Classical RASSF Proteins
RASSF1 Family, Ras Association Domain Family Member 1
Loss of heterozygosity (LOH) studies in lung, breast, and kidney tumors identified several loci in chromosome 3p likely to harbor one or more tumor suppressor genes (TSGs). Particularly, the 3p21.3 region was suspected to be the home for a TSG because instability of this region is the earliest and most frequently detected deficiency in lung cancer (Hung et al. 1995; Kok et al. 1997; Sekido et al. n.d.; Sundaresan et al. 1992; Wistuba et al. 1999; Wistuba et al. 2000). This 120 kb region was found to be gene rich and eight genes were identified: CACNA2D2, PL6/placental protein 6, CYB561D2/101F6, TUSC4/NPRL2/G21, ZMYND10/BLU, RASSF1/123F2, TUSC2/FUS1, and HYAL2/LUCA2. However, none of these candidate genes are frequently mutated in cancer (Agathanggelou et al. 2005; Lerman and Minna 2000). The RASSF1 gene was identified in a yeast two-hybrid screen as an interacting partner of the DNA damage repair protein Xeroderma Pigmentosum Complementation Group A (XPA) (Dammann et al. 2003a). The RASSF1 gene became of interest because it mapped in the critical 3p21.3 region, according to Genebank database, it shared high sequence homology with the Ras effector in mouse Nore1 (human RASSF5) and also because expression of one of its isoforms, RASSF1A, was lost in most lung tumor cell lines (Dammann et al. 2003a; Agathanggelou et al. 2005). The name Ras-association domain family member 1 was assigned after identification of a Ras association domain (RA) in the RASSF1 sequence.
The RASSF1 Gene Locus
The RASSF1 locus spans about 11 kb of the human genome and it includes eight exons. Differential promoter usage and alternative splicing generate eight isoforms (A-H). Two CpG islands are associated with the promoter regions of RASSF1. The smaller of the two spans the promoter region of RASSF1A, 1D, 1E, 1F, 1G, 1H, and methylation at this CpG island has been shown to be one of the most common events in human cancers (Donninger et al. 2007; Hesson et al. 2007; Jones and Baylin 2007). The second CpG island surrounds the promoter regions for RASSF1B and 1C. Isoforms A and C are the major transcripts of the gene and are ubiquitously expressed (Richter et al. 2009). The B isoform is expressed mainly in hematopoietic cells and D and E are present in cardiac and pancreatic tissue, respectively. Little information is available regarding the functions of splice variants B, D, E, F, G, and H (Donninger et al. 2007).
RASSF1A, Ras Association Domain Family Member 1A
The 1.9 kb RASSF1A transcript initiates from a promoter located in the first CpG island of the RASSF1 gene and translates to a 340 amino acids protein. Transcription initiates with exon 1α followed by exon 2αβ. These exons encode for a diacylglycerol/phorbol ester domain (DAG), also called protein kinase C conserved region (C1) which encodes for amino acids 51–101 in RASSF1A (Newton 1995). In vitro studies have revealed a sequence within exon 3 that is an ataxia telangiectasia mutated (ATM) kinase phosphorylation site, conserved in RASSF1A, 1C, 1D, 1E, and 1H (Kim et al. 1999).
A Ras association (RA) or RalGDS/AF-6 domain is encoded by exon 4 and 5, and it generates residues 194–288 of RASSF1A. This domain is located at the COOH terminus of isoforms A-E. The RA domain mediates interactions with Ras and Ras-like small GTPases. It shares similarity in structure with the RasGTP binding domain of Raf1 kinase, a well-established RasGTP effector (Ponting and Benjamin 1996; Yamamoto et al. 1999).
The last 47 amino acids on the COOH terminus of RASSF1A are part of the so-called SARAH (Sav/RASSF/Hpo) domain. The SARAH domain mediates protein-protein interactions and facilitates the formation of both homo- and heterodimers of RASSF isoforms and other Hippo pathway components such as protein salvador homolog 1 (SAV1) and mammalian sterile 20-like protein kinase (MST) (Hwang et al. 2007; Scheel and Hofmann 2003).
The isoforms A, D-H are expressed from the same promoter within the same CpG island. Therefore, these isoforms are commonly missing in various tumors due to epigenetic inactivation.
RASSF1A has been categorized by many studies as a bona fide tumor suppressor gene (TSG), and the biological functions that RASSF1A performs in the cell are tightly related to its TSG role. The most studied functions of RASSF1A include involvement in cell cycle, microtubule organization, induction of apoptosis, control of tissue size/proliferation via the Hippo pathway, and maintenance of genomic stability.
RASSF1A and Apoptosis
The apoptotic cascade plays a pivotal role in determining cell fate. Notably, one of the hallmarks of cancer cells is their ability to escape apoptotic processes and develop high-grade malignancy and resistance to therapy (Hanahan and Weinberg 2011). Many RASSF family members have been described as proapoptotic (Khokhlatchev et al. 2002; Eckfeld et al. 2004; Vos et al. 2000, 2003a, b). RASSF1A has been documented to regulate apoptosis via the death receptor signaling pathway as well as to work as an effector protein of KRas-driven induction of apoptosis. In the first scenario, RASSF1A has been shown to be required for death receptor-induced Bax conformational change and mediation of apoptosis (Baksh et al. 2005). Briefly, death receptor stimulation by tumor necrosis factor α (TNFα) or TNFα-related apoptosis-inducing ligand (TRAIL) resulted in recruitment of RASSF1A and the modulator of apoptosis 1 (MOAP1) proteins to the receptors and formation of a complex between RASSF1A and MOAP1. This complex allowed MOAP1 to associate with the proapoptotic protein Bax and eventually resulted in apoptotic signaling activation (Baksh et al. 2005; Foley et al. 2008; Vos et al. 2006). Concomitant loss of RASSF1A and MOAP1 has been recently correlated to tumorigenesis (Law et al. 2015).
In the KRas-driven apoptosis scenario, RASSF1A (and its homologue NORE1/RASSF5) has been proposed to be a driver of the proapoptotic Hippo pathway component MST1 (Praskova et al. 2004). Various publications also demonstrated RASSF1A-mediated apoptosis with engagement of the Hippo signaling pathway, with RASSF1A enhancing MST2 interaction with its substrate, large tumor suppressor homolog 1 (LATS1). Subsequently, activated LATS1 phosphorylates and releases the transcriptional regulator Yes associated protein 1 (YAP1), allowing YAP1 to translocate to the nucleus and associate with p73, resulting in transcription of the proapoptotic target gene puma (Matallanas et al. 2007). A later study documented the importance of Ras signaling in the activation of the apoptotic MST2-LATS1 cascade. Interestingly, mutant but not wild-type KRas directly binds to the tumor suppressor RASSF1A to activate the MST2-LATS1 pathway. In this pathway LATS1 binds to and sequesters the ubiquitin ligase MDM2 causing stabilization of the tumor suppressor p53 and apoptosis (Matallanas et al. 2011). Always in the mutant KRas scenario, another scaffold protein, the connector enhancer of KSR (CNK1) has been shown to participate to the apoptotic response and to require a direct interaction with RASSF1A (and MST1) in order to enhance cell death (Rabizadeh et al. 2004).
