Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi


  • Christian R. Robinson
  • Venkateswarlu Kanamarlapudi
Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_101964


Historical Background

The ADP-ribosylation factor (Arf) family of small GTP-binding proteins regulate many cellular events by cycling between active GTP- and inactive GDP-bound forms. They depend on GTP-exchange factors (GEFs) for activation and GTPase-activating proteins (GAPs) for inactivation (Donaldson and Jackson 2011). Mammalian cells express six ARF isoforms (ARF1–ARF6), ARF1 and ARF6 are the best characterized. ARF1 regulates the membrane trafficking mainly at the Golgi, whereas ARF6 regulates endocytosis, exocytosis, and actin reorganization at the plasma membrane (Donaldson and Jackson 2011).

In the human genome, more than 30 genes are predicted to encode proteins with an ARF-GAP domain that have been classified into ten subfamilies on the basis of sequence similarities of their ARF-GAP domains (Kahn et al. 2008). The ARF GAP with dual pleckstrin homology (PH) domains (ADAP) subfamily consists two isoforms (ADAP1 and ADAP2) in human, rat, and mouse, which both evolved from a common ancestor gene (Hanck et al. 2004). ADAP2 is 381 amino acids in length and has a deduced molecular mass of 44.4 kDa. It was originally named centaurin-α2 but renamed as ADAP2 to be consistent with the ARF GAP family classification (Kahn et al. 2008). ADAP2 was initially identified from screening a rat adipocyte cDNA library and found to be closely related to ADAP1, sharing a 58% identical and 75% similar amino acid sequence (Whitley et al. 2002; Hanck et al. 2004). ADAP2 has the same domain structure (see Fig. 1) as ADAP1, an ARF-GAP domain followed by two PH domains (one in the middle [N-PH] and the other at the C-terminus [C-PH]), which is predicted to bind phosphatidyl inositol (PI) lipids (Whitley et al. 2002; Hanck et al. 2004). Due to these structural similarities, ADAP2, like ADAP1, is expected to be involved in cytoskeletal organization, vesicular trafficking, differentiation, and development (Hanck et al. 2004; Venkateswarlu et al. 2007). However, unlike ADAP1, ADAP2 lacks the N-terminal nuclear localization signal.
ADAP2, Fig. 1

The domain structure of ADAP2

Protein Function and Regulation of Activity

ADAP2 displays a widespread cytoplasmic distribution in unstimulated cells. Treatment of transfected PC12 cells with epidermal growth factor (EGF) results in recruitment of ADAP2 to the plasma membrane, through binding to activated PI 3-kinase products (Venkateswarlu et al. 2007). To confirm this, PI 3-kinase inhibitors were used (wortmannin and LY294002), which prevented EGF-induced ADAP2 relocalization. Further, the inactive form of PI 3-kinase similarly inhibited EGF-induced ADAP2 relocation. Moreover, the C-PH domain is sufficient and necessary for ADAP2 recruitment to the plasma membrane (Whitley et al. 2002; Hanck et al. 2004). Thus, PI 3-kinase activation is crucial for the localization of ADAP2 at the plasma membrane. It has been revealed that both PI (3,4,5)-trisphosphate (PI(3,4,5)P3 or PIP3) and PI (3,4)-bisphosphate (PI(3,4)P2), which are the plasma membrane–localized second messengers produced by PI 3-kinase, specifically interact with the C-PH domain of ADAP2 and are responsible the for PI 3-kinase-dependent recruitment of ADAP2, whose localization is maintained at the plasma membrane by PI(3,4)P2 binding (Venkateswarlu et al. 2007). To confirm this, inositol polyphosphate phosphatase (PIPP) and type 1α inositol polyphosphate 4-phosphatase (4-phosphatase) were used to hydrolyze PIP3 to PI(3,4)P2 and PI(3,4)P2 to PI(3)P, respectively. ADAP2 localization in EGF-stimulated cells is inhibited following treatment with 4-phosphatase only (Venkateswarlu et al. 2007). This confirms that sustained ADAP2 recruitment to the plasma membrane is due to ADAP2 specific binding to PI(3,4)P2. However, another study (Hanck et al. 2004) showed that ADAP2 binds PIP3 (PI(3,4)P2 binding was not analyzed in this study) and associates constitutively to the membrane. The constitutive membrane location of ADAP2 in cells disappears upon serum starvation or treatment with wortmannin, indicating that ADAP2 retains at the plasma membrane by binding to the PIP3/PI(3,4)P2 produced by serum-activated PI 3-kinase (Hanck et al. 2004). This may explain why ADAP2 localizes constitutively to the plasma membrane in serum-unstarved cells. ADAP2 has also been shown to bind to inosiotol 1,3,4,5-tetrakisphosphate (IP4), the water soluble derivative of PIP3, and PI(4,5)P2 (Whitley et al. 2002).

