Hitherto, more and more proteins have been found to play important physiological functions in various life processes after being phosphorylated. Therefore, protein phosphorylation, as a fundamental regulatory modification of proteins, has gradually become an important research content and has come into people’s sight (Cohen, 2002; Taliabracci et al., 2012; Klement and Medzihradszky, 2017). The first sign of protein kinases’ existence was in 1883, when stoichiometric amounts of phosphate were found in the secreted protein casein. This casein was therefore identified as the first phosphoprotein, but the responsible kinase responsible for it was not clear until 2013 (Tagliabracci et al., 2012). The emergence of a novel atypical protein kinase family, the Fam20 family, solved these problems to a certain extent. These protein kinases are secreted in the Golgi apparatus and seem to be able to phosphorylate many extracellular proteins including casein (Tagliabracci et al., 2013). Protein kinase phosphorylates the substrate through phosphorylation, thus changing the activity of the substrate and mediating most of the signal transduction and cell processes, including transcriptional regulation and metabolic regulation (Manning et al., 2002). Therefore, abnormal protein phosphorylation is the cause of many diseases.

Fam20C, this kinase, and its family members Fam20A and Fam20B define a new family of secretory proteins that collectively regulate a diverse network of secretory pathway components (Palma-Lara et al., 2021). As the most widely studied member of the Fam20 family, Fam20C is a secreted protein with kinase activity. In the process of evolution, Fam20C not only obtained new substrate specificity, but also gained the ability to form evolutionarily conserved homodimers or heterodimers with Fam20A to regulate protein conformation and kinase activity. Although the other two members have a high degree of sequence homology, their biochemical activities are completely different. Fam20B has a unique active site that can recognize Galβ1-4Xylβ1 and is able to regulate the synthesis of proteoglycans by acting as a glycan kinase phosphorylating xylose residues and triggers peptidoglycan biosynthesis. At the same time, all Fam20B subfamily members function as monomers and do not need to form homologous or heterodimers. Fam20A, which did not appear until spinal animals, is a pseudokinase because it lacks active site residues that are essential for kinase activity and binds ATP in a catalytically incapable way, but it also obtains a more optimized ability to form dimers with Fam20C and activate Fam20C (Xiao et al., 2013; Zhang et al., 2018a; Worby et al., 2021).

Fam20C, originally named as Golgi-enriched-fraction casein kinase (GEF-CK) and discovered in 1972, is confirmed by Tagliabracci et al. and Ishikawa et al. as a protein kinase that can phosphorylate the S-x-E/pS (S is serine, x is any one amino acid, E/pS is glutamate or phosphoserine) motif of secreted proteins (Bingham et al., 1972; Tagliabracci et al., 2012). In addition, Fam20C can also recognize amino acid sequences other than the typical S-x-E/pS motif, such as S-x-Q-x-x-D-E-E motif, which indicates that Fam20C-dependent phosphorylation is more extensive than indicated by only those substrates that are phosphorylated within the canonical motif (Brunati et al., 2000). Moreover, FAM20C, expressed in a variety of tissues, has been shown by functional annotation of substrates that it not only is involved in biomineralization but also plays an important role in cell migration and adhesion, wound healing, endopeptidase inhibitor activity, and lipid metabolism disorders (Tagliabracci et al., 2015). Such rich and useful functions have established Fam20C as a major protein kinase. Here, we have summarized the structure and function of Fam20C, discussed its important role in various diseases, and further discussed its potential diagnosis and therapeutic effect. Through an in-depth understanding of these relationships, the biological functions of Fam20C will be further developed, laying a new foundation for studying the role of protein phosphorylation in life processes and diseases.

The Fam20C gene is located on chromosome 7, with a full length of 72,410 bp and contains 10 exons (Nalbant et al., 2005). The Fam20C protein sequence includes 584 amino acids, of which the C-terminal contains 350 amino acids that are highly conserved regions and is called conserved C-terminal domain (CCD), and the N-terminal has a signal peptide targeting the secretory pathway, a total of 22 amino acids. Its kinase domain (KD) spans 222 amino acids, from residue 354 to 565 (Cozza and Pinna, 2016). Fam20C is considered an atypical protein kinase and is therefore not included in the “typical” group of human kinases. The kinase core of Fam20C has a two-lobe structure (N-lobe and C-lobe), which is characteristic of all protein kinases. The crystal structure of the C. elegans homolog of human Fam20C (ceFam20) displays a large kinase domain and a protein kinase-like fold characterized by being contained in a shell-like structure formed by an N-terminal segment and a novel insertion domain. The N-terminal segment wraps around the lower half of the molecule, forming the base of the C-lobe. The insertion domain has a novel fold and forms a cap-like structure that covers the N-lobe. Therefore, sequence analyses usually failed to identify the Fam20 family as kinases because this domain is hidden in the N-lobe (Xiao et al., 2013). In addition, the Mn/ADP-bound structure is important for nucleotide binding and catalysis as the critical residues. Therefore, the unique architecture of the kinase suggests that Fam20C is an efficient catalyst as opposed to a dynamically regulated enzyme (Cozza and Pinna, 2016). Activation of Fam20C requires the formation of either an evolutionarily conserved homodimer or a heterodimer with Fam20A. Compared with Fam20C itself, Fam20A has a more effective Fam20C binding surface and is a special Fam20C allosteric activator. For example, Ile214A, Ile255A, and Leu365A are unique to Fam20A, and contribute to the formation of an optimized hydrophobic surface for interacting with Fam20C (Zhang et al., 2018a). In a word, the structure of Fam20C is distinct to other “typical” kinases, which makes it more druggable. Therefore, further studying of Fam20c structure can help us understand their biological functions, which can promote the development of more effective new drugs to treat diseases.

Fam20C is not a specific kinase dedicated to phosphorylation of casein, but a ubiquitous protein kinase, which mostly acts on the phosphorylation of many secreted proteins within the SxE/pS motif (Zhou et al., 2009; Salvi et al., 2010). Chromosome 4 harbors a cluster of genes encoding small integrin binding ligand-N-linked glycoproteins (SIBLINGs). The proteins encoded by these genes are involved in binding calcium and are the substrates of Fam20C, mainly five members of the secretory calcium binding phosphoprotein (SCPP) family, namely, dentin matrix protein 1 (DMP1), matrix extracellular phosphoglycoprotein (MEPE), osteopontin (OPN), bone sialoprotein (BSP), and dentin sialophosphoprotein (DSPP), indicating that Fam20C plays an important role in the process of biomineralization (Tagliabracci et al., 2012). Histidine-rich calcium binding protein (HRC) was the first sarcoplasmic reticulum (SR) protein to be discovered as a substrate of Fam20 (Pollak et al., 2017). In addition, other proteins involved in Ca²⁺ signaling include calsequestrin 2 (CSQ2) and matrix interacting molecule 1 (STIM1), as well as fibroblast growth factor 23 (FGF23), which directly causes cardiovascular problems in patients, and proprotein convertase subtilisin 9 (PCSK9), which is related to LDL-cholesterol disorders, which are all key links in the process of Fam20C affecting heart disease (Ezumba et al., 2014; Pollak et al., 2017; Pollak et al., 2018; Ben Djoudi Ouadda et al., 2019). Sortilin is not limited to the expression of a single tissue and is another unique target for Fam20C because of its versatility and its important role in the pathogenesis of a variety of diseases, especially neurovascular diseases. (Xu et al., 2019). Besides, in phosphorylated proteomics screening, the main proteins in plasma and serum such as vertebrate clotting factor fibrinogen, von Willebrand factor (VWF), and collagen were determined as potential substrates for Fam20C. Therefore, the role of Fam20C and its key substrate phosphorylation in the coagulation pathway is also worthy of further study (Tagliabracci et al., 2015; Qiu et al., 2018; Da et al., 2019). Studies have also found that the endoplasmic reticulum (ER) sulfhydryl oxidase ER oxidoreductin 1α (Ero1α), which can regulate the redox homeostasis of the ER, can also be phosphorylated by Fam20C, thereby establishing a new connection between phosphorylation modification and oxidative folding (Zhang et al., 2018b). Finally, many Fam20C phosphorylated substrates are also related to tumor cell apoptosis and migration, including insulin-like growth factor binding proteins (IGFBPs), OPN, and serine protease inhibitors (Rangaswami et al., 2006; Baxter, 2014; Tagliabracci et al., 2015). In short, Fam20C plays an important role in various life processes through phosphorylation of multiple substrates (Figure 1; Supplementary Table S1).

Given the important role of FAM20C in biomineralization and phosphorylating secretory proteins, mutations of FAM20C gene and aberrant function of fam20c kinase are responsible for many diseases, including Raine Syndrome (RS), cancer, and other diseases.

Considering the important role of Fam20C in various diseases, especially its role in tumorigenesis, targeting Fam20C is a potential treatment for diseases including cancer in the future (Wu et al., 2021). Some attempts to design inhibitors of Fam20C for treating cancer, focused in TNBC, have emerged. For the first time, FL-1607 was identified through virtual screening and interacted well with Fam20C, with better stability (RSMD reached its dynamic equilibrium after 2 ns simulation) and compactness (RG of 1.462 ± 0.0073 nm), as well as lower flexibility (RMSF of 0.0805 ± 0.0353 nm; mean ± SD) when binding with Fam20C. Meanwhile, it showed good antiproliferative property in human breast cancer MCF-7, MDA-MB-231 and MDA-MB-468 cells, with the best IC50 of 7.89 μM in MDA-MB-468 cells. Moreover, FL-1607 was shown to induce apoptosis and suppress migration of MDA-MB-468 cells. The researchers found that Fam20C harbored unique amino acids residues in ATP-binding sites, which makes Fam20C more possible to be selectively target by small molecules (Qin et al., 2016). Encouraged by this study and the x-ray crystal structure of Fam20C (Zhang et al., 2018a), a new inhibitor of Fam20C was identified in 2021 (Zhao et al., 2021). Optimized from F2078-0064, compound 3r showed good kinase inhibitory ability against Fam20C (6.243 μM), with IC50 of 5.986 μM in MDA-MB-231 cells. 3r killed cancer cells through apoptosis induction and could significantly block migration of MDA-MB-231 cells. In addition, 3r inhibited tumor growth in MDA-MB-231 xenograft mouse model through inducing apoptosis (Zhao et al., 2021) (Figure 2).

Two inhibitors of Fam20C mentioned above shed light on the therapeutic potential for targeting Fam20C in cancer, providing successful examples for future discovery of new Fam20C inhibitors. However, in-depth investigations of therapies targeting Fam20C remain limited, and no studies illuminated its therapeutic effects in heart diseases and diabetes, in which experimental evidence confirmed the important pathogenetic role of Fam20C. The two existing inhibitors may be good tools for studying biological and pathogenetic functions of Fam20C while their application for treating other diseases associated with Fam20C warrants further investigations. Additionally, the wide distribution of Fam20C in bone, heart, kidney, etc. may make therapeutics targeting Fam20C easily cause side effects in unexpected organs or tissues, thus requiring more comprehensive understanding of Fam20C functions in different tissues and calling for precise delivery of therapeutics. Moreover, molecular subtyping of diseases, especially of cancers with aberrant expression of Fam20C, represents a future direction.

Of note, the Fam20 family is composed of three secreted proteins, including Fam20A, Fam20B, and Fam20C; among them, Fam20C could phosphorylate a series of downstream secreted proteins. Fam20C is a typical member of Fam20 family, which has been well-known as a Golgi casein kinase that can be identified during hematopoietic differentiation (Nalbant et al., 2005). Interestingly, the other two members Fam20A and Fam20B have a high sequence homology with Fam20C, but their biological activities are completely different. Thus, Fam20C has its unique biological function in several cellular processes (Worby et al., 2021).

Recently, Fam20C, which is closely associated with RS, has been reported to phosphorylate more than 100 types of secreted proteins and more multiple substrates, thereby playing a key role in regulation of their complicated biological function (Tagliabracci et al., 2015). Accordingly, accumulating evidence has been demonstrating that Fam20C has several links to many types of human diseases, such as RS, cancer, and others. More importantly, Fam20C has also been found to enhance the metastasis of several types of human cancers, such as breast cancer and CRC, indicating that Fam20C may be a promising therapeutic target for the current drug development. More recently, some small-molecule inhibitors of Fam20C (e.g., FL-1607 and 3r) have been reported to induce apoptosis as well as to prevent metastasis in breast cancer, which may reveal Fam20C as a druggable target for cancer therapy (Qin et al., 2016; Zhao et al., 2021).

With the rapid development of structure biology and artificial intelligence (AI), the tertiary structure of Fam20C and even the quaternary structure of Fam20C-Fam20A have been elucidated (Zhang et al., 2018a). Therefore, the structure-guided design of Fam20C inhibitors will provide more precisely small-molecule inhibitors for potential therapy. Also, some AI technologies, such as Alphafold2 would also provide more potential selectively small-molecule inhibitors for possible pharmaceutical applications. Especially when the experimental structure is not available, Alphafold2 may provide more accurate model for virtual screening (Cramer, 2021). In a nutshell, these inspiring findings would shed new light on exploiting Fam20C as a potential therapeutic target and inhibiting Fam20C with small-molecule compounds will provide a new clue on discovery of more candidate drugs in the treatment of human diseases in the future.