The Role of ARF Family Proteins and Their Regulators and Effectors in Cancer Progression: A Therapeutic Perspective.

In some cases, the expression of membrane traffic regulators, namely ARF family proteins or their effectors, GEFs or GAPs is found downregulated in tumor cells ( Tables 1 , 2 ). Besides its behavior as an oncogene in breast cancer, ARF3 has been found downregulated in gastric cancer ( Chang et al., 2009 ). In fact, ARF3 expression is significantly decreased in gastric cancer stages I-III, when compared with paired normal gastric mucosa tissues, indicating that this protein could be a marker for gastric cancers without metastasis. The clinical significance of these results remains to be elucidated. SMAP1 or ARFGAP1 is a member of the ARF GAP family that is involved in clathrin-dependent endocytosis of the Transferrin receptor and E-cadherin ( Kon et al., 2008 ; Kobayashi et al., 2014 ). In colorectal cancers with microsatellite instability, short deletions or insertions frequently occur in SMAP1 , generating a premature termination codon. This results in reduced or abolished SMAP1 protein levels in colorectal tumors ( Kon et al., 2014 ). Unlike other phosphotyrosine proteins that are usually overexpressed or hyperphosphorylated in gastric tumor cells, the ARF GAP ARAP3 is downregulated in gastric cancer tissues ( Yagi et al., 2011 ). Furthermore, GIT2 stabilizes FAs by reducing Rac1 activity in the breast cancer cell line MDA-MB-231 ( Frank et al., 2017 ). Also, in a gene expression profile analysis of breast cancer patient samples, GIT2 was found downregulated in a group of lymph node-positive breast cancer patients ( Sirirattanakul et al., 2015 ). Finally, EFA6 GEFs are downregulated in breast, brain and ovarian cancers ( Pils et al., 2005 ; Van Den Boom et al., 2006 ; Zangari et al., 2014 ). ARL2 has been shown to directly influence α/β-tubulin polymerization in the breast cancer cell line MCF-7 ( Beghin et al., 2007 ). Moreover, MCF-7 cells expressing higher levels of ARL2 are more sensitive to cytotoxic agents, while cells with reduced expression of ARL2 show enhanced resistance to the same agents ( Beghin et al., 2008 ). This resistance is mediated by Protein Phosphatase 2A (PP2A), whose activity is regulated by ARL2. When ARL2 is decreased, impaired dephosphorylation of p53 by PP2A occurs, leading to an increase of phosphorylated p53, which alters PP2A localization and causes a chemo-resistant phenotype ( Beghin et al., 2008 ). Moreover, in vitro assays using breast cancer cells depleted for ARL2 show less contact inhibition, an enhanced clonogenic potential and increased proliferation than control cells ( Beghin et al., 2009 ). Furthermore, using orthotopic mouse models, depletion of ARL2 was shown to impair cancer progression ( Beghin et al., 2009 ). Additionally, ARL2 downregulation was recently correlated with more aggressive cases of glioma and a lower survival of the patients ( Wang et al., 2018 ). Finally, ARL2 overexpression inhibits proliferation, as well as migration and tumorigenicity of glioma cells, through regulation of the receptor tyrosine kinase AXL, a known regulator of glioma tumorigenesis ( Wang et al., 2018 ). ARL3 mRNA and protein expression were shown to be downregulated in gliomas ( Wang et al., 2019b ). Furthermore, an extensive bioinformatics analysis suggested that ARL3 plays a role in angiogenesis and immune cell infiltration in the tumor microenvironment ( Wang et al., 2019b ). ARL4C was associated with reduced metastatic potential of ovarian cancer cells, in which it inhibits cell motility but not cell proliferation ( Su et al., 2015 ). Furthermore, ARL4C mRNA expression is lower in ovarian cancer samples of patients with a poor treatment response, while patients with higher ARL4C expression show increased overall survival ( Su et al., 2015 ). Finally, SAR1B was identified as a potential metastatic suppressor in colorectal cancer, through a targeted proteomic approach ( Huang and Wang, 2019 ). Also, migration and invasion assays showed that SAR1B silencing leads to an increase in colorectal cancer cell motility and invasive capacity ( Huang and Wang, 2019 ). ARL11 , also known as ADP Ribosylation factor-Like Tumor Suppressor gene 1 ( ARLTS1 ) was described as a potential low-penetrance tumor suppressor gene in different types of cancers, such as breast cancer, melanoma and chronic lymphocytic leukemia ( Calin et al., 2005 ). Different variants of ARLTS1 have been associated with familial and sporadic cancers, where the mutations Trp149Stop and Cys148Arg are the most studied ( Yendamuri et al., 2008 ). The nonsense mutation Trp149Stop leads to the production of a truncated protein unable to bind GTP, which results in decreased apoptotic potential of the cell ( Petrocca et al., 2006 ). Both variants were found to be associated with predisposition to familial breast cancer and, more recently to familial hematological malignancies ( Calin et al., 2005 ; Frank et al., 2006b ; Hamadou et al., 2017 ). Additionally, the Cys148Arg variant was associated with melanoma and both familial and sporadic colorectal cancers ( Frank et al., 2006a , c ; Castellví-Bel et al., 2007 ). Furthermore, ARLTS1 expression was found to be decreased in different types of tumors, including ovarian, lung and prostate cancer, as well as chronic lymphocytic leukemia ( Yendamuri et al., 2008 ; Siltanen et al., 2013 ). More recently, a study in ovarian cancer suggested that ARLST1 increases tumor cell sensitivity to chemotherapeutic agents by regulating apoptosis ( Yang et al., 2011 ).

All authors conceived, wrote, reviewed, and edited the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Members of the RAS superfamily of small GTPases are master regulators of all the steps involved in membrane traffic. Thus, it is not surprising that many of them are hijacked by cancer cells to enhance their capacity to form a tumor and spread to other organs. In particular, ARF family proteins, their GEFs, GAPs and effectors are often upregulated in expression and/or activity in several types of cancer. Moreover, upregulated expression/activity can be linked to enhanced cancer progression and aggressiveness. Therefore, these proteins are good candidates to serve as therapeutic targets and, indeed several strategies have already been proposed and tested. These include the targeting of ARF proteins themselves or their GEFs, GAPs or effectors. While our knowledge of the GEFs, GAPs and effectors of ARFs is fairly complete, much less is known about the functions and identity of GEFs, GAPs and effectors of ARL subfamily members. Hence, the knowledge about these molecular players should be developed in order to find new therapeutic strategies for cancer types where ARLs or their regulators/effectors are subverted. Since most ARF family proteins are ubiquitous and required for essential cellular functions, the targeting of specific effectors and GEFs/GAPs could ensure tissue/function specificity. Nevertheless, specificity could also be achieved through targeted delivery of vectors/drugs. In conclusion, the study of the mechanisms subverted by cancer cells involving ARF family proteins and their regulators of activity and effectors can shed light on the functions of these proteins and simultaneously provide clues about new therapeutic targets and strategies, which continue to be a pressing need in the cancer field.

As can be concluded from Tables 1 , 2 , several ARFs and ARF GEFs and GAPs are overexpressed in different types of cancers. Therefore, therapeutic strategies aiming to inhibit the expression of these proteins can be proposed. Other approaches like the use of small GTPase inhibitors that impair GTP binding or the binding to membranes, the blockade of GEF activity or ARF-GEF interaction, should also be considered. Furthermore, the stimulation of GAP activity/expression and the inhibition of the interaction with downstream effectors or the function/expression of these effectors can also be envisaged ( Figure 1 ). For instance, the inhibitor LM11 can abolish specifically ARF1 activation through the blockade of the binding of the ARF GEF Cytohesin 2/ARNO ( Flisiak et al., 2008 ; Xie et al., 2016 ). Indeed, it has been shown that the aggressiveness of breast tumors that overexpress ARF1 is reduced after treatment with this inhibitor through the decrease in cell invasion and proliferation and increased apoptosis ( Schlienger et al., 2015 ; Xie et al., 2016 ). Also, the small inhibitor EXO2 reduces ARF1 activation and effectively impairs the proliferation of prostate cancer cells by blocking ERK1/2 activation ( Lang et al., 2017 ). Moreover, EXO2 inhibits invasion of prostate cancer cells and induces their apoptosis. Furthermore, the same study shows that the simultaneous blockade of ARF1 and RAS activation in prostate cancer is a potential targeted strategy to prevent the development of this type of tumor ( Lang et al., 2017 ). Putative therapeutic strategies to target ARF proteins, GEFs, GAPs and effectors. (A) In the case of ARF family members that can act as oncogenes, their expression or activity could be downregulated (1); GEF activity or expression downregulated (2); ARF-GEF binding blocked or nucleotide binding blocked (3); active ARF binding to membranes (4) or effectors blocked (5); effector function impaired (6); GAP expression or activity upregulated (7). (B) Regarding ARF proteins that can act as tumor suppressors, their expression or activity could be upregulated (1); GEF activity or expression upregulated (2); GAP activity or expression downregulated (3). Since some ARF proteins like ARF6 and ARF1, are ubiquitously expressed and perform essential functions in all cell types ( D’Souza-Schorey and Chavrier, 2006 ), targeting the proteins themselves could have dramatic and unwanted consequences. In alternative, targeting their regulators, such as ARF GEFs or GAPs, might represent a viable strategy for the development of specific anti-cancer therapies. Regarding the targeting of ARF GEFs, it has been shown that SecinH3, an ARF GEF inhibitor that impairs both ARF1 and ARF6-dependent signaling, is effective in decreasing the growth of breast cancer xenografts and reducing lung metastasis ( Zhao et al., 2016 ), while suppressing angiogenesis of melanoma and lung carcinoma tumors ( Grossmann et al., 2014 ; Hongu et al., 2015 ). Thus, inhibitors of the ARF6-dependent signaling pathway could be useful to control specifically tumor invasion and angiogenesis. It has been observed that several ARF GAPs are overexpressed in cancer ( Table 2 ), even though overexpression of ARF GAPs does not imply increased GAP activity. For instance, AGAP2 expression in chronic myeloid leukemia cells and prostate cancer is regulated by Specific Protein 1 (SP1) and ATRA ( Doush et al., 2019 ). Additionally, the authors observed that the treatment of cells of these types of cancer with the polyphenol curcumin, leads to a decrease in ATRA-mediated AGAP2 expression ( Doush et al., 2019 ; Giordano and Tommonaro, 2019 ). This data illustrates the relevance of regulating ARF GAP expression levels in cancer. On the other hand, it was observed that QS11, the only inhibitor of ARF GAPs known, binds to ARF GAP1 and inhibits the activity of this GAP on ARF1 and ARF6 ( Zhang et al., 2007 ; Zhu et al., 2012 ). Interestingly, it was observed that QS11 blocks migration of metastatic breast cancer cells, in vitro ( Zhang et al., 2007 ). Thus, inhibitors of ARF GAP activity could also be effective in controlling cancer cell migration and invasion. Inhibition of the expression or function of downstream effectors of ARF family proteins is also a plausible strategy to impair the oncogenic potential of ARFs and ARLs. An interesting candidate is NMII. Indeed, several types of cancer exhibit differential expression and/or activation of NMII isoforms, leading to alterations in cell migration and invasion that are involved in tumorigenesis ( Newell-Litwa et al., 2015 ). Moreover, we found that NMIIA is an effector of ARL13B ( Casalou et al., 2014 , 2019 ). Furthermore, blebbistatin inhibits the ATPase activity of NMIIA and has been found to block invasiveness of both breast cancer cells ( Derycke et al., 2011 ) and pancreatic adenocarcinoma cells ( Duxbury et al., 2004 ), and is phototoxic in human cancer cells under exposure to blue light ( Mikulich et al., 2012 ).

The Adenosine diphosphate-Ribosylation Factor (ARF) family of proteins belongs to the RAS superfamily of small GTPases and comprises around 30 members in mammals ( Sztul et al., 2019 ). This family includes 6 ARFs (5 in humans since ARF2 is absent), 21 ARLs, 2 Secretion-Associated RAS-related (SARs) and the TRIpartite Motif-containing protein 23 (TRIM23). ARF1-5 regulate vesicle budding at the Golgi apparatus by recruiting coat complexes ( Li et al., 2004 ; Kahn et al., 2006 ). ARF6 localizes to the plasma membrane, as well as endosomes and is involved in actin cytoskeleton dynamics and endocytic recycling ( Donaldson, 2003 ). The functions of ARL proteins are more heterogeneous and currently unknown for several of them. ARL2 and ARL3 interact with microtubules and function in tubulin assembly and cytokinesis, respectively, while ARL4C and ARL4D are involved in actin remodeling and regulate cell migration ( Li C.-C. et al., 2007 ; Chiang et al., 2017 ). Our laboratory has shown that ARL13B binds actin and regulates cell migration ( Barral et al., 2012 ; Casalou et al., 2014 ). Interestingly, several ARLs, namely ARL3, ARL6 and ARL13B are associated with the cilium and play different roles in ciliary biology and signaling pathways associated with this organelle ( Marwaha et al., 2019 ). ARL8B is well characterized and has been shown to localize to lysosomes and regulate several aspects of lysosome biology, such as positioning and motility ( Khatter et al., 2015 ). Finally, SARs play a well-described role in the budding of COPII-coated vesicles from the ER, while TRIM23 was implicated in antiviral defense and adipocyte differentiation ( Arimoto et al., 2010 ; Watanabe et al., 2015 ; Saito et al., 2017 ). Like other GTPases, ARF family proteins switch between an active state, in which proteins are GTP-bound and an inactive state, in which proteins are GDP-bound. For this reason, they are referred to as “molecular switches.” Nucleotide exchange is catalyzed by GEFs and GTP hydrolysis is promoted by GAPs. When they are active, ARF proteins associate with membranes via lipid modifications, namely myristoylation, palmitoylation or acetylation and bind effectors. These are responsible for the downstream functions of ARF family proteins and are highly diverse. Among the effectors identified are coat complexes and adaptors, cytoskeleton-binding proteins and tethering factors ( Donaldson and Jackson, 2012 ). The functions of ARF and ARL proteins, as well as their GEFs and GAPs are thoroughly reviewed in two excellent recent reviews ( Marwaha et al., 2019 ; Sztul et al., 2019 ). Since ARFs and their regulators play essential functions in cell cycle, cytoskeleton remodeling, cell migration and adhesion, it is not surprising that they can be subverted by cancer cells for proliferation, migration and invasion. Indeed, the expression and/or activity of several ARF family proteins and their GEFs and GAPs has been shown to be modulated in several types of cancer ( Tables 1 , 2 ). Moreover, the amplification and overexpression of ARF family genes, as well as the overexpression of their GEFs and GAPs, and variance in post-translational modifications are the most commonly detected alterations thought to be implicated in cancer. Here, we review the members of the ARF family and their activity regulators and effectors that have been implicated in cancer, and can either function as oncogenes or tumor suppressors and propose possible therapeutic approaches to target ARF family proteins or their effectors, GEFs and GAPs. Expression of ARF family members in human neoplastic tissues and cancer cells. Expression of ARF GEFs in human neoplastic tissues and cancer cells.