pathogenesis of nut midline carcinomapm07ch10-french ari 10 october 2011 15:29 table 2 initial...

PM07CH10-French ARI 10 October 2011 15:29

RE V I E W

S

IN

AD V A

NC

E

Pathogenesis of NUTMidline CarcinomaChristopher A. FrenchDepartment of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts 02215;email: [email protected]

Annu. Rev. Pathol. Mech. Dis. 2012. 7:247–65

The Annual Review of Pathology: Mechanisms ofDisease is online at pathol.annualreviews.org

This article’s doi:10.1146/annurev-pathol-011811-132438

Copyright c© 2012 by Annual Reviews.All rights reserved

1553-4006/12/0228-0247$20.00

Keywords

BRD4-NUT, BET inhibitor, histone deacetylase inhibitor, histoneacetyl-transferase, http://www.NMCRegistry.org, differentiationtherapy

Abstract

NUT midline carcinoma (NMC), an aggressive form of squamous cellcarcinoma, is defined by the presence of acquired chromosomal re-arrangements involving NUT, usually BRD4-NUT fusion genes and,less commonly, NUT-variant fusion genes involving BRD3 or still-uncharacterized genes. Improved diagnostic tests reveal that althoughrare, NMCs occur in people of any age and may be indistinguishablefrom more common squamous cell carcinomas of adulthood. NMCshave simple karyotypes whose hallmark is genomic instability, suggest-ing that NMC arises through a distinct pathogenic pathway represent-ing a genetic shortcut to the phenotype of squamous cell carcinoma.Mechanistically, BRD-NUT fusion proteins appear to act by blockingdifferentiation, possibly by sequestering histone aceyltransferase activ-ity. Accordingly, histone deacetylase inhibitors or BET inhibitors, thelatter of which inhibit binding of BRD-NUT proteins to chromatin,induce terminal differentiation of NMC cells. These insights providea rationale for targeted therapy of NMC, which is almost uniformlyrefractory to conventional chemotherapy and radiotherapy.

247


NMC: NUT midlinecarcinoma

Carcinoma:malignant neoplasm ofepithelial cells

Chromosomalrearrangement:structural aberrationof a chromosome;includes deletions,duplications,inversions, andtranslocations

Ewing sarcoma:a rare, malignant,round-cell tumorarising within bone orin soft tissue

Squamous: literally,flat or scale-like;describes, for instance,the epidermis of skin,which is composed ofsquamous epithelium

Epigenetic: refers toheritable alterations ingene expression causedby mechanisms otherthan DNA sequencechanges; histoneacetylation andmethylation arewell-known examples

CLINICAL FEATURES OF NUTMIDLINE CARCINOMA

NUT midline carcinoma (NMC) is a rare,aggressive cancer. It can affect any gender orage group (Table 1) and occurs equally inmales and females from the neonatal period (1)through the eighth decade of life (2). The cellof origin is not known: No in situ epithelialcomponent has ever been observed, and thetumor is usually locally invasive and widelymetastatic at diagnosis. Because NMCs arisefrom various anatomical sites (Table 1), theycannot be categorized by the same systemof nomenclature as most other carcinomas,which are traditionally defined by the tissue oforigin. Instead, NMC is defined genetically bythe presence of chromosomal rearrangementsinvolving the NUT gene (3). The term midlineis applied because of NMC’s tendency to arisefrom midline anatomical sites, most commonlythe upper aerodigestive tract (50%) and the me-diastinum (41%) (Table 1) (4). However, it hasbeen diagnosed within such varied tissues as theparotid gland (5, 6), pancreas (1), adrenal gland(M. O’Sullivan & C.A. French, unpublisheddata), subcutis (C.A. French & E. Salman, un-published data), bladder (3), and iliac bone (7).

Apart from these unique features, NMC isbest known for its devastating clinical course.Of the 62 cases known to this author (C.A.French, unpublished observations), only onepatient has been cured (7). NMC is ofteninitially responsive to chemotherapy and ra-diation, but it invariably recurs rapidly anddoes not respond to subsequent therapeuticinterventions. The average survival is only9.5 months (4), despite its frequent occur-rence in previously healthy children or youngadults without comorbid conditions. The onecured case was in many ways atypical. It arosewithin the iliac bone of a 10-year-old boy andwas pathologically and clinically indistinguish-able from Ewing sarcoma, except that it hadthe BRD4-NUT translocation and lacked EWSrearrangement. It is possible, therefore, thatthis singular case was a Ewing sarcoma asso-ciated with BRD4-NUT rearrangement. In line

with this possibility, several subsequent typicalNMCs were treated with Ewing sarcoma regi-mens, but they did not show any improved ther-apeutic response (8).

CYTOGENETICS

A unique feature of NMC is its cytogenet-ics. Most carcinomas, including typical squa-mous cell carcinomas of adulthood, havevery complex karyotypes and are aneuploid(Figure 1) (9). NMC, by contrast, usuallyhas a single chromosomal translocation [clas-sically, a reciprocal t(15;19)(q14;p13.1)] andfew, if any, other cytogenetic abnormalities(Figure 1). The cytogenetic features ofNMCs and their occurrence even in veryyoung children suggest that, as is the casefor translocation-associated hematopoietic can-cers, the genesis of NMC may involve relativelyfew mutations. This finding stands in contrastto mutagen-associated squamous cell carcino-mas of adulthood, which arise over a periodof many years and are associated with theaccumulation of thousands to tens of thousandsof mutations (10).

The basis for this genetic distinction isunclear. In hematopoietic tumors, the cell oforigin is often a stem-like cell with the capac-ity to circulate widely between tissues and theblood (11); thus, spread from tissue to tissue isan ingrained property of most hematopoietictumors. In contrast, acquisition of the abilityof carcinomas to invade tissue and metastasizemay require a complex series of genetic and epi-genetic changes (12, 13), which may be drivenby a high level of genomic instability. NMCmay bypass the requirement for genomic insta-bility due to origination from a stem-like cellwith migratory properties or because NUT fu-sion proteins modify the epigenome of the cellof origin such that it acquires these properties.

MOLECULAR GENETICS

Approximately two-thirds of NMCs have a re-ciprocal chromosomal translocation involvingNUT (nuclear protein in testis, chr15orf55) on

248 French


Table 1 Clinical characteristics of reported NUT midline carcinomasa

Casenumber

Age(years) Sex Location Diagnosis

Survival(months)

BRD4-NUT

NUTvariant Reference(s)

1 3 M Bladder PD squamous cellcarcinoma

8.5 Present Absent 3

2 5 M Mediastinum Mucoepidermoidcarcinoma


3 10 M Iliac bone Ewing sarcoma/ PNET AWOD180b

Present Absent 7

4 11 F Thorax Undifferentiatedcarcinoma


5 12 F Nasopharynx PD squamous cellcarcinoma

3.25 Present Absent 17, 23

6 13 F Epiglottis Undifferentiatedcarcinoma

9 Present Absent 23

7 15 M Mediastinum Thymic PD squamouscell carcinoma

6 Present Absent 24

8 15 F Orbit PD squamous cellcarcinoma

7 Present Absent 3

9 16 M Lung Squamous cellcarcinoma

37 Absent Presentc 3, 16

10 16 F Trachea PD carcinoma 31.5 Absent Present 311 21 F Nasopharynx Nasopharyngeal

carcinoma10.25 Absent Present 3

12 22 F Thymus PD carcinoma 3.5 Present Absent 2013 26 M Sinonasal Undifferentiated

carcinoma16.75 Present Absent 3

14 30 F Mediastinum PD carcinoma 3.25 Present Absent 815 31 M Nasal cavity Sinonasal

undifferentiatedcarcinoma

— Present Absent 2

16 34 F Thorax NK NK Present Absent Dang et al. 200017 34 M Mediastinum PD carcinoma 2 Absent Presentc 1618 35 F Mediastinum PD carcinoma 2 Present Absent 319 39 F Nasal cavity

and sinusPD squamous cellcarcinoma


20 40 F Nasal cavityand sinus

PD squamous cellcarcinoma


21 47 M Nasal cavityand sinus

Sinonasalundifferentiatedcarcinoma


22 78 F Supraglotticlarynx

Undifferentiatedcarcinoma


Average 25.14 — — — 9.53 — — —

aData used with permission from Reference 4.bNot included in average survival.cThese variants are BRD3-NUT. Abbreviations: AWOD, alive without disease; PD, poorly differentiated; NK, not known; PNET, primitiveneuroectodermal tumor.

www.annualreviews.org • Pathogenesis of NUT Midline Carcinoma 249


1 12 23 3

6 67 78 89 910 1011 1112 12

1313

1414

1515

1616

1717

1818

19 1920 2021 2122 22 X YX

4 45 5

Squamous cell carcinoma NUT midline carcinoma

Figure 1(Left) Karyotypes of typical non–NUT midline carcinoma (NMC) squamous cell carcinoma, compared with (right) NMC. The redarrows denote chromosomal translocations between chromosomes 15 and 19. Left panel adapted with permission from Dr. SusanGollin and Elsevier, Ltd. (9).

Bromodomain andextraterminaldomain–containingprotein (BETprotein): bindsacetylated lysineresidues; often termeda reader

Oncogene: a genethat can cause cancer

Histoneacetyltransferase(HAT): an enzyme,often termed a writer,that acetylates histoneresidues, mostcommonly lysine

Differentiation:a maturation processwhereby a lessspecialized cellbecomes a morespecialized cell

chromosome 15q14 and the BET family geneBRD4 (also known as MCAP and HUNK1) onchromosome 19p13.1, which forms an in-frameBRD4-NUT fusion oncogene whose product isdriven by the BRD4 promoter (Figures 1 and2). The BRD4-NUT fusion gene contains thefirst half of BRD4, which includes all of itsknown functional domains: two paired bromod-omains, an extraterminal (ET) domain, and abipartite nuclear localization signal (NLS). TheNUT portion of the fusion gene contains es-sentially the entire coding region for NUT,a largely unstructured polypeptide highly ex-pressed in normal spermatocytes that has anacidic binding domain for the histone acetyl-transferase (HAT) p300 (14). Although thet(15;19) is reciprocal, the other product of thetranslocation, NUT-BRD4, is not expressed, inaccordance with the testis-restricted activity ofthe NUT promoter (15).

Less commonly, NMC harbors a differentrearrangement involving NUT (Figure 2) (3). Asubset (approximately 25%) of these cases haveBRD3-NUT rearrangements (16). The BRD3-NUT fusion gene also drives the expressionof a chimeric protein consisting of the two

bromodomains, an ET domain, and NLSof BRD3, fused to the entirety of NUT(Figure 2). The remaining genes involved inother NUT-variant rearrangements are uniden-tified; it will be of interest to determine whetherthese genes encode proteins that serve simi-lar functions or whether they are part of thesame functional complex as BRD4 and BRD3(described below). On the basis of outcomes ina small number of cases, it has been suggested(3) that NUT-variant NMCs may be associatedwith longer survival.

HISTOPATHOLOGY

The histopathology of NMC is characteristic,but not diagnostic. The most common appear-ance is that of a poorly differentiated carcinomawith focal, abrupt squamous differentiation(Figure 3). In contrast to many other poorlydifferentiated carcinomas, which consist ofhighly pleomorphic large cells (Figure 3),NMC cells are usually medium sized, round,and often monomorphic in appearance. Overtareas of squamous differentiation are seen inapproximately half of cases (3; C.A. French,

250 French


Bromodomain

ET domain

NLS

AD1

AD2

NES

N

N

Chromosome 19p13.1

Chromosome 9q34.2

Chromosome 15q14 N

N

N BRD4

BRD3

NUT

BRD3-NUT

BRD4-NUT

Figure 2Domain structures encoded by BRD-NUT fusions and native component genes. The black arrows indicate the locations oftranslocation-associated break points. Abbreviations: AD1, acidic domain 1; AD2, acidic domain 2; ET, extraterminal; NES, nuclearexport signal; NLS, nuclear localization signal.

unpublished observations) but may not alwaysbe present, particularly in small biopsies. Thehistopathologic features of NMC overlap withthose of several other poorly differentiated

cancers (Table 2), including poorly differ-entiated squamous cell carcinoma, sinonasalundifferentiated carcinoma (2), Ewing sarcoma,Epstein-Barr virus–associated nasopharyngeal

BRD4-NUT BRD4-NUT BRD4-NUTBRD4-NUT

BRD4-NUT

BRD3-NUT Non-NMC Non-NMC

NUT- variant NUT- variantBRD4-NUT

BRD3-NUT Non-NMC Non-NMC

NUT- variant NUT- variant

BRD4-NUT BRD4-NUT

Figure 3Histomorphology of NUT midline carcinomas (NMCs) and non-NMCs. Slides are stained with hematoxylinand eosin. 400× magnification is identical for all images. Adapted with permission from Reference 4.



Table 2 Initial diagnoses in cases of NUT midline carcinomas

DiagnosisNumber of

casesSquamous cell carcinoma 12Poorly differentiated carcinoma 8Sinonasal undifferentiated carcinoma 3Malignant neoplasm, not otherwise specified 2Thymic carcinoma 2Ewing sarcoma 2Leukemia 2Neuroblastoma 1Pancreatoblastoma 1Carcinoma expleomorphic adenoma 1Nasopharyngeal carcinoma 1Total 35

Leukemia: malignantneoplasm ofhematopoietic cells(originating from bonemarrow, lymph node,and thymus) thatcirculates in the blood

carcinoma, thymic carcinoma, neuroblastomaand pancreatoblastoma (1), and even primarysalivary gland carcinoma (5).

NMC is commonly misdiagnosed; ithas even been mistaken for acute leukemia(due to the occasional expression of CD34)(Table 2). Also contributing to the failure to di-agnose NMC are its rarity, physicians’ failure toconsider it in adult patients, and until recently,the absence of widely available diagnostictests (described in the following section). Tocounter the lack of awareness of NMC amongpathologists and oncologists, an Internet-basedinternational NMC registry (http://www.NMCRegistry.org) has been established.This registry provides access to (a) pathologicreview, (b) updated information about the dis-ease, (c) treatment guideline suggestions, (d ) arepository for clinical data, and (e) educationalinformation for physicians and patients aboutthis disease.

DIAGNOSIS

The diagnosis of NMC depends on thedemonstration of NUT gene rearrangementor misexpression. Although cytogeneticdemonstration of a t(15;19) by karyotyping issufficient for a presumptive diagnosis of NMC(Figure 1), most diagnoses are based on

formalin-fixed, paraffin-embedded (FFPE)tissues because few suspected carcinomasare submitted for karyotyping. To fill thisdiagnostic need, we developed a fluorescencein situ hybridization assay that uses dual-color,break-apart bacterial artificial chromosomeprobes flanking the NUT and BRD4 loci(Figure 4a) (15, 17, 18). This assay can be usedon virtually any specimen preparation (exceptfor those with heavy-metal fixation), includingfrozen tissue, acetic acid–fixed cytogeneticpreparations, thin (4–5-μm) sections of FFPEtissues or disaggregated cells, and air-driedor ethanol-fixed slides. Because this approachworks on archival, formalin-fixed tissue, retro-spective analysis can be performed on samplesthat are decades old (3).

If fresh or frozen tissue is available, afaster approach, reverse-transcriptase poly-merase chain reaction, can be employed by useof primers flanking the known BRD4 and NUTbreak points (Figure 4b) (8). However, thismethod overlooks NUT-variant fusion genes.

To develop a diagnostic test for NMCthat can be used routinely in the commu-nity, we raised and characterized a monoclonalantibody to NUT (clone C52) that detectsNUT expression by immunohistochemistry(19). In a study that included a large number ofcontrol tissues, such as other forms of car-cinoma, nuclear reactivity with this antibodywas 100% specific and 87% sensitive for thediagnosis of NMC. Interestingly, the nuclearNUT staining appeared speckled in most cases(Figure 4c), which may have implicationsfor the function of BRD-NUT, as discussedfurther below.

DEMOGRAPHICS ANDPREVALENCE

Improvements in diagnosis have led to in-creased recognition of NMC, but its true preva-lence remains unknown. Notably, the antibodystudy described above (19) revealed that vir-tually none of the control tumors screenedon tissue microarrays (TMAs) (N = 674)were discovered to be NMCs. However, most

252 French

http://www.NMCRegistry.org



BRD4

Direct sequencing

NUT

We

igh

t m

arke

r

Inn

er

con

tro

l

BRD4-NUT

EWSR

1-FLI1

EWSR

1-ERG

EWSR

1-WT1

BRD4-NUT

EWSR

1-FLI1

EWSR

1-ERG

EWSR

1-WT1

Present case Positive control

a b RT-PCR c

d e

d Non-NMC e NMC

Figure 4Various molecular methods used to diagnose NUT midline carcinoma (NMC). (a) Fluorescence in situhybridization using dual color probes flanking the NUT gene. The split-apart red and green signals indicaterearrangement of the NUT locus. Reproduced with permission from Reference 4. (b) (Left) Reverse-transcriptase polymerase chain reaction (RT-PCR) using primers that flank the coding sequence of theBRD4-NUT break point and appropriate controls. (Right) sequence of the product on the left. Reproducedcourtesy of Yukichi Tanaka, Mio Tanaka, Toru Horisawa, and Yutaka Saikawa of Kanazawa MedicalUniversity, Ishikawa, Japan. (c) Immunohistochemistry on formalin-fixed, paraffin-embedded sections ofcarcinoma with the anti-NUT antibody. Produced by Cell Signaling Technologies, Inc., Danvers,Massachusetts. Reproduced with permission from Reference 19.

tumors on TMAs are at least partially re-sectable, whereas most NMCs are unresectableat presentation; thus, TMAs may be biasedagainst inclusion of NMCs. Other screens havebeen intentionally biased toward tumor groupsthat are suspected of being enriched for NMCs.For example, screening of 98 visceral carcino-mas in young patients (average age, 32.5 years)identified seven NMCs (7%) (3), and a second

study that focused only on an uncommon groupof undifferentiated carcinomas of the upperaerodigestive tract (N = 28) identified fiveNMCs (18%) (2). Interestingly, four of thosefive patients were over the age of 40 years, andone patient was a 78-year-old who carried a di-agnosis of laryngeal squamous cell carcinoma.

The description of NMC in older pa-tients (2) and the development of an



05

1015202530354045

11-200-10 21-30 31- 40 41-50 51- 60 ≥61

%%

%

Age (years)

11-200-10 21-30 31- 40 41-50 51- 60 ≥61

Age (years)

11-200-10 21-30 31- 40 41-50 51- 60 ≥61

Age (years)

0

10

20

30

40

50

0

5

10

15

20

25

a Age distribution, all cases b Age distribution <2008

c Age distribution ≥2008 d

Figure 5Demographics of NUT midline carcinoma (NMC). (a–c) Age distribution of NMC. (d ) Distribution throughout the world of allknown NMC cases since 1991.

immunohistochemical test (19) have beenassociated with an increase in the number ofNMC cases diagnosed. This increase, in turn,has provided a more complete picture of thedemographics of NMC. For example, the firstdescriptions of this disease in children (3, 15,17, 20–23) led to the perception that NMC wasa pediatric cancer that affected mostly peoplein their first or second decade of life. In thecohort of patients that is known to us, thereremains a pediatric-weighted age distribution(Figure 5a), but analysis of cases before2008 (Figure 5b) versus those from 2008 on(Figure 5c) suggests that the prevalence ofNMC is more evenly distributed amongpatients of all ages.

Given the age distribution of NMC, it isunlikely that smoking plays a pathogenic role;this hypothesis is supported by the absence of asmoking history in most, but not all, NMC pa-tients (C.A. French, unpublished observations).None of the cases tested to date have beenassociated with Epstein-Barr virus or with hu-man papilloma virus infection (2, 5, 17, 23;

C.A. French, unpublished observations). NMChas been encountered in many parts of the de-veloped world (Figure 5d ) but is undoubt-edly underrecognized in China and many otherdeveloping countries.

BET PROTEINS

The critical initiating oncogenic event in NMCis hypothesized to be the formation of NUT fu-sion oncogenes, most commonly BRD4-NUT.BRD4 is a ubiquitously expressed memberof the family of double-bromodomain-containing, ET domain–containing BETproteins. BRD4 encodes two major isoforms.The short isoform is nearly entirely includedin BRD4-NUT, whereas the long isoformencodes a long C-terminal domain (Figure 1)that is not present in BRD4-NUT and is impli-cated in the binding of papilloma viruses to hostcell genomes (25). The paired bromodomainsof BRD3/BRD4, both of which are present inBRD-NUT fusion proteins (Figure 2), bindspecifically to acetylated histones associated

254 French


with DNA (26–28). BRD4, originally termedmitotic chromosome-associated protein (29),binds acetylated chromatin throughout allphases of the cell cycle, including mitosis. Thisproperty is one basis for the hypothesis thatBRD4 confers a sort of epigenetic memory topostmitotic daughter cells (29). The ability ofBRD4 to facilitate transcriptional elongationof early postmitotic genes, such as cyclinD1, through its interaction with the positivetranscriptional elongation factor P-TEFbsupports this idea (30–32). In the context ofNMC, the bromodomains of BRD3 and BRD4probably play an important role in tetheringNUT to chromatin (16). Notably, normalNUT shuttles between the cytoplasm andthe nucleus in a CRM1-dependent fashiondue to the presence of a nuclear export signal(Figure 2) (16), whereas BRD-NUT fusionproteins do not shuttle, instead remaining asso-ciated with chromatin throughout the cell cycle.

The ET domain is also present in BRD-NUT fusion proteins. It is structured (33)and binds the Kaposi sarcoma virus LANA-1antigen (34), suggesting that it functions as aprotein-protein interaction module.

NUT PROTEIN

NUT is a mammal-specific protein that isonly moderately conserved. Human NUT has66% amino acid identity with mouse NUT,compared with the 96% identity betweenhuman and mouse BRD4. NUT expression isnormally restricted to postmeiotic spermatids.It is predicted to be a largely unstructuredpolypeptide, but contains two acidic, potentialprotein-binding domains (ADs) (Figure 2).One of these domains, AD1, is contained withina region that binds and activates p300 (35).On the basis of this association, Reynoird et al.(14) suggested that within spermatids, NUTbinds and activates the HAT activity of p300,thereby leading to a global increase in histoneacetylation, an epigenetic change that precedesthe removal of histones and subsequent chro-matin compaction during spermatogenesis.When fused to BRD, NUT appears to retain

the capacity to bind p300, but it may do so ina way that impairs the expression of genes thatare needed to drive the terminal differentiationof squamous epithelial cells (14, 16, 18). Thisidea is discussed in more detail below.

BRD-NUT FUNCTION

BRD-NUT proteins are toxic when expressedin heterologous cell lines (36; C.A. French,unpublished observations), and standardcolony-forming assays (e.g., in NIH 3T3 cells)are not useful for the study of its transformingactivities. However, long-term cultures ofNMC cells obtained from patients are readilyestablished, thereby providing a substrate forstudies of BRD-NUT function. The strikingobservation is that knockdown of BRD4-NUTor BRD3-NUT induces rapid terminal squa-mous differentiation accompanied by arrestedproliferation (Figure 6) (16, 18). Followingknockdown, NMC cells begin to accumulateproteins associated with squamous differentia-tion, such as cytoplasmic keratin and involucrin,and demonstrate an increase in nuclear size andnuclear euchromatin. With additional time,these cells arrest their growth and terminallydifferentiate into flat, resting squamous cells(Figure 6). These responses to BRD-NUTwithdrawal suggest that one of the majorpathogenic consequences of BRD-NUTfusion proteins is to block differentiation,highlighting the therapeutic potential oftargeting these proteins.

Given the known association of BRD4-NUT with acetylated chromatin and p300,a logical question was whether its expres-sion alters histone acetylation globally. Theanswer is yes, but paradoxically, rather thanincreasing histone acetylation, BRD-NUTdecreases it, a change that is associated withdecreased expression of many genes (18).These findings suggested a model whereinBRD-NUT sequesters HAT activity or otherfactors that are required for expression of genesneeded for terminal squamous differentiation.The sequestration model is supported by theobservation that BRD-NUT localizes in a



Control Knockdown: 48 h Knockdown: 3 weeks

Figure 6Small interfering RNA (siRNA) knockdown of BRD4-NUT in patent-derived NUT midline carcinomacells using an siRNA directed against NUT (16), TC-797 (scale bar, 50 μm). Control consists of TC-797cells 72 h after transfection with scrambled siRNA (produced by Ambion, Austin, Texas). Reproduced withpermission from Reference 4.

HDACi: histonedeacetylase inhibitor

bromodomain-dependent fashion to discretenuclear foci (16) that are enriched for p300and acetylated histones (Figure 7a) (14), yettranscriptionally inactive (Figure 7b) (14).Whether BRD-NUT interferes with transcrip-tion by physically sequestering HAT activityfrom differentiation-specific genes or throughmore subtle mechanisms remains to be clari-fied. Note that there is a testis-specific BRDfamily member, BRDT, that has been impli-cated in chromatin compaction in spermatids,which suggests that BRD-NUT fusion proteinsmay in some way mimic the normal activitiesof BRDT and NUT in the testis (26, 37, 38).

Assuming that the major effect of BRD4-NUT proteins is on differentiation, it followsthat these oncoproteins are likely to collabo-rate with other, currently unknown mutationsthat increase growth and survival. The best-characterized and best-validated examples ofthis type of “oncogene conspiracy” are variouscombinations of transcription factor mutationsand activating tyrosine kinase mutations inacute leukemia (39–41). With the advent ofaffordable deep sequencing (42), it is antici-pated that the identification of collaboratingmutations that are required for the growthof NMC cells will soon become possible.

Such studies will rigorously assess the relativegenomic stability of NMCs (as compared withother squamous cell carcinomas) and may leadto additional therapeutic opportunities.

PATHOGENIC INSIGHTS FROMSMALL-MOLECULE INHIBITORS

A nonneoplastic process in which HAT seques-tration appears to play a role is Huntingtondisease, wherein expanded polyglutamine re-peats sequester CBP/p300, resulting in globalhistone hypoacetylation and transcriptionalrepression (43–46). In Huntington disease,transcription can be restored by treatment withhistone deacetylase inhibitors (HDACi) (44,46), which restore the overall balance betweenHAT and HDAC activity. On the basis ofthe sequestration model for BRD-NUTproteins, we hypothesized that treatment withHDACi may tip the balance in favor of HATactivity outside of BRD-NUT foci, therebyrestoring acetylation and transcription globally(Figure 8). In an initial test of this idea,we noted that treatment of cells transientlyexpressing green fluorescent protein (GFP)-BRD4-NUT with HDACi resulted inrapid dispersion of GFP-BRD4-NUT foci

256 French


BRD4-NUT H3K9ac Merge BRD4-NUT RNApolII Merge

BRD4-NUT CTD S2Ph Merge

BRD4-NUT CTD S5Ph Merge

BRD4-NUT H3K9ac Merge



COS

COS

Figure 7Immunofluorescence localization of BRD4-NUT and other proteins. (a) BRD4-NUT colocalizes with a number of acetylation marksassociated with transcriptionally active DNA, including histone 3 acetyl lysine 9 (H3K9ac), acetyl lysine 14 (H3K14ac), acetyl lysine 5(H3K5ac), and acetyl lysine 12 (H3K12ac). (b) BRD4-NUT localization relative to RNA polymerase II (RNApolII), and its activemarks: C-terminal domain phosphoserine 2 (CTD S2Ph) and C-terminal domain phosphoserine 5 (CTD S5Ph). Scale bars, 10 μm.Reproduced with permission from Reference 14.

