Ral proteins, RalA and RalB, are members of the Ras family of small GTPases. Ral proteins are activated by guanine nucleotide exchange factors (GEFs) that catalyze the exchange of GDP for GTP and facilitate the binding of Ral to its various downstream effector proteins. GTPase-activating proteins (GAPs) stimulate the hydrolysis of GTP to GDP, which inactivates Ral.
The discovery that many tumors contained a transforming Ras allele (HRAS, KRAS and NRAS) in the 1980s spurred interest in identifying new members of the Ras family. In 1986, Pierre Chardin and Armand Tavitian synthesized a 20-mer oligonucleotide probe corresponding to a conserved region of Ras proteins to identify novel Ras genes by screening a simian B-cell line cDNA library (Chardin and Tavitian 1986). The screen resulted in the discovery of an open reading frame that shared a high degree of homology with the three Ras genes and was consequently named Ral (Ras-like). The probe identified two RNAs that differed in length, resulting from alternative processing (Chardin and Tavitian 1986). The cDNA of this simian Ral was subsequently used as a probe to isolate human RalA cDNA from a human pheochromoyctoma library (Chardin and Tavitian 1989). Additionally, this screen identified cDNA that encoded for a related protein, RalB.
RalA and RalB are closely related, sharing 85% amino acid sequence identity. Moreover, RalA and RalB share 50% amino acid sequence identity with Ras proteins. Although mammals encode both Ral isoforms, invertebrates such as Drosophila melanogaster and Caenorhabditis elegans harbor only one Ral gene (Mirey et al. 2003). Ral genes have only been identified in multicellular organisms, with no Ral orthologs in yeast. In humans, RalA and RalB are located on chromosomes 7 and 2, respectively (Hsieh et al. 1990).
Geranylgeranyltransferase-I (GGTaseI) transfers a 20-carbon geranygeranyl moiety on the first cysteine residue. Following prenylation, Ral localizes to the endoplasmic reticulum where Ras-converting endopeptidase 1 (RCE1) cleaves the AAX tripeptide (Leung et al. 2007). Isoprenylcysteine carboxymethyltransferase (ICMT) then methylates the lipid-modified cysteine. An alternative posttranslational modification exists where the CAAX motif bypasses proteolysis and carboxymethylation, ultimately resulting in palmitoylation at the second cysteine (Nishimura and Linder 2013). The precise role of these two different posttranslational modifications remains unclear. Both RalA and RalB require RCE1 to localize to the plasma membrane. However, RalB requires ICMT for targeting to the plasma membrane, whereas RalA requires ICMT for localization to recycling endosomes (Gentry et al. 2015). Moreover, palmitoylation is important for RalB, but not RalA, localization to the plasma membrane.
The most significant sequence divergence between RalA and RalB is in their 30 amino acid hypervariable C-terminal domain (Chardin and Tavitian 1989). These C-terminal sequences contain distinct sites for posttranslational modifications that allow Ral to associate with different membranes. To investigate how differences in the C-terminus affect the biological functions of RalA and RalB, the C-terminal amino acids of RalA were fused to the N-terminal portion of RalB. These studies revealed that the C-terminus of RalA is sufficient to promote RalA-driven polarized delivery of membrane proteins and to facilitate RalA-mediated tumorigenesis (Shipitsin and Feig 2004; Lim et al. 2005).
It is hypothesized that the C-terminus facilitates different biological functions for Ral proteins by mediating changes in subcellular localization. RalA and RalB are phosphorylated on different sites within their C-terminus. Aurora A phosphorylates RalA on Ser194, promoting a shift in RalA subcellular distribution and potentially enhanced RalA activity, as ectopically expressed Aurora A increases RalA activity in a GST-RalBP1-RalBD pull-down assay (Wu et al. 2005). RalA is also phosphorylated on Ser183, possibly by PKA (Wang et al. 2010).
The tumor suppressor, PP2Aβ, dephosphorylates RalA on both Ser183 and 194, inactivating it (Sablina et al. 2007). In contrast to RalA, PKCα phosphorylates RalB on Ser198 and promotes its relocalization to perinuclear regions and anchorage-independent growth in bladder cancer cells (Wang et al. 2010). PKCα also phosphorylates RalB on Ser192 (Martin et al. 2012). The phosphatase that dephosphorylates RalB on Ser198 and 192 has not yet been discovered. It is not understood exactly how phosphorylation changes RalA and RalB subcellular localization. However, it has been hypothesized that the addition of a negatively charged phosphate group neutralizes the charge of the polybasic region, enabling dissociation from plasma membrane and relocalization to other membranes, as was previously described for KRas (Bivona et al. 2006).
In addition to phosphorylation, Ral proteins can undergo nondegradative ubiquitination (Neyraud et al. 2012). Monoubiquitination of RalA (but not RalB) results in enrichment of RalA in lipid rafts within the plasma membrane. Furthermore, de-ubiquitination of RalA in raft microdomains dissociates RalA from the plasma membrane. The discovery that Ral proteins experience reversible ubiquitination prompted inquiry into the enzymes that regulate Ral ubiquitination. Consequently, it was found that the ubiquitin-specific protease USP33 interacts with both RalA and RalB, which appear to be direct targets of USP33 (Simicek et al. 2013). However, RalB demonstrates higher affinity for USP33 than RalA when RalB is GDP-bound. Ubiquitination of RalB at Lys47 modulates the ability of RalB to interact with two of its downstream effectors: Sec5 and Exo84. Therefore, nondegradative ubiquitination differently regulates the functions of RalA and RalB.
Ral activity is not only affected by GDP/GTP cycling and posttranslational modifications but is also altered by calcium and calmodulin. RalA and RalB contain a calmodulin-binding site in their C-terminus and a calcium-independent binding site in their N-terminus (Clough et al. 2002). High intracellular calcium levels promote Ral activity as demonstrated by increased in vitro Ral GTP-binding (Wolthuis et al. 1998; Wang and Roufogalis 1999). How calcium levels activate Ral has not been elucidated. However, one possibility is that phosphorylation by calcium-dependent kinases may regulate RalGEF or RalGAP activity.
Ral binding to calmodulin requires prenylation of the C-terminal CAAX motif (Sidhu et al. 2005). Calmodulin-binding to Ral is important for thrombin-induced activation of Ral in human platelets. Interestingly, RalA has a higher binding affinity for calmodulin than RalB, which possibly indicates the functional differences between RalA and RalB in calcium- and calmodulin-mediated signaling pathways.
Ral in Development
Given that RalA and RalB are members of the Ras family of small GTPases, most research has focused on understanding the role of Ral proteins in cancer. However, we cannot appreciate the role of Ral proteins in cancer without understanding their contributions during development. Ral is expressed ubiquitously in tissues, but levels are highest in the brain, testis, and platelets (Olofsson et al. 1988; Bhullar et al. 1990; Wildey et al. 1993). Zhao and Rivkees characterized the spatial and temporal patterns of RalA and RalB expression in mice during embryogenesis from E9.5 to E16 and showed that RalA and RalB expression first emerges in the brain and gut (Zhao and Rivkees 2000).
RalA null mice are embryonic lethal, while RalB null mice display no apparent phenotype (Peschard et al. 2012). To determine why RalA null mice are embryonic lethal, embryos from RalA+/− crosses were examined and several embryos between E10.5 and E19.5 exhibited exencephaly: a disorder where the brain resides outside of the skull. Failure of neural tube closure causes exencephaly, indicating that RalA plays a role in neural tube closure. Moreover, RalA and RalB appear to share some functions during development because RalA−/−; RalB−/− mice exhibited a more severe phenotype than the RalA single knockout.
