Julie D.R. Reimann

The lore of the RINGs: substrate recognition and catalysis by ubiquitin ligases

Recently, many new examples of E3 ubiquitin ligases or E3 enzymes have been found to regulate a host of cellular processes. These E3 enzymes direct the formation of multiubiquitin chains on specific protein substrates, and - typically - the subsequent destruction of those proteins. We discuss how the modular architecture of E3 enzymes connects one of two distinct classes of catalytic domains to a wide range of substrate-binding domains. In one catalytic class, a HECT domain transfers ubiquitin directly to substrate bound to a non-catalytic domain. Members of the other catalytic class, found in the SCF, VBC and APC complexes, use a RING finger domain to facilitate ubiquitylation. The separable substrate-recognition domains of E3 enzymes provides a flexible means of linking a conserved ubiquitylation function to potentially thousands of ubiquitylated substrates in eukaryotic cells.

Emi1 is a mitotic regulator that interacts with Cdc20 and inhibits the anaphase promoting complex.

We have discovered an early mitotic inhibitor, Emi1, which regulates mitosis by inhibiting the anaphase promoting complex/cyclosome (APC). Emi1 is a conserved F box protein containing a zinc binding region essential for APC inhibition. Emi1 accumulates before mitosis and is ubiquitylated and destroyed in mitosis, independent of the APC. Emi1 immunodepletion from cycling Xenopus extracts strongly delays cyclin B accumulation and mitotic entry, whereas nondestructible Emi1 stabilizes APC substrates and causes a mitotic block. Emi1 binds the APC activator Cdc20, and Cdc20 can rescue an Emi1-induced block to cyclin B destruction. Our results suggest that Emi1 regulates progression through early mitosis by preventing premature APC activation, and may help explain the well-known delay between cyclin B/Cdc2 activation and cyclin B destruction.

Prophase Destruction of Emi1 by the SCFβTrCP/Slimb Ubiquitin Ligase Activates the Anaphase Promoting Complex to Allow Progression beyond Prometaphase

Progression through mitosis occurs because cyclin B/Cdc2 activation induces the anaphase promoting complex (APC) to cause cyclin B destruction and mitotic exit. To ensure that cyclin B/Cdc2 does not prematurely activate the APC in early mitosis, there must be a mechanism delaying APC activation. Emi1 is a protein capable of inhibiting the APC in S and G2. We show here that Emi1 is phosphorylated by Cdc2, and on a DSGxxS consensus site, is subsequently recognized by the SCF(betaTrCP/Slimb) ubiquitin ligase and destroyed, thus providing a delay for APC activation. Failure of betaTrCP-dependent Emi1 destruction stabilizes APC substrates and results in mitotic catastrophe including centrosome overduplication, potentially explaining mitotic deficiencies in Drosophila Slimb/betaTrCP mutants. We hypothesize that Emi1 destruction relieves a late prophase checkpoint for APC activation.

E2F-dependent accumulation of hEmi1 regulates S phase entry by inhibiting APC(Cdh1).

Emi1 promotes mitotic entry in Xenopus laevis embryos by inhibiting the APCCdc20 ubiquitination complex to allow accumulation of cyclin B. We show here that human Emi1 (hEmi1) functions to promote cyclin A accumulation and S phase entry in somatic cells by inhibiting the APCCdh1 complex. At the G1–S transition, hEmi1 is transcriptionally induced by the E2F transcription factor, much like cyclin A. hEmi1 overexpression accelerates S phase entry and can override a G1 block caused by overexpression of Cdh1 or the E2F-inhibitor p105 retinoblastoma protein (pRb). Depleting cells of hEmi1 through RNA interference prevents accumulation of cyclin A and inhibits S phase entry. These data suggest that E2F can activate both transcription of cyclin A and the hEmi1-dependent stabilization of APCCdh1 targets, such as cyclin A, to promote S phase entry.

