2.2. Mitogen-Activated Protein Kinases (MAPKs)

• c-Jun N-terminal kinases are highly homologous (>85%) but have a distinct tissue distribution. JNK1 and 2 are ubiquitously expressed, while JNK3 expression is mainly limited to the brain. In contrast to ERKs, JNKs are primarily activated by stress signals such as oxidative stress, radiation, and DNA-damaging agents. JNKs are mainly localized in the cytoplasm, but their identified substrates are mostly TFs, including c-Jun, p53, STAT3, and c-Myc. The phosphorylation of c-Jun leads to AP-1 complex formation and thus the transcription of cyclin D1, promoting cell cycle progression, similar to ERK1/2 [54]. To date, only a few cytoplasmic interaction partners of JNKs have been identified [50,55].
• The p38 subfamily consists of four members (α, β, γ, δ), which respond to various environmental stress stimuli and cytokines such as interleukin-1 and tumor necrosis factor α(TNF). Interestingly, p38 both regulates the production of cytokines and responds to them. Other targets of p38 regulation are TFs and other protein kinases. Based on observations of p38 activation, it plays a role in inflammation, cell cycle regulation, and apoptosis [50,56].
• ERK5 (BMK1 or big MAP kinase 1) has a kinase domain similar to ERK1/2, sharing 51% similarity with ERK2. ERK5 is essential during normal embryogenesis [57]. An upstream activator of ERK5 is MEK5, whose expression is elevated in metastatic prostate cancer [58]. Similar to ERK1/2 and JNK, ERK5 also promotes cyclin D1 expression and cell cycle progression [59], as well as plays a crucial role in the maintenance of mitochondrial function and neuronal survival [60]. ERK5 is involved in various cellular processes, including cell survival, differentiation, and angiogenesis. Its activation has been linked to growth factors, oxidative stress, and other extracellular stimuli.
• The MAPK pathway has a critical role in cancer biology, extending beyond the extensively studied BRAF mutation in melanoma. The MAPK/ERK pathway, for instance, has been implicated in colorectal cancer, with mutations in KRAS and NRAS genes leading to its persistent activation, promoting uncontrolled cell proliferation and tumor growth [61]. These mutations, unfortunately, render the tumors resistant to EGFR-targeted therapies, highlighting the need for novel therapeutic strategies [62].
• Additionally, the JNK MAPK pathway, associated primarily with responses to stress signals and apoptosis, has shown links to cancer biology. Aberrations in JNK signaling can lead to an imbalance between cell proliferation and death, thereby contributing to oncogenesis. For instance, overactive JNK signaling has been found in several cancers, including breast and gastric cancer, often correlating with a worse prognosis [63]. Furthermore, the p38 MAPK pathway, typically associated with inflammation and cell differentiation, is also relevant in cancer research. Its complex, dual role in tumorigenesis is being unraveled; while its activation can suppress tumor growth by promoting cell cycle arrest and apoptosis, chronic activation can also enhance cancer cell survival, contributing to chemoresistance [64].

2.3. Glycogen Synthase Kinase-3 (GSK-3)

2.4. Dual-Specificity Tyrosine (Y)-Phosphorylation-Regulated Kinases (DYRKs)

2.5. Cdc2-like Kinase (CLK) and Other Less-Studied Kinases

2.6. SR-Specific Protein Kinase (SRPK)

2.7. Tyrosine Kinase Gene v-Ros Cross-Hybridizing Kinase (RCK)

• MAK (male germ cell-associated kinase) is mainly expressed in testicular germ cells during spermatogenesis and in the retina. In the retina, MAK localizes to connecting cilia in photoreceptor cells, negatively regulates the length of their cilia, and is essential for the survival of these cells [113]. Not surprisingly, MAK mutations are associated with retinitis pigmentosa, a photoreceptor degeneration disease in the retina [114,115].
• ICK (intestinal cell kinase) is highly conserved, constitutively, and widely expressed. Similarly to MAK, it negatively regulates ciliary length and is identified as an essential component of sonic hedgehog signaling [116]. These two factors seem to be the underlying cause of human ECO syndrome, a multi-organ illness affecting the endocrine, cerebral, and skeletal systems, caused by a missense mutation in the ICK gene [117]. In 2017, a study reported the role of ICK in colorectal cancer progression and its potential as a therapeutic target for the treatment of colorectal cancer [118].

2.8. Cyclin-Dependent Kinase-like (CDKL)

3. Protein Kinase Therapeutics—Kinase Inhibitors

Table 1. Prevalence and pattern of CMGC kinase subfamily dysregulation in cancer.

