Apoptosis defines a programmed form of cell death that leads to caspase activation, controlled specific cellular modifications such as cell shrinkage and formation of apoptotic bodies that are later removed by phagocytes (Li et al., 2021). This serves the homeostasis of multicellular organisms, where old cells are replaced by new ones in order to eliminate degenerated cells, ensure stable tissue size and furthermore plays a critical role during development (Plati et al., 2008; Bucur, 2016). While there are two pathways that initiate apoptosis (Plati et al., 2011), the goal of both is the same: the activation of effector caspases (cysteine-aspartic proteases), enzymes that cleave cellular proteins and activate endonucleases that degrade the DNA in the nucleus (Aitken and Baker, 2013; Bucur, 2016). The extrinsic pathway of apoptosis is initiated by the activation of death receptors such as Fas or death receptors (DR4, DR5), which can polymerize and recruit adaptor proteins when ligands such as FasL and Apo2L/TRAIL are present to trigger the downstream activation of caspase 8 (Kumar et al., 2005; Bucur et al., 2006, 2015; Elmore, 2007). On the other hand, the intrinsic pathway of apoptosis is induced by endogenous DNA damage, which leads to the accumulation of p53, a tumor suppressor protein that acts as a transcription factor for pro-apoptotic genes, e.g., BAX and BAK and many others (Elmore, 2007; Pennarun et al., 2013; Bucur, 2016).

Tyrosine kinases regulate many critical cellular processes and can be divided into two subgroups (Okada, 2012): 1) receptor tyrosine kinases, such as the EGF or VEGF receptors, that typically span the cell membrane and have the ability to autophosphorylate upon ligand binding and dimerization; and 2) cytoplasmic tyrosine kinases, such as CSK and the SRC family tyrosine kinases (SFKs), that can be found in the cytoplasm.

C-terminal Src kinase (CSK) is a cytosolic tyrosine-protein kinase with an important role in regulating critical cellular decisions, such as cellular apoptosis, survival, proliferation, cytoskeletal organization and many others. CSK is an endogenous inhibitor of the SRC family of tyrosine kinases (SFKs) by phosphorylating them on a conserved C-terminal tyrosine residue (Chong et al., 2009). By doing so, CSK regulates important cellular functions, such as cellular apoptosis and survival (Okada, 2012). Since our current knowledge of the CSK mechanisms of action is limited, the known functions that CSK modulates are generally dependent on CSK inhibition of SFKs.

Until now, there are only a few reviews that focused on the tyrosine protein kinase CSK. For instance, Okada gives a great overview over the structure, function and discovery of CSK (Okada, 2012). Also, the review by Kim et al. (Ia et al., 2010) and a different review by Chong et al. (Chong et al., 2009) focused on the interaction of SFKs with CSK-family kinases, describing the structural inactivating mechanism CSK exerts on SFKs.

In this Review, we are focusing on revealing the apoptosis regulation mechanisms by CSK, CSK inhibition of the SFKs, while further exploring the clinical relevance of CSK in important pathologies, such as cancer, autoimmune, autoinflammatory, neurologic diseases, hypertension and HIV/AIDS.

SFKs are a family of cytoplasmic tyrosine kinases that can be found in mammalian cells (Dodd et al., 2014).

They can be activated by different transmembrane receptors, such as the EGF receptor, T cell receptor or integrins and constitute an integral part of signaling cascades regulating cell proliferation, differentiation and cell motion (Thomas and Brugge, 1997).

There are currently ten known types of SFKs (c-Src, c-Yes, Fyn, c-Fgr, Lyn, Hck, Lck, Blk, Frk, Hck), c-Src, c-Yes and c-Fgr being the human analogues of the historically earlier discovered viral SFKs (Okada, 2012; Amata et al., 2014). Some SFK types, such as c-Src, Yes and Fyn, can be ubiquitously found, while other SFK types are only expressed in specific cell types. For instance, Lck is particularly located in T-cells and plays a critical role in lymphocyte T-cell activation (Engen et al., 2008; Manz et al., 2015).