ARF6 is in the inactive GDP-form and displays intracellular localization under basal conditions, regardless of ADAP2 expression. EGF treatment of ARF6-transfected cells results in the activation of ARF6 (ARF6-GTP) and its relocation to the plasma membrane. However, EGF treatment of ADAP2 and ARF6 cotransfected cells results in the plasma membrane recruitment of ADAP2 but not ARF6 (Venkateswarlu et al. 2007). Furthermore, ADAP2 requires both the plasma membrane association and the ARF-GAP activity to inhibit ARF6 activation and thereby its localization. Interestingly, ADAP2 requires PI(3,4,5)P3 and PI(3,4)P2 for plasma membrane association but not for its ARF-GAP activity.

ADAP2 has also been observed to interact with β-tubulin. A yeast two-hybrid assay first identified the interaction between β-tubulin and ADAP2. Further coimmunoprecipitation studies in transfected HeLa cells confirm the interaction in mammalian cells, revealing that ADAP2 specifically interacts with the C-terminal of β-tubulin (Zuccotti et al. 2012). After analyzing cytoskeletal fractions of transfected HeLa cells, 80% of ADAP2 association occurred with insoluble microtubules and any remaining ADAP2 associated with the soluble β-tubulin dimers. ADAP2 has also found to induce microtubule formation, without increasing total β-tubulin levels. Further, cells overexpressing ADAP2 show enhanced acetylation and tyrosination of tubulin. Since acetylation provides microtubule breakdown resistance, ADAP2 seems to protect microtubule stability by regulating posttranslational modifications (Zuccotti et al. 2012). Further research shows that the downregulation of ADAP2 in HeLa cells enhances abnormal mitotic spindle formation, whereas ADAP2 upregulation significantly decreases the number of centrosomes following treatment with EGF. Together, these suggest that ADAP2 plays a role in the correct assembly of the mitotic apparatus, possibly by maintaining microtubule stability and playing a role in centrosome duplications (Zuccotti et al. 2012).

ADAP2 is also able to inhibit cortical actin formation during cytoskeletal rearrangement in EGF-treated cells by inhibiting ARF6 activation, which is required for cortical actin formation (Venkateswarlu et al. 2007). ADAP2-transfected cells also showed enhanced formation of actin-rich membrane ruffles. Ten percent of ADAP2-transfected cells showed increased formation and accumulation of actin-rich vesicles. Vesicles expressing ADAP2 display three types of movement: (1) internalization, (2) free movement of preformed vesicles, and (3) restricted movements of large vesicles (Shu et al. 2015). These results indicate that ADAP2 is crucial in membrane ruffling and vesicle internalization. ADAP2 may also be implicated during macropinocytosis. Cells treated with an inhibitor of micropinocytosis (EIPA) led to ADAP2 accumulation at the plasma membrane and reduced vesicle formation. Further evidence suggests that expression of the wild-type GTPase Rab34 increases the size and number of ADAP2-rich vesicles. As expected, mutant Rab34 expression inhibited vesicle formation, and ADAP2 remained localized at the plasma membrane (Shu et al. 2015). Therefore, ADAP2 may regulate macropinocytosis in a Rab34-dependent manner. Furthermore, ADAP2 localizes to Rab8a-rich endosomes and macropinosomes, suggesting that ADAP2 plays a role in endocytic recycling. Following internalization, macropinosomes begin to fuse; however, they do not usually fuse with lysosomes or endosomes. ADAP2-rich vesicles are also excluded from vesicles with both early and late endosomal markers. However, ADAP2-rich vesicles are actually associated with lysosomal vesicles during maturation. The ARF-GAP activity of ADAP2 is also implicated during micropinocytosis. The expression of ADAP2 mutants with the ARF-GAP domain either mutated (R53Q) or missing (ΔARF GAP) similarly result in an inhibition of micropinocytosis, thus, highlighting the importance of the ARF-GAP domain of ADAP2 for inducing cytoplasmic vesicles and macropinocytosis (Shu et al. 2015).

