partial antitumor response of CDK5 inhibition in tumor patients. Evaluations in other pathologies are still lacking. What is more, mechanistic roles of CDK5 in different pathogenesis deserve intensive investigations because of the versatility of this kinase in different cell types and tissues. How does CDK5 complex switch from physiological to pathological events precisely? Are there any other mechanisms contributing to AD apart from cleavage of p35 to p25? Does CDK5 signaling participate in innate and acquired immune response in neurodegeneration? In nonneuronal cells, does CDK5 still suppress cell cycle reentry? Is this mechanism also involved in endothelial senescence or tumor cell growth? Why is CDK5 expression increased in many cancer cases but maintained stable under other pathological conditions? Which mechanism is responsible for different transcriptional/translational regulation of CDK5 in different cell types? Future studies will be of great significance to clarify the detailed CDK5-related mechanisms in different pathological pathways, which will supply more cues for therapeutic approach development.
- Amato AA, Rajagopalan S, Lin JZ, Carvalho BM, Figueira ACM, Lu J, et al. GQ-16, a novel peroxisome proliferator-activated receptor gamma (PPAR gamma) ligand, promotes insulin sensitization without weight gain. J Biol Chem. 2012;287:28169–79. doi: 10.1074/jbc.M111.332106.CrossRefPubMedPubMedCentralGoogle Scholar
- Benson C, White J, De Bono J, O’Donnell A, Raynaud F, Cruickshank C, et al. A phase I trial of the selective oral cyclin-dependent kinase inhibitor seliciclib (CYC202; R-Roscovitine), administered twice daily for 7 days every 21 days. Br J Cancer. 2007;96:29–37. doi: 10.1038/sj.bjc.6603509.CrossRefPubMedGoogle Scholar
- Bisht S, Nolting J, Schutte U, Haarmann J, Jain P, Shah D, et al. Cyclin-dependent kinase 5 (CDK5) controls melanoma cell motility, invasiveness, and metastatic spread-identification of a promising novel therapeutic target. Transl Oncol. 2015;8:295–307. doi: 10.1016/j.tranon.2015.06.002.CrossRefPubMedPubMedCentralGoogle Scholar
- Hsu FN, Chen MC, Lin KC, Peng YT, Li PC, Lin E, et al. Cyclin-dependent kinase 5 modulates STAT3 and androgen receptor activation through phosphorylation of Ser(7)(2)(7) on STAT3 in prostate cancer cells. Am J Physiol Endocrinol Metab. 2013;305:E975–86. doi: 10.1152/ajpendo.00615.2012.CrossRefPubMedGoogle Scholar
- Lalioti V, Muruais G, Dinarina A, van Damme J, Vandekerckhove J, Sandoval IV. The atypical kinase Cdk5 is activated by insulin, regulates the association between GLUT4 and E-Syt1, and modulates glucose transport in 3T3-L1 adipocytes. Proc Natl Acad Sci USA. 2009;106:4249–53. doi: 10.1073/pnas.0900218106.CrossRefPubMedPubMedCentralGoogle Scholar
- Le Tourneau C, Faivre S, Laurence V, Delbaldo C, Vera K, Girre V, et al. Phase I evaluation of seliciclib (R-roscovitine), a novel oral cyclin-dependent kinase inhibitor, in patients with advanced malignancies. Eur J Cancer. 2010;46:3243–50. doi: 10.1016/j.ejca.2010.08.001.CrossRefPubMedGoogle Scholar
- Li BS, Ma W, Jaffe H, Zheng YL, Takahashi S, Zhang L, et al. Cyclin-dependent kinase-5 is involved in neuregulin-dependent activation of phosphatidylinositol 3-kinase and Akt activity mediating neuronal survival. J Biol Chem. 2003;278:35702–9. doi: 10.1074/jbc.M302004200.CrossRefPubMedGoogle Scholar
- Ohshima T, Ogawa M, Veeranna, Hirasawa M, Longenecker G, Ishiguro K, et al. Synergistic contributions of cyclin-dependant kinase 5/p35 and Reelin/Dab1 to the positioning of cortical neurons in the developing mouse brain. Proc Natl Acad Sci U S A. 2001;98:2764–9. doi: 10.1073/pnas.051628498.CrossRefPubMedPubMedCentralGoogle Scholar
- Takahashi S, Ohshima T, Hirasawa M, Pareek TK, Bugge TH, Morozov A, et al. Conditional deletion of neuronal cyclin-dependent kinase 5 in developing forebrain results in microglial activation and neurodegeneration. Am J Pathol. 2010;176:320–9. doi: 10.2353/ajpath.2010.081158.CrossRefPubMedPubMedCentralGoogle Scholar
- Zheng YL, Li CY, Hu YF, Cao L, Wang H, Li B, et al. Cdk5 Inhibitory Peptide (CIP) Inhibits Cdk5/p25 Activity Induced by High Glucose in Pancreatic Beta Cells and Recovers Insulin Secretion from p25 Damage. Plos One. 2013;8:e63332. doi: 10.1371/journal.pone.0063332.CrossRefPubMedPubMedCentralGoogle Scholar