Emi1 is required for cytostatic factor arrest in vertebrate eggs

Vertebrate eggs are arrested at metaphase of meiosis II with stable cyclin B and high cyclin B/Cdc2 kinase activity. The ability of the anaphase-promoting complex/cyclosome (APC), an E3 ubiquitin ligase, to trigger cyclin B destruction and metaphase exit is blocked in eggs by the activity of cytostatic factor (CSF) (reviewed in ref. 1). CSF was defined as an activity in mature oocytes that caused mitotic arrest when injected into dividing embryos. Fertilization causes a transient increase in cytoplasmic calcium concentration leading to CSF inactivation, APC activation, cyclin B destruction and mitotic exit. The APC activator Cdc20 is required for APC activation after fertilization. We show here that the APCcdc20 inhibitor Emi1 (ref. 6) is necessary and sufficient to inhibit the APC and to prevent mitotic exit in CSF-arrested eggs. CSF extracts immunodepleted of Emi1 degrade cyclin B, and exit from mitosis prematurely in the absence of calcium. Addition of Emi1 to these Emi1-depleted extracts blocks premature inactivation of the CSF-arrested state. Emi1 is required to arrest unfertilized eggs at metaphase of meiosis II and seems to be the long-sought mediator of CSF activity.

The Evi5 Oncogene Regulates Cyclin Accumulation by Stabilizing the Anaphase-Promoting Complex Inhibitor Emi1

The anaphase-promoting complex/cyclosome (APC/C) inhibitor Emi1 controls progression to S phase and mitosis by stabilizing key APC/C ubiquitination substrates, including cyclin A. Examining Emi1 binding proteins, we identified the Evi5 oncogene as a regulator of Emi1 accumulation. Evi5 antagonizes SCF(betaTrCP)-dependent Emi1 ubiquitination and destruction by binding to a site adjacent to Emi1s DSGxxS degron and blocking both degron phosphorylation by Polo-like kinases and subsequent betaTrCP binding. Thus, Evi5 functions as a stabilizing factor maintaining Emi1 levels in S/G2 phase. Evi5 protein accumulates in early G1 following Plk1 destruction and is degraded in a Plk1- and ubiquitin-dependent manner in early mitosis. Ablation of Evi5 induces precocious degradation of Emi1 by the Plk/SCF(betaTrCP) pathway, causing premature APC/C activation; cyclin destruction; cell-cycle arrest; centrosome overduplication; and, finally, mitotic catastrophe. We propose that the balance of Evi5 and Polo-like kinase activities determines the timely accumulation of Emi1 and cyclin, ensuring mitotic fidelity.

Carcinoma Showing Thymus-like Differentiation of the Thyroid (castle): A Comparative Study: Evidence of Thymic Differentiation and Solid Cell Nest Origin

Carcinoma showing thymus-like differentiation (CASTLE) is a rare intrathyroidal neoplasm, a member of a tumor family probably arising from ectopic thymus or branchial pouch remnants. Thyroid solid cell nests (SCNs) may also be derived from branchial pouch remnants. SCNs express p63, carcinoembryonic antigen (CEA), and high molecular weight keratin (HMWK). To determine whether CASTLE and SCNs derive from similar embryologic origins/lines of differentiation, and to better differentiate CASTLE from other thyroid neoplasms, we compared p63, CD5, HMWK, and CEA staining of CASTLE and SCNs with other thyroid and thymic lesions. Seven CASTLE, 11 SCNs, 10 thymic carcinoma, 11 invasive thymoma, 12 thymoma, 28 papillary thyroid carcinoma, 4 thyroid squamous cell carcinoma, 2 childhood sclerosing carcinoma, 4 follicular adenoma, 6 follicular carcinoma, 4 poorly differentiated carcinoma, and 20 lymphocytic thyroiditis cases were analyzed. In normal thyroid, only SCNs stained for p63, HMWK, and CEA. The only CD5-positive cells in normal thyroid were T cells. Thymomas and normal thymus stained similarly to SCNs. All CASTLE and thymic carcinomas exhibited diffuse p63 and HMWK staining and all CASTLE cases and the majority of thymic carcinomas were positive for CEA and CD5. In contrast, none of the other thyroid neoplasms examined exhibited consistent staining for all 4 markers studied. These findings provide further evidence that CASTLE is distinct from other thyroid neoplasms, is probably of thymic origin, and may arise from branchial pouch remnants, the thyroid SCNs. Moreover CD5, HMWK, CEA and p63 can be used to help distinguish CASTLE from other thyroid neoplasms.

Identification of HRAS mutations and absence of GNAQ or GNA11 mutations in deep penetrating nevi

HRAS is mutated in ∼15% of Spitz nevi, and GNAQ or GNA11 is mutated in blue nevi (46–83% and ∼7% respectively). Epithelioid blue nevi and deep penetrating nevi show features of both blue nevi (intradermal location, pigmentation) and Spitz nevi (epithelioid morphology). Epithelioid blue nevi and deep penetrating nevi can also show overlapping features with melanoma, posing a diagnostic challenge. Although epithelioid blue nevi are considered blue nevic variants, no GNAQ or GNA11 mutations have been reported. Classification of deep penetrating nevi as blue nevic variants has also been proposed, however, no GNAQ or GNA11 mutations have been reported and none have been tested for HRAS mutations. To better characterize these tumors, we performed mutational analysis for GNAQ, GNA11, and HRAS, with blue nevi and Spitz nevi as controls. Within deep penetrating nevi, none demonstrated GNAQ or GNA11 mutations (0/38). However, 6% revealed HRAS mutation (2/32). Twenty percent of epithelioid blue nevi contained a GNAQ mutation (2/10), while none displayed GNA11 or HRAS mutation. Eighty-seven percent of blue nevi contained a GNAQ mutation (26/30), 4% a GNA11 mutation (1/28), and none an HRAS mutation. Within Spitz nevi, none demonstrated GNAQ or GNA11 mutations (0/30). Seventeen percent contained an HRAS mutation (5/30). All GNAQ and GNA11 mutations were p.Q209L (c.626A>T) point mutations, except 2 GNAQ mutations, which contained novel c.625_626CA>TT double mutations. Four HRAS mutations were in exon 2, and three in exon 3. This is the first study to identify HRAS mutations in deep penetrating nevi. The presence of HRAS mutations and absence of GNAQ or GNA11 mutations in deep penetrating nevi suggests classification of these unusual nevi within the Spitz nevus category of melanocytic tumors, rather than the blue nevus category.

Human papillomavirus-associated adenocarcinoma of the base of the tongue.

Human papillomavirus (HPV) is a major cause of oropharyngeal squamous cell carcinoma with characteristic clinical and pathologic features relative to their non-HPV-associated counterparts. Here we describe 2 cases of HPV-associated adenocarcinoma of the oropharynx. Both cases arose at the base of the tongue, and neither had the histologic or immunohistochemical features of a primary salivary gland tumor or metastasis from another location. One patient had metastases to neck lymph nodes and the lungs and died of disease 37 months after diagnosis. Evidence for an HPV association consisted of strong diffuse expression of p16, polymerase chain reaction-based detection of HPV16 DNA sequences, and localization of HPV by in situ hybridization within tumor cells of both primary and metastatic lesions. These results further expand the spectrum of HPV-associated head and neck malignancy. This rare entity should be distinguished from primary salivary gland adenocarcinoma and may be a candidate for HPV-specific targeted therapies.

Importance of universal mismatch repair protein immunohistochemistry in patients with sebaceous neoplasia as an initial screening tool for Muir-Torre syndrome☆☆☆

Muir-Torre syndrome, a Lynch syndrome variant, is characterized by sebaceous neoplasia plus one or more malignancies, typically colon cancer. The significance of DNA mismatch repair (MMR) deficiency detection by immunohistochemistry (IHC) in colorectal carcinomas is well established and is recommended as a screening tool for Lynch syndrome in newly diagnosed colorectal carcinomas. In comparison, literature on IHC application to detect MMR proteins (MLH1, MSH2, MSH6, and PMS2) in sebaceous neoplasia has been less studied and has been derived almost exclusively from tertiary care centers. Herein we describe the largest series to date characterizing MMR deficiency in sebaceous neoplasms, as well as the relative frequencies of each deficiency. Two hundred sixteen consecutive sebaceous neoplasms (216 patients) were analyzed from a community practice setting (133 sebaceous adenomas, 68 sebaceomas, 15 sebaceous carcinomas). One hundred forty-three were MMR deficient (66%), of which 90 were MSH2/MSH6 deficient (63%), 27 MLH1/PMS2 deficient (19%), 22 MSH6 deficient (15%), and 4 PMS2 deficient (3%). MMR deficiency was significantly associated with site, with tumors off of the head and neck more likely to be MMR deficient (specificity 96%). In contrast to prior reports, no significant trend in MMR-deficient versus -nondeficient tumors was seen in age at presentation (median age, 68 versus 66), tumor-infiltrating lymphocytes, or tumor type. Given the low sensitivity of age < 60 years (30%), location off of the head and neck (41%), or presence of tumor-infiltrating lymphocytes (29%) in MMR deficiency detection, IHC screening programs should test all sebaceous neoplasms for MMR deficiency, regardless of their clinicopathological features.