(Supplemental Video 1; follow the Supple-mental Material link from the Annual Reviewshome page at http://www.annualreviews.org). We then evaluated whether redistribut-ing BRD-NUT proteins with HDACi couldovercome the differentiation block that is char-acteristic of NMC cells. Indeed, we found thatunlike non-NMC squamous cell carcinomalines, treatment of NMC cells with low concen-trations (25 nM) of the tool HDACi trichostatinA restored acetylation and led to the rapiddifferentiation and growth arrest of NMC cells(Figure 9a). The transcriptional profile and ex-pression of select, differentiation-specific genesinduced by HDACi closely mimicked thoseinduced by small interfering RNA knockdown

of BRD-NUT (18), suggesting that HDACiinterferes with BRD-NUT function andreverses its blockade on differentiation.

With an eye toward potential therapy forNMC with HDACi, a panel of structurally un-related HDACi was screened on multiple NMCcell lines; all were found to all be potently an-tiproliferative. Importantly, the antiprolifera-tive activity of these HDACi on NMC cellsclosely correlated with increases in nuclearacetylation (R2 = 0.96), which suggests thatthe inhibitory effects of the HDACi were on-target (18). Studies were then undertaken inthree murine xenograft models of NMC. Inall of these models, the investigational HDACiLBH-589 significantly inhibited growth and


http://www.annualreviews.org

http://www.annualreviews.org


AcAc AcAc AcAc

HATHAT

HAT HAT

HAT

HAT

HATHAT

HATHAT

HATHAT

HATHAT

HAT

HATHAT HAT

HAT

HAT

HATHAT

HAT

HAT

HAT

Co-F

Co-FCo-F Co-F

Co-FCo-FCo-F

Co-FCo-F

Co-FCo-F Co-F

Co-F

Co-F

Co-FCo-F

Co-F

Co-F

Co-F

Co-F

Co-FCo-F

HDAC HDAC

HDACHDAC

HAT

Co-FCo-F

Co-FCo-F

Co-FCo-F

Co-FCo-F

Co-F

DNA of unknown identity GeneHyperacetylated Hypoacetylated

AcAc

HATHATHAT

Co-F

Co-F

AcAc

AcAc

AcAc

HATHAT

HAT

HAT

HATHAT

HAT

HAT

HAT

Co-F

Co-F

Co-F

Co-F Co-F

Co-F

Co-F

Co-F

DNA of unknown identity GeneAcetylated

AcAc

Figure 8Proposed model of how BRD-NUT may repress transcription and block differentiation by sequesteringhistone acetyltransferases (HATs) and cofactors (Co-Fs). Abbreviation: HDAC, histone deacetylase.

Suberoylanilidehydroxamic acid(SAHA): a histonedeacetylase inhibitoralso known asvorinostat or Zolinza R©

induced differentiation (18) and, in extendedstudies, improved survival (Figure 9b).

During these investigations, a 10-year-oldmale with mediastinal BRD4-NUT NMCinvading the left atrium and pulmonaryvein was transferred to Children’s Hospi-tal Boston. Because NMC is refractory to

conventional therapies, treatment of thispatient with SAHA (vorinostat), the only clini-cally approved HDACi used to date in children,was considered. Culture of patient-derivedNMC cells confirmed that vorinostat inducedsquamous differentiation and growth arrest,and vorinostat given to the patient as a single

258 French


Survival

0 10 20 30 40 50 60 700

10

20

30

40

50

60

70

80

90

100

110

Day of treatment

Pe

rce

nt

surv

iva

l

p = 0.001

Pretreatment 5 weeks 10 weeks

b

c

EtOH TSAHTBa TC-797

EtOH TSA 1.8E+09

1.6E+09

1.4E+09

1.2E+09

1.0E+09

8.0E+08

6.0E+08

4.0E+08

2.0E+08

0.0E+000 10 20 30 40

Vehicle

LBH589

Vehicle

LBH589

LBH589 treatment of Per403 xenografts

Days of treatment

Bio

lum

ine

sce

nce

(p

ho

ton

s s–1

RO

I–1)

Figure 9Effect of histone deacetylase inhibition on NUT midline carcinoma (NMC) in vitro and in vivo. (a) Morphologic andimmunophenotypic changes consistent with differentiation (stained by hematoxylin and eosin and antikeratin AE1/AE3, respectively)and growth arrest (Ki-67 fraction) following trichostatin A (72 h, 25 nM) treatment of the NMC cell line TC-797 and the non-NMCsquamous cell carcinoma cell line HTB-43. Magnification (scale bar, 25 μm) is identical for all panels. (b) (Top) Growth inhibition ofxenograft model of NMC (PER-403) by the histone deacetylase inhibitor LBH-589. (Bottom) Survival analysis of PER-403 xenograft(n = 5 per group) treated with LBH589. (c) P18PF-fluorodeoxyglucose–positron emission tomography and computed tomographyscan of the patient’s mediastinal tumor (arrow). Reproduced with permission from Reference 18. Abbreviation: ROI, roipoly polygonalregion of interest.

APL: acutepromyelocyticleukemia

agent for five weeks produced an objectiveresponse, as assessed by positron emissiontomography (Figure 9c) (18). Unfortunately,SAHA therapy was complicated by markedthrombocytopenia, a well-known dose-limitingtoxicity (47), and severe nausea and emesis,which became intolerable. Withdrawal ofSAHA was followed by rapid recurrence oftumor growth, which was refractory to com-bination chemotherapy and led to the patient’sdeath 11 months after the initial diagnosis.

The glimmer of hope provided by the initialencouraging preclinical and clinical response to

HDACi suggests that NMC may be amenableto therapy designed to induce differentiation.The best example of successful differentiationtherapy at present is in the treatment of acutepromyelocytic leukemia (APL) with all-transretinoic acid (48–50), which overcomes a blockin differentiation induced by the PML-RARα

fusion oncoprotein. However, further work isneeded to understand in detail how BRD-NUTblocks differentiation and how HDACi over-comes this blockade.

A second lesson from the experience in APLis that targeting of an oncoprotein through



BET INHIBITORS IN INFLAMMATION

BETi show promise as anti-inflammatory agents (57). In thisstudy, investigators demonstrated that disruption of BRD-mediated transcription specifically interfered with the transcrip-tion of inflammation-associated genes in activated macrophages,thereby preventing the sequelae of bacterial sepsis.

several different approaches yields betterresponses. In APL, this has been achieved bycombining retinoids with arsenic salts, whichpromote the degradation of the PML-RARα

fusion protein (48–50). With this finding inmind, we considered other ways to target BRD-NUT. Most effective antiproliferative small-molecule inhibitors bind to and competitively

inhibit highly structured, catalytic componentsof enzymatic oncoproteins, such as the tyrosinekinases c-Kit (51), EGFR (52), BCR-ABL (53),and B-Raf (52, 54). The bromodomains andET domain of BRD4 are highly structured(Figure 2), which prompted Bradner andcoworkers (55) to test a new class of inhibitorsthat bind competitively to the acetyl-histonebinding pocket of BET-family protein bro-modomains (i.e., BRD3, BRD4, BRDT)(Figure 10a). These BET inhibitors (BETi;see the sidebar) caused rapid dissolution ofBRD4-NUT nuclear speckles, presumably byinterfering with its ability to bind acetylatedchromatin (14), and resulted in rapid differ-entiation and growth arrest of cultured NMCcells (Figure 10b). This prodifferentiativeand antiproliferative activity of BETi was

a b

c

797 (keratin)

t = 0 t = 48 h

(+)-JQ1

O ON N

N

NS

Cl

Tu

mo

r v

olu

me

(m

m3

)