Kinase Family	Deregulation Mechanism	Examples of Cancer Types
CDKs	Mutations, amplifications, deletions, and altered expression levels	Breast cancer [127,128,129], lung cancer [130], pancreatic cancer [131]
MAPKs	Mutations in pathway components (e.g., RAS, RAF)	Melanoma [132,133,134], colorectal cancer [135], lung cancer [48,136]
DYRK	Overexpression	Pancreatic [137,138], ovarian cancer [139,140], osteosarcoma [141], rhabdomyosarcoma [142]
GSKs	Altered expression levels, post-translational modifications	Glioblastoma [143], pancreatic cancer [144], colon cancer [145]
CLKs	Overexpression, cancer-associated splicing alterations	Breast cancer [146], prostate cancer [147], ovarian cancer [148]

4. Protein–Protein Interactions of the CMGC Kinases

CMGC kinases participate in various protein–protein interactions that are crucial for their activity, functions, and signaling pathways. Understanding these interactions is essential to elucidate the molecular mechanisms governing CMGC-kinase-mediated cellular processes. In this section, we will discuss some of the known interactions involving CMGC kinases and their importance (Figure 2).

Figure 2. (A) A network of known high-confidence protein–protein interactions of CMGC kinases obtained from the IntAct Molecular Interaction Database are grouped according to their simplified GO process annotation. (B) Distribution of the number of known interactors of the CMGC kinases based on the affinity purification and proximity labeling detection method. The visualization of the protein–protein interaction (PPI) network was generated by importing the publicly available interaction dataset from https://www.ebi.ac.uk/intact (accessed on 8 May 2023) protein interaction data into Cytoscape v3.9.

CMGC kinases contribute to the development and progression of various diseases. Therefore, understanding the intricate network of protein–protein interactions involving CMGC kinases can provide valuable insights into the molecular mechanisms underlying their functions and offer potential therapeutic targets for the treatment of diseases associated with their dysregulation. For instance, the development of small molecule inhibitors that specifically target the protein–protein interactions of CMGC kinases could represent a promising approach to modulate their activity and restore normal cellular functions in disease settings [180].

The development of small molecule inhibitors that target the protein–protein interactions (PPIs) of these kinases has emerged as a promising strategy in cancer therapeutics. Traditional kinase inhibitors target the ATP-binding site of kinases. However, this site’s high conservation across different kinases challenges the development of inhibitors with high specificity. In contrast, the PPI interfaces are often less conserved, presenting an opportunity for creating more specific inhibitors [183,184]. Small molecule inhibitors act by obstructing the critical protein interactions necessary for the activation or functioning of these kinases. For instance, inhibitors may preclude a kinase from interacting with its activator or substrate, consequently inhibiting the kinase’s activity and downstream signaling [185]. In cancer, these inhibitors can restore normal cellular functions by blocking the aberrant signaling engendered by dysregulated kinases. For example, if a kinase is overactive in a cancer cell, promoting uncontrolled cell proliferation, an inhibitor that obstructs this kinase’s interactions could impede or slow down the proliferation [186]. Considerable progress has been made in the development of such inhibitors. High-throughput screening techniques have enabled researchers to identify small molecules that can bind to the PPI interfaces of CMGC kinases [185]. Furthermore, advances in structural biology have provided intricate details of these kinases’ 3D structures and their complexes with interacting proteins, aiding in the design of more effective inhibitors [187]. Nonetheless, challenges persist in developing PPI inhibitors. The interfaces involved in PPIs are often expansive and flat, complicating the ability of small molecules to bind with high affinity. Also, these inhibitors must navigate cellular barriers to reach their intracellular targets [186].

Most of the studies regarding kinase interactions have focused on a single kinase and its function or dysfunction in a disease. Additionally, the names of the CDK kinases were unified as late as in 2009 [10], and that might have contributed to the slowness of studies of some these kinases. It is, after all, rather difficult to assign a kinase to this interesting family when the name is very different, such as PFTK1 or CRKRS.

5. Assessment of Kinase Activity

As the function of protein kinases is to transfer a phosphate group, Western blotting with antibodies against a substrate’s phosphoserine, -threonine or -tyrosine is the most common and widely used method to indirectly identify kinase activity. This approach is very labor-intensive, highly dependent on the specificity and sensitivity of the available antibodies, and not directly comparable with high-throughput approaches.

A more sensitive, direct approach to quantitatively measure kinase activity in vitro is to measure the incorporation of radiolabeled phosphate from γ-³³P ATP into the substrate [205]. A prerequisite to the solution-based in vitro kinase assays is a purified kinase, which in most cases is compared to the kinase dead mutant that abolishes ATP binding. An extensive study of 84 pairs of wild-type kinases and their kinase-dead counterparts [9] confirmed, using in vitro kinase assays, that indeed the mutated “VAIK” or “HRD” motif is essential for kinase activity. The lysine in the “VAIK” motif is among the most conserved amino acids in all kinases as it is a central catalyzer of the phosphate transfer [5].

Functional protein microarrays are an efficient ‘omics’-scale tool to detect kinase-substrate interactions [206,207]. The commercially available protein microarrays have thousands of purified proteins spotted onto a glass slide, providing an efficient means to identify the phosphorylation of candidate substrates for the kinases of interest when incubated together with radiolabeled ATP [208]. To date, the only large-scale study utilizing protein microarrays in kinase substrate identification has been performed with yeast protein kinases [209], although human protein microarrays are nowadays commercially available.