SFKs exist in an active and inactive state, depending on the three-dimensional phenotype of the protein. In order to understand SFK function and regulation, it is helpful to visualize the general structure of SFKs (Figure 1 uses c-Src as a representative SFK).

c-Src (as an example for SFK) can be divided into the following functional domains (Okada, 2012):

• N-terminal domain (also called SH4): it contains an acylation site (myristoylation or palmitoylation) responsible for plasma membrane adherence of SFKs.

In order to inactivate a SFK, a tyrosine residue at the C-terminus of the SFK can be phosphorylated by kinases, causing an intramolecular interaction of the tail with the SH2 domain (Engen et al., 2008). In addition, an interaction between the SH3 domain and an amino acid sequence connecting the SH2 with the tyrosine-kinase domain is formed (Engen et al., 2008). These changes in protein conformation downregulate SFK activity.

In contrast, SFK activation can be achieved by autophosphorylation of one or more conserved tyrosine residues in the autophosphorylation-loop, in the kinase domain (Lin et al., 2003; Chong et al., 2005). Simultaneously, phosphatases dephosphorylating the C-terminal tyrosine residue are needed as well, to undo the conformation change that occurs when inactivating the SFK.

Regarding SFK regulation at the C-terminal tyrosine, phosphorylation can be carried out by C-terminal Src kinase (CSK) (Figure 2) or by CSK homologous kinase (CHK), whereas phosphatases identified to remove the phosphate group from the tyrosine can be found in the cytoplasm (PTP1B, Shp1, Shp2) or in the plasma membrane, where they reside as transmembrane enzymes (CD45, PTPɑ, PTPγ, PTPε) (Roskoski, 2005).

In addition to this catalytic SFK regulation mechanism, SFKs can also be regulated by non-catalytic binding of an inhibitor to the SFK protein. Examples for such non-catalytic inhibitors are CHK, WASP, caveolin and RACK1 (Chong et al., 2005).

Meticulous control of SFK activity is very important in order to ensure proper receptor signaling. Moreover, increased activation of SFK has an oncogenic effect, and thus it is associated with cancer development and progression (Guarino, 2010). T cells from mice with a mutated CSK, displayed longer and elevated T-cell receptor signaling and stronger T-cell proliferation (Manz et al., 2015). Similarly, IgE receptor-mediated mast cell signaling and degranulation increased when knocking down CSK, while simultaneously lessening production of proinflammatory cytokines (Potuckova et al., 2018).

C-terminal Src kinase (CSK) is an endogenous inhibitor of SFKs by phosphorylating the C-terminal tyrosine residue of the SFK, as described above (Sondhi et al., 1998).

A CSK molecule weighs 50 kDa and is built of three functional domains: SH3, SH2 and kinase domain (Okada, 2012). In comparison to SFK structure, CSK does not contain an N-terminal acylation group and the regulatory tyrosine residues in the kinase and C-terminal domain. Furthermore, SFKs and CSK differ in the position of the SH2 domain (Figure 1) (Roskoski, 2004).

As CSK, unlike SFKs, does not contain an acylation site, it is a cytoplasmic protein and hence needs to be recruited to the plasma membrane in order to exert its inhibitory function (Manz et al., 2015; Potuckova et al., 2018). Therefore, adaptor proteins are needed. Csk-binding protein (Cbp), also called phosphoprotein associated with glycosphingolipid microdomains (PAG), is such a transmembrane protein that can be found in lipid rafts (Davidson et al., 2003). In T cells, Cbp binds the SH2 domain of CSK, which inhibits SFKs (Potuckova et al., 2018). When Cbp gets dephosphorylated, e.g. upon activation of a T-cell receptor, it dissociates from CSK, thus enabling SFK activity, since the SFKs are no longer inhibited by CSK (Torgersen et al., 2001). Other adaptor proteins include Lck-interacting molecule (LIME) (Brdičková et al., 2003), paxillin (Schaller and Parsons, 1995), Dok-1 and Dok-2 (Yasuda et al., 2007), VE-cadherin (Baumeister et al., 2005), caveolin-1 (Lee et al., 2000) and ZO-1 (Saito et al., 2008).

Similarly, CSK has been found to bind to the protein-tyrosine phosphatases (PTP) PEP, PTP-PEST and PTP-HSCF in hematopoietic cells, whereby association of CSK with PTP-PEST can also be observed in non-hematopoietic cells (Davidson et al., 1997; Wang et al., 2001). Whereas PEP and PTP-PEST use a proline-rich region outside the catalytic region to attach to the SH3 domain of CSK, PTP-HSCF binds to the SH2 domain of CSK (Cloutier and Veillette, 1996; Davidson et al., 1997; Wang et al., 2001). The complex of CSK with a PTP serves a more powerful inhibition of SFKs. On the one hand, PTPs exert their dephosphorylation of the tyrosine in the activation loop to inactivate the SFK (Cloutier and Veillette, 1999), and on the other hand, binding of a PTP to a CSK improves CSK function (Wang et al., 2001). Thus, PTPs do not only serve as recruiters of CSK to activated SFKs, but simultaneously support the inhibitory function of CSK.

Recently, Cui et al. found that CSK expression can be altered by SUMOylation, which is a post-translational modification of proteins that occurs in cells, where a small ubiquitin-like modifier (SUMO) is attached to a lysine of the target protein by the SUMO E1, E2 and E3 enzymes (Cui et al., 2019). CSK is SUMOylated on lysine 53, which negatively affects its function (Cui et al., 2019). Thus, less CSK is recruited to the lipid rafts after SUMOylation, which comes along with an increase in SFK activity (Cui et al., 2019).

Regarding the regulation of the CSK-SFK interaction, a negative feedback exists, activating CSK during SFK hyperactivity, although further investigation is required to elucidate the mechanism in detail (Jiang et al., 2006; Chen et al., 2014).

Furthermore, apart from its inhibitory function of SFKs, CSK has been observed to play an important role in gliotactin protein regulation, by promoting its endocytosis independently from its effect on SFKs (Samarasekera and Auld, 2018). Gliotactin is a protein that can be found at tricellular junctions, i.e. the location where tight junctions from three epithelial cells meet. There, it is responsible for orderly cell adhesion. Hence, CSK participates in maintaining cellular structure in a pathway that does not require SFKs, but of course, signaling cascades involving SFKs are critical for proper cell morphology. Inactivation of CSK in squamous epithelia in mice led to a compromise in cell-cell attachment, because of a change in cytoskeletal organization, which was mediated through CSK’s effect on SFKs (Yagi et al., 2007).

CSK regulates a wide variety of cellular functions, such as apoptosis, proliferation, cytoskeletal reorganization, cell migration, invasion, angiogenesis, mainly through the modulation of SFKs, while also being involved in preventing cancer progression, since SFKs have been associated with cell differentiation, cancer stage and development of metastases (Kim et al., 2009; Chen et al., 2014). As illustrated in Figures 3, 4, these functions are modulated and controlled by the influence exerted on specific pathways, such as mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinases (ERK) pathway, signal transducer and activator of transcription 3 (STAT3) pathway and phosphatidylinositol-3-kinase (PI3K)/AKT signaling pathway, alongside the direct influence over other specific substrates and/or pathways, such as beta-actin, paxillin, rhoA or integrins, mostly involved in cytoskeletal organization and other functions (Chen et al., 2014).

The main results of these interactions are: stimulating apoptosis (as opposed to cell survival), inhibition of angiogenesis, interrupting the pro-oncogenic effects of SFKs, such as cancer cell migration and invasion, while also interfering with cytoskeletal reorganization (Figure 3) (Kim et al., 2009; Chen et al., 2014).

As shown in Figure 4, the three main signaling pathways that are heavily influenced by SFKs in a positive manner (promoting cell survival/proliferation) as opposed to the negative influence of CSK which could promote apoptosis (by preventing the effects of SFKs) are MAPK/ERK, STAT3 and PI3K/AKT pathways (Chen et al., 2014).

The MAPK/ERK enzymatic cascade is crucial for cellular division. ERK itself is a member of the MAPK family with mitogenic abilities provided by regulating transcription factor activity and gene expression in the cell nucleus. The effects of this pathway are ensured by the activation of ERK, which is achieved through an activating sequence starting from Ras and continuing with the activation of Raf, MEK1/MEK2 and subsequently, ERK. When the latter is activated, it migrates from the cytoplasm to the nucleus, phosphorylating various transcription factors, such as c-Fos, c-Jun, Elk-1, ATF-2 and gene expression regulators such as CREB and c-Myc, in order to achieve its functions (Chen et al., 2014; Guo et al., 2020).

The STAT3 signaling pathway, which is a JAK-STAT pathway, is initiated when ligands, such as cytokines (interferons and interleukins; e.g. interleukin-6, involved in breast cancer progression) bind to its receptor which is coupled with a Janus kinase (JAK) that further activates STAT transcription factors, such as STAT3, through phosphorylation. This will further activate the expression of cyclin D1, Bcl-2, Bax, c-Myc, hypoxia-inducible factor (Hif) and other genes, therefore stimulating cell growth and ensuring increased cell survival (Chen et al., 2014; Ma et al., 2020).

The PI3K/AKT pathway is initiated when a growth factor binds to a tyrosine kinase receptor, thus activating PI3K which is able to phosphorylate AKT. AKT further phosphorylates TSC2 (within a TSC1/TSC2 dimer), stimulating Rheb and ultimately activating mTOR (Fang et al., 2013). Both mTOR and AKT have the ability to regulate protein synthesis in the nucleus in order to promote cell division, by modulating the activity of IκB kinase alpha (IKK-α), MDM2, telomerase reverse transcriptase (TERT), glycogen synthase kinase-3β (GSK-3β) and nuclear factor κB (NF-κB) (Chen et al., 2014; Vidal et al., 2021). AKT also phosphorylates and inactivates Forkhead Box O (FOXO) transcription factors proteins, which will remain sequestrated in the cytoplasm. When the PI3K/AKT pathway is inhibited (by CSK, for example), FOXO family of transcription factors could potentially be activated through dephosphorylation, becoming able to translocate to the nucleus and regulate the expression of various genes involved in the control of apoptosis, proliferation, cell differentiation and many other critical cellular processes (Singh et al., 2010; Bucur et al., 2014). Through this mechanism, FOXOs can promote apoptosis, which can be initiated either by the intrinsic pathway, if FOXO proteins upregulate pro-apoptotic members of the Bcl-2 family (such as Bim, Puma, NOXA, BNIP3) and downregulate the expression of the anti-apoptotic members of the Bcl-2 family (such as Bcl-xL), or by the extrinsic pathway through upregulation of TRADD, FasL, Apo2L/TRAIL, cFLIP and other proteins (Dumitrascu and Bucur, 2013). FOXO proteins are also involved in cell cycle arrest by upregulating p21 and p27 expression, while also downregulating cyclin D1 and cyclin D2 expression (Bucur et al., 2014).

Furthermore, inhibition of mTOR by modulation of SFK activity is not only a stimulator for apoptosis, but also for autophagy. For example, tumor-suppressing STF cDNA 3 (TSSC3) can interact with an SFK-bound RanBP9 protein, thereby inhibiting SFK function and the PI3K/AKT pathway, and thus promoting apoptosis and autophagy (Dai et al., 2016; Zhao et al., 2018). Loss of TSSC3 expression has been found to be involved in the pathogenesis of different cancers, e.g. osteosarcoma (Dai et al., 2012). However, increased SFK activity can also lead to autophagy. That is the case with interleukin IL-17A expression, that arises in neural tissue under ischemic conditions (Liu et al., 2019). SFKs get activated by IL-17 receptors, negatively regulating mTOR via PP2B activation to induce autophagy (Liu et al., 2019). Interestingly, SFKs can be targeted to influence deregulated autophagy in diseases. For instance, Kim et al. described 2-cyano-3,12-dioxo-oleana-1,9(11)-dien-28-oic acid methyl ester (CDDO-Me), a substance that activates nuclear factor-erythroid 2-related factor 2 (Nrf2), to also be able to inhibit SFKs and the PI3K/AKT pathway in order to regulate autophagy and clasmatodendrosis in astrocytes in the epileptic hippocampus (Kim and Kang, 2021).

Another interesting aspect of apoptosis regulation by CSK is a change in actin dynamics in order to enable morphologic changes of cells during apoptosis, leading to the formation of apoptotic bodies which can be removed by phagocytes (Bucur et al., 2001). Notably, apoptosis comes with dramatic changes in the organization of the cytoskeleton (Franklin-Tong and Gourlay, 2008; Lavoie et al., 2010): whereas microtubules and intermediate filaments dissemble, actin dynamics undergo huge changes that include actin formation, as well as actin destruction in the course of apoptosis.

Interestingly, in a study using transforming growth factor (TGF)-beta1 to stimulate the extrinsic apoptosis pathway, SFK activity was observed to shortly increase before dramatically declining (Park et al., 2004). Although not investigated in the study, this might suggest a delayed inhibition of SFKs by CSK, which eventually could lead to SFK cleavage into fragments that are translocated to the nucleus, as described by Park et al. (Park et al., 2004).

Since SFK activation plays an important role in regulating actin polymerization and thus, intracellular transport of proteins, initially increasing SFK activity might trigger a sudden change in actin organization, resulting in deranged traffic in the cell (Peterson and Chernoff, 2006; Bharti et al., 2007; Lavoie et al., 2010). These effects are obtained through phosphorylation of small GTP-binding proteins from the Rho family (e.g. RhoA, Rac1 and Cdc42) by SFKs, that in turn initiate other signaling cascades that regulate the disruption of cytoskeletal dynamics (Robert et al., 2006):

• RhoA activates Rho-associated protein kinase (ROCK), that among others is able to phosphorylate myosin light chain (MLC) contributing to contraction of actin-myosin and the formation of stress fibers, or it can for instance phosphorylate LIM kinase, which leads to an inhibition of actin depolymerization because of inhibited cofilin activity (Hartmann et al., 2015).

The subsequent decrease in SFK activity mediated by CSK is necessary to suppress survival signaling of the apoptotic cells as described above. Hence, CSK cleverly influences SFK activity to ensure the proper sequence of cellular events needed for apoptosis.

Aberrant CSK and/or SFK activity can be associated with a large number of diseases, such as cancer, autoimmune, autoinflammatory and neurologic diseases, hypertension and HIV/AIDS (Engen et al., 2008; Szilveszter et al., 2019; Xu et al., 2021). This highlights the importance of CSK as a key regulator of SFKs and its potential as a therapeutic target in the treatment of these diseases.

