Serine-Arginine Protein Kinase 1 (SRPK1) as a Prognostic Factor and Potential Therapeutic Target in Cancer: Current Evidence and Future Perspectives

Cancer, a heterogeneous disease composed of tumor cells and microenvironment, is driven by deregulated processes such as increased proliferation, invasion, metastasis, angiogenesis, and evasion of apoptosis. Alternative splicing, a mechanism led by splicing factors, is implicated in carcinogenesis by affecting any of the processes above. Accumulating evidence suggests that serine-arginine protein kinase 1 (SRPK1), an enzyme that phosphorylates splicing factors rich in serine/arginine domains, has a prognostic and potential predictive role in various cancers. Its upregulation is correlated with higher tumor staging, grading, and shorter survival. SRPK1 is also highly expressed in the premalignant changes of some cancers, showing a potential role in the early steps of carcinogenesis. Of interest, its downregulation in preclinical models has mostly been tumor-suppressive and affected diverse processes heterogeneously, depending on the oncogenic context. In addition, targeting SRPK1 has enhanced sensitivity to platinum-based chemotherapy in some cancers. Lastly, its aberrant function has been noted not only in cancer cells but also in the endothelial cells of the microenvironment. Although the aforementioned evidence seems promising, more studies are needed to reinforce the use of SRPK1 inhibitors in clinical trials.


Introduction
Cancer pathogenesis is driven by deregulated signaling pathways that result in uncontrolled proliferation, enhancement of angiogenesis, epithelial-mesenchymal transition (EMT), invasion, metastasis, and evasion of apoptosis. Of these hallmarks, metastasis to distant sites is the main cause of cancer death [1,2]. Each cancer is composed not only of tumor cells but also of diverse components that collectively form the tumor microenvironment. The endothelial cells and pericytes of blood vessels, cancer-associated fibroblasts (CAFs), various immune and inflammatory cells, and extracellular matrix (ECM) all interact with the tumor cells and influence cancer progression, survival, and response to therapy [3,4]. Notably, cancer is not a single disease yet appears heterogeneous among individuals (intertumor heterogeneity), also within its own tumor mass in each affected individual (intratumor heterogeneity) [5,6]. Apart from the tumor microenvironment, cancer stem cells (CSCs) are also a

SRPK1 as a Prognostic Factor
SRPK1 has been investigated in clinical material of various cancers, and its expression has been correlated with prognostic factors (e.g., staging, grading, and molecular subtypes) and survival. SRPK1 high expression is found in most human cancers; this is shown in the analysis of publicly available TCGA data accessed through the UALCAN website (http://ualcan.path.uab.edu/cgi-bin/Pan-cancer.pl? genenam=SRPK1) [27]. A summary of all published gene expression studies on human material that highlight the prognostic role of SRPK1 in various cancers is shown in Table 1. Table 1. Gene expression studies on human material that highlight the prognostic role of SRPK1 in various cancers. Cancers are listed according to their global estimated number of new cases (highest to lowest; GLOBOCAN 2018 estimates) [28].

Cancer type (Primary Location)
Approach Followed (Testing on Histology Samples/Blood; Data Mining) Level Tested (mRNA; Protein) Clinical Significance Reference Lung data mining; histology samples mRNA; protein ↑SRPK1 was found in NSCLC and was associated with higher tumor stage and shorter survival [29] histology samples mRNA; protein ↑SRPK1 was found in NSCLC [30] data mining mRNA ↑SRPK1 was found in lung adenocarcinoma (a NSCLC histologic type) [31] Breast histology samples mRNA; protein ↑SRPK1 was found in breast cancer and was associated with higher tumor stage and shorter survival [32] histology samples; data mining mRNA; protein ↑SRPK1 was associated with shorter survival in breast cancer and metastatic disease to the lungs and brain [33] data mining mRNA ↑SRPK1 was associated with shorter survival in HER2-negative breast cancer patients treated with chemotherapy [34] histology samples protein ↑SRPK1 was found in breast cancer [35] histology samples protein ↑SRPK1 was found in breast cancer [22] histology samples protein ↑SRPK1 was found in breast cancer and was associated with higher tumor grade [36] Prostate histology samples protein ↑SRPK1 was found in prostate cancer and its precursor (PIN) [26] histology samples protein ↑SRPK1 was found in prostate cancer and was associated with higher stage [37] Colorectal histology samples protein ↑SRPK1 was found in colon cancer and its precursor (adenoma) [22] histology samples protein ↑SRPK1 was found in colon cancer and was associated with higher tumor grade [36] histology samples mRNA; protein ↑SRPK1 was found in colorectal cancer and was associated with higher tumor stage and shorter survival [38] histology samples mRNA; protein ↑SRPK1 was found in colorectal cancer and was associated with higher tumor stage and shorter survival [39] histology samples mRNA ↑SRPK1 was found in colorectal cancer and its precursor (adenoma) [31] histology samples protein ↑SRPK1 and p-AKT were found in colon cancer [40] Stomach data mining mRNA ↑SRPK1 was found in gastric cancer [31] histology samples protein ↑SRPK1 was found in gastric cancer and was associated with higher tumor stage and shorter survival [41] histology samples protein ↑SRPK1 was found in gastric cancer [42] histology samples mRNA; protein ↑SRPK1 was found in gastric cancer and was associated with higher tumor stage and shorter survival [43]  Liver histology samples mRNA; protein ↑SRPK1 was associated with higher stage and shorter survival; SRPK1 expression was inversely correlated with miR-1296 expression [45] histology samples mRNA; protein ↑SRPK1 was found in liver cancer and was associated with higher stage and shorter survival [46] histology samples mRNA; protein ↑SRPK1 was found in liver cancer [47] Esophagus histology samples mRNA; protein ↑SRPK1 was found in esophageal cancer and was associated with higher grade, stage, and shorter survival [48] Pancreas histology samples protein ↑SRPK1 was found in pancreatic cancer and its precursor (dysplasia) [22] Leukemia blood mRNA; protein ↑SRPK1 was found in acute ATL [49] Kidney histology samples protein ↑SRPK1 was found in renal cell carcinoma [50] Glioma histology samples protein ↑SRPK1 was found in gliomas and was associated with lower rather than higher tumor grade [51] histology samples protein ↑SRPK1 was found in glioblastomas (Grade IV gliomas) and was associated with shorter survival [52] Ovary histology samples protein ↑SRPK1 was found in ovarian cancer [53] histology samples mRNA ↑SRPK1 was found in ovarian cancer [54] Testis histology samples protein ↓SRPK1 was associated with resistance to chemotherapy in NSGCTs [23] Eye histology samples protein ↓SRPK1 was found in advanced retinoblastomas, especially the ones that recurred or metastasized [55] SRPK1, serine-arginine protein kinase 1; NSCLC, non-small cell lung carcinoma; HER2, human epidermal growth factor receptor 2; PIN, pancreatic intraepithelial neoplasia; miR-1296, microRNA 1296; ATL, adult T-cell leukemia; p-AKT, phosphorylated-protein kinase B; NSGCTs, non-seminomatous germ cell tumors.
Except its reported overexpression in non-small cell lung cancer (NSCLC) compared to normal lung tissue [29][30][31], evidence shows that SRPK1 expression status could be a prognostic factor in this cancer type. Gong et al. performed SRPK1 immunohistochemistry (IHC) on histology samples of NSCLC patients and noted that SRPK1 overexpression was correlated with higher staging-a significant association with all components of the TNM System (T, N, and M) was found-and shorter survival at both univariate and multivariate levels. Thus, SRPK1 overexpression was found to be an independent poor prognostic factor in NSCLC [29]. Of interest, SRSF1 was also upregulated in NSCLC and its overexpression was associated with higher staging and increased metastatic potential [56]. Testing in human samples hence showed that the SRPK1/SRSF1 axis is implicated in NSCLC progression and survival.
Similar to NSCLC, SRPK1 overexpression was also found in breast cancer clinical samples compared to normal breast tissue [22,32,35,36] and was correlated with higher grading [36], staging-once more, a significant association with all components of the TNM System (T, N, and M) was noted-and shorter survival at both univariate and multivariate levels [32]. SRRPK1 upregulation seemed to favor a specific distant metastatic pattern to the lungs and brain, rather than bones and liver [33]. Furthermore, SRPK1 expression was investigated in relation to the breast cancer molecular subtypes and resistance to chemotherapy. Van Roosmalen et al. reported that SRPK1 overexpression in ER-positive breast cancer patients correlated with shorter metastasis-free survival and higher proliferation status, as shown with Ki-67 IHC. When they sub-categorized luminal breast cancer patients into luminal A and B, they found a significant association between SRPK1 overexpression and the more aggressive of the two luminal B subtype [33]. In another study, Cheng et al. performed a data-mining study using the GEO database and showed that SRPK1 is overexpressed in highly aggressive TNBCs. In addition, they linked SRPK1 overexpression with a shorter distance-relapse free survival in HER2-negative patients treated with chemotherapy, suggesting SRPK1 might have a role in the development of resistance to therapy [34]. SRPK1 upregulation has also been described in other common epithelial cancers and has been correlated with various prognostic factors and survival. In prostate cancer, SRPK1 overexpression was noted in both cancer and its precursor, PIN (prostatic intraepithelial neoplasia), and was associated with a higher tumor stage [26,37]. Similarly, in colorectal cancer, SRPK1 was overexpressed in both cancer and its precursor, adenoma, compared to the normal colonic tissue [22,31,36,[38][39][40]; its high expression was also linked with higher grade and stage colorectal cancers, presence of lymph node metastases, in addition to shorter overall and recurrence-free survival [31,36,38,39]. In stomach cancer, high SRPK1 was also found compared to the normal stomach histology [31,[41][42][43][44] and was associated with higher grade, stage, larger tumor size, lymph node metastases, and shorter overall survival [41,43,44]. Liver cancer was also characterized by SRPK1 overexpression, while the latter was correlated with higher stage, larger tumor size, and shorter survival [45][46][47]. In esophageal cancer, high expression of SRPK1 was associated with higher grade, stage, the capacity to metastasize, and shorter overall survival at both univariate and multivariate levels [48].
In contrast to epithelial malignancies, the results of SRPK1 expression seemed less predictable in non-epithelial cancers. For instance, while SRPK1 was highly expressed in gliomas, this was correlated with low rather than high-grade gliomas [51]. However, SRPK1 overexpression was linked with high proliferative activity (as measured with Ki67 IHC) and shorter survival in glioblastomas (WHO grade IV gliomas) [52]. In patients with NSGCTs (non-seminomatous germ cell tumors) of the testis, low SRPK1 expression was associated with resistance to chemotherapy, whereas high SRPK1 expression characterized the tumors that responded to therapy [23]. Lastly, patients with advanced retinoblastomas, especially the ones that recurred or metastasized, were found to have low rather than high expressions of SRPK1 [55].

SRPK1 as a Potential Therapeutic Target in Cancer
Up to date, no clinical trials targeting SRPK1 as a cancer treatment have been performed or are currently running (https://clinicaltrials.gov/). However, several preclinical studies on cell lines and animal models have tested the effect of aberrant SRPK1 expression on distinct oncogenic processes highlighting the potential role of SRPK1 as a therapeutic target in various cancers. A summary of these studies is shown in Table 2. Of interest, existing evidence supports that SRPK1 acts heterogeneously among various cancer types. Different regulatory pathways are disrupted hence distinct oncogenic processes (e.g., proliferation, angiogenesis, apoptosis) are primarily influenced depending on the cellular context [57]. Targeting SRPK1 in cancer could simultaneously affect different diverse processes in every cancer type, as shown in Figure 1, making it a potentially attractive treatment strategy. This chapter mainly focuses on the processes affected in each cancer when SRPK1 is downregulated (e.g., by inhibitors or gene silencing), while it briefly describes the effects of SRPK1 upregulation and the main pathways involved. Table 2. Published preclinical studies (cell lines/in vitro; animal models/in vivo) that summarize the effect of aberrant SRPK1 expression on distinct oncogenic processes (e.g., proliferation, angiogenesis, apoptosis) via a plethora of mechanisms/pathways, highlighting the potential role of SRPK1 as a therapeutic target in various malignancies. Cancers are listed according to their global estimated number of new cases (highest to lowest; GLOBOCAN 2018 estimates) [28].

SRPK1-Mediated Mechanism(s)/ Pathway(s) Involved Reference
Lung ↑SRPK1 promoted cell growth, invasion, CSCs aggregation, and the expression of various stem cell markers in vitro; it also promoted tumor growth and tumorigenicity in vivo ↓SRPK1 suppressed cell growth and CSCs in vitro; it also suppressed tumor growth and tumorigenicity in vivo SRPK1 interacts with the Wnt/β-catenin pathway [29] ↑SRPK1 promoted NSCLC cell growth and migration in vitro ↓SRPK1 suppressed cell growth and migration in vitro, also tumorigenicity in vivo SRPK1 interacts with the β-catenin/TCF pathway [30] ↑SRPK1 was found in NSCLC endothelial cell lines and promoted cell growth, migration, and invasion in vitro ↓SRPK1 suppressed cell growth, migration and invasion and induced apoptosis in vitro; it also suppressed tumor growth and metastasis and prolonged survival in vivo SRPK1 interacts with the β-catenin/TCF pathway and the GSK3-β [58] ↑SRPK1 and proangiogenic VEGF were found in tumor endothelial cells in vivo ↓SRPK1 induced the expression of antiangiogenic VEGF in tumor endothelial cells and suppressed angiogenesis in vivo The SRPK1/SRSF1 axis interacts with WT-1 and regulates VEGF alternative splicing (proangiogenic vs. antiangiogenic isoforms) [59] Breast ↑SRPK1 was found in BLBC cell lines ↓SRPK1 suppressed the migration of BLBC cells in vitro and breast cancer metastasis to the lungs in vivo SRPK1 interacts with the NF-κB pathway [33] ↑SRPK1 was found in hormone-positive and triple-negative breast cancer cell lines and promoted the localization of RBM4 in the cytoplasm; this enhanced the expression of the antiapoptotic IR-A and MCL-1L isoforms in hormone-positive breast cancer cell lines ↓SRPK1 in hormone-positive breast cancer cells promoted apoptosis by reducing the phosphorylation of RBM4 and restoring its localization in the nucleus; this enhanced the expression of the proapoptotic IR-B and MCL-1s isoforms SRPK1 regulates the alternative splicing of IR and MCL-1 by modulating the localization of RBM4 inside the cell (cytoplasm vs. nucleus) [35] ↑SRPK1 was found in hormone-positive and triple-negative breast cancer cell lines ↓SRPK1 induced apoptosis and sensitivity to chemotherapy in hormone-positive breast cancer cells SRPK1 interacts with the AKT and MAPK pathways and modulates MAP2K2 alternative splicing [36] ↑SRPK1 was found in breast cancer cell lines [32] Prostate ↑SRPK1 was found in prostate cancer cell lines ↓SRPK1 suppressed angiogenesis (as shown by the reduced MVD) and thus tumor growth in vivo by inducing the expression of the antiangiogenic VEGF-165b The SRPK1/SRSF1 axis regulates VEGF-A alternative splicing (proangiogenic vs. antiangiogenic VEGF isoforms) [26] Colorectal ↑SRPK1 was found in colorectal cancer cell lines ↓SRPK1 induced apoptosis and sensitivity to chemotherapy in vitro SRPK1 interacts with the AKT and MAPK pathways and modulates MAP2K2 alternative splicing [36] ↑SRPK1 was found in colorectal cancer cell lines and induced proliferation in vitro ↓SRPK1 suppressed growth, angiogenesis, migration and induced apoptosis in vitro [38] ↑SRPK1 was found in colorectal cancer cell lines ↓SRPK1 suppressed proliferation, migration and invasion in vitro SRPK1 interacts with the miR-216b [39]  ↓SRPK1 suppressed tumor growth and apoptosis in vivo, also angiogenesis in vitro and in vivo The plexin-B1/SRPK1 pathway regulates cell motility, apoptosis, and angiogenesis [69] ↑SRPK1 was found in glioma cell lines ↓SRPK1 suppressed cell viability and response to chemotherapy in vitro [52] ↑SRPK1 was found in glioma cell lines  In NSCLC, a main oncogenic process affected by SRPK1 was reported to be the acquisition of a CSC (cancer stem cell) phenotype. SRPK1 upregulation promoted cell growth, CSCs accumulation, and the expression of various stem cell markers, while its downregulation suppressed cell growth, CSCs, and tumorigenicity [29]. Gong et al. also found that SRPK1 regulates CSCs in NSCLC by interacting with the Wnt/β-catenin pathway [29]. Other teams reported that SRPK1 interacts with the β-catenin/TCF (T-cell factor) pathway, also the GSK3-β (glycogen synthase kinase 3-β), and promotes cell growth, migration and invasion, while its downregulation suppresses metastasis and induces apoptosis [30,58]. Except for NSCLC cells, SRPK1 acts on the endothelial cells of the tumor microenvironment. WT-1 (Wilms tumor-1) interacts with the SRPK1/SRFS1 axis and regulates the alternative splicing of VEGF (vascular endothelial growth factor) [59]. VEGF alternative splicing can result either in proangiogenic or antiangiogenic isoforms, regulating angiogenesis in cancer [19]. Wagner et al. noted that, whereas SRPK1 overexpression induced the expression of proangiogenic VEGF164a isoform, its reduced expression resulted in a shift towards the expression of the antiangiogenic VEGF120, suppressing angiogenesis in vivo [59]. To summarize, SRPK1 downregulation affects both tumor epithelial cells and endothelial cells of the microenvironment, whilst it suppresses the acquisition of CSC phenotype, migration, invasion, metastasis, and angiogenesis and promotes apoptosis in preclinical NSCLC models.
In breast cancer, SRPK1-induced alternative splicing plays a crucial role in oncogenesis as well. SRPK1 regulates apoptosis by interacting with RBM4 (RNA-binding motif protein 4) and modulating the alternative splicing of IR (insulin receptor) and MCL-1 (myeloid cell leukemia-1) [35]. SRPK1 downregulation enhanced apoptosis in hormone-positive breast cancer models by reducing RBM4 phosphorylation; the latter enhanced the expression of the proapoptotic IR-B and MCL-1s rather than the antiapoptotic IR-A and MCL-1L isoforms [35]. Similarly, SRPK1 downregulation also induced hormone-positive breast cancer sensitivity to cisplatin [36]. In contrast, a different SRPK1-driven oncogenic mechanism was noted in BLBC. Van Roosmalen et al. reported that reduced SRPK1 expression suppressed the migration of BLBC cells in vitro and breast cancer metastasis to the lungs in vivo through interacting with the NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) pathway [33]. All in all, SRPK1 downregulation enhances apoptosis and sensitivity to chemotherapy in hormone positive, while it suppresses migration and metastasis in BLBC preclinical models.  on distinct oncogenic processes, highlighting its potential role as a therapeutic target in various malignancies. Fourteen different cancer types are shown. Arrows and inhibition symbols represent promotion or suppression, respectively, of the pointed processes after SRPK1 downregulation in each one of the listed cancer types. Red color means the specific interaction between SRPK1 and the pointed process is discussed in the literature. Gray color could mean that the interaction is not described, described but not activated (e.g., ↓SRPK1 doesn't involve proliferation, migration, and invasion in prostate cancer or proliferation in melanoma) or described with an opposite result in the literature (e.g., ↓SRPK1 enhances resistance, rather than sensitivity to chemotherapy in gliomas).
Likewise, SRPK1-driven alternative splicing is also vital in prostate carcinogenesis and particularly affects VEGF-induced angiogenesis. SRPK1 reduced expression-induced either through gene knockdown or the SRPK1 inhibitor SPHINX-suppressed angiogenesis thus tumor growth in vivo through a switch in VEGF alternative splicing. In particular, expression of the antiangiogenic VEGF165b splicing isoform was favored over the proangiogenic VEGF165. Of interest, SRPK1 downregulation did not alter other key oncogenic processes in vitro such as proliferation, migration, and invasion [26]. To summarize, targeting SRPK1 could treat prostate cancer by suppressing the VEGF-induced angiogenesis.
Several oncogenic pathways and processes were reported to be deregulated after targeting SRPK1 in colorectal cancer. First, SRPK1 downregulation suppressed proliferation, migration, and invasion by interacting with the microRNA miR-216b or the long noncoding RNA MALAT1 (metastasis-associated lung adenocarcinoma transcript 1) [39,63]. Second, like prostate cancer, it also suppressed angiogenesis through a switch in VEGF alternative splicing resulting in higher expression of the VEGF165b isoform [62]. Third, it enhanced apoptosis and resistance to chemotherapy [36,38,61]. Of interest, deregulated alternative splicing was also found to play a role in colorectal carcinogenesis. In the case of the GTPase Rac1 (Ras-related C3 botulinum toxin substrate 1), it resulted in the expression of the isoform Rac1b which sustains tumor survival; SRPK1 downregulation was tumor suppressive through downregulating the expression of Rac1b [60]. Likewise, deregulated alternative splicing of SLC39A14 (zinc transporter ZIP14) resulted in low 4A/4B exon ratio both in precursor lesion (adenoma) and colorectal cancer, when it was high in normal colonic tissue. SRPK1 inhibition increased the 4A/4B exon ratio in colorectal cancer cell lines [31].
Concerning other epithelial cancer types, SRPK1 downregulation suppressed migration and invasion in skin basal cell carcinoma [64]. It also suppressed proliferation, migration, and invasion in stomach cancer preclinical models through various mechanisms, including interactions with the AKT, ERK (extracellular signal regulated kinase) and small nucleolar RNA-mediated pathways, also the microRNA miR-126 [41][42][43][44]. In liver cancer, it suppressed proliferation, migration, and invasion [45,47]. In esophageal cancer, it suppressed proliferation, migration, invasion and enhanced apoptosis [48], while in pancreatic cancer it suppressed proliferation and also enhanced apoptosis and sensitivity to chemotherapy [22]. In kidney cancer, SRPK1 low expression suppressed proliferation, migration, and invasion [50]. Lastly, targeting SRPK1 suppressed proliferation, migration, and invasion whilst it enhanced apoptosis and sensitivity to chemotherapy in ovarian cancer preclinical models [53,54]. Of interest, an older study had suggested that SRPK1 inhibition enhances resistance, rather than sensitivity, to chemotherapy [71].
Similar to prostate and colorectal cancer, the SRPK1/SRSF1-driven VEGF alternative splicing also plays a key role in the regulation of angiogenesis in melanoma. Gammons et al. noted that SRPK1 downregulation (achieved either through gene knockdown or the SRPK1 inhibitor SPRIN340) suppressed angiogenesis by decreasing the expression of proangiogenic VEGF165; of interest, it did not have a significant effect on proliferation [72]. Apart from melanoma tumor cells, VEGF alternative splicing is critical in regulating the oncogenic role of tumor-associated endothelial cells, just like in NSCLC. Low expression of SRPK1 resulted in a shift towards the expression of the antiangiogenic VEGF120, suppressing angiogenesis in vivo [59].
In AML (acute myeloid leukemia), SRPK1 low expression suppressed proliferation while it enhanced cell cycle arrest, apoptosis and survival. SRPK1-mediated aberrant alternative splicing was implicated in the pathogenesis of AML and its inhibition switched BRD4 (bromodomain-containing protein 4) splicing, favoring the production of its long isoform. Of interest, SRPK1 pharmacologic inhibition with SPHINX31 did not impact normal hematopoiesis in the long term [65]. In other studies, SRPK1 downregulation suppressed proliferation and enhanced apoptosis in various leukemia types [66][67][68]. Siqueira et al. also noted that the effect on apoptosis took place in a synergistic fashion with chemotherapy [67].
The most conflicting results from SRPK1 downregulation were derived from glioma studies. Although it suppressed proliferation, migration, invasion, angiogenesis, and cell cycle progression, SRPK1 downregulation also suppressed apoptosis and at the same time induced resistance to chemotherapy in preclinical glioma studies [51,52,69,70]. Therefore, SRPK1 inhibition would require more caution if applied in brain cancers, as it could result in both tumor-suppressive and -promotive effects.

Discussion and Future Perspectives
SRPK1 is upregulated in the majority of human cancers, where its overexpression was associated with higher TNM staging, grading, and shorter survival in various studies that tested clinical samples (Table 1). Therefore, SRPK1 has the potential to be used as a prognostic biomarker in common malignancies (Table 1) [28], especially the epithelial cancers (e.g., lung, breast, prostate, colorectal, stomach, liver, and esophageal). Apart from cancers themselves, it was also overexpressed in the precursor lesions of prostatic, colorectal, and pancreatic cancers; this suggests that SRPK1 could be involved in early oncogenic steps of these epithelial malignancies [22,26,31]. There were only rare studies that challenged the aforementioned link between SRPK1 expression and cancer promotion; characteristically, these studies investigated non-epithelial cancers. Wu et al. correlated high SRPK1 expression with lower grading in gliomas [51], while Krishnakumar et al. correlated low SRPK1 with higher staging in retinoblastomas, especially the ones that recurred or metastasized [55]. In conclusion, current evidence suggests a prognostic role of SRPK1 in cancer.
Contrary to SRPK1, the other two kinases of the SRPK family, SRPK2 and SRPK3, are much less studied. In normal histology, both SRPK2 and SRPK3 show a tissue-specific expression pattern and, whereas the former is overexpressed in the brain [74], the latter is highly expressed in the skeletal and cardiac muscles [75,76]. Concerning cancer tissues, SRPK2 was upregulated in colon cancer [77], prostate cancer [78], NSCLC [56,79], pancreatic cancer [80], and acute myeloid leukemia [81]. In addition, SRPK2 overexpression was associated with poor survival in NSCLC and pancreatic cancer [56,79,80], also with grading and staging in prostate cancer [78]. In contrast, SRPK3 was downregulated in rhabdomyosarcoma, a mesenchymal pediatric cancer [82].
Although not yet tested in human trials, evidence from preclinical studies supports a potential role of SRPK1 in cancer treatment. Indeed, SRPK1 presents as an attractive therapeutic target. First, SRPK1 is implicated in the oncogenesis of multiple cancers and has been targeted with promising results, at the preclinical level, in most of them (Table 2, Figure 1). Second, targeting SRPK1 could simultaneously affect diverse oncogenic processes-proliferation, migration, invasion, metastasis, angiogenesis, apoptosis, acquisition of CTC phenotype, and sensitivity to chemotherapy-through various mechanisms depending on the cancer type (Table 2, Figure 1). For instance, deregulation of alternative splicing was reported to impact angiogenesis in lung [59], prostate [26], colorectal cancer [62] and melanoma [72], also apoptosis in breast [35,36], colorectal cancer [36], and leukemia [65,66] (Figure 2). Of interest, apart from cancer cells, SRPK1 inhibition influences tumor microenvironment and more specifically the tumor-associated endothelial cells, suppressing angiogenesis in vivo [59]. In this context, the SRPK1/SRSF1 axis is regulated by WT-1 (Wilms Tumor-1) [59]. WT1 is upregulated in the endothelium of diverse tumors such as lung, breast, pancreatic, ovarian, bladder cancers, and glioblastoma [59,83], while its endothelial-specific knockout suppresses tumor growth and metastasis [84]. As, in contrast to the endothelium, WT-1 suppresses SRPK1 expression in human podocytes, unidentified co-factors of WT1 might be involved in the regulation of the SRPK1/SRSF1 axis [62].
Targeting SRPK1 effect on cancer appears heterogeneous and depends on the oncogenic context. For instance, SRPK1 inhibition suppresses angiogenesis, metastasis, and the acquisition of a CSC phenotype, and also enhances apoptosis in lung cancer [29,30,58,59]. In contrast, it primarily suppresses angiogenesis albeit causes no significant alteration in proliferation, migration, and invasion in prostate cancer [26] (Figure 1). Molecular subtype also seems to play a critical role; SRPK1 inhibition suppresses metastasis in BLBCs [33], while enhancing apoptosis and sensitivity to platinum-based chemotherapy in hormone-positive breast cancers [36]. Various inhibitors have been used in preclinical studies. SRPIN340, a selective SRPK1 and SRPK2 inhibitor, induced the expression of the antiangiogenic isoform VEGF-165b and suppressed angiogenesis in melanoma and colorectal cancer models [62,72]. SRPIN340 also exhibited a cytotoxic effect, whilst it reduced the expression of the proangiogenic VEGF isoform and enhanced apoptosis in various leukemia cell lines [66]. SPHINX, an inhibitor with preference for SRPK1 [85], suppressed angiogenesis and inhibited tumor growth when it was administered in prostate cancer animal models [26]. SPHINX31, an inhibitor highly selective for SRPK1, enhanced cell cycle arrest and apoptosis in AML [65]. Notably, Hatcher et al. developed the first irreversible SRPK inhibitor; SRPKIN-1, which is selective for both SRPK1 and SRPK2, induced the expression of the antiangiogenic isoform VEGF-165b much more efficiently than SRPIN340 [86].
Resistance to chemotherapy, through diverse mechanisms, is a main cause of treatment failure and death in cancer patients [87,88]. Targeting SRPK1 in preclinical models was linked with sensitivity to platinum-based chemotherapy in breast [36], colorectal [36], pancreatic [22], and ovarian cancer [53]. One study by Schenk et al. noted that SRPK1 inhibition enhanced chemoresistance rather than sensitivity in ovarian cancer [71], contradicting with the previously mentioned ones. Of interest, Schenk et al. used the A2780 ovarian cancer cell line for their experiments [71], which was later found not to represent the most common histology type of ovarian cancer, the high-grade serous carcinoma [89,90]. In contrast, the cell line OVCAR3 used by Odunci et al. [53] has been characterized as a TP53-mutant, a high-grade serous carcinoma cell line [91]. In accordance with their ovarian cancer study, Schenk et al. noted that low SRPK1 expression was associated with resistance to chemotherapy in NSGCTs [23]. SRPK1 inhibition was also reported to promote resistance to chemotherapy in gliomas [51,52]. In summary, current evidence points that SRPK1 inhibition promotes sensitivity to chemotherapy in epithelial cancers such as breast, colorectal, pancreatic, and ovarian, while it promotes resistance in gliomas and NSGCTs.
Even if results from preclinical studies are encouraging, SRPK1 has not yet reached clinical trials. Our opinion is that future research on the role of SRPK1 in cancer should first focus more on personalized medicine. More studies that investigate the distinct molecular subtypes of each cancer type (e.g., breast cancer: hormone positive, HER2-enriched, and BLBCs) rather than regard each cancer as single entity, should be performed. In BLBCs, in contrast to hormone and HER2 positive breast cancers, targeted therapies are still ineffective and toxic chemotherapy is the only viable option [16,92]. Van Roosmalen et al. directed their study to the highly aggressive BLBCs [33], setting an example for future research in this field. In addition, although studies that show that SRPK1 downregulation restores sensitivity to platinum-based chemotherapy have been performed [22,36,53], reports on resistance to targeted therapies (e.g., hormone and anti-HER2 therapies in breast cancer) or immunotherapy are still lacking. Cancer stem cells are associated with cancer recurrence, metastasis, and resistance to chemotherapy [7] yet no relevant study, except the one presented by Gong et al. [29], has been performed.
A deeper understanding of the SRPK1 role in oncogenesis and tumor evolution is still needed. For instance, SRPK1 is overexpressed in colorectal, pancreatic, and prostatic precursor lesions [22,26,31]. However, what happens with the precursor lesions of other cancer types? Likewise, we still lack the knowledge about SRPK1 expression status or its interplay with already established oncogenic pathways in metastatic cancers compared to their primary counterparts. Wagner et al. highlighted the link between SRPK1 and tumor associated endothelial cells in lung cancer and melanoma in vivo models [59]. Nevertheless, how does SRPK1 act upon other critical elements of tumor microenvironment such as CAFs or immune cells? Future preclinical studies could also use better models that recapitulate more accurately specific cancer types or their molecular subtypes. PDX (patient-derived xenograft) models could be a promising approach that we have not yet seen in SRPK1-related cancer research except in the study of Tzelepis et al. [65]. In addition, interacting networks affected by SRPK1 downregulation could be better studied using transcriptomics or proteomics accompanied by big data analysis. Lastly, more evidence concerning the side effects of SRPK1 targeting is needed before its potential implementation into clinical trials. Indeed, studies like the one performed by Tzelepis et al., which showed that SRPK1 inhibition does not significantly affect normal hematopoiesis in the long term [65], offer promise.
In conclusion, the presented evidence shows that SRPK1 has a prognostic role in various cancers, while it could pose as an attractive target for cancer therapy. More effective preclinical models (e.g., PDXs) could give a more personalized direction in SRPK1-related cancer research. The goal would be to assess its use as a potential cancer treatment target in specific scenarios, such as treating specific molecular subtypes of diverse cancers or therapy-resistant malignancies.