RASSF1A in Mitosis and Cell Cycle Control
RASSF1A plays a major role in the regulation and progression of mitosis and the cell cycle. Perhaps the mitotic role most frequently attributed to RASSF1A is its association with microtubules (MTs). RASSF1A interacts with MTs via microtubule associated proteins (MAPs) (Dallol et al. 2004) and localizes with the spindle and centrosomes during mitosis (Liu et al. 2003). RASSF1A’s localization with tubulin has been shown to influence MT stability via RAN-GTPase (Dallol et al. 2009), as well as increasing stability by inhibiting HDAC6 (histone deacetylase 6) activity (Jung et al. 2013). The ability of RASSF1A to associate with the MTs correlates with its tumor suppressive functions, and loss of this property can lead to apoptotic defects (Matallanas et al. 2007), centrosomal/spindle errors (Liu et al. 2003), and tubulin instability (Vos et al. 2004) (which has been linked to cancer cell invasion/metastasis (Kaverina and Straube 2011)). Of note, the MT binding sites of RASSF1A have been found to interact specifically with nondynamic regions of MTs which are crucial for Golgi integrity and the establishment of polarity during cell migration (Arnette et al. 2014).
Given its importance at the mitotic spindle, it is unsurprising that RASSF1A is involved in regulating the cell cycle and has been shown to induce cell cycle arrest in all its phases (Rong et al. 2004). Importantly, the cell cycle control role of RASSF1A is impaired upon loss of MT association (Donninger et al. 2014). Various mechanisms by which RASSF1A exerts its cell cycle regulatory functions have been described, including its role as a death domain-associated protein 6 (DAXX) dependent mitotic stress checkpoint (Giovinazzi et al. 2012), its inhibition of cyclin D accumulation at the G1/S transition checkpoint (Shivakumar et al. 2002), and as an S-phase progression (and inducing senescence) regulator via a distinct p21Cip1/Waf1-dependent pathway (Thaler et al. 2009). It has been proposed that RASSF1A can modulate the activity of the APC-Cdc20 complex (which is necessary for mitotic progression). Conflicting reports have explored RASSF1A’s inhibition of, and interaction with, this complex (Sup Song et al. 2004; Liu and Katrin Baier RHD & GPP 2007) with the most recent studies showing that Aurora A phosphorylates RASSF1A on Ser203 and uncouples it from APC/Cdc20 at the centrosomes (Rong et al. 2007; Song et al. 2009). Interestingly, Ser203 is located within the microtubule binding domain of RASSF1A and phosphorylation of this site disrupts the microtubule binding ability of RASSF1A and abrogates RASSF1A-mediated mitotic arrest (Rong et al. 2007) concomitant with the need for MT association in RASSF1A-mediated cell cycle regulation.
RASSF1A and Stem Cells
RASSF1A interacts with and induces activation of the core kinases MST1 and MST2 of the Hippo pathway via its SARAH domain. The Hippo pathway is a highly conserved developmental signaling pathway, first described in Drosophila melanogaster, that regulates organ growth (Heallen et al. 2011), stem cell function and tissue regeneration in the mammalian system (reviewed in Wang et al. (2017)). Crosstalk between the Hippo and TGFβ/BMP pathway has been found to regulate self-renewal and differentiation in mouse (Lian et al. 2010) and human embryonic stem cells (James et al. 2005) by mediating SMAD signaling via the Hippo effectors YAP/ TAZ (Varelas et al. 2008). RASSF1A has been shown to be targeted for degradation in response to TGFβ signaling allowing nuclear localization of YAP/SMAD and transcription of TGFβ target genes (Pefani et al. 2016).
Interestingly, epigenetic inactivation of RASSF1A in addition to hypermethylated in cancer 1 protein (HIC1) was sufficient to induce a tumorigenic phenotype of human mesenchymal stem cells (hMSCs) (Teng et al. 2011) suggesting that loss of function of RASSF1A contributes to the transitioning of stem to cancer-like stem cells.
RASSF1A and Genome Stability
In 2009, Hamilton et al. identified the ATM-dependent phosphorylation of RASSF1A at Ser131 in response to DNA damage (Hamilton et al. 2009). Upon phosphorylation RASSF1A dimerizes and stimulates the kinase activity of MST2, resulting in stabilization of p73 (Hamilton et al. 2009) and expression of YAP/p73 apoptotic target genes (Strano et al. 2005). In 2014, Pefani et al. established an MST2-LATS1-dependent role for RASSF1A in response to replication stress. Phosphorylation of RASSF1A by ataxia telangiectasia and Rad3-related protein (ATR) promotes replication fork stability by restricting cyclin-dependent kinase 2 (CDK2) binding to LATS1 and therefore decreasing breast cancer type 2 susceptibility protein 2 (BRCA2) phosphorylation to enhance DNA repair protein RAD51 homolog 1 (RAD51) nucleofilament formation at stalled forks (Pefani et al. 2014). Additionally, Donninger et al. showed that RASSF1A has a direct role in DNA repair by modulating sirtuin 1 (SIRT1)-dependent deacetylation of XPA and therefore enhancing its nucleotide excision repair (NER) activity in response to UV-induced DNA damage (Donninger et al. 2015a). However, the RASSF1A single nucleotide polymorphism variant (RASSF1 c.397G > T, p.Ala133Ser, rs2073498) fails to get phosphorylated and is associated with early cancer onset (Kanzaki et al. 2006; Bergqvist et al. 2010) particularly in patients with BRCA mutations (Gao et al. 2008), supporting that RASSF1A loss of function contributes to genomic instability and subsequently to tumorigenesis. Notably, RASSF1A has been shown to promote p53 activity and stability by disrupting the MDM2-DAXX-ubiquitin carboxyl-terminal hydrolase 7 (HAUSP) complex and therefore destabilizing MDM2, the p53 E3 ubiquitin ligase (Song et al. 2008).
RASSF1A in Cancer
Epigenetic Inactivation of RASSF1A
Silencing of putative tumor suppressor genes (TSG) is one of the hallmarks of carcinogenesis (Jones and Baylin 2007). Downregulation of TSGs may occur via genetic mutation or epigenetic silencing (e.g., through gene methylation). RASSF1A is one of the most frequently epigenetically inactivated tumor suppressor genes in sporadic human malignancies (Donninger et al. 2007; Hesson et al. 2007; van der Weyden and Adams 2007; Grawenda and O’Neill 2015). Methylation of the RASSF1A gene is rare in normal tissues, whereas the frequency of methyl-cytosine in the promoter spanning CpG island increases in tumor tissue and is one of the highest described. It has been reported over the years in a number of malignancies with the highest frequencies being reported in lung, breast, and prostate cancers (Donninger et al. 2007; Grawenda and O’Neill 2015).
Given the link between RASSF1A and Ras signaling pathways, several studies have investigated a correlation between KRas mutation and RASSF1A inactivation in cancer. An inverse, and thus synergetic for the tumor, correlation has been shown for colorectal (van Engeland et al. 2002), pancreatic (Dammann et al. 2003b), nonsmall cell lung cancer (NSCLC) (Li et al. 2003), and melanoma (Reifenberger et al. 2004), even though conflicting results are proposed in a different study on NSCLC (Kim et al. 2003).
RASSF1A Epigenetic Inactivation as a Marker for Tumor Diagnosis and Prognosis
Given the strong correlation between RASSF1A promoter methylation and cancer onset, RASSF1A methylation status has a compelling clinical utility potential as a biomarker for cancer risk and prognosis (Grawenda and O’Neill 2015). An apparent correlation of RASSF1A methylation with clinical characteristics of invasive tumors is evident in both breast and lung cancer (Donninger et al. 2007; Grawenda and O’Neill 2015). Correlation of RASSF1A methylation with cancer risk is also best validated in gastrointestinal (GI) cancer (Grawenda and O’Neill 2015). Intriguingly, the fact that RASSF1A promoter methylation occurs rarely in normal tissues also makes RASSF1A a candidate diagnostic marker (Agathanggelou et al. 2005). Furthermore, detection of methylated DNA in samples obtained with noninvasive techniques such as plasma, sputum, urine, throat washing, and nipple aspirates (Chang et al. 2003; de Caceres et al. 2004; Topaloglu et al. 2004; Hoque et al. 2004; Fiegl et al. 2004) makes this approach clinically relevant.
Loss of RASSF1A expression is mainly attributed to promoter methylation, as somatic mutations are uncommon. Analysis of lung, breast, kidney, nasopharyngeal carcinomas, and related cell lines identified only one frame-shift mutation (at codon 277 in the RA domain) and one missense mutation (at codon 201 in the RA domain) (Dammann et al. 2000; Burbee et al. 2001; Astuti et al. 2001; Dreijerink et al. 2001; Lo et al. 2001). However, various somatic polymorphisms have been reported in tumors. Most of them localize in RASSF1A functional domains. Namely, five reside in the C1 domain, four in the ATM phosphorylation site, and five in the RA domain, the most commonly reported actually being a germline Ala133Ser SNP (Dammann et al. 2003a; Gordon et al. 2012). The functional significance of these alterations is not yet fully understood, but data suggests that they are defective mutants. For instance, the Ser131Phe and Ala133Ser mutants reside on the ATM phosphorylation site. Both mutants are unable to induce cell cycle arrest by blocking cyclin D1 accumulation and the Ser131Phe mutant shows reduced phosphorylation which causes a less efficient inhibition of cell proliferation (Shivakumar et al. 2002). Moreover, the Ala133Ser mutant leads to a disrupted conformation of the ATM phosphorylation site and this event leads to a failure in activating p73-dependent apoptotic response upon DNA damage (Yee et al. 2012). Interestingly, the findings on the Ala133Ser mutant correlated with poor outcome and early age onset of sarcoma in patients bearing this SNP (Yee et al. 2012). An independent study also found a correlation between the Ala133Ser polymorphism and early onset of breast cancer in patients carrying a BRCA1/2 mutation (Gao et al. 2008).
RASSF1A in Heart Disease
RASSF1A is a candidate gene in heart disease. As in other tissues, RASSF1A activates MST1 in the heart (Delre and Clark 2010); however, it seems to play contradicting roles depending on which cell type the activation occurs in. In cardiomyocytes, RASSF1A activation of MST1 promotes apoptosis, a detrimental event in the heart; however, in cardiac fibroblasts RASSF1A elicits a positive response by downregulating TNFα expression, which drives cardiac failure via hypertrophy and fibrosis (Bryant et al. 1998; Yokoyama et al. 1997; Sun et al. 2007). Furthermore, studies performed in hypertrophic mouse hearts show that RASSF1A expression is significantly reduced (Oceandy et al. 2009) and RASSF1A expression shows a protective role during pressure overload in cardiac fibroblasts (Delre and Clark 2010). Overall, RASSF1A is a potential cell-type specific therapeutic target in heart disease, but more research is necessary to elucidate its utility.
RASSF1 Knockout Mouse Studies
Knockout mice have been generated for total Rassf1 (all isoforms), as well as for Rassf1A specifically. Rassf1 knockout mice have been shown to be both viable and fertile and to date no other obvious phenotypes have been reported (Liu et al. 2003). However, Rassf1-/- MEFs isolated from the mice displayed smaller overall size as well as an increased sensitivity to MT destabilizing agents, confirming the role of Rassf1 in MT stability. In 2005, two groups separately generated Rassf1A knockout mice (van der Weyden et al. 2005; Tommasi et al. 2005). Once again, the Rassf1A-/- mice were healthy with no obvious abnormalities. Following analysis of the Rassf1A-/- MEFs and lymphocyte populations, it was reported that Rassf1A absence did not grossly affect genome stability and the data further supported the MT stabilizing effects of Rassf1A (van der Weyden et al. 2005). It has also been shown that Rassf1A-/- MEFs have delayed mitotic progression and cytokinesis defects (Guo et al. 2007) displaying the importance of RASSF1A in maintaining mitotic integrity.
The absence of major phenotypes in these mice is surprising given the widespread anti-tumorigenic roles that have been proposed for RASSF1(A). However, when the effect on tumor development was analyzed, the ablation of Rassf1A expression led to a higher susceptibility of malignant disease (van der Weyden et al. 2005; Tommasi et al. 2005) which is in line with the view of Rassf1A as a tumor suppressor. Concomitant with a role for RASSF1A in the heart, it was reported that Rassf1a null mice showed a higher susceptibility to cardiac failure (van der Weyden et al. 2005). Furthermore, there is speculation that the activities of Rassf1A may be compensated for by the Rassf5A/Nore1A protein which has highly similar sequence identity (Dammann et al. 2000; Tommasi et al. 2002), but further study is necessary to fully understand the physiology.
RASSF1B, Ras Association Domain Family Member 1B
The RASSF1B transcript is transcribed from the same promoter as RASSF1A but uses a different 5’ exon resulting in a minor variant lacking the C1 domain. RASSF1B is specifically expressed in hematopoietic cells (Dammann et al. 2003a).
RASSF1C, Ras Association Domain Family Member 1C
RASSF1C is the second ubiquitously expressed isoform of the RASSF1 gene. RASSF1C is transcribed from a different CpG island promoter, 3.5 kb distant from the RASSF1A one. RASSF1C transcription starts at exon 2γ and then continues with four C-terminal exons (3–6) shared with RASSF1A. The main difference to RASSF1A is the lack of the N-terminal C1 domain resulting in a shorter isoform of 270 amino acids. RASSF1C, however, retains the other two C-terminal domains, the RA and SARAH domains (van der Weyden and Adams 2007) (Fig. 1). Like RASSF1A, RASSF1C contains an ATM phosphorylation consensus sequence at Ser61, but data on the phosphorylation has not been shown yet (Kim et al. 1999). Conversely to RASSF1A, a clear role for RASSF1C in tumor suppression has not yet been defined, as reports on its activity show varied outcomes and potential tissue-specificity (Li et al. 2004). RASSF1C does not show promoter hypermethylation and is expressed in almost all lung, breast, pediatric and pancreatic endocrine tumors and cancer cell lines analyzed to date (Dammann et al. 2003a; Burbee et al. 2001; Harada et al. 2002; Malpeli et al. 2011; Reeves et al. 2013). However, in the renal cell carcinoma cell line KRC/Y, RASSF1C was almost undetectable (Dreijerink et al. 2001) suggesting a tissue-specific expression.
One of RASSF1A’s characteristics is to be associated with and stabilize microtubules (Liu et al. 2003; Vos et al. 2004). RASSF1C has also been found to colocalize with tubulin structures in the cell but is not able to stabilize them to the same extent as RASSF1A (Vos et al. 2004).
RASSF1C has been shown to associate with active Ras. In the embryonic kidney cell line HEK293T and in the human lung cancer cell line NCI-H1299, RASSF1C overexpression suppresses the destabilizing effects on genomic integrity of overexpressed mutant RasG12V. The aforementioned protective effect of wild-type RASSF1C was lost when its defective mutant (RASSF1C Ser61Phe) was expressed (Vos et al. 2004). Collectively these data show a potential protective role for RASSF1C in DNA damage response.
The RASSF1C binding to active KRas was also reported to elicit cell death via canonical apoptotic pathway activation (Vos et al. 2000).
Moreover, RASSF1C has been recently reported as a DAXX binding partner (Kitagawa et al. 2006; Escobar-Cabrera et al. 2010), and this could potentially define a role for RASSF1C in apoptosis in response to stress stimuli. In ovarian cancer cells, RASSF1C was found to sensitize cells to cisplatin treatment and induce apoptosis and its activity could be halted by expression of the cellular apoptosis susceptibility CAS/CSE1L gene (Lorenzato et al. 2012). Conversely, other reports supported the evidence of RASSF1C being phosphorylated by glycogen synthase kinase 3 (GSK3) and being targeted for degradation under stress conditions in a AKT-dependent fashion (Zhou et al. 2012). Since AKT promotes RASSF1C upregulation, this would characterize RASSF1C as an oncogene. These findings are in agreement with reports showing that RASSF1C increased cell proliferation in osteoblasts and lung cancer cells and migration in breast cancer cell lines (Amaar et al. 2006; Reeves et al. 2010).
RASSF1C in Cancer
The aforementioned role of RASSF1C as a promoter of cell proliferation in osteoblasts is mediated by the extracellular signal-regulated kinase (ERK) 1/2 and the insulin-like growth factor binding partner 5 (IGFBP5) (Amaar et al. 2005). In lung cancer cells, IGFBP5 seems to play a negative role in RASSF1C promotion of cell proliferation (Reeves et al. 2014).
The variety of downstream genes upregulated upon RASSF1C overexpression also suggests an oncogenic role for the protein. For example, overexpression of RASSF1C in breast cancer cell lines leads to C-X-C chemokine receptor type 4 (CXCR4) upregulation, which enhances cell proliferation and metastasis (Reeves et al. 2010). Interestingly, it has been suggested that RASSF1C modulates the expression of the stem cell renewal gene Piwi-like protein 1 (PIWIL1) in lung cancer (Reeves et al. 2012).
Interplay between the two major isoforms RASSF1A and RASSF1C has been reported in breast cancer cells. In the report, both isoforms alternatively regulate SRC activity and β-catenin/YAP-mediated invasion (Vlahov et al. 2015).
Other reports confirm that RASSF1C induces nuclear β-catenin translocation and transcriptional activation via inhibition of the βTrCP receptor subunit of SCFβ-TrCP, a ubiquitin ligase known to target β-catenin for proteasomal degradation (Estrabaud et al. 2007).
Furthermore, the oncogenic role of RASSF1C has been supported by clinical evidence. Several groups have shown RASSF1C upregulation in a variety of cancers, such as pancreatic endocrine tumors, large cell neuroendocrine carcinomas, small cell lung carcinomas, and esophageal squamous cell carcinoma (Malpeli et al. 2011; da Costa et al. 2011; Pelosi et al. 2010; Guo et al. 2014).
RASSF1D, RASSF1E, Ras Association Domain Family Member 1D and 1E
The RASSF1D and RASSF1E isoforms are both considered RASSF1A splice variants specifically expressed in the cardiac and pancreatic tissue, respectively (van der Weyden and Adams 2007). Both isoforms are encoded by exon 1α. In RASSF1D four additional amino acids 5’ of exon 2αβ are encoded, whereas the RASSF1E transcript has additional four amino acids 3’ of exon 2αβ, for a total of 344 amino acids in both cases.
RASSF1F, Ras Association Domain Family Member 1F
RASSF1F is produced by alternative splicing from RASSF1D where the 2αβ exon is excluded. The RASSF1F protein differs from the canonical sequence by exchange of the amino acids 85–91 (LSADCKF → RACGVGD) and is missing amino acids 93–344. The truncated protein of 92 amino acids terminates within the C1 domain.
RASSF1G, Ras Association Domain Family Member 1G
RASSF1G transcript skips exon 2αβ and 3. The protein differs from the canonical sequence by exchange of the amino acids 84–149 (RLSADCKFTC…EQKIKEYNAQ → QQGRFLHRLH …PACAVTHKGT) and is missing amino acids 150–344 resulting in a short protein lacking the RA and SARAH domain.
RASSF1H, Ras Association Domain Family Member 1H
The RASSF1H transcript variant misses the alternate coding exon from isoform D resulting in a frameshift that translates into isoform RASSF1B. The protein is missing amino acids 1–74 and 140–344. It differs from the canonical sequence by exchange of the amino acids 75–123 ( VVRKGLQCAR…EPAVERDTNV → MGEAEAPSFE…SLARRPRRDQ).
RASSF2, Ras Association Domain Family Member 2, Rasfadin
RASSF2, located at human chromosomal region 20p13, was originally called Rasfadin and was first identified as a novel gene in close proximity to the bovine prion gene (Comincini et al. 2001). The first report showed high nucleotide (88%) and amino acid similarity (95%) with a previously described human cDNA, KIAA0168. In silico characterization of RASSF2 reported three isoforms (namely RASSF2A, RASSF2B, and RASSF2C). All three isoforms contain predicted RA domains, although RASSF2B mRNA produces a much shorter protein with a truncated RA domain. RASSF2A and RASSF2C contain a C-terminal coiled coil SARAH domain that is absent in RASSF2B and share identical sequence (Hesson et al. 2005) (Fig. 1).
The longest isoform RASSF2A contains a 5’ CpG island and a predicted promoter region (Hesson et al. 2005). It is a 326 amino acids protein with the RA domain sharing 28% identity to that of RASSF1A and 31% to that of RASSF5 (van der Weyden and Adams 2007)
RASSF2 has been described as a nuclear protein (Cooper et al. 2008). Interaction has been shown with KRas in a GTP-dependent manner via the RA domain (Vos et al. 2003a) and this interaction could explain at least partially the RASSF2 tumor suppressive role. The idea that silencing of RASSF2 plays a key role in KRas-mediated transformation is supported by reports that KRAS/BRAF mutations are found more frequently in colorectal carcinomas (CRCs) with RASSF2 methylation than in those without (Akino et al. 2005; Nosho et al. 2007; Park et al. 2007). Yeast two-hybrid screens have also proved an association between RASSF2 and RASSF5, MST1/2 and RASSF3. Particularly, interaction with the core Hippo pathway kinase MST2 leads to its stabilization, enhancing MST2 proapoptotic potential (Cooper et al. 2009). Moreover, another report demonstrated a role for MST1 in maintaining RASSF2 protein stability and a proapoptotic activation of MST1 after complexing with RASSF2 (Song et al. 2010). Further links to apoptotic signaling have been made with the observation that RASSF2 was required for the prostate apoptotic response protein 4 (PAR4) to translocate to the nucleus and promote apoptosis (Donninger et al. 2010). Independent groups also observed reduced cell proliferation upon RASSF2 overexpression in lung and colorectal cancer cells and the mechanistic basis for growth inhibition have been related to apoptosis and cell cyce arrest (Vos et al. 2003a; Akino et al. 2005.