Pe

r ce

nt

surv

iva

l

Day of treatment Day of treatment

2,000

1,500

1,000

500

0

100

50

P < 0.0001

P < 0.00010

0 10 20 300 10 20 30

–10 kT e –1

+10 kT e –1

Figure 10BET inhibitor (BETi) structure and activity in NUT midline carcinoma (NMC). (a) (Left) Structure of the BETi JQ1, an activeenantiomer that binds directly to the active site of BRD4, competitively with the acetylated histone tail. (Right) A space-fillingrepresentation of the high-resolution crystal structure of the first bromodomain of BRD4 bound to JQ1 ( yellow). (b) Differentiation ofNMC cells cultured in vitro with JQ1 over 48 h, as demonstrated by immunocytochemistry with an antikeratin (AE1/AE3) antibody.Images are shown at identical magnifications. (c) Tumor volume and survival are shown over time for cohorts of NMC tumor-bearingmice treated with JQ1 (50 mg kg−1 daily for 18 days) or vehicle. Adapted from Reference 55.

260 French


also observed in vivo in two NMC xenograftmodels (Figure 10c), demonstrating thepotential of BETi as a targeted NMC therapy(55). The prodifferentiative effect of BETi,as with HDACi and BRD-NUT knockdown,indicates that it is also likely to interfere withBRD-NUT function, but how is unclear. Inparticular, BETi are not expected to decreasethe association of BRD4-NUT with p300;if this hypothesis proves true, it throws thesequestration model of BRD-NUT functioninto question and suggests that other modesof action will need to be considered. In thisregard, it has not yet been determined whetherthe effect of BETi on NMC cells is purely dueto interference with BRD4-NUT, or whetherinhibition of normal BRD4 (expressed from anintact allele) also plays a role. If BRD4 inhi-bition does contribute to the response, NMCcells may be more sensitive than non-NMCcells or normal tissue because these cells arehaploinsufficient for BRD4.

Another potential drawback of BETi ther-apy is that currently available molecules of thistype bind all members of the BET family, in-cluding BRD2, BRD3, BRD4, and BRDT (55).Despite this concern about “off-BRD4-NUT”activity, however, initial studies with BETi haveproduced no overt toxicity in mice (55).

As in APL, in which arsenic salts andretinoids are a very potent and effective ther-apeutic combination, it will be of interest todetermine the effects of combined HDACi andBETi therapy. Lessons from other cancers havetaught us that such combination therapies willprobably be essential if NMC is to be treated ef-fectively. For example, even APL, whose fusiononcoprotein, PML-RARα, is targeted with all-trans retinoic acid, cannot be cured without theaddition of other agents such as arsenic (48–50,56). In the near future, additional molecular in-sight into the pathogenesis of NMC will hope-fully yield new clinical trials of rational targetedinhibitors.

SUMMARY POINTS

1. NMC is a genetically defined, highly aggressive, and incurable squamous cell carci-noma associated with chromosomal rearrangements of NUT, most commonly resultingin BRD4-NUT fusion oncogenes or, less commonly, BRD3-NUT or NUT-variant fusiononcogenes.

2. The karyotype of NMCs is uniquely simple compared with that of other squamous cellcarcinomas, suggesting that it arises through a genetic or epigenetic shortcut to thephenotype of squamous cell carcinoma.

3. Once a difficult diagnosis to make, NMC is now diagnosable in most cases by immuno-histochemical staining with an anti-NUT monoclonal antibody.

4. BRD-NUT chimeric oncoproteins include the acetyl histone–binding bromodomainsand the protein-binding ET domain of BRDs, as well as an acidic domain in NUT thatbinds p300.

5. BRD-NUT fusion proteins block differentiation, possibly by sequestering p300 andthereby repressing the expression of genes that are required for squamous differentiation.

6. HDACi and BETi reverse the effects of BRD-NUT fusion proteins through mechanismsthat remain to be elucidated.

7. Because it is newly recognized and rare, NMC is a so-called orphan cancer for which thereis now a resource for doctors, families, and patients: the International NMC Registry(http://www.NMCRegistry.org).




FUTURE ISSUES

1. The advent of targeted inhibitors of NMC may lead to more effective therapy, and clinicaltrials with these drugs are anticipated to occur in the near future.

2. Because the genomic localization of BRD-NUT is critical to its function, as evidencedby its inhibition by BETi, identification of the DNA with which BRD-NUT associatesis imperative.

3. Cytogenetic analysis of NMC indicates that there may be relatively few collaborativebackground mutations, which makes NMC unique among aggressive carcinomas and,potentially, an outstanding example of a genetic shortcut to squamous cell carcinoma.Whole-genome or -exome sequencing offers a newly available means of ascertainingexactly how many, and which, genomic mutations exist in the background of NMCs.

4. Some evidence suggests that NUT-variant NMC patients may live longer than BRD-NUT NMC patients. Segregation of NUT-variant tumors into good-prognosis andpoor-prognosis groups, by the identification of NUT-variant partner genes, will be keyto improving our clinical and biologic understanding of these variants.

5. Because precedent predicts that NMC, like most aggressive carcinomas, cannot be curedwith a single drug, clinical trials that employ a combinatorial approach with one or moretargeting agents and chemotherapy may be the only way to determine its cure.

DISCLOSURE STATEMENT

The author is not aware of any affiliations, memberships, funding, or financial holdings that mightbe perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

Many special thanks go to our collaborators, Drs. Nicolas Reynoird and Saadi Khochbin ofINSERM, Universite Joseph Fourier, Institut Albert Bonniot, Grenoble, France, for providingthe time-lapse movie. Additional thanks go to Drs. Yukichi Tanaka, Mio Tanaka, Toru Horisawa,and Yutaka Saikawa of Kanazawa Medical University, Ishikawa, Japan, for providing Figure 4b.This work was supported by a Dana-Farber/Harvard Cancer Center 521 Nodal Award(5P30CA06516-44), the U.S. National Institutes of Health (NIH) (grant 1R01CA124633),the Stanley L. Robbins Memorial Award, the NIH Institutional National Research Service(grant T32-HLO7627), the NIH National Cancer Institute Mentored Clinical Scientist (award1K08CA92158-01), the NIH National Cancer Institute Mentored Clinical Scientist (award1K08CA128972), the Burroughs-Wellcome Foundation, and funds from the National CancerInstitute’s Initiative for Chemical Genetics (contract number N01-CO-12400).

LITERATURE CITED

1. Shehata BM, Steelman CK, Abramowsky CR, Olson TA, French CA, et al. 2009. NUT midline carcinomain a newborn with multiorgan disseminated tumor and a 2-year-old with a pancreatic/hepatic primary.Pediatr. Dev. Pathol. 13:481–85

2. Stelow EB, Bellizzi AM, Taneja K, Mills SE, Legallo RD, et al. 2008. NUT rearrangement in undiffer-entiated carcinomas of the upper aerodigestive tract. Am. J. Surg. Pathol. 32:828–34

262 French


3. French CA, Kutok JL, Faquin WC, Toretsky JA, Antonescu CR, et al. 2004. Midline carcinoma of childrenand young adults with NUT rearrangement. J. Clin. Oncol. 22:4135–39

4. French CA. 2010. Demystified molecular pathology of NUT midline carcinomas. J. Clin. Pathol. 63:492–96

5. den Bakker MA, Beverloo BH, van den Heuvel–Eibrink MM, Meeuwis CA, Tan LM, et al. 2009. NUTmidline carcinoma of the parotid gland with mesenchymal diffentiation. Am. J. Surg. Pathol. 33:1253–58

6. Ziai J, French CA, Zambrano E. 2010. NUT gene rearrangement in a poorly differentiated carcinoma ofthe submandibular gland. Head Neck Pathol. 4:163–68

7. Identifies the onlylong-term survivor withNMC.

7. Mertens F, Wiebe T, Adlercreutz C, Mandahl N, French CA. 2007. Successful treatment of achild with t(15;19)-positive tumor. Pediatr. Blood Cancer 49:1015–17

8. Engleson J, Soller M, Panagopoulos I, Dahlen A, Dictor M, Jerkeman M. 2006. Midline carcinomawith t(15;19) and BRD4-NUT fusion oncogene in a 30-year-old female with response to docetaxel andradiotherapy. BMC Cancer 6:69

9. Liebertz DJ, Lechner MG, Masood R, Sinha UK, Han J, et al. 2010. Establishment and characterizationof a novel head and neck squamous cell carcinoma cell line USC-HN1. Head Neck Oncol. 2:5

10. Bielas JH, Loeb KR, Rubin BP, True LD, Loeb LA. 2006. Human cancers express a mutator phenotype.Proc. Natl. Acad. Sci. USA 103:18238–42

11. Bonnet D, Dick JE. 1997. Human acute myeloid leukemia is organized as a hierarchy that originates froma primitive hematopietic cell. Nat. Med. 3:730–37

12. Huber MA, Kraut N, Beug H. 2005. Molecular requirements for epithelial-mesenchymal transition duringtumor progression. Curr. Opin. Cell Biol. 17:548–58

13. Thiery JP, Sleeman JP. 2006. Complex networks orchestrate epithelial-mesenchymal transitions.Nat. Rev. Mol. Cell Biol. 7:131–42

14. Reynoird N, Schwartz BE, Delvecchio M, Sadoul K, Meyers D, et al. 2010. Oncogenesis by sequestrationof CBP/p300 in transcriptionally inactive hyperacetylated chromatin domains. EMBO J. 29:2943–52

15. Characterizes theBRD4-NUT fusiononcogene in NMC.

15. French CA, Miyoshi I, Kubonishi I, Grier HE, Perez-Atayde AR, Fletcher JA. 2003. BRD4-NUTfusion oncogene: a novel mechanism in aggressive carcinoma. Cancer Res. 63:304–7

16. Demonstrates thatBRD-NUToncoproteins blockdifferentiation andmaintain proliferationof NMC cells.

16. French CA, Ramirez CL, Kolmakova J, Hickman TT, Cameron MJ, et al. 2008. BRD4-NUToncoproteins: a family of closely related nuclear proteins that block epithelial differentiation andmaintain the growth of carcinoma cells. Oncogene 27:2237–42

17. French CA, Miyoshi I, Aster JC, Kubonishi I, Kroll TG, et al. 2001. BRD4 bromodomain gene rearrange-ment in aggressive carcinoma with translocation t(15;19). Am. J. Pathol. 159:1987–92

18. Identifies thetherapeuticprodifferentiative andantiproliferative effectsof HDACi on NMC.

18. Schwartz BE, Hofer MD, Lemieux ME, Bauer DE, Cameron MJ, et al. 2011. Differentiation ofNUT midline carcinoma by epigenomic reprogramming. Cancer Res. 71:2686–96

19. Describes thediagnostic use ofmonoclonal NUTantibody as a sensitiveand specificimmunohistochemicaltest for NMC.

19. Haack H, Johnson LA, Fry CJ, Crosby K, Polakiewicz RD, et al. 2009. Diagnosis of NUT midlinecarcinoma using a NUT-specific monoclonal antibody. Am. J. Surg. Pathol. 33:984–91

20. Kubonishi I, Takehara N, Iwata J, Sonobe H, Ohtsuki Y, et al. 1991. Novel t(15;19)(q15;p13) chromosomeabnormality in a thymic carcinoma. Cancer Res. 51:3327–28

21. Kees UR, Mulcahy MT, Willoughby ML. 1991. Intrathoracic carcinoma in an 11-year-old girl showinga translocation t(15;19). Am. J. Pediatr. Hematol. Oncol. 13:459–64

22. Lee AC, Kwong YI, Fu KH, Chan GC, Ma L, Lau YL. 1993. Disseminated mediastinal carcinoma withchromosomal translocation (15;19). A distinctive clinicopathologic syndrome. Cancer 72:2273–76

23. Vargas SO, French CA, Faul PN, Fletcher JA, Davis IJ, et al. 2001. Upper respiratory tract carcinoma withchromosomal translocation 15;19: evidence for a distinct disease entity of young patients with a rapidlyfatal course. Cancer 92:1195–203

24. Toretsky JA, Jenson J, Sun CC, Eskenazi AE, Campbell A, et al. 2003. Translocation (11;15;19): a highlyspecific chromosome rearrangement associated with poorly differentiated thymic carcinoma in youngpatients. Am. J. Clin. Oncol. 26:300–6

25. You J, Croyle JL, Nishimura A, Ozato K, Howley PM. 2004. Interaction of the bovine papillomavirus E2protein with Brd4 tethers the viral DNA to host mitotic chromosomes. Cell 117:349–60

26. Moriniere J, Rousseaux S, Steuerwald U, Soler-Lopez M, Curtet S, et al. 2009. Cooperativebinding of two acetylation marks on a histone tail by a single bromodomain. Nature 461:664–68



27. Dey A, Chitsaz F, Abbasi A, Misteli T, Ozato K. 2003. The double bromodomain protein Brd4binds to acetylated chromatin during interphase and mitosis. Proc. Natl. Acad. Sci. USA 100:8758–63

26–28. Demonstratethat bromodomainsbind acetylatedhistones.

28. Dhalluin C, Carlson JE, Zeng L, He C, Aggarwal AK, Zhou MM. 1999. Structure and ligand of ahistone acetyltransferase bromodomain. Nature 399:491–96

29. Dey A, Ellenberg J, Farina A, Coleman AE, Maruyama T, et al. 2000. A bromodomain protein, MCAP,associates with mitotic chromosomes and affects G2-to-M transition. Mol. Cell Biol. 20:6537–49

30. Jang MK, Mochizuki K, Zhou M, Jeong HS, Brady JN, Ozato K. 2005. The bromodomain proteinBrd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II–dependenttranscription. Mol. Cell 19:523–34

31. Mochizuki K, Nishiyama A, Jang MK, Dey A, Ghosh A, et al. 2008. The bromodomain proteinBrd4 stimulates G1 gene transcription and promotes progression to S phase. J. Biol. Chem.283:9040–48

30–32. Show that BRD4binds P-TEFb tofacilitate transcriptionalelongation of earlypostmitotic genes.

32. Yang Z, Yik JH, Chen R, He N, Jang MK, et al. 2005. Recruitment of P-TEFb for stimuation oftranscriptional elongation by the bromodomain protein Brd4. Mol. Cell 19:535–45

33. Lin YJ, Umehara T, Inoue M, Saito K, Kigawa T, et al. 2008. Solution structure of the extraterminaldomain of the bromodomain-containing protein BRD4. Protein Sci. 17:2174–79

34. Ottinger M, Christalla T, Nathan K, Brinkmann MM, Viejo-Borbolla A, Schulz TF. 2006. Kaposi’ssarcoma–associated herpesvirus LANA-1 interacts with the short variant of BRD4 and releases cells froma BRD4- and BRD2/RING3-induced G1 cell cycle arrest. J. Virol. 80:10772–86

35. Reynoird N, Schwartz BE, Delvecchio M, Sadoul K, Meyers D, et al. 2010. Oncogenesis by sequestra-tion of CBP/p300 in transcriptionally inactive hyperacetylated chromatin domains. EMBO J. 29:2943–52

36. Haruki N, Kawaguchi KS, Eichenberger S, Massion PP, Gonzalez A, et al. 2005. Cloned fusion productfrom a rare t(15;19)(q13.2;p13.1) inhibit S phase in vitro. J. Med. Genet. 42:558–64

37. Govin J, Lestrat C, Caron C, Pivot-Pajot C, Rousseaux S, Khochbin S. 2006. Histone acetylation–mediated chromatin compaction during mouse spermatogenesis. Ernst Schering Res. Found. Workshop2006:155–72

38. Pivot-Pajot C, Caron C, Govin J, Vion A, Rousseaux S, Khochbin S. 2003. Acetylation-dependent chro-matin reorganization by BRDT, a testis-specific bromodomain-containing protein. Mol. Cell Biol. 23:5354–65

39. Grisolano JL, O’Neal J, Cain J, Tomasson MH. 2003. An activated receptor tyrosine kinase,TEL/PDGFβR, cooperates with AML1/ETO to induce acute myeloid leukemia in mice. Proc. Natl.Acad. Sci. USA 100:9506–11

40. Nanri T, Matsuno N, Kawakita T, Suzushima H, Kawano F, et al. 2005. Mutations in the receptortyrosine kinase pathway are associated with clinical outcome in patients with acute myeloblastic leukemiaharboring t(8;21)(q22;q22). Leukemia 19:1361–66

41. Nakao M, Yokota S, Iwai T, Kaneko H, Horiike S, et al. 1996. Internal tandem duplication of the flt3gene found in acute myeloid leukemia. Leukemia 10:1911–18

42. Maher B. 2009. Exome sequencing takes centre stage in cancer profiling. Nature 459:146–4743. Helmlinger D, Hardy S, Abou-Sleymane G, Eberlin A, Bowman AB, et al. 2006. Glutamine-

expanded ataxin-7 alters TFTC/STAGA recruitment and chromatin structure leading to pho-toreceptor dysfunction. PLoS Biol. 4:e67

44. McCampbell A, Taye AA, Whitty L, Penney E, Steffan JS, Fischbeck KH. 2001. Histone deacety-lase inhibitors reduce polyglutamine toxicity. Proc. Natl. Acad. Sci. USA 98:15179–84

45. McCampbell A, Taylor JP, Taye AA, Robitschek J, Li M, et al. 2000. CREB-binding proteinsequestration by expanded polyglutamine. Hum. Mol. Genet. 9:2197–202

43–46. Identify themechanism ofp300/CBPsequestration in disease.

46. Steffan JS, Bodai L, Pallos J, Poelman M, McCampbell A, et al. 2001. Histone deacetylaseinhibitors arrest polyglutamine-dependent neurodegeneration in Drosophila. Nature 413:739–43

47. Duvic M, Talpur R, Ni X, Zhang C, Hazarika P, et al. 2007. Phase 2 trial of oral vorinostat (suberoylanilidehydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL). Blood 109:31–39

264 French


48. Redner RL, Corey SJ, Rush EA. 1997. Differentiation of t(5;17) variant acute promyelocyticleukemic blasts by all-trans retinoic acid. Leukemia 11:1014–16

49. Soignet SL, Maslak P, Wang ZG, Jhanwar S, Calleja E, et al. 1998. Complete remission aftertreatment of acute promyelocytic leukemia with arsenic trioxide. N. Engl. J. Med. 339:1341–48

48–50. Present the firstexample ofdifferentiation therapyin cancer.

50. Hu J, Liu YF, Wu CF, Xu F, Shen ZX, et al. 2009. Long-term efficacy and safety of all-transretinoic acid/arsenic trioxide–based therapy in newly diagnosed acute promyelocytic leukemia.Proc. Natl. Acad. Sci. USA 106:3342–47

51. Tuveson DA, Willis NA, Jacks T, Griffin JD, Singer S, et al. 2001. STI571 inactivation of the gastroin-testinal stromal tumor c-KIT oncoprotein: biological and clinical implications. Oncogene 20:5054–58

52. Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, et al. 2004. EGFR mutations in lung cancer: correlationwith clinical response to gefitinib therapy. Science 304:1497–500

53. Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, et al. 1996. Effects of a selective inhibitor ofthe Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat. Med. 2:561–66

54. Bollag G, Hirth P, Tsai J, Zhang J, Ibrahim PN, et al. 2010. Clinical efficacy of a RAF inhibitor needsbroad target blockade in BRAF-mutant melanoma. Nature 467:596–99

55. Identifies a newclass of bromodomain-inhibitor drugs, BETi,with potentialtherapeutic utility inNMC.

55. Filippakopoulos P, Qi J, Picaud S, Shen Y, Smith WB, et al. 2010. Selective inhibition of BETbromodomains. Nature 468:1067–73

56. Adya N, Stacy T, Speck NA, Liu PP. 1998. The leukemic protein core binding factor β (CBFβ)-smooth-muscle myosin heavy chain sequesters CBFα2 into cytoskeletal filaments and aggregates. Mol. Cell Biol.18:7432–43

57. Nicodeme E, Jeffrey KL, Schaefer U, Beinke S, Dewell S, et al. 2010. Suppression of inflammation by asynthetic histone mimic. Nature 468:1119–23


pathogenesis of nut midline carcinomapm07ch10-french ari 10 october 2011 15:29 table 2 initial...

Documents