ADP-ribosylation factors (ARFs) are a family of Ras-related small GTPases, which are ubiquitously expressed and known to regulate many cellular processes including intracellular trafficking and membrane-related processes (D’Souza-Schorey and Chavrier 2006; Donaldson and Jackson 2011). Out of the six-member ARFs (ARF1–ARF6) in mammalian cells, ARF1 and ARF6 are well characterized. ARF1 regulates membrane trafficking between the endoplasmic reticulum (ER) and the Golgi, while ARF6 has been shown to modulate clathrin-dependent and independent endocytosis, exocytosis, and organization of actin cytoskeleton at the plasma membrane (Donaldson and Honda 2005). Like other Ras-related small GTPases, the ARF family of GTPases acts as binary switches by cycling between an inactive GDP-bound and an active GTP-bound conformations. The active/inactive cycle of ARFs is tightly regulated by guanine exchange factors (GEFs) and GTPase-activating proteins (GAPs). GEFs activate ARFs by catalyzing the exchange of bound GDP for GTP. Upon activation of downstream signaling events, the active ARFs revert back to the inactive state through the hydrolysis of bound GTP to GDP by their intrinsic GTPase activity, which is stimulated by GAPs.
The ARF GAPs are a family of proteins that mainly control intracellular trafficking, including vesicle formation. The stromal membrane-associated protein (SMAP) subfamily of ARF GAPs consists of SMAP1 and SMAP2 in mammalian cells. SMAP1 preferentially interacts with ARF6 over ARF1 through its ARFGAP domain, and it interacts directly with clathrin heavy chain (CHC) through its clathrin-binding motif. In addition to these domains, SMAP2 contains a clathrin assembly lymphoid myeloid (CALM) domain, which allows it to interact with CALM and regulate clathrin coat formation. Although no experimental evidence yet exists for SMAP1-CALM interaction, the amino acid sequence alignment of SMAP1 and SMAP2 shows conserved regions in the CALM interaction region, an indication of similar functional role in SMAP1 (Tanabe et al. 2005). The murine SMAP1 was originally identified by a study looking into the erythropoietic activity of stromal cells (Yanai et al. 1997). SMAP1 expression in distinct hematopoietic organs of mice correlated with the predominance of erythropoiesis, while its expression decreased following the ceasing of erythropoiesis (Sato et al. 1998). The human SMAP1 protein is the putative product of the SMAP1 gene, which composed of 11 exons and localized to the 6q13 chromosomal region. It consists of 468 amino acids (aa; 50.4 kDa) and shares 86% homology to its murine counterpart.
Sites of Expression, Domains, and Splice Variants of SMAP1
SMAP1 is a type II integrin membrane glycoprotein expressed in a variety of mammalian tissues, including the bone marrow, heart, brain, placenta, lung, liver, muscle, kidney, pancreas, spleen, thymus, prostate, ovary, and colon, but not in the intestine (Meyer et al. 2005). The amino acid sequence between 171 and 195 of SMAP1 has repetitive lysine- and glutamine-rich regions – similar to human microtubule-associated proteins MAP1A and MAP1B, indicative of possible roles in microtubule assembly (Marcos et al. 2002). Through alternative splicing, four splice variants are predicted for the SMAP1 protein, and only splice variant 1 has been the main focus of many studies. To our knowledge, no functional analysis has been done so far for SMAP1 splice variants 2–4.
The Role of SMAP1 in ARF6-Mediated Endocytosis and Beyond
The clathrin-coated vesicle transport system mediates the trafficking of membrane proteins to appropriate organelles, and this process is tightly regulated by the ARF family of small GTPases (Donaldson and Jackson 2011). SMAP1 has been shown to be a GAP for ARF6 rather than ARF1, and its overexpression inhibits transferrin receptor (TfnR) internalization at the plasma membrane (Tanabe et al. 2005; Kobayashi et al. 2014). Endocytosis of cell surface membrane proteins, such as TfnR and E-cadherin, is initiated by ARF6 activation at the plasma membrane (Tanabe et al. 2006). The active form of ARF6 recruits an adaptor coat protein AP-2, promoting the formation of clathrin-coated pits (Tanabe et al. 2006). ARF6 has also been shown to regulate, in its active form, the actin reorganization and clathrin-independent endocytosis (Tanabe et al. 2005). However, it has been shown that SMAP1 is involved in clathrin-dependent endocytosis but not in clathrin-independent endocytosis or actin reorganization (Tanabe et al. 2005). This is consistent with the notion that different GEFs and GAPs contribute the site-specific regulation of ARF6 activation/inaction. SMAP1 can interact with the CHC via its “clathrin box” (a 5-amino acid motif, LLGLD), and as a result mutations within this sequence abolish the clathrin-binding activity of SMAP1 (Tanabe et al. 2006). The CHC-SMAP1 interaction brings SMAP1 into the vicinity of activated ARF6, where a putative quaternary complex comprising ARF6, AP-2, clathrin, and SMAP1 is proposed to form to facilitate clathrin assembly into nascent vesicles (Tanabe et al. 2006). Finally, the ARF GAP domain of SMAP1 regulates the budding of vesicles from the plasma membrane by promoting the ARF6 inactivation (Tanabe et al. 2006). SMAP1 overexpression prolongs ARF6 inactivation and thereby inhibits the recruitment of AP-2 and clathrin to form coated pits (Tanabe et al. 2005). The ARF6 GAP activity of SMAP1 is dependent on the presence of a conserved arginine residue at position 61. A GAP-negative mutant (R61Q) of SMAP1 overexpression in Hela cells prevents AP-2 recruitment, indicating that the ARF GAP domain of SMAP1 is important for inhibition of clathrin-dependent endocytosis (Tanabe et al. 2005). Consisting with this, siRNA-mediated downregulation of SMAP1 expression in Hela cells has shown to inhibit endocytosis of TfnR (Tanabe et al. 2005; Kobayashi et al. 2014).
In in vivo also, SMAP1 has been shown to function exclusively as an ARF6 GAP (Kon et al. 2013). Unlike in in vitro studies, the absence of SMAP1 in erythroblast cells in in vivo has been shown to enhance TfnR endocytosis. This is because SMAP2 plays a compensatory role in regulating ARF6-mediated endocytosis in those cells (Sakakura et al. 2011). SMAP1 and SMAP2 proteins not only colocalize (in HeLa cells) but also directly interact with one another (in COS-7 cells) (Kobayashi et al. 2014). Their in vivo absence decreases TfnR endocytosis; however, the presence of either is sufficient for TfnR internalization (Kobayashi et al. 2014). The possible domains of SMAPs 1 and 2 required for their interaction have been identified: a 166–349 or 350–440aa region in SMAP1 and a 339–428aa region in SMAP2 (Kobayashi et al. 2014). A recent study has further narrowed down the SMAP2-binding region in SMAP1 to 368–386aa (Sumiyoshi et al. 2015).
In Mardin-Darby canine kidney (MDCK) epithelial cells, SMAP1 localizes to the cytoplasm and the site of adherent junctions where the adhesion protein E-cadherin is present (Kon et al. 2008). E-cadherin internalization has been shown to occur prior to the epithelial-mesenchymal transition (EMT) (Kalluri and Weinberg 2009). Overexpression of SMAP1 in these cells inhibits E-cadherin endocytosis, resulting in reduced EMT. This implicates SMAP1 as a key regulator of the EMT, an event intrinsic to the metastatic ability of many cancers. The SMAP1 has also been shown to effect c-Kit internalization and sorting to lysosomes in in vivo. C-Kit – a member of tyrosine kinase receptor family and tumor marker – is integral to the growth of numerous progenitor cells such as stem, mast, and germ cells (Sattler et al. 1997). Although the internalization of c-Kit is unaffected, the degradation of c-Kit in lysosomes is delayed in SMAP1 knockout mice bone marrow-derived mast cells (Kon et al. 2013). This results in the rise of c-Kit levels in the cytoplasm and thereby increases in the ERK activation and cell growth. Finally, apoptosis has been observed at the distal regions of embryos at E7.5 in SMAP1 and SMAP2 double-knockout mice, suggesting their importance in normal embryogenesis (Sumiyoshi et al. 2015).
The Role of SMAP1 in Disease
The role of SMAP1 in disease has been documented in several studies. The frameshift mutations in the SMAP1 gene have been associated with high-risk colorectal cancer (El-Bchiri et al. 2008; Wachter et al. 2014). In colorectal cancers displaying microsatellite instability (MSI CRC), SMAP1 mutations have been observed in 73% (32/44) of primary tumors (Sangar et al. 2014). In human MSI CRC, SMAP1 mutation frequencies have been shown to be highest among stage II CRC MSI while being significantly lower in stage IV CRC MSI, suggesting that the mutation of SMAP1 inversely correlates with the disease progression (Sangar et al. 2014). In addition, 78% of all cell lines featuring MSI CRC have shown SMAP1 mutations due to deletion or insertion in the ten-adenine repeat region located at the end of the ARFGAP domain (Sangar et al. 2014). In subcutaneous xenograft experiments in nude mice, SMAP1 overexpression has shown to inhibit tumor growth. In contrast, SMAP1 mutations increase cell proliferation by shortening the G2/M cell cycle phase, whereas the restoring of SMAP1 expression reverses cell proliferation to normal (Sangar et al. 2014). SMAP1 overexpression in colorectal carcinoma HCT116 cells, similar to its expression in MDCK cells, inhibits E-cadherin and beta-catenin internalization, while an increase in key EMT markers such as snail, slug, and vimentin has been seen. These results suggest that SMAP1 may be an emerging key regulator of EMT (Kon et al. 2013; Sangar et al. 2014).