RASSF2 in Cancer
Similarly to its homolog RASSF1A, RASSF2 undergoes promoter methylation in a variety of cancers such as breast (Cooper et al. 2008), lung (Cooper et al. 2008; Kaira et al. 2007), colorectal (Vos et al. 2003a; Hesson et al. 2005; Akino et al. 2005), and gastric cancer (Endoh et al. 2005), among others. Particularly strong evidence in CRC shows that RASSF2 promoter methylation is an early event during CRC development as it has been reported in a high proportion of colon adenomas and it is a specific marker of tumor status, since the normal mucosa is found unmethylated (Hesson et al. 2005; Akino et al. 2005). RASSF2A promoter methylation has been positively correlated in some of these studies to KRAS, BRAF, or PIK3CA mutations (Akino et al. 2005; Nosho et al. 2007; Park et al. 2007). It has also been detected in nasopharyngeal carcinoma, in which it positively correlates to lymph node metastasis (Zhang et al. 2007).
RASSF3, Ras Association Domain Family Member 3
The RASSF3 gene, located at 12q14.1, is predicted to produce three transcripts (RASSF3A, RASSF3B, and RASSF3C) due to alternative splicing of the exons. Particularly, RASSF3A contains three exons and translates a 238-residue protein. The last four exons encode a consensus RA domain with a 44% identity (59% homology) to the C-terminus of RASSF1 (both A and C isoforms) and 46% identity to the mouse Rassf5 protein (Tommasi et al. 2002) (Fig. 1). The protein sequence translated from the first exon of RASSF3 shares high homology with the translated sequence of the first exon of RASSF1C (van der Weyden and Adams 2007; Tommasi et al. 2002).
Although there are indications that RASSF3 may act as a tumor suppressor, to date no systematic functional characterization has been performed and mechanisms supporting its tumor suppressor role are still lacking. Recently, RASSF3 has been reported to stabilize p53 and induce p53-dependent apoptosis, thus playing a role both in cell cycle and DNA repair mechanisms (Kudo et al. 2012) In agreement with this data, RASSF3 overexpression in HER2/Neu positive breast cancer cell lines (both mouse and human) inhibited cell proliferation (Jacquemart et al. 2009) and RASSF3 knockdown in NSCLC cells increased migration rate (Fukatsu et al. 2014), thus reinforcing the hypothesis that RASSF3 acts a tumor suppressor gene.
RASSF3 in Cancer
The protein has been found both in normal and cancer cells (Tommasi et al. 2002) Interestingly, RASSF3 has been reported to be downregulated in NSCLC patients, even though the downregulation was not due to promoter methylation, unlike other RASSF family members (Fukatsu et al. 2014). Consistent with the NSCLC data, two independent studies on gliomas and colorectal tumor cell lines also reported no evidence for RASSF3 silencing via promoter methylation (Hesson et al. 2005; Hesson et al. 2004).
RASSF4, Ras Association Domain Family Member 4, AD037
The RASSF4 gene is located on chromosome 10q11.21. It has 12 transcripts variants resulting from alternative splicing from which 5 transcripts are protein coding. RASSF4 contains a RA and a SARAH domain but lacks the ATM phosphorylation site (Fig. 1). It shows 60% identity with RASSF2 and 25% with RASSF1A.
RASSF4 is a potential tumor suppressor that binds the KRas effector protein in a GTP-dependent manner via its association domain (Eckfeld et al. 2004), and it has been shown to bind also other Ras family members. RASSF4 is expressed in various human tissues. It has been shown to play a role in apoptosis (Eckfeld et al. 2004) and cell cycle arrest and is frequently downregulated in numerous cancers (Han et al, 2016). Eckfeld et al. reported that with exogenous expression RASSF4 interacts with MST1 of the Hippo pathway, but data were not presented in the publication (Eckfeld et al. 2004).
RASSF4 in Cancer
Downregulation or loss of RASSF4 is often correlated with promoter methylation (Eckfeld et al. 2004; Chow et al. 2004). Interestingly, no RASSF4 promoter methylation was observed in pheochromocytomas (Richter et al. 2015). RASSF4 has been shown to inhibit growth by suppression of the MAPK pathway in human oral squamous cell carcinoma (Michifuri et al. 2013). Its overexpression inhibits proliferation, migration, invasion, epithelial-mesenchymal transition (EMT), and Wnt signaling in osteosarcoma cells (Zhang et al. 2017).
RASSF5, Ras Association Domain Family Member 5, NORE1, RAPL, Maxp1
RASSF5 is often considered the founding member of the RASSF family. It was originally identified in a yeast-2-hybrid system as an interaction partner of mutant RasG12V and therein named and classified as NORE1 (Novel Ras Effector 1) (Vavvas et al. 1998). It is located on chromosome 1q32.1 and has four transcript variants (RASSF5A-D) which are expressed via differential promoter usage and alternative splicing. RASSF5A/NORE1A is the canonical RASSF5 sequence encoding a 418 amino acid protein (Fig. 1). The two predominant isoforms RASSF5A and RASSF5C (NORE1A and NORE1B respectively) are expressed by differential promoter expression (RASSF5A and C have separate CpG islands) leading to a shorter RASSF5C isoform. RASSF5B is expressed from the same promoter as RASSF5A but does not have the SARAH domain due to exon skipping. Its function has not been demonstrated yet (Hesson et al. 2003). RASSF5 interacts via its RA domain with active (GTP-bound) Ras and various other Ras family/subfamily GTPases (mostly with KRas) (Avruch et al. 2006; Rodriguez-Viciana et al. 2004) and its heterodimerization with RASSF1A is essential for RASSF1A association with RasG12V (Ortiz-Vega et al. 2002). Interestingly, the RASSF5 transcripts display 40–60% homology to RASSF1A (Dammann et al. 2000; Tommasi et al. 2002), suggesting functional overlap between the encoded proteins.
RASSF5A and Apoptosis
RASSF5A and Cell Cycle/Growth
Like RASSF1A, RASSF5A is localized at the centrosomes and binds MTs via its RA domain (Moshnikova et al. 2006) The cytoskeletal interactions of RASSF5A are considered essential for RASSF5’s cell cycle inhibitory roles, which were found to be mediated via negative ERK signaling (Moshnikova et al. 2006) (Fig. 2). RASSF5A can induce MT nucleation via direct interaction with tubulin in a mechanism that is negatively regulated by Aurora A phosphorylation and the binding of RASSF5 to active Ras (Bee et al. 2010). The growth suppression roles of RASSF5 are well documented, with its nuclear localization reported to be essential for this function (Kumari et al. 2007). RASSF5A has been shown to restrict tumor cell line growth in an apoptotic (Kumari et al. 2010) Ras and MST1/2 independent manner (Vos et al. 2003b; Aoyama et al. 2004).
Other Biological Roles of RASSF5A
The role of RASSF proteins in cell fate decisions has been gaining considerable interest. Of note, RASSF5A was found to be important for the specification of axons, allowing neuronal polarization through the small GTPase, Rap1B (Nakamura et al. 2013). Interestingly, RASSF5A displays antiviral activity (Arora et al. 2017) which is specifically targeted and inhibited by the hepatitis C virus (HCV) infection.
RASSF5C (NORE1B, RAPL), like RASSF1A, has been shown to have tumor suppressor functions and is epigenetically silenced in various tumors (Macheiner et al. 2006; Lee et al. 2010). Additionally, it modulates the immune system in response to T-cell receptor (TCR) activation. RASSF5C binds to activated RAP1 to increase lymphocyte-function-associated antigen 1 (LFA1) activity and induce lymphocyte polarization through integrin activation (Katagiri et al. 2003). Furthermore, it has been reported that RASSF5C binds to MST1 in a RAP1-GTP dependent manner. It regulates the activity and localization of MST1 to induce cell polarity and lymphocyte adhesion (Katagiri et al. 2006). Interestingly, Praskova et al. showed that RASSF5A inhibits MST1 activity (Praskova et al. 2004) suggesting different functions for the two isoforms. RASSF5C is predominantly expressed in lymphoid tissue where activation of MST1 is required for lymphocyte adhesion and polarization (Katagiri et al. 2006). Upon TCR activation, RASSF5C mediates NFkB signaling by recruiting active Ras to the plasma membrane and Carma1 (an essential lipid raft-associated regulator of TCR) into the immune synapse (Ishiguro et al. 2006). RASSF5C also localizes to microtubules (Fujita et al. 2005).
RASSF5 and Cancer
Typical of several of the RASSF family members, RASSF5 has been found to be downregulated in a variety of tumors (Vos et al. 2003b; Tommasi et al. 2002) and is a proven bona fide tumor suppressor. Like other RASSF family members, epigenetic inactivation of RASSF5 has been reported in several cancers, including neuroblastoma (Djos et al. 2012; Geli et al. 2008), thyroid (Nakamura et al. 2005), lung (Vos et al. 2003b), colorectal (Lee et al. 2010), and kidney (Steiner et al. 1996; Morris et al. 2003a). However, such silencing is absent in several tumor types (Tommasi et al. 2002; Chow et al. 2004; Nakamura et al. 2005; Foukakis et al. 2006) and deletion is rare (Steiner et al. 1996). RASSF5 has subsequently been reported to be downregulated by proteasomal degradation in transformed cells. The E3 ubiquitin ligase, ITCH, has been shown to interact with RASSF5 and regulate its tumor suppressive activities by downregulation of the protein (Suryaraja et al. 2013). Given the association of HCV infection with the onset of liver cancer, the inhibition of RASSF5A by the HCV protein NS5B (Arora et al. 2017) may also be considered a contributing factor to tumorigenesis in this context.
A number of studies have investigated the consequence of RASSF5 downregulation in tumorigenesis. RASSF5A has been shown to be an inhibitor of the oncoprotein HIPK1 (Lee et al. 2012) by inducing its MDM2-mediated proteasomal degradation, thus displaying further tumor suppressive activities independent of its apoptotic and cell cycle roles. Rassf5-null mice were found to be resistant to TNFα/TRAIL-mediated apoptosis as they could not activate Mst1, and MEFs isolated from the animals displayed more tumorigenic traits than their Rassf5 expressing counterparts (Park et al. 2010). Notably, RASSF5 downregulation has also been linked to the presence of other alterations in specific cancer types. For example, the reduction of RASSF5A in follicular thyroid tumors was shown to be dramatically reduced in cases with a PAX8-PPAR11 translocation as well as being mutually exclusive with the presence of Ras mutations (Foukakis et al. 2006), while in other tumor types there was no correlation (Vos et al. 2003b).
RASSF6, Ras Association Domain Family Member 6
The RASSF6 gene is located on chromosome 4q13.3. There are five known splice variants of the RASSF6 gene, four of them are protein coding. All RASSF6 isoforms contain a RA and a SARAH domain. Additionally, they have a C-terminal PDZ-binding motif, which distinguishes RASSF6 from other RASSF proteins (Fig. 1).
Conflicting data exists regarding RASSF6 and its binding to Ras. Ikeda et al. showed that RASSF6, despite harboring a RA domain, does not bind to Ras proteins under the same conditions as RASSF5 (Ikeda et al. 2007). However, Allen et al. found that RASSF6 specifically binds to activated KRas via its effector domain and that cotransfection with KRas leads to increased apoptosis (Allen et al. 2007). RASSF6 has been shown to induce apoptosis dependent on MDM2 and p53 (Iwasa et al. 2013) and MST1/2, but not the canonical Hippo pathway (Ikeda et al. 2009). Interestingly, it inhibits MST2 activation and therefore Hippo signaling. Upon dissociation from MST2, RASSF6 binds to MOAP1 (Allen et al. 2007) to induce apoptosis. RASSF6 expression is frequently epigenetically silenced in cancer cells and primary tumors (Djos et al. 2012; Hesson et al. 2009; Mezzanotte et al. 2014) and has been shown to have prognostic value (Guo et al. 2016; Wen et al. 2011).
Additionally, RASSF6 has been shown to inhibit NFκB in response to respiratory syncytial virus (RSV) therefore mediating an inflammatory response.
N-Terminal RASSF Proteins
The N-terminus RASSF proteins are a distinct group, as they contain their RA/Ub folds at the N-terminus and none contain a C1 or SARAH domain (Fig. 1).
RASSF7, Ras Association Domain Family Member 7, C11orf13, HRC1
RASSF7 is the most studied member of the N-terminal members of the RASSF proteins. RASSF7 is located at 11p15.5 and the RA/Ub fold of the protein is near the N-terminus and there are two coiled coil regions at the C-terminus (Fig. 1). There are three known splice variants of the protein, with isoform 1 referred to as the canonical sequence. RASSF7 was originally discovered in 1992, when the genome close to the HRas gene was screened for nearby genes (Weitzel et al. 1992) and a new gene was identified and named HRC1 (HRas cluster 1). The gene has since been officially renamed to RASSF7.
The role of RASSF7 in the cell is only just beginning to be uncovered. RASSF7 expression has been recorded in several tissues, including the developing Xenopus (Sherwood et al. 2008) and mouse (Recino et al. 2010) embryos and in many human tissues/cells (Recino et al. 2010). Given its embryonic expression, the developmental role of RASSF7 has been investigated and was recently found to work in concert with DISC1 (Disrupted in schizophrenia 1 protein) in the regulation of astrogenesis (Morris et al. 2003b; Wang et al. 2016), which may suggest a role for RASSF7 in brain development.
RASSF7 has been reported to localize with the centrosome via its coiled coil domain (Sherwood et al. 2008; Gulsen et al. 2016) and its interaction with DISC1 in yeast also indicates a centrosomal role (Porteous et al. 2011). It has been found to be essential for the completion of mitosis (Sherwood et al. 2008; Recino et al. 2010), due to its role as an activator of Aurora B and a regulator of microtubule dynamics (Recino et al. 2010). The protein has also been shown to have an antiapoptotic role via its interactions with NRas and MKK7, to negatively regulate proapoptotic JNK signaling (Takahashi et al. 2011), a role which may be developmentally important in human invertebral disc degeneration (Liu et al. 2015).
RASSF7 in Cancer
Recently, RASSF7 involvement in cancer has only begun to be uncovered. It has been shown to be upregulated in several cancers including pancreatic ductal adenocarcinomas (Vasseur et al. 2003; Logsdon et al. 2003; Friess et al. 2003), pancreatic islet cell tumors (Lowe et al. 2007), endometrial carcinomas (Mutter et al. 2001), malignant thyroid neoplasms (Li et al. 2013), and ovarian clear cell carcinomas (Tan et al. 2009), while a truncated form has been identified and suggested to act as an oncogene (Gulsen et al. 2016). Notably, RASSF7 has been suggested as potential diagnostic marker for islet cell (Lowe et al. 2007) and endometrial (Mutter et al. 2001) tumor identification, given its specific upregulated expression in these cancers. Furthermore, RASSF7 expression is increased in hypoxic conditions (Camps et al. 2008; Liang et al. 2009), a common feature of solid tumors. In contrast, RASSF7 has been shown to be epigenetically silenced in neuroblastomas (Djos et al. 2012).
RASSF8, Ras Association Domain Family Member 8, HoJ-1, C12ORF2
Similarly to RASSF7, which maps close to the HRAS1 gene, RASSF8 is located in close proximity to the KRAS2 locus, 70.8 kb from KRAS2 on chromosome 12p11 (van der Weyden and Adams 2007). RASSF8 has been implicated in a complex type of synpolydactyly by the reciprocal chromosomal translocation t(12;22) (p11.2;q13.3), which involves genes RASSF8 and FBLN1 (Debeer et al. 2002). RASSF8 contains an N-terminal RA domain and lacks the SARAH domain (Lock et al. 2010). Bioinformatic programs (Vega, Ensembl (van der Weyden and Adams 2007)) primarily predict transcripts which differ due to premature truncation at the C-terminus (exons 4–6). To date, seven transcripts have been predicted from the RASSF8 locus (most of which remain to be experimentally confirmed). RASSF8A and RASSF8B share high similarity, differing only in the C-terminal exon. RASSF8C-E transcripts prematurely terminate at the 5’ end of exon 4 and translate shorter proteins that consist almost entirely of the RA domain. RASSF8F and RASSF8G translate for much shorter proteins with no identifiable functional domains (van der Weyden and Adams 2007).
RASSF8 has been showed to inhibit cell growth and to promote adherent junction formation in lung cancer and particularly it has been found to interact with E-cadherin and β-catenin at the adherent junctions (Lock et al. 2010). RASSF8 has been also reported to interact with the ubiquitous scaffolding protein 14-3-3 (Jin et al. 2004) which regulates various molecular mechanisms, such as cell cycle progression and apoptosis (van Hemert et al. 2001).
RASSF8 in Cancer
RASSF8’s initial tumor suppressor role was proposed due to decreased expression in lung adenocarcinomas (Falvella et al. 2006). Moreover, a recent study on melanoma has shown positive correlation between RASSF8 methylation and tumor progression (Wang et al. 2015). In the aforementioned report, RASSF8 significantly inhibited cell growth, cell migration and invasion, whereas its knockdown had an opposing effect by increasing expression of p65 and its downstream target IL6. Additionally, RASSF8 was found to induce apoptosis in melanoma cells by activating the p53-p21 pathway (Wang et al. 2015). A report on childhood leukemia, however, highlighted infrequent methylation of the RASSF8 promoter, thus proving that the methylation status may vary among cancer types (Hesson et al. 2009). A contrasting report on breast cancer identified, among others, RASSF8 mRNA to be enriched in patients’ blood, thus suggesting a role for RASSF8 in cancer progression in this scenario (Rykova et al. 2008).
RASSF9, Ras Association Domain Family Member 9, PAMCI, PCIP1
RASSF9 is located at 12q21.31. It has two isoforms and is the only member of the N-terminal RASSF family that lacks a coiled coil domain and the only one not to be transcribed from a CpG island (Richter et al. 2009) (Fig. 1). It was originally identified in a yeast two-hybrid screen as an interactor of peptidylglycine alpha-amidating monooxygenase (PAM) and was thus named PCIP1 (PAM C-terminal interactor 1) (Alam et al. 1996). In 2008, PCIP1 was renamed RASSF9 given its structural similarities to RASSF7/8 (Sherwood et al. 2008). Interestingly, RASSF9 has been implicated in regulating the trafficking of PAM as it associates with recycling endosomes (Chen 1998). It is also the only member of the N-terminal RASSF proteins that has been shown to interact with Ras proteins (N-,K- and RRas) (Rodriguez-Viciana et al. 2004). While the precise molecular functions of RASSF9 remain unclear, studies suggest that it may be important in epidermal homeostasis. Rassf9-null mice show signs of accelerated aging and defects in epidermal epithelial cell proliferation and differentiation (Lee et al. n.d.). Genetic variants of RASSF9 have been identified that are associated with UV-exposure levels, suggesting that the gene may have recently been evolutionarily selected for (Kita and Fraser n.d.).
RASSF9 in Cancer
To date, there is no clear evidence that RASSF9 has an effect on cancer development. In fact, it was considered a good candidate for colorectal cancer based on its chromosomal localization but was recently ruled out as an influential gene (Sánchez-Tomé et al. 2015). However, there is generally very little data on RASSF9 and given its emerging role in skin development and differentiation; it may yet be uncovered that RASSF9 can play a role in tumorigenesis.
RASSF10, Peptidylglycine Alpha-Amidating Monooxygenase COOH-Terminal Interactor-Like
RASSF10 was discovered by the Chalmers group at locus 11p15.2 and named at the same time as RASSF9 (Sherwood et al. 2008); it contains a RA domain and one coiled coil region (Fig. 1). The molecular functions of RASSF10 remain relatively unknown but the expression of the gene has been found in several tissues (Hesson et al. 2009; Schagdarsurengin et al. 2009; Reeves and Posakony 2005; Hill et al. 2011) and reduction of RASSF10 function has been shown to impair Hedgehog signaling in Drosophila (Nybakken et al. 2005).
RASSF10 in Cancer
Like several of the RASSF genes, the RASSF10 gene has a large CpG island which appears to be the most frequently methylated of the N-terminal family (Underhill-Day et al. 2011) and, following the discovery of its epigenetic silencing in childhood acute lymphocytic leukemia (Hesson et al. 2009), it has been considered a putative tumor suppressor. More recent studies have further supported this notion, as RASSF10 has been found to be epigenetically inactivated in a plethora of cancers including, thyroid (Schagdarsurengin et al. 2009), melanoma (Helmbold et al. 2012), glioma (Hill et al. 2011), lung (Richter et al. 2012), head and neck (Richter et al. 2012), sarcoma (Richter et al. 2012), and pancreatic (Richter et al. 2012).
The RASSF family of proteins is involved in a plethora of physiological functions, such as apoptosis, cytoskeleton dynamics, mitosis, genome stability, and tissue homeostasis. Loss of function, specifically via promoter methylation, of various members of the family has been reported in different tumor types, thus suggesting that some RASSF members could become valuable diagnostic and prognostic markers for cancer onset. However, it is fundamental to further elucidate the roles of RASSFs in physiological and pathological conditions.
- Amaar YG, Minera MG, Hatran LK, Strong DD, Mohan S, Reeves ME. Ras association domain family 1C protein stimulates human lung cancer cell proliferation. Am J Physiol Lung Cell Mol Physiol. 2006;291(6):1185–1190Google Scholar
- Bryant D, Becker L, Richardson J, et al. Cardiac failure in transgenic mice with myocardial expression of tumor necrosis factor-α. Circulation. 1998;97(14):1375–1381Google Scholar
- Chang HW, Chan A, Kwong DLW, Wei WI, Sham JST, Yuen APW. Evaluation of hypermethylated tumor suppressor genes as tumor markers in mouth and throat rinsing fluid, nasopharyngeal swab and peripheral blood of nasopharygeal carcinoma patient. Int J Cancer. 2003;105(6):851–5. doi:10.1002/ijc.11162.PubMedCrossRefGoogle Scholar
- Chow LS, Lo KW, Kwong J, Wong AY, Huang DP. Aberrant methylation of RASSF4/AD037 in nasopharyngeal carcinoma. Oncol Rep. 2004;12(4):781–7. https://www.ncbi.nlm.nih.gov/pubmed/15375500.
- Dammann R, Schagdarsurengin U, Strunnikova M, et al. Epigenetic inactivation of the Ras-association domain family 1 (RASSF1A) gene and its function in human carcinogenesis. Histol Histopathol. 2003a;18(2):665–77. http://www.ncbi.nlm.nih.gov/pubmed/12647816.
- Delre D, Clark GJ. Proapoptotic Rassf1A/Mst1 signaling in cardiac fibroblasts is protective against pressure overload in mice. J Clin Invest. 2010;120. doi:10.1172/JCI43569.Google Scholar
- Fujita H, Fukuhara S, Sakurai A, et al. Local activation of Rap1 contributes to directional vascular endothelial cell migration accompanied by extension of microtubules on which RAPL, a Rap1-associating molecule, localizes. J Biol Chem. 2005;280(6):5022–31. doi:10.1074/jbc.M409701200.PubMedCrossRefGoogle Scholar
- Kita R, Fraser HB. Local adaptation of sun-exposure-dependent gene expression regulation in human skin. Tishkoff SA, ed. PLOS Genet. 2016;12(10):e1006382. doi:10.1371/journal.pgen.1006382.Google Scholar
- Lee C-M, Yang P, Chen L-C, et al. A novel role of RASSF9 in maintaining epidermal homeostasis. Beier F, ed. PLoS One. 2011;6(3):e17867. doi:10.1371/journal.pone.0017867.Google Scholar
- Lerman MI, Minna JD. The 630-kb lung cancer homozygous deletion region on human chromosome 3p21.3: identification and evaluation of the resident candidate tumor suppressor genes. The International Lung Cancer Chromosome 3p21.3 Tumor Suppressor Gene Consortium. Cancer Res. 2000;60(21):6116–33.PubMedGoogle Scholar
- Li X, Zhao G, Wang Y, Zhang J, Duan Z, Xin S. RASSF7 and RASSF8 proteins are predictive factors for development and metastasis in malignant thyroid neoplasms. J Cancer Res Ther. 2013;9 (Suppl 7):S173–7. doi:10.4103/0973–1482.122519.Google Scholar
- Michifuri Y, Hirohashi Y, Torigoe T, et al. Small proline-rich protein-1B is overexpressed in human oral squamous cell cancer stem-like cells and is related to their growth through activation of MAP kinase signal. Biochem Biophys Res Commun. 2013;439(1):96–102. doi:10.1016/j.bbrc.2013.08.021.PubMedCrossRefGoogle Scholar
- Morris JA, Kandpal G, Ma L, Austin CP. DISC1 (Disrupted-In-Schizophrenia 1) is a centrosome-associated protein that interacts with MAP1A, MIPT3, ATF4/5 and NUDEL: regulation and loss of interaction with mutation. Hum Mol Genet. 2003b;12(13):1591–608. doi:10.1093/hmg/ddg162.PubMedCrossRefGoogle Scholar
- Nybakken K, Vokes SA, Lin TY, McMahon AP, Perrimon N. A genome-wide RNA interference screen in Drosophila melanogaster cells for new components of the Hh signaling pathway. Nat Genet. 2005;37. doi:10.1038/ng1682.Google Scholar
- Pelosi G, Fumagalli C, Trubia M, et al. Dual role of RASSF1 as a tumor suppressor and an oncogene in neuroendocrine tumors of the lung. Anticancer Res. 2010;30(10):4269–81. http://www.ncbi.nlm.nih.gov/pubmed/21036752.
- Ponting CP, Benjamin DR. A novel family of Ras-binding domains. Trends Biochem Sci. 1996;21(11):422–5. http://www.ncbi.nlm.nih.gov/pubmed/8987396.
- Reeves N, Posakony JW. Genetic programs activated by proneural proteins in the developing Drosophila PNS. Dev Cell. 2005;8. doi:10.1016/j.devcel.2005.01.020.Google Scholar
- Rykova EY, Skvortsova TE, Hoffmann AL, et al. Breast cancer diagnostics based on extracellular DNA and RNA circulating in blood. Biochem Suppl Ser B Biomed Chem. 2008;2(2):208–13. doi:10.1134/S1990750808020133.Google Scholar
- Sekido, Y., Ahmadian, M., II, W., Latif, F., Bader, S., Wei, M. H., … Minna, J. D. Cloning of a breast cancer homozygous deletion junction narrows the region of search for a 3p21.3 tumor suppressor gene. Oncogene, 1998;16(24):3151–3157. http://doi.org/10.1038/sj.onc.1201858.
- Shivakumar L, Minna J, Sakamaki T, Pestell R, White MA. The RASSF1A tumor suppressor blocks cell cycle progression and inhibits cyclin D1 accumulation. Mol Cell Biol. 2002;22(12):4309–18. http://www.ncbi.nlm.nih.gov/pubmed/12024041.
- Sun M, Chen M, Dawood F, et al. Tumor necrosis factor-α mediates cardiac remodeling and ventricular dysfunction after pressure overload state. Circulation. 2007;115(11):1398–1407Google Scholar
- Sup Song M, Sook Chang J, Jeong Song S, Hong Yang T, Lee H, Lim D-S. The centrosomal protein RAS association domain family protein 1A (RASSF1A)-binding protein 1 regulates mitotic progression by recruiting RASSF1A to spindle poles. J Biol Chem. 2004. doi:10.1074/jbc.M409115200.Google Scholar
- Tommasi S, Dammann R, Zhang Z, et al. Tumor susceptibility of Rassf1a knockout mice. Cancer Res. 2005;65(1):92 LP-98.Google Scholar
- Vlahov N, Scrace S, Soto MS, et al. Alternate RASSF1 transcripts control SRC activity, E-cadherin contacts, and YAP-mediated invasion. Curr Biol. 2015;25:1–16. doi:10.1016/j.cub.2015.09.072.Google Scholar
- Vos MD, Martinez A, Elam C, et al. A role for the RASSF1A tumor suppressor in the regulation of tubulin polymerization and genomic stability. Cancer Res. 2004;64(12):4244–4250.Google Scholar
- Vos MD, Dallol A, Eckfeld K, et al. The RASSF1A tumor suppressor activates Bax via MOAP-1. J Biol Chem. 2006;281. doi:10.1074/jbc.M512128200.Google Scholar
- Wang Y, Yu A, Yu F-X. The Hippo pathway in tissue homeostasis and regeneration. Protein Cell. 2017. doi:10.1007/s13238-017-0371-0.Google Scholar
- Wen Y, Wang Q, Zhou C, et al. Decreased expression of RASSF6 is a novel independent prognostic marker of a worse outcome in gastric cancer patients after curative surgery. Ann Surg Oncol. 2011;7–11. doi:10.1245/s10434–011–1668-5.Google Scholar
- Wistuba II, Behrens C, Virmani AK, et al. High resolution chromosome 3p allelotyping of human lung cancer and preneoplastic/preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. Cancer Res. 2000;60(7):1949–60.PubMedGoogle Scholar
- Yamamoto T, Taya S, Kaibuchi K. Ras-induced transformation and signaling pathway. J Biochem. 1999;126(5):799–803. http://www.ncbi.nlm.nih.gov/pubmed/10544270.
- Yokoyama T, Nakano M, Bednarczyk JL, McIntyre BW, Entman M, Mann DL. Tumor necrosis factor-α provokes a hypertrophic growth response in adult cardiac myocytes. Circulation. 1997;95(5):1247–1252Google Scholar
- Zhou X, Li T-T, Feng X, et al. Targeted polyubiquitylation of RASSF1C by the Mule and SCFβ-TrCP ligases in response to DNA damage. Biochem J. 2012;441(1):227–236Google Scholar