USP8 (Ubiquitin-Specific Protease 8)
Deubiquitinating enzymes (DUBs) are proteases which specifically hydrolyze the isopeptide bond between ubiquitin and target proteins in ubiquitin-protein conjugates or between ubiquitin molecules in ubiquitin chains. Ubiquitin-specific protease 8 (USP8), also known as ubiquitin isopeptidase Y (UBPY), was originally identified as a DUB which is upregulated in proliferating cells and promotes the entry of cells into the S-phase of the cell cycle (Naviglio et al. 1998). Subsequent studies, however, have accumulated evidence that this DUB plays a major role in regulating the lysosomal traffic/degradation of plasma membrane proteins upon endocytosis from the cell surface.
Downregulation of Growth Factor Receptors
Endosomal Sorting of Growth Factor Receptors
Ubiquitination is a posttranslational protein modification in which a 76-amino-acid protein called ubiquitin is conjugated through its C-terminal carboxyl group to the ε-amino group of Lys residues of various intracellular proteins (specifically referred to as monoubiquitination). Ubiquitin can also be conjugated via its C-terminus to one of the seven Lys residues of another ubiquitin. Therefore, repeated conjugation of ubiquitins to the ubiquitin moiety of ubiquitinated proteins leads to the formation of a ubiquitin chain on substrate proteins (specifically referred to as polyubiquitination). Because ubiquitin has seven Lys residues, the tertiary structure of the polyubiquitin chain differs depending on which Lys residue is used for ubiquitin conjugation (Kulathu and Komander 2012). And importantly, it has been shown that different types of polyubiquitination mediate different events in the cell (Kulathu and Komander 2012). For instance, Lys48-linked polyubiquitination is well known as a tag that targets the ubiquitinated proteins for degradation in the proteasome.
Upon ligand binding on the cell surface, growth factor receptors undergo Lys63-linked polyubiquitination at specific Lys residues in their cytoplasmic regions by the E3 ligase c-Cbl (Mohapatra et al. 2013). This ubiquitination serves as a tag which directs plasma membrane proteins to the lysosome for degradation. Nutrient receptors such as the LDL receptor do not normally undergo ubiquitination and are sorted to the recycling pathway to the cell surface. On the endosomal membrane, ubiquitinated receptors, but not unmodified receptors, are sequentially recognized by several ubiquitin-binding endosomal sorting complex required for transport (ESCRT) protein complexes (Fig. 1) (Henne et al. 2011). ESCRT-0 is a complex of two proteins, Hrs and STAM, which are both ubiquitin-binding proteins harboring the ubiquitin-interacting motif (UIM) and the VHS domain. ESCRT-0 is localized on the early endosomal membrane via the FYVE domain in Hrs which specifically binds to an early endosome-specific phospholipid, phosphatidylinositol 3′-phosphate. ESCRT-0 recognizes and traps ubiquitinated receptors on the early endosome through the UIMs and the VHS domains and hands them over to ESCRT-I, a complex of three proteins including Tsg101 which harbors another ubiquitin-binding domain, ubiquitin E2 variant (UEV) domain. The ESCRT-II complex also has a component, Eap45 in mammals and Vps36 in yeast, which harbors a ubiquitin-binding domain (GLUE domain in Eap45 and NZF domain in Vps36). ESCRT-II interacts with both ESCRT-I and ESCRT-III and is believed to link the two ESCRT complexes in the endosomal sorting of endocytosed plasma membrane proteins. Eventually, ubiquitinated cargoes sorted on the early endosome are incorporated into the endosomal luminal vesicles formed by the action of oligomerized ESCRT-III protein complex.
Domain Structure of USP8
Based on the amino acid sequences of the catalytic domains, DUBs are classified to five subfamilies: ubiquitin C-terminal hydrolase (UCH), ubiquitin-specific protease (USP), ovarian tumor protease (OTU), Josephin, and JAMM subfamilies (Komander et al. 2009). Except for the JAMM enzymes, which are Zn2+-binding metalloproteases, DUBs are Cys proteases. The catalytic domain of the USP subfamily is composed of two peptide motifs/domains called the Cys-box and His-box. Although often far apart from each other in the entire primary sequences of USP enzymes, the Cys- and His-boxes form a single catalytic core in their tertiary structures.
