Mutation signature of adenoid cystic carcinoma: evidence for transcriptional and epigenetic reprogramming

Abstract

Adenoid cystic carcinoma (ACC), a relatively rare malignancy usually of salivary gland origin, has a signature v-myb avian myeloblastosis viral oncogene homolog–nuclear factor I/B (MYB-NFIB) gene fusion that activates MYB transcriptional regulatory activity. A new study in this issue by Stephens et al. is a comprehensive genomic mutation profiling analysis of this neoplasm and documents a common theme of alteration in chromatin regulatory genes. Also, mutations in SPEN (split ends, homolog of Drosophila), which encodes an RNA-binding coregulatory protein, suggest that other changes in transcriptional regulation may involve the NOTCH, FGFR, or other signaling pathways in which SPEN participates. Since there is a low prevalence of mutations in common oncogenes and tumor-suppressor genes, it is likely that alterations primarily in specific transcriptional regulatory genes, augmented by changes in chromatin structure, drive the neoplastic process in ACC.

In this issue of JCI, Stephens et al. (1) report the results of exome sequencing of 24 cases of adenoid cystic carcinoma (ACC), a relatively rare tumor, but one that is among the most common malignancies arising in salivary glands. ACC has distinctive clinical and pathologic features, including an often lengthy clinical course before the majority of patients succumb to their disease (2), a proclivity for tumor cells to invade nerves, which may lead to incomplete surgical resection and recurrence, and distinct myoepithelial/luminal epithelial cellular differentiation. The elucidation of the specific molecular events that underlie ACC may lead to targeted therapies for patients who have distant metastases for whom there currently are no effective chemotherapeutic agents.

MYB-NFIB is the signature molecular alteration

The study by Stephens et al. (1) confirms the presence of activation of v-myb avian myeloblastosis viral oncogene homolog (MYB) (on chromosome 6) in the majority of ACC (19/24, 79%); this occurs chiefly by chromosomal translocation and fusion to nuclear factor I/B (NFIB) (on chromosome 9). This key oncogenic event, first discovered in 2009 by Persson et al. (3) in Goran Stenman’s laboratory, appears to result in increased concentration and activity of the MYB transcriptional regulatory protein domains. The overexpression of MYB, which may be dysregulated by other mechanisms in ACC that lack MYB-NFIB fusion (4), leads to altered expression of its putative target genes involved in cell-cycle control, apoptosis, cell growth, angiogenesis, and cell adhesion (3). Which of these genes is the most critical for the growth and maintenance of ACC remains to be proven experimentally.

Aside from MYB alterations, Stephens et al. (1) report a mean of 13 mutations per exome in ACC, a mutation rate lower than that reported in comprehensive sequencing analyses of the most common types of carcinoma. The relative stability of the ACC genome at the nucleotide level is in keeping with comparative genomic hybridization (CGH) and array CGH studies that have revealed relatively few copy number alterations per genome (4–8), including the absence of high copy number amplifications. These data suggest that MYB activation is the primary driver event in ACC and that genomic instability appears to be a less important mechanism of tumorigenesis. It is also noteworthy that some of the most commonly altered oncogenes and tumor suppressor genes in cancer were underrepresented in this study; mutations in PIK3CA and CDKN2A occurred in only one ACC, and mutations in TP53, RB1, ERBB2, BRAF, EGFR, KRAS, PTEN, and KIT were absent, again indicating the unique and limited mutation signature of this neoplasm (Figure 1).

Figure 1

The neoplastic transformation of normal salivary gland (upper histologic figure) to adenoid cystic carcinoma (lower histologic figure) is the culmination of genetic alterations including translocation, deletions, and mutations, catalogued from comprehensive array CHG (4) and whole exome sequencing (1) analyses. Key targets include the activation of MYB, most commonly by a t(6;9) chromosomal translocation (karyotype shown) and mutations in genes that regulate chromatin structure (shown graphically). Original magnification, ×400 (upper panel); ×200 (lower panel).

Secondary ACC mutations target other transcriptional and chromatin regulators

While no other single gene has been found to be mutated at a frequency approaching that of the MYB-NFIB fusion, a combination of genes whose protein products are involved in chromatin regulation were mutated in approximately 50% of ACC. These include ARID1A, a member of the switch/sucrose non-fermentable (SWI/SNF) chromatin remodeling complex (9), CREBBP, a histone acetylase and transcriptional coactivator (10), EP300, a histone acetylase (11) and KDM6A, a histone lysine demethylase (12). While it is not known whether or how the chromatin modifications brought about by these mutations interact with the altered transcriptional activity caused by MYB activation (as well as perhaps other transcriptional regulatory pathways), it is interesting to speculate that changes in histone structure may either permit or augment reprogramming of transcriptional regulatory networks that drive the neoplastic cellular phenotype. Indeed, investigations in model systems suggest that joint action by a series of chromatin remodeling and transcriptional activation complexes is required for a more robust alteration in transcriptional regulation (13).