The tyrosine protein kinase CSK is able to efficiently inhibit SFKs in vitro and in vivo (Hui and Vale, 2014). SFK regulation mainly occurs through phosphorylation of conserved tyrosine residues of the protein. Phosphorylation of the C-terminal regulatory tyrosine, as performed by CSK, inhibits SFK activity, whereas phosphorylation of tyrosine residues in the activation loop enhances SFK activity. Simultaneously, the dephosphorylation on these respective sites has an opposing effect on SFKs. Although CSK and SFKs resemble in protein structure regarding their domains, the lack of a C-terminal regulatory tyrosine of CSK should be emphasized on the one hand, and of the N-terminal acylation on the other hand, making CSK a cytoplasmic protein, unlike SFKs that are membrane-bound. This difference in cellular location especially presents an obstacle for CSK activity, as other recruiter proteins such as PAG/Cbp, LIME, paxillin and Dok-1 and Dok-2 are needed to enable spatial proximity of CSK with SFKs. SFKs regulate a great variety of cellular functions, by multiple signaling cascades, that include MAPK/ERK, STAT3, PI3K/AKT, VEGF, IL-8 pathways. While active, SFKs promote cell survival, proliferation, angiogenesis and cytoskeletal reorganization amongst others. SFK suppression by CSK in different settings most likely induces apoptosis and inhibits uncontrolled cell growth and migration. Thus, CSK has a pro-apoptotic effect by inhibiting activation of SFKs, which, in an active state, would lead to the downstream activation of MAPK/ERK, STAT3 and PI3K/AKT signaling pathways. By inhibiting the latter, AKT-mediated inhibition of FOXO can cease, and FOXO is hence potentially able to exert its pro-apoptotic effect by regulating the transcription of its target genes in the nucleus. As a result, apoptotic pathways can get activated leading to the activation of effector caspases that induce the molecular and structural changes for cell death.

At the same time, short SFK activation followed by CSK-mediated SFK inhibition controls actin reorganization required for apoptosis as it ensures the morphologic cellular changes during apoptosis, resulting in the fragmentation of the cell into apoptotic bodies.

The clinical importance of SFKs and CSK is underlined by the various diseases they have been associated with. These include cancer, autoinflammatory, autoimmune and neurologic diseases, hypertension and HIV/AIDS.

Indeed, inhibitors of SFKs such as dasatinib, PP2, saracatinib and SI221 have already been approved for treating different types of hematologic cancers, such as chronic myelogenous leukemia (U.S. Food and Drug Administration, 2017) and acute lymphoblastic leukemia (Drugs.com, 2019) in both adult and pediatric patients. Their uses have also been discussed in treating solid tumors, with positive effects obtained by Redin et al. by administering dasatinib alongside immune checkpoint inhibitors in NSCLC which accelerated tumor regression, while clinical studies are also in progress regarding the use of SFK inhibitors, such as SI221, for treating rhabdomyosarcoma (Redin et al., 2021). SI221 is a pyrazolo-pyrimidine derivative which exerts inhibitory effects on Yes (a member of SFKs) which is overexpressed in rhabdomyosarcoma, thus reducing proliferation of cancer cells (Bagella and Marchesi, 2016). Other such molecules have been studied for treating different types of solid tumors, namely SAB298, which has shown promising effects of tumor growth inhibition in vivo for patients with melanoma. However, clinical trials need to be performed in order to certify its utility (Halaban et al., 2019).

Furthermore, the linkage of CSK to a wide variety of diseases might have potential to be used in the clinical field, as the search for potent CSK inhibitors/activators is ongoing (O’Malley, 2020). Until now, pyridazinone and pyrazolopyridine have proved efficient in significantly reducing SFK phosphorylation by CSK in vivo (O’Malley et al., 2019). In this study, the substances were then tested in T-cells and produced greater T-cell activation in response to a stimulating antigen. Thus, the optimization and application of pyridazinone and pyrazolopyridine could be promising for cancer immunotherapy. The challenge in the development of such a CSK inhibitor relies in obtaining substrate specificity of the agent to CSK, an objective that is made difficult by the structural resemblance of CSK and SFKs (O’Malley, 2020). Hence, more research is needed to further investigate the implication of CSK in these various diseases, its potential as a target for therapy and the influence this might have on cell survival and apoptosis.