ADAP2 plays a vital role in cell proliferation and migration. Microarray data revealed that 8505c cells transfected with superoxide dismutase (SOD3), an enzyme that regulates cell growth, exhibit increased ADAP2-mRNA levels (Laukkanen et al. 2015). Similarly, an increase in ADAP2 expression enhanced SOD3 expression. Since low SOD3 expression levels enhance cell proliferation and migration, this suggests that SOD3 levels regulate ADAP2 expression to control cell proliferation and migration (Laukkanen et al. 2015). Interestingly, ADAP2 is upregulated in a STAT1-dependent manner following treatment with interferon-β (IFNβ) during endocytosis and intracellular trafficking (Shu et al. 2015). ADAP2 expression is able to prevent Dengue virus (DENV) and vesicular stomatitis virus (VSV) infection by restricting viral replication, suggesting that ADAP2 exerts non-IFN-based antiviral activity. The ARF-GAP activity of ADAP2 is further required for restricting DENV and VSV replication, as well as preventing viral entry by altering transferrin internalization and impairing viral uptake (Shu et al. 2015). Sequence analysis reveals that ADAP2 contains a conserved IFN-stimulated response element within its promotor sequence and is upregulated following type I IFN-signaling. This allows ADAP2 to alter intracellular trafficking and prevent RNA viral entry (Shu et al. 2015).

Using a protein microarray screen, ADAP2 has been isolated as one of the NF-kappaB essential modulator (NEMO) binding protein (Fenner et al. 2010). Like ADAP1, ADAP2 is also seen to interact with addition molecules such as nardilysin (NRDc), a member of the M16 family of zinc metalloendopeptidases, and Ran-binding protein in microtubule-organizing center (RanBPM) (Stricker et al. 2006; Haase et al. 2008). The functional significance of these interactions remains to be elucidated.

Major Sites of Expression and Subcellular Localization

ADAP2 is widely expressed in various tissue types, expected to act as an ARF GAP in nonneuronal cell types (Venkateswarlu et al. 2007). Western blot analysis reveals that ADAP2 exhibits a widespread tissue distribution, with the highest expression in membrane fractions of adipocytes, heart and skeletal muscle, brain, and liver. Lower levels were also found in the membrane factions of the liver and lung (Whitley et al. 2002). However, unlike ADAP1, ADAP2 expression was absent in any cytoplasmic fractions, providing the first evidence that ADAP2 is tightly membrane associated. Following membrane subfractionation, ADAP2 expression was observed in the mitochondrial membrane fraction and dense granules of rat adipocytes. However, ADAP2 expression does not respond to insulin signaling or PIP3 and PI 3-kinase activation in rat adipocytes (Whitley et al. 2002). Further, Northern blot analysis revealed strong ADAP2 expression in the adrenal gland, placenta, spleen, kidney, and skeletal muscle and weak ADAP2 expression in the heart, lung, thyroid, liver, small intestine, and peripheral blood leukocytes. ADAP2 expression was absent in the brain, spinal cord, lymph node, trachea, stomach, colon, and bone marrow (Hanck et al. 2004). When overexpressed in HEK-293 cells, ADAP2 has been shown to exhibit the plasma membrane localization but absent from the nucleus (Hanck et al. 2004). Interestingly, treatment of ADAP2 transfected HEK-293 cells with wortmannin, an inhibitor of PI 3-kinase, results in the relocalization of membrane-bound ADAP2 to the cytoplasm. Consistent with this, ADAP2 shows the cytoplasmic localization in serum-starved cells and translocates to the plasma membrane upon stimulation with EGF (Venkateswarlu et al. 2007). Taken together, these results suggest that ADAP2 retains at the plasma membrane by binding to the PIP3/PI(3,4)P2 produced by serum-activated PI 3-kinase (Hanck et al. 2004; Venkateswarlu et al. 2007).