Of note, the single highest mutation frequency after MYB-NFIB occurred in SPEN (split ends, homolog of Drosophila) (7/66, 11%), whose gene product is an RNA-binding protein that acts as a coregulator with other transcriptional regulatory proteins. SPEN (also known as SHARP and MINT), a downstream effector of NOTCH signaling, generally represses the transcription of specific genes in the absence of NOTCH signaling (14); it also participates in the transcriptional response in other signaling pathways, including those of the FGFR family (15). Mutations in NOTCH1/2 and FGFR2 were also identified, further suggesting that these pathways play a role in ACC tumorigenesis and may have some overlap in regulating the effects of SPEN in this particular cellular phenotype.

Model systems and potential downstream therapeutic targets

For patients with ACC who develop distant metastases, new modes of therapy are essential, as both conventional single and combination chemotherapeutic agents have largely been ineffective (16). The investigation of biochemical mechanisms in cell signaling in ACC and preclinical studies of potential therapies would be aided by the analysis of model systems, but, to our knowledge, no validated and available ACC cell lines currently exist (17). However, xenografts derived from primary and metastatic ACC and serially passaged in mice have been found to be histopathologically identical and have gene expression profiles similar to those of their corresponding original tumors (18). Indeed, we have found these models to be extremely useful in screening compounds for their activity in ACC (unpublished observations).

Although MYB is clearly the driver oncogene in most ACC, targeting this transcription factor is not currently possible therapeutically. A reasonable approach would be to identify a pathway often activated in ACC for which kinase inhibitors are in current clinical usage. Indeed, our group has found that FGFR1 is overexpressed and activated in many ACC compared with normal salivary glands (unpublished observations). FGFR1 has MYB consensus binding sites just upstream of its promoter (19), and the major ligand of FGFR1, FGF2, is regulated by MYB in melanoma cells (20). These findings have led to the testing of various chemotherapeutic agents in ACC xenografts The multi–tyrosine kinase inhibitor Dovitinib shows significant activity in growth suppression (unpublished observations), prompting a clinical trial of Dovitinib for patients with progressive, metastatic ACC (21). The identification of likely activating mutations in FGFR2 in a few cases of ACC (1) supports the disruption of FGFR signaling as a rational therapeutic approach and suggests that mutational analysis of this gene in any type of neoplasm responsive to Dovitinib should be considered.

Lessons from a rare disorder

With a fairly stable genome, ACC provides a remarkably clear genetic fingerprint that points to specific molecular alterations likely responsible for neoplastic transformation. The identification of the signature MYB-NFIB fusion (3) together with the comprehensive mutation profiling by Stephens et al. (1) are key events in the unraveling of the molecular landscape of ACC. As MYB-NFIB fusion is the obvious driver for this cancer type, the transcriptional reprogramming that it causes will be a key focus of future research. Modeling MYB activation in cell culture and transgenic animal models, especially in combination with disruption of chromatin remodeling complexes, will help us more fully understand the secondary changes that are required for full neoplastic transformation. The mutational signature in ACC suggests that activated transcriptional regulators require the participation of altered chromatin structure to fully effect durable changes in cell phenotype, and it appears that signal transduction events that culminate in complexes with the SPEN coregulatory factor are likely targets. It is intriguing, however, that ACC has so few mutations in the “upstream” portion of signal transduction pathways, which are so dominant in other cancers. Deciphering the specific transcriptional control elements in ACC may help highlight the mechanistic differences that underlie the diversity of cellular differentiation and clinical behavior of human neoplasms.

Note added in proof. Ho and colleagues recently reported the exome or whole-genome sequences for 60 ACC and found a low exonic somatic mutation rate, MYB translocations in 57%, and a 35% frequency of mutations targeting chromatin-remodeling genes (22).

Acknowledgments

Henry F. Frierson Jr. is supported by National Institute of Dental and Craniofacial Research (NIDCR) grant RFP NHLBI-DE-09-10. Christopher A. Moskaluk is supported by the Adenoid Cystic Research Foundation and NIDCR grant RC1-DE020687.

Address correspondence to: Henry F. Frierson Jr., Department of Pathology, Box 800214, University of Virginia Health System, Charlottesville, Virginia 22908, USA. Phone: 434.982.4404; Fax: 434.924.8767; E-mail: hff@virginia.edu.

Footnotes

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest. doi:10.1172/JCI69070.

See the related Brief Report Whole exome sequencing of adenoid cystic carcinoma.