The nuclear respiratory factor 1 (NRF-1) is a leading transcription factor involved in many cellular processes such as protein synthesis, proliferation, and cell cycle (Efiok and Safer 2000). In human neuroblastoma IMR-32 cells, NRF-1 enhances neurite length when overexpressed (Chang et al. 2005). The shRNA-mediated knockdown of SMAP1 has been shown to increase the levels of NRF-1, resulting in uncontrolled neurite outgrowth (Tong et al. 2013). This highlights the role of SMAP1 in regulating the differentiation of neuroblastoma cells. Due to SMAP1 expression in retina and its close chromosomal proximity to RP25, a locus for autosomal recessive retinitis pigmentosa (arPR), it has been identified to have a potential role in the arPR disease pathogenesis (Barragan et al. 2005). Finally, SMAP1 has also been identified as a translocation partner of the human mixed-lineage leukemia (MLL) gene in 8% (3/36) of patients with high-risk acute leukemia (Meyer et al. 2005). Chromosomal translocations involving the MLL gene are associated with high-risk acute leukemia (Schoch et al. 2003). During abnormal recombination events, SMAP1 (a translocation partner gene) fuses with the MLL allele to confer oncogenic potentials (Schoch et al. 2003).
The main functions of SMAP1 discovered so far include the regulation of clathrin-dependent endocytosis, specifically of E-cadherin, TfnR, and beta-catenin, and sorting of cytoplasmic c-Kit. Both E-cadherin and TfnR are transported in clathrin-coated vesicles, and overexpression of SMAP1 – which results in ARF6 inactivation – impairs endocytosis of these molecules. SMAP1 preferentially interacts with ARF6 over ARF1 via its ARFGAP domain and clathrin proteins via its CHC and CALM domains and binds to the carboxy-terminal of SMAP2. However, the functional relevance of these interactions is yet to be fully understood. It is likely that both SMAPs 1 and 2 play a complementary and compensatory role in regulating ARF-mediated endocytosis. Indeed, in vivo it has been shown that SMAP2 compensates for the absence of SMAP1 in TfnR endocytosis. In addition to its involvement in embryogenesis, SMAP1 also regulates EMT. SMAP1-deficient mice showed the inhibition of E-cadherin endocytosis and increased levels of key EMT markers such as snail, slug, and vimentin. The cumulative effect of these changes is an uncontrolled growth as a result of uncontrolled cell cycle, which represents some of the underlying reasons behind cancer progression. In cutaneous xerograph models, the overexpression of SMAP1 wild type has shown to inhibit tumor growth, while its loss-of-function mutant expression increases tumor growth. SMAP1 has been implicated in MSI CRC where its mutation frequency inversely correlates with disease progression. In neuroblastoma cells, SMAP1 knockdown results in abnormal neurite outgrowth due to increased levels of NRF-1. Its association with the MLL gene implicates SMAP1 in high-risk acute leukemia. In summary, SMAP1 is a key regulator of ARF6-mediated clathrin-dependent endocytosis in both in vitro and in vivo, and its emerging role in disease makes it as a potential therapeutic target.
- J. El-Bchiri, Guilloux, A., Dartigues, P., Loire, E., Mercier, D., Buhard, O., et al. Nonsense-mediated mRNA decay impacts MSI-driven carcinogenesis and anti-tumor immunity in colorectal cancers. PLoS One 2008;3:e2583. doi: 10.1371/journal.pone.0002583.
- Schoch C, Schnittger S, Klaus M, Kern W, Hiddemann W, Haferlach T. AML with 11q23/MLL abnormalities as defined by the WHO classification: incidence, partner chromosomes, FAB subtype, age distribution, and prognostic impact in an unselected series of 1897 cytogenetically analyzed AML cases. Blood. 2003;102:2395–402. doi: 10.1182/blood-2003-02-0434.PubMedCrossRefGoogle Scholar