Interestingly, ADAP2 is highly expressed in both adult and fetal heart tissue (Venturin et al. 2004, 2005). During mouse embryo development, ADAP2 expression occurs before and during chamber formation, implicating an early role for ADAP2 during heart development (Venturin et al. 2005). Using in situ hybridization, ADAP2-mRNA expression was visualized during vital phases of heart development, such as heart looping, endocardial cushion formation, and heart septation. Expression was strongest between 8 and 10.5 days postcoitum (dpc) and maintained for up to 15.5 dpc (Venturin et al. 2014). ADAP2 spatio-temporal expression in zebrafish revealed the presence of ADAP2 transcript for up to 120 h postfertilization. ADAP2 mRNA has also been found in zebrafish oocytes, suggesting that ADAP2 is zygotically expressed. Additionally, knockout studies performed in zebrafish suggest impaired heart looping (65%), blood circulation defects (61%), blood cell accumulation in the trunk region (48%), or blood stases (13%) at 2 dpc. All defects in mutant zebrafish were a result of impaired heart development and function, not vascular development. As such, ADAP2 is crucial during cardiac morphogenesis. At 5 dpc, three phenotypes were revealed: mild, intermediate, and severe. Mild phenotype showed disorganization of cells and defective atrioventricular (AV) valves. Intermediate phenotype consisted immature and irregular valve formation. Severe phenotypes showed impaired heart morphology, lacking separation of the heart chambers (Venturin et al. 2014). Together, these data suggest that ADAP2 may be involved in the regulation of AV valve development.

Phenotypes and Splice Variants

There is one known isoform of ADAP2 (see Fig. 1). There has been one additional isoform predicted, consisting of 380 amino acids, with the amino acid 269 missing due to a competing acceptor splice site. Also, seven splice variants have been predicted to be formed through alternative splicing of ADAP2, which differ from each other in length and amino acid sequence. However, there is no experimental evidence available to confirm these isoforms existence. The human ADAP2 gene is located on chromosome 17, at region 29 12 842 – 29 550 556. The ADAP2 gene contains 11 exons to give a cDNA of 2747 base pairs (Hanck et al. 2004). Both ADAP1 and ADAP2 evolved from a single common ancestor that diverged to produce two proteins with varying cellular distributions and binding specificities (Stricker et al. 2006). ADAP2 orthologs were also found in mouse, rat, and puffer fish (Hanck et al. 2004).

The ADAP2 gene is located in the vicinity of the neurofibromatosis gene. Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder, where 5–20% of patients exhibit large deletions of at least 11 functional genes, including NF1 and ADAP2. Classical NF1 patients display less severe symptoms than microdeleted patients. NF1 microdeletions are a result of haploinsufficiency of the NF1 gene and its neighboring regions, whereby patients are at an increased risk of neurofibroma, cardiovascular malformations, and learning disabilities (Hanck et al. 2004; Venturin et al. 2004). Furthermore, NF1-microdeleted patients may go on to develop malignant peripheral nerve sheath tumors (MPNSTs). ADAP2 expression in malignant peripheral nerve sheath tumor (MPNST) cell lines is significantly reduced, suggesting that ADAP2 may also be a tumor suppressor gene (Pasmant et al. 2011).

Whitley et al. (2002) raised an antibody against the 14-amino-acid peptide sequence (CPTEKEQREWLENL) from the C-terminus of ADAP2. The purified antibody was used in Western blot purposes for the detection of ADAP2 antigen (Whitley et al. 2002). Various anti-ADAP2 antibodies are commercially available. A goat unconjugated polyclonal antibody, raised against the C-RLTASTESGRSSR sequence from the C-terminus of ADAP2, is available from Novus Biologicals (NB100-1319). This antibody is suitable for use in immunohistochemistry (IHC), with or without paraffin. Similarly, a goat polyclonal antibody to human ADAP2, raised against the C-RLTASTESGRSSR sequence, is available from MyBioSource (MBS241949) and is suitable for IHC-paraffin and ELISA.

A rabbit unconjugated polyclonal antibody reactive against human, mouse, and rat is available from Proteintech Group (13706-1-AP) and is suitable for use in ELISA, Western blot, and IHC. A further rabbit unconjugated polyclonal antibody is also available from Bioorbyt (orb182535) that is reactive with human, mouse, and rat, suitable for use in ELISA, Western blot, and IHC-paraffin. Life Span Biosciences, Inc. offer a wide range of anti-ADAP2 antibodies, including a goat unconjugated polyclonal antibody to human ADAP2 (aa375-387) for use in IHC-paraffin and ELISA (LS-B2279), a rabbit biotin-conjugated polyclonal antibody to human ADAP2 (aa49-361) for use in IHC (paraffin and frozen), immunocytochemistry, Western blot, and ELISA (LS-C423874), as well as a rabbit biotin-conjugated polyclonal to mouse ADAP2 (aa49-361) for use in Western blot and ELISA.


ADAP2 is a protein that functions as a GAP for ARF6 and binds specifically with second messengers PI(3,4)P2 and PIP3. It has a broad tissue distribution, with high-expression levels found in the adrenal gland, placenta, spleen, kidney, and skeletal muscle. ADAP2 structurally contains an Arf GAP domain followed by two PH (N-PH and C-PH) domains. It is mostly localized in the cytosol of unstimulated cells. Binding to PI 3-kinase lipid products, PIP3/PI(3,4)P2, through its PH domains results in ADAP2 translocation to the plasma membrane, where it has been shown to regulate actin cytoskeleton reorganization by inhibiting ARF6 activation. ADAP2 also interacts with β-tubulin and is involved in increasing microtubule stability as well as duplicating centrosomes during assembly of mitotic apparatus. ADAP2 is also seen to interact with addition molecules such as NEMO, NRDc, and RanBPM but the physiological significance of these interactions is currently unknown. SOD3 controls cell migration and proliferation by regulating ADAP2 expression. Furthermore, ADAP2 expression induces macropinocytosis by affecting ARF6-dependent vesicle trafficking. ADAP2 is also able to alter transferrin uptake and exert RNA-antiviral activity by disrupting DENV and VSV entry. ADAP2 expression is observed in adult and fetal heart tissue, indicating that it may a play role in regulation of heart development. Consistent with this, gene knockout studies in zebrafish have highlighted the importance of ADAP2 during heart development, especially during heart looping and AV valve formation. Since the gene for ADAP2 is located on human chromosome 17q11.2 in the vicinity of NF1, it may be a vital candidate gene affecting NF1-microdeleted patients with cardiovascular malformations.


  1. Donaldson JG, Jackson CL. ARF family G proteins and their regulators: roles in membrane transport, development and disease. Nat Rev Mol Cell Biol. 2011;12:362–75. doi: 10.1038/nrm3117.PubMedCentralCrossRefPubMedGoogle Scholar
  2. Fenner BJ, Scannell M, Prehn JH. Expanding the substantial interactome of NEMO using protein microarrays. PLoS One. 2010;5:e8799. doi: 10.1371/journal.pone.0008799.PubMedCentralCrossRefPubMedGoogle Scholar
  3. Haase A, Nordmann C, Sedehizade F, Borrmann C, Reiser G. RanBPM, a novel interaction partner of the brain-specific protein p42IP4/centaurin alpha-1. J Neurochem. 2008;105:2237–48.CrossRefPubMedGoogle Scholar
  4. Hanck T, Stricker R, Sedehizade F, Reiser G. Identification of gene structure and subcellular localization of human centaurin alpha 2, and p42IP4, a family of two highly homologous, Ins 1,3,4,5-P4-/PtdIns 3,4,5-P3-binding, adapter proteins. J Neurochem. 2004;88:326–36.CrossRefPubMedGoogle Scholar
  5. Kahn RA, Bruford E, Inoue H, Logsdon Jr JM, Nie Z, Premont RT, et al. Consensus nomenclature for the human ArfGAP domain-containing proteins. J Cell Biol. 2008;182:1039–44.PubMedCentralCrossRefPubMedGoogle Scholar
  6. Laukkanen MO, Cammarota F, Esposito T, Salvatore M, Castellone MD. Extracellular superoxide dismutase regulates the expression of small gtpase regulatory proteins GEFs, GAPs, and GDI. PLoS One. 2015;10:e0121441.PubMedCentralCrossRefPubMedGoogle Scholar
  7. Pasmant E, Masliah-Planchon J, Lévy P, Laurendeau I, Ortonne N, Parfait B, et al. Identification of genes potentially involved in the increased risk of malignancy in NF1-microdeleted patients. Mol Med. 2011;17:79–87.CrossRefPubMedGoogle Scholar
  8. Shu Q, Lennemann NJ, Sarkar SN, Sadovsky Y, Coyne CB. ADAP2 is an interferon stimulated gene that restricts RNA virus entry. PLoS Pathog. 2015;11:e1005150.PubMedCentralCrossRefPubMedGoogle Scholar
  9. Stricker R, Chow KM, Walther D, Hanck T, Hersh LB, Reiser G. Interaction of the brain-specific protein p42IP4/centaurin-alpha1 with the peptidase nardilysin is regulated by the cognate ligands of p42IP4, PtdIns(3,4,5)P3 and Ins(1,3,4,5)P4, with stereospecificity. J Neurochem. 2006;98:343–54.CrossRefPubMedGoogle Scholar
  10. Venkateswarlu K, Brandom KG, Yuh H. PI 3-kinase-dependent membrane recruitment of centaurin-alpha2 is essential for its effect on ARF6-mediated actin cytoskeleton reorganisation. J Cell Sci. 2007;120:192–801.CrossRefGoogle Scholar
  11. Venturin M, Guarnieri P, Natacci F, Stabile M, Tenconi R, Clementi M, et al. Mental retardation and cardiovascular malformations in NF1 microdeleted patients point to candidate genes in 17q11.2. J Med Genet. 2004;41:34–41.CrossRefGoogle Scholar
  12. Venturin M, Bentivegna A, Moroni R, Larizza L, Riva P. Evidence by expression analysis of candidate genes for congenital heart defects in the NF1 microdeletion interval. Ann Hum Genet. 2005;69:508–16.CrossRefPubMedGoogle Scholar
  13. Venturin M, Carra S, Gaudenzi G, Brunelli S, Gallo GR, Moncini S, et al. ADAP2 in heart development: a candidate gene for the occurrence of cardiovascular malformations in NF1 microdeletion syndrome. J Med Genet. 2014;51:436–43.CrossRefPubMedGoogle Scholar
  14. Whitley P, Gibbard AM, Koumanov F, Oldfield S, Kilgour EE, Prestwich GD, et al. Identification of centaurin-alpha2: a phosphatidylinositide-binding protein present in fat, heart and skeletal muscle. Eur J Cell Biol. 2002;81:222–30.CrossRefPubMedGoogle Scholar
  15. Zuccotti P, Cartelli D, Stroppi M, Pandini V, Venturin M, Aliverti A, et al. Centaurin-α2 interacts with β-tubulin and stabilizes microtubules. PLoS One. 2012;7:e52867.PubMedCentralCrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Christian R. Robinson
    • 1
    • 2
  • Venkateswarlu Kanamarlapudi
    • 1
  1. 1.Institute of Life Science 1, School of MedicineSwansea UniversitySwanseaUK
  2. 2.Calon Cardio-Technology Ltd, Institute of Life Science 2Medical School, Swansea UniversitySwanseaUK