References

1. Stephens PJ, et al. Whole exome sequencing of adenoid cystic carcinoma. J Clin Invest. doi: 10.1172/JCI67201 .
   View this article via: JCI Google Scholar
2. Lloyd S, Yu JB, Wilson LD, Decker RH. Determinants and patterns of survival in adenoid cystic carcinoma of the head and neck, including an analysis of adjuvant radiation therapy. Am J Clin Oncol. 2011;34(1):76–81.
   View this article via: PubMed CrossRef Google Scholar
3. Persson M, Andren Y, Mark J, Horlings HM, Persson F, Stenman G. Recurrent fusion of MYB and NFIB transcription factor genes in carcinomas of the breast and head and neck. Proc Natl Acad Sci U S A. 2009;106(44):18740–18744.
   View this article via: PubMed CrossRef Google Scholar
4. Persson M, et al. Clinically significant copy number alterations and complex rearrangements of MYB and NFIB in head and neck adenoid cystic carcinoma. Genes Chromosomes Cancer. 2012;51(8):805–817.
   View this article via: PubMed CrossRef Google Scholar
5. Bernheim A, et al. High-resolution array comparative genomic hybridization analysis of human bronchial and salivary adenoid cystic carcinoma. Lab Invest. 2008;88(5):464–473.
   View this article via: PubMed CrossRef Google Scholar
6. Rao PH, et al. Deletion of 1p32-p36 is the most frequent genetic change and poor prognostic marker in adenoid cystic carcinoma of the salivary glands. Clin Cancer Res. 2008;14(16):5181–5187.
   View this article via: PubMed CrossRef Google Scholar
7. El-Rifai W, Rutherford S, Knuutila S, Frierson HF Jr, Moskaluk CA. Novel DNA copy number losses in chromosome 12q12--q13 in adenoid cystic carcinoma. Neoplasia. 2001;3(3):173–178.
   View this article via: PubMed CrossRef Google Scholar
8. Yu Y, Baras AS, Shirasuna K, Frierson HF Jr, Moskaluk CA. Concurrent loss of heterozygosity and copy number analysis in adenoid cystic carcinoma by SNP genotyping arrays. Lab Invest. 2007;87(5):430–443.
   View this article via: PubMed CrossRef Google Scholar
9. Nie Z, et al. A specificity and targeting subunit of a human SWI/SNF family-related chromatin-remodeling complex. Mol Cell Biol. 2000: 20(23);8879–8888.
   View this article via: PubMed Google Scholar
10. Das C, Lucia MS, Hansen KC, Tyler JK. CBP/p300-mediated acetylation of histone H3 on lysine56. Nature. 2009;459(7259):113–117.
    View this article via: PubMed CrossRef Google Scholar
11. Gayther SA, et al. Mutations truncating the EP300 acetylase in human cancers. Nat Genet. 2000;24(3):300–303.
    View this article via: PubMed CrossRef Google Scholar
12. Lan F, et al. A histone H3 lysine 27 demethylase regulates animal posterior development. Nature. 2007;449(7163):689–694.
    View this article via: PubMed CrossRef Google Scholar
13. Urnov FD, Wolffe AP. Chromatin remodeling and transcriptional activation: the cast (in order of appearance). Oncogene. 2001;20(24):2991–3006.
    View this article via: PubMed CrossRef Google Scholar
14. Oswald F, et al. SHARP is a novel component of the Notch/RBP-J-kappa signaling pathway. EMBO J. 2002;21(20):5417–5426.
    View this article via: PubMed CrossRef Google Scholar
15. Sierra OL, Cheng S-L, Loewy AP, Charlton-Kachigian N, Towler DA. MINT, the Msx2 interacting nuclear matrix target, enhances Runx2-dependent activation of the osteocalcin fibroblast growth factor response element. J Biol Chem. 2004;279(31):32913–32923.
    View this article via: PubMed CrossRef Google Scholar
16. Ross PJ, et al. Epirubicin, cisplatin and protracted venous infusion 5-Fluorouracil chemotherapy for advanced salivary adenoid cystic carcinoma. Clin Oncol (R Coll Radiol). 2009;21(4):311–314.
    View this article via: PubMed CrossRef Google Scholar
17. Phuchareon J, Ohta Y, Woo JM, Eisele DW, Tetsu O. Genetic profiling reveals cross-contamination and misidentification of 6 adenoid cystic carcinoma cell lines: ACC2, ACC3, ACCM, ACCNS, ACCS and CAC2. PloS One. 2009;4(6):e6040.
    View this article via: PubMed CrossRef Google Scholar
18. Moskaluk CA, et al. Development and characterization of xenograft model systems for adenoid cystic carcinoma. Lab Invest. 2011;91(10):1480–1490.
    View this article via: PubMed CrossRef Google Scholar
19. Deng Q-L, Ishii S, Sarai A. Binding site analysis of c-Myb: Screening of potential binding sites by using the mutation matrix derived from systematic binding affinity measurements. Nucl Acids Res. 1996;24(4):766–774.
    View this article via: PubMed CrossRef Google Scholar
20. Miglarase MR, Halaban R, Gibson NW. Regulation of fibroblast growth factor 2 expression in melanoma cells by the c-MYB proto-oncogene. Cell Growth Differ. 1997;8(11):1199–1210.
    View this article via: PubMed Google Scholar
21. http://clinicaltrials.gov/ct2/show/NCT01524692.
22. Ho AS, et al. The mutational landscape of adenoid cystic carcinoma [published online ahead of print May 19, 2013]. Nat Genet. doi: 10.1038/ng.2643 .