The role of Ral in development has also been studied in Drosophila melanogaster. Drosophila only has one Ral gene, which is an essential gene for viability (Balakireva et al. 2006). The development of Drosophila bristles and hairs is commonly used as a model for studying regulation of the cytoskeleton because these phenotypes are easy to observe. Consequently, one group sought to identify the role of Ral in development by examining the effects of different Ral mutants in transgenic Drosophila on bristle and wing hair formation. A dominant negative Drosophila homolog of Ral, DRalS25N, results in loss of bristles on the nota by impairing bristle shaft initiation (Sawamoto et al. 1999). Moreover, expression of DRalS25N results in multiple wing hairs shorter than those in wild-type Drosophila, suggesting disturbed organization of the actin cytoskeleton. Single large bundles of F-actin (referred to as prehair) form in each wing cell. Drosophila expressing DRalS25N exhibits an abnormal number of prehairs and irregular prehair morphology, suggesting that Ral is involved in the initiation of hair development (Sawamoto et al. 1999).
Interestingly, mutating genes encoding JNKK and JNK suppresses the DRalS25N induced bristle phenotype and expression of constitutively active Ral, DRalG20V, inhibits JNK phosphorylation (Sawamoto et al. 1999). Consequently, these data suggest that Ral controls changes in cell shape by inhibiting the JNK pathway. Consistent with the observed JNK phosphorylation in Drosophila, depletion of RalA in HeLa cells increases JNK phosphorylation, further supporting that Ral is a negative regulator of JNK (Balakireva et al. 2006). Given that JNK and p38 MAP kinase are often coactivated in mammals in response to various stresses such as TNF, the effect of Ral on p38 MAP kinase phosphorylation was investigated. RalA depletion in HeLa cells impairs TNF-induced p38 MAP kinase activation, indicating that RalA positively regulates p38 MAP kinase. While the full molecular details of the RalA-JNK signaling axis remain to be fully elucidated, epistasis experiments demonstrated that Ral regulates cell death though the JNK pathway during embryogenesis (Balakireva et al. 2006). Furthermore, the exocyst was identified as a point of cross talk between Ral and JNK signaling.
Ral’s function depends upon the numerous effector proteins to which it binds in a GTP-dependent manner via its effector-binding loop. This chapter highlights several Ral effectors, though it is not a comprehensive list.
The first Ral binding partner identified was RalBP1 (synonyms: RLIP76 and Rip1) (Cantor et al. 1995; Jullien-Flores et al. 1995; Park and Weinberg 1995). RalBP1 is a large protein that has a vast array of functions ranging from receptor-mediated endocytosis to facilitating mitochondrial fission during mitosis (Jullien-Flores et al. 2000; Kashatus et al. 2011). RalBP1 binds GTP-bound Ral (Ikeda et al. 1998). Even though RalA and RalB are 100% identical in their contact sites for binding RalBP1, the binding surfaces on RalBP1 that contact RalA and RalB are not identical (Campbell et al. 2015). For example, an L492A mutation reduces RalA binding more than RalB. Functionally, RalBP1 exhibits GAP activity toward CDC42 and Rac1, which stimulates filopodia and lamellipodia formation, respectively. Therefore, a RalBP1-Ral interaction implicates Ral in regulating actin cytoskeleton changes.
The Ral-RalBP1 interaction has also been implicated in receptor-mediated endocytosis. Specifically, RalBP1 binds two Eps homology (EH) domain-containing proteins: POB1 and Reps15. RalBP1 also binds to the μ2 subunit of AP2, which recruits clathrin to the sites of endocytosis (Yamaguchi et al. 1997; Nakashima et al. 1999; Jullien-Flores et al. 2000). However, it remains unclear whether Ral promotes or inhibits endocytosis. Overexpression of wild-type RalA and the EGF receptor (EGFR) in rat fibroblasts results in reduced EGF-induced EGFR internalization, suggesting that RalA negatively regulates receptor-mediated endocytosis (Shen et al. 2001). However, expression of either constitutively active Ral (RalG23V) or inactive Ral (RalS28N) in two different cell lines inhibits EGFR internalization, suggesting GDP/GTP cycling may be important for RalA’s regulation of EGFR internalization (Nakashima et al. 1999). Additional studies will be required to determine the precise role of the Ral-RalBP1 interaction in endocytosis.
In addition to mediating endocytosis, RalBP1 functions as an ATP-dependent transporter, potentially contributing to multidrug resistance (Awasthi et al. 2000). RalBP1 is overexpressed in different human cancers such as colorectal and breast cancer. However, it remains unclear how RalBP1 contributes to Ral’s tumorigenic potential.
The exocyst is an octameric protein complex that tethers vesicles to the plasma membrane and regulates exocytosis. Both RalA and RalB interact with two subunits of the exocyst, Sec5 and Exo84, in a GTP-dependent manner (Moskalenko et al. 2002, 2003). However, RalA has a higher affinity for the exocyst than RalB due to distinct amino acid sequences in RalA that are distal to the effector-binding domain. The exocyst targets the glucose transporter, Glut4, to the plasma membrane and insulin-induced activation of RalA stimulates Glut4 trafficking to the plasma membrane in adipocytes (Chen et al. 2007). The exocyst is not only involved in exocytosis, but also regulates cell polarity and membrane remodeling. RalA’s association with the exocyst enhances basolateral delivery of membrane components in epithelial cells (Shipitsin and Feig 2004).
The Ral-Exocyst complex also plays a role in innate immunity. Knockdown of Sec5 or RalA results in impaired cytotoxicity in NK cells, suggesting that Ral-mediated exocyst assembly is required for NK cells to efficiently kill target cells (Sánchez-Ruiz et al. 2011). The RalB and Sec5 complex has also been implicated in innate immunity. A RalB-Sec5 interaction recruits and activates the IκB family member kinase, TBK1, which in turn promotes activation of the immune response to virus exposure (Chien et al. 2006).
Additionally, RalB interactions with the exocyst have functions outside of regulating exocytosis. For example, RalB and its effector, Exo84 help activate autophagosome assembly to mediate the cellular starvation response (Bodemann et al. 2011).
Phospholipase D (PLD)
Ral interacts with PLD1 in a nucleotide-independent manner, but it is not known if Ral binds PLD2. PLD is a lipase that catalyzes the hydrolysis of phosphatidylcholine to choline and phosphatidic acid. Phosphatidic acid acts as a second messenger to facilitate vesicle budding and transport. The Ral-PLD1 complex is associated with a small GTPase, Arf, which in turn promotes PLD1 activity (Luo et al. 1998). Arf-induced PLD1 activation is involved in receptor-mediated endocytosis and exocytosis (Shen et al. 2001; Vitale et al. 2002).
RalA interacts with the Y-box transcription factor, ZO-1-associated nucleic-acid-binding protein (ZONAB) in a GTP and cell-density-dependent manner (Frankel et al. 2005). High cell density and cell cycle arrest cause RalA to relocalize to the plasma membrane where it colocalizes with ZONAB. Relocalization of ZONAB from the nucleus to the plasma membrane relieves its transcriptional repression.
RalA binds filamin in a GTP-dependent manner (Ohta et al. 1999). RalA-induced filopodia formation requires filamin binding. Moreover, a constitutively active form of RalA changes the cellular distribution of filamin and recruits it into the cytoskeleton.
RalA binds to the C-terminus of PKCη, which is a kinase required for keratinocyte differentiation (Shirai et al. 2011). PKCη binding to RalA results in activation of RalA and actin depolymerization. However, the mechanism behind how RalA is activated during keratinocyte differentiation remains unclear.
Ral and Cancer
Given that Ral proteins function downstream of Ras and that four RalGEFs bind directly to activated Ras, a lot of research has focused on the role of Ral in Ras-driven tumorigenesis. Interestingly, activating mutations of Ral have not been identified in human cancers. Despite this, several studies point to aberrant activation of Ral in human cancer tissues and cell lines. For example, increased Ral activation has been found in bladder, colon, pancreatic, and brain cancers (Lim et al. 2006; Smith et al. 2007; Martin et al. 2011; Ginn et al. 2016). The mechanism of Ral activation in cancer remains an ongoing investigation. However, aberrant activation of Ral in different cancers can be caused by altered expression of its regulators such as GAPs, GEFs, and deregulation of known Ral-regulating kinases.
RalA and RalB have both distinct and redundant roles in cancer, which reflects the complexity of their function. RalA is required for cellular proliferation and anchorage-independent proliferation in human tumor cell lines, while RalB is required for survival of transformed cells (Chien and White 2003). Chris Counter’s group used siRNA to deplete RalA or RalB levels in human cells and found that constitutively GTP-bound RalA, but not RalB, is sufficient to transform human cells (Lim et al. 2005). Further highlighting the importance of RalA in the initiation of tumorigenesis, RalA, but not RalB, is required for the establishment of K-Ras mutation-positive human pancreatic tumors in immunocompromised mice (Lim et al. 2006).
In contrast to much of the work that has been performed in human cell lines, analysis of RalA and RalB knockout mice demonstrated that elimination of RalA and RalB in a mouse model of non-small cell lung carcinoma (NSCLC) did not impair KRasG12D-driven tumors, suggesting that both RalA and RalB are dispensable for tumorigenesis, at least in this model (Peschard et al. 2012). The apparent discrepancy in these data may reflect species-specific differences in Ral’s function in tumorigenesis, as it was previously demonstrated that human and mouse cells exhibit distinct effector requirements during Ras-driven oncogenesis (Hamad et al. 2002). Alternatively, the differences may reflect variations in experimental design, as much of the previous analysis of Ral function in cancer has been done with siRNA-mediated knockdown or expression of dominant negative mutants, both of which approaches have inherent off-target effects that can make data interpretation more complicated.
It is well established that the RalGEF pathway promotes tumor invasion and metastasis. RalA is predominantly associated with anchorage-independent survival and cellular proliferation, but RalB appears to play more of a role in metastasis. RalB mediates trafficking of multiple myeloma cells and promotes invasion and metastasis of glioma cells (de Gorter et al. 2008; Song et al. 2015). Although RalB appears to be more involved in metastasis, this effect may be tumor-type specific. For instance, in human prostate cancer cells, RalA, but not RalB, is required for bone metastasis (Yin et al. 2007). Knockdown of neither RalA nor RalB affects metastasis of MDA-MB-231 cells, a human breast cancer cell line in athymic nude mice (Yin et al. 2007). Collectively, these data suggest that the role of RalA or RalB in metastasis is tumor-type specific and caution should be used in assigning particular functions in cancer to either Ral protein.
Therapeutic Targeting of Ral
In 2014, Yan et al. discovered two chemical inhibitors of Ral that inhibited tumor xenograft growth similar to depleting Ral in human lung cancer cell lines: RBC8 and BQU57. Importantly, these Ral inhibitors are selective for Ral over other GTPases such as RhoA and Ras. However, RBC8 and BQU57 are first generation tools and it is not known how effective they will be in clinical trials.
Helen Mott’s group recently developed a stapled peptide approach for inhibiting Ral (Thomas et al. 2016). Stapled peptides contain a chemically synthesized staple, which stabilizes the α-helical confirmation of the peptides. The introduction of a staple improves the ability of the peptide to penetrate the cell and helps protect the peptide from protease degradation. The future of stapled peptides as cancer therapeutics appears promising as stapled peptides designed to reactivate p53 are currently being examined in phase I clinical trials, while another stapled peptide has already passed phase I trials. Stapled peptides were designed to inhibit Ral based on the RalB-RalBP1 Ral-binding domain, so the peptides exhibit a higher affinity for RalB than RalA (Thomas et al. 2016). Consequently, the ability of the stapled peptides to inhibit Ral activity was only tested for RalB and not RalA. RalB becomes active in response to nutrient starvation and is both necessary and sufficient for autophagy. Mott’s group found that one of their stapled peptides against Ral (referred to as SP1), inhibited GFP-LC3 turnover in HeLa cells that stably express GFP-LC3. Nutrient starvation failed to reverse SP1-inhibition of GFP-LC3 turnover, demonstrating that SP1 inhibits RalB. Further investigation is required to examine how SP1 will function as a possible cancer therapeutic in a mouse model of Ras-driven tumorigenesis.
In addition to targeting Ral for use as a cancer therapy, it is also possible that Ral may be used as a biomarker for certain cancers such as hepatocellular carcinoma and prostate cancer (Chen et al. 2010; Li et al. 2016). Patients with cancer often mount an immune response, which is likely driven by tumor-associated antigens (TAAs). Consequently, there is a change in autoantibody production against TAAs, enabling TAAs to be used as biomarkers for cancer. Studies have found that patients with prostrate cancer or hepatocellular carcinoma tend to have elevated autoantibody production against RalA (Chen et al. 2010; Li et al. 2016). Moreover, examining both RalA autoantibody and PSA levels significantly increases prostate cancer detection. Therefore, combining Ral with other established biomarkers might further increase our ability to detect cancer.
Our understanding of RalA and RalB function and regulation has steadily increased since the discovery of Ral in 1986. However, further research is required to formally determine why RalA and RalB have the same set of effectors and yet function differently. Moreover, it remains to be understood exactly how RalA and RalB contribute to tumor growth downstream of Ras activation. There is mounting evidence that Ral proteins function as drivers in Ras-mediated tumorigenesis. However, is the role of different Ral proteins tumor-type specific and if so, what drives this preference for one Ral protein over another? With recent advancements in Ral inhibition such as stapled peptides and small molecule inhibitors, we should be able to distinguish RalA and RalB functions in Ras-driven cancers. Additionally, advancements in genetic modification such as CRISPR will aid our understanding of RalA and RalB function and regulation. Other questions remain as well, such as, what distinct roles are played by the multiple RalGEFs and RalGAPs and do other yet to be discovered kinases and phosphatases regulate Ral activity? As our knowledge about Ral regulation and function expands, we can hopefully develop new methods to target the Ral pathway in Ras-driven cancers.
- Chen X-W, Leto D, Chiang S-H, Wang Q, Saltiel AR. Activation of RalA is required for insulin-stimulated Glut4 trafficking to the plasma membrane via the exocyst and the motor protein Myo1c. DEVCEL. 2007;13(3):391–404.Google Scholar
- Falsetti SC, Wang D-A, Peng H, Carrico D, Cox AD, Der CJ, et al. Geranylgeranyltransferase I inhibitors target RalB to inhibit anchorage-dependent growth and induce apoptosis and RalA to inhibit anchorage-independent growth. Mol Cell Biol. 2007;27(22):8003–14.PubMedPubMedCentralCrossRefGoogle Scholar
- Gentry LR, Nishimura A, Cox AD, Martin TD, Tsygankov D, Nishida M, et al. Divergent roles of CAAX motif-signaled posttranslational modifications in the regulation and subcellular localization of Ral GTPases. J Biol Chem. 2015;290(37):22851–61. doi: 10.1074/jbc.M115.656710.PubMedPubMedCentralCrossRefGoogle Scholar
- Kashatus DF, Lim K-H, Brady DC, Pershing NLK, Cox AD, Counter CM. RALA and RALBP1 regulate mitochondrial fission at mitosis. Nat Cell Biol. 2011;13(8):1–10.Google Scholar
- Li J, Dai L, Lei N, Xing M, Li P, Luo C, et al. Evaluation and characterization of anti-RalA autoantibody as a potential serum biomarker in human prostate cancer. Oncotarget. 2016;7(28):43546–43556.Google Scholar
- Vitale N, Chasserot-Golaz S, Bailly Y, Morinaga N, Frohman MA, Bader M-F. Calcium-regulated exocytosis of dense-core vesicles requires the activation of ADP-ribosylation factor (ARF)6 by ARF nucleotide binding site opener at the plasma membrane. J Cell Biol. 2002;159(1):79–89.PubMedPubMedCentralCrossRefGoogle Scholar