The SBMs and the MIT domain participates in the recruitment of USP8 to the endosomal sorting machinery of ubiquitinated receptors. The three SBMs interact with the Src homology (SH) 3 domain of STAM, a component of ESCRT-0. The MIT domain interacts with CHMP1 and CHMP7 proteins, which are both components of ESCRT-III. The RH domain of USP8 interacts with its substrate, the E3 ligase RNF41/Nrdp1. This E3-DUB interaction regulates the stability of each other: RNF41 destabilizes USP8 through ubiquitination, while USP8 stabilizes RNF41 through deubiquitination. Finally, 14-3-3 proteins bind to the 14-3-3 binding motif, RSY*SSP, of USP8 when *S (S718 in humans and S680 in mice) is phosphorylated. The 14-3-3 protein binding to USP8 suppresses its catalytic activity (Mizuno et al. 2007). It has been shown that USP8 is dephosphorylated at the 14-3-3 binding motif and undergoes catalytic activation in the M phase. A recent study has revealed that upon dissociation of 14-3-3 proteins, USP8 undergoes cleavage just N-terminal to the 14-3-3 binding motif, and the generated C-terminal 40-kDa fragment, composed of the SBM3 and the catalytic domain, acquires elevated catalytic activity (Reincke et al. 2015). In addition, genetic mutations in the 14-3-3 binding motif, which results in catalytic hyperactivation of USP8, were identified in a specific type of human pituitary tumor (see below).
Endosomal Function of USP8
The molecular basis of how USP8 specifies its substrates is not understood. In addition to EGF receptors, a number of plasma membrane proteins have been shown to undergo deubiquitination by USP8. They include not only receptor tyrosine kinases (e.g., ErbB2, nerve growth factor receptor TrkA, vascular endothelial growth factor receptor VEGFR2) but also different types of transmembrane proteins (e.g., amyloid precursor protein-cleaving protease, BACE1; epithelial Na+ channel, ENaC; Wnt receptor, Frizzled; Ca2+-activated K+ channel, KCa3.1; growth factor receptor-regulating protein, LRIG1; G protein-coupled receptor, PAR2; Hedgehog signaling component, Smoothened), suggesting that USP8 has a broad substrate specificity toward endocytosed plasma membrane proteins. In addition, endosomal substrates of USP8 are not restricted to transmembrane cargo proteins. USP8 has also been shown to regulate receptor downregulation by deubiquitinating Hrs and STAM and stabilizing ESCRT-0 by preventing their proteasomal degradation (Row et al. 2006; Niendorf et al. 2007). Finally, despite the above mentioned accumulating evidence for the endosomal function of USP8, it has also been reported to deubiquitinate several nonendosomal proteins (e.g., DNA damage response factor, BRIT1; circadian transcription factor, CLOCK; extrinsic apoptosis mediator, FLIPL; hypoxia-induced transcription factor, HIF-1α; mitophagy-related E3 ligase, Parkin; amyotrophic lateral sclerosis-associated RNA-binding protein, TDP-43) at different subcellular sites, suggesting the involvement of this DUB in a wide variety of cellular activities.
Homozygous mutation in the USP8 gene is lethal at midgestation in mice. While the heterozygous mutants are normal, homozygous mutants die by embryonic day 10 before the mesoderm induction due to severe growth defect (Niendorf et al. 2007). In liver-specific conditional USP8 mutants, hepatocytes exhibit apoptosis, and the protein levels of growth factor receptors (i.e., EGF receptor, ErbB3, and c-Met) are significantly reduced, supporting the conclusion in vitro that USP8 negatively regulates receptor downregulation (Niendorf et al. 2007).
Activating USP8 Mutations in Cushing’s Disease
In 2015, whole exome sequencing in adrenocorticotropic hormone (ACTH)-secreting pituitary tumors in the German population identified a single somatic mutational hotspot in the 14-3-3 binding motif of the USP8 gene in 4 out of 10 cases (Reincke et al. 2015). The same mutational hotspot was independently discovered in the Asian population in 8 out of 12 ACTH-secreting pituitary tumors (Ma et al. 2015). Targeted sequencing in larger patient cohorts showed that the prevalence of somatic USP8 mutations in ACTH tumors ranges from ∼35% to 60% (Reincke et al. 2015; Ma et al. 2015; Perez-Rivas et al. 2015; Hayashi et al. 2016). Similarly, whole exome sequencing in a large set of pituitary tumors picked somatic USP8 mutations in 11 out of 20 ACTH tumors (Song et al. 2016).
ACTH-secreting pituitary tumors are the most frequent cause of Cushing’s disease (CD), a devastating condition caused by excessive glucocorticoid production that is triggered by aberrant ACTH stimulation of the adrenals (Valassi et al. 2011). CD is characterized by metabolic syndrome and central obesity, hypertension, muscle wasting, osteoporosis, abnormal skin conditions, depression, and other psychiatric comorbidities. Patients with CD present with low quality of life and increased mortality.
Pathogenesis of USP8-Mutated Cushing’s Disease
EGFR is highly expressed in ACTH-secreting pituitary tumors, and EGF increases the level of ACTH precursor proopiomelanocortin (POMC) transcription as well as ACTH secretion in murine, canine, and human ACTH-secreting pituitary tumor cells (Fukuoka et al. 2011; Theodoropoulou et al. 2004, 2015; and references therein). Overexpression of the mutant USP8 forms in the murine ACTH-producing pituitary tumor cell line AtT-20 potentiated the stimulatory action of EGFR on Pomc transcription and ACTH secretion (Reincke et al. 2015; Perez-Rivas et al. 2015). A brief analysis of signaling cascades that may mediate this action indicated that the effect is dependent on the Erk1/2, but not Akt, pathway. The mutant as well as wild-type USP8 proteins amplified EGF-induced Erk1/2 phosphorylation without affecting Akt phosphorylation (Reincke et al. 2015). USP8 did not induce Erk1/2 phosphorylation in the absence of EGF stimulation, indicating that its effect is downstream to impaired downregulation of ligand-activated receptor. Contrary to what was observed on ACTH synthesis, USP8 mutants did not further potentiate the cell-proliferative action of EGFR compared to what was observed with wild-type USP8. This divergence between ACTH-producing and cell-proliferative actions is unclear at the moment.
In consistence with the potent stimulatory action of the USP8 mutants on Pomc expression in vitro, USP8 mutant tumors contain higher POMC transcript levels (Hayashi et al. 2016). However, no significant changes were observed in plasma ACTH levels between USP8 mutant and wild-type cases in a multicenter European-based study (Perez-Rivas et al. 2015) or they were significantly lower in the mutant cases in an Asian study (Hayashi et al. 2016). In terms of tumor size, the clinical presentation of the CD patients with USP8 mutant tumors reflects the in vitro finding (i.e., lack of significant cell-proliferative effect): maximum tumor size did not differ significantly between USP8 mutant and wild-type tumors in a multicenter study (Perez-Rivas et al. 2015), while the two studies in Asian populations reported that USP8 mutant tumors are smaller than the wild-type ones (Ma et al. 2015; Hayashi et al. 2016).
Regarding the EGFR protein levels in patients with CD, the first study showed increased EGFR immunoreactivity in USP8 mutant tumors (Ma et al. 2015). However, a follow-up study using a validated antibody (clone 31G7) did not find increased EGFR levels in tumors with mutant USP8 forms, despite their higher deubiquitinase activity and retention of EGFR to the plasma membrane in in vitro assays (Hayashi et al. 2016).
Patients with USP8 mutation-positive tumors are mainly adult female (very rarely pediatric). A multicenter study showed that they present with worsened outcome after curative transsphenoidal surgery for the removal of tumors, as indicated by the high postoperative urinary cortisol levels (Perez-Rivas et al. 2015). Similarly, these patients had smaller probability to develop adrenal insufficiency, which is considered as an indicator of long-term remission, after surgery, suggestive of worsened operative outcome (Perez-Rivas et al. 2015). In contrast, the monocentric study in the Asian population showed surgical remission in the majority of patients with USP8 mutant tumors (Hayashi et al. 2016).
USP8 mutant tumors had increased gene expression and immunoreactivity of somatostatin receptor 5 (SSTR5), while no differences were observed in SSTR2 transcript and protein levels (Hayashi et al. 2016). This is of particular interest since SSTR5 is highly expressed in ACTH-secreting pituitary tumors and is a pharmacological target of the second-generation somatostatin analog pasireotide that is approved for the pharmacological management of CD (Pivonello et al. 2015).
The high prevalence (∼50%) of USP8 mutations, which are expected to contribute to deregulated ACTH synthesis, in CD tumors showcases USP8 as a potential diagnostic and therapeutic target for CD. The mutant USP8 forms lose binding to 14-3-3 proteins, leading to proteolytic activation of USP8. These findings suggest that the activity of wild-type USP8 is also regulated by 14-3-3 protein-regulated proteolytic cleavage in physiological conditions. Because 14-3-3 protein binding to USP8 is dependent on phosphorylation of Ser718 in the 14-3-3 binding motif, the USP8 activity is expected to be regulated by phosphorylation and dephosphorylation of Ser718. Then, what are the protease, kinase, and phosphatase responsible for the USP8 regulation? When (by what kind of stimulation) and where (in which cell types) does USP8 undergo Ser718 dephosphorylation and proteolytic activation? These are important questions to understand the regulatory mechanism of USP8 activation. The other major question is on the substrate specificity of USP8. While many proteins, mostly plasma membrane proteins, have been reported to be deubiquitinated by USP8, it has been unknown how the USP8 specificity toward these proteins is determined. In particular, identifying the USP8 substrate(s) in pituitary ACTH-producing cells is of significant medical importance to elucidate the pathogenesis of CD. Although EGFR has been suggested as a potential USP8 substrate responsible for CD, the presence of other ACTH-producing cell-specific substrate(s) is also suggested because the hotspot mutations in the 14-3-3 binding motif of USP8 are rarely found in other tumors/cancers (Forbes et al. 2010). Identification of such USP8 substrate(s) is important for the development of anti-CD drugs.
- Forbes SA, Tang G, Bindal N, Bamford S, Dawson E, Cole C, Kok CY, Jia M, Ewing R, Menzies A, Teague JW, Stratton MR, Futreal PA. COSMIC (the catalogue of somatic mutations in cancer): a resource to investigate acquired mutations in human cancer. Nucleic Acids Res. 2010;38:D652–7.PubMedCrossRefGoogle Scholar
- Hayashi K, Inoshita N, Kawaguchi K, Ibrahim Ardisasmita A, Suzuki H, Fukuhara N, Okada M, Nishioka H, Takeuchi Y, Komada M, Takeshita A, Yamada S. The USP8 mutational status may predict drug susceptibility in corticotroph adenomas of Cushing’s disease. Eur J Endocrinol. 2016;174:213–26.PubMedGoogle Scholar
- Ma ZY, Song ZJ, Chen JH, Wang YF, Li SQ, Zhou LF, Mao Y, Li YM, Hu RG, Zhang ZY, Ye HY, Shen M, Shou XF, Li ZQ, Peng H, Wang QZ, Zhou DZ, Qin XL, Ji J, Zheng J, Chen H, Wang Y, Geng DY, Tang WJ, Fu CW, Shi ZF, Zhang YC, Ye Z, He WQ, Zhang QL, Tang QS, Xie R, Shen JW, Wen ZJ, Zhou J, Wang T, Huang S, Qiu HJ, Qiao ND, Zhang Y, Pan L, Bao WM, Liu YC, Huang CX, Shi YY, Zhao Y. Recurrent gain-of-function USP8 mutations in Cushing’s disease. Cell Res. 2015;25:306–17.PubMedPubMedCentralCrossRefGoogle Scholar
- Mohapatra B, Ahmad G, Nadeau S, Zutshi N, An W, Scheffe S, Dong L, Feng D, Goetz B, Arya P, Bailey TA, Palermo N, Borgstahl GE, Natarajan A, Raja SM, Naramura M, Band V, Band H. Protein tyrosine kinase regulation by ubiquitination: critical roles of Cbl-family ubiquitin ligases. Biochim Biophys Acta. 2013;1833:122–39.PubMedCrossRefGoogle Scholar
- Niendorf S, Oksche A, Kisser A, Löhler J, Prinz M, Schorle H, Feller S, Lewitzky M, Horak I, Knobeloch KP. Essential role of ubiquitin-specific protease 8 for receptor tyrosine kinase stability and endocytic trafficking in vivo. Mol Cell Biol. 2007;27:5029–39.PubMedPubMedCentralCrossRefGoogle Scholar
- Perez-Rivas LG, Theodoropoulou M, Ferraù F, Nusser C, Kawaguchi K, Stratakis CA, Faucz FR, Wildemberg LE, Assié G, Beschorner R, Dimopoulou C, Buchfelder M, Popovic V, Berr CM, Tóth M, Ardisasmita AI, Honegger J, Bertherat J, Gadelha MR, Beuschlein F, Stalla G, Komada M, Korbonits M, Reincke M. The gene of the ubiquitin-specific protease 8 is frequently mutated in adenomas causing Cushing’s disease. J Clin Endocrinol Metab. 2015;100:E997–1004.PubMedPubMedCentralCrossRefGoogle Scholar
- Reincke M, Sbiera S, Hayakawa A, Theodoropoulou M, Osswald A, Beuschlein F, Meitinger T, Mizuno-Yamasaki E, Kawaguchi K, Saeki Y, Tanaka K, Wieland T, Graf E, Saeger W, Ronchi CL, Allolio B, Buchfelder M, Strom TM, Fassnacht M, Komada M. Mutations in the deubiquitinase gene USP8 cause Cushing’s disease. Nat Genet. 2015;47:31–8.PubMedCrossRefGoogle Scholar
- Theodoropoulou M, Arzberger T, Gruebler Y, Jaffrain-Rea ML, Schlegel J, Schaaf L, Petrangeli E, Losa M, Stalla GK, Pagotto U. Expression of epidermal growth factor receptor in neoplastic pituitary cells: evidence for a role in corticotropinoma cells. J Endocrinol. 2004;183:385–94.PubMedCrossRefGoogle Scholar
- Valassi E, Santos A, Yaneva M, Tóth M, Strasburger CJ, Chanson P, Wass JA, Chabre O, Pfeifer M, Feelders RA, Tsagarakis S, Trainer PJ, Franz H, Zopf K, Zacharieva S, Lamberts SW, Tabarin A, Webb SM, ERCUSYN Study Group. The European registry on Cushing’s syndrome: 2-year experience baseline demographic and clinical characteristics. Eur J Endocrinol. 2011;165:383–92.PubMedCrossRefGoogle Scholar