One repressor to rule them all: ANCO1 links YAP and AIB1

The transcriptional co‐activators YAP and AIB1 individually promote breast cancer progression, but are not known to be mechanistically linked. A study published in this issue of EMBO Reports [1] now shows that YAP‐AIB1 form a physical complex in breast epithelial cells that cooperates in both activation and, unexpectedly, repression of key breast cancer genes. The repressive effect is due to the recruitment of ANCO1, a previously defined AIB1 interactor [2] that binds and inhibits the YAP‐AIB1 complex. These data identify ANCO1 as a candidate tumor suppressor through YAP‐AIB1 inhibition and could hint at a broader crosstalk between pathways that utilize YAP and AIB1 to control epithelial homeostasis.

The co‐activators YAP and AIB1 individually promote breast cancer progression. A study in this issue now shows that YAP‐AIB1 form a physical complex that cooperates in both activation and repression of key breast cancer genes.

Despite progress in early diagnosis and therapy, breast cancer remains the second leading cause of cancer death in women 3. Improved understanding of molecular mechanisms that drive breast cancer emergence and spread could improve diagnosis and enable more precise therapies. Amplification of the p160 nuclear hormone receptor co‐activator amplified in breast cancer‐1 (AIB1; also known as NCOA3, SRC3, or TRAM3) correlates with poor prognosis and reduced survival in breast cancer 4. AIB1 binds the estrogen receptor (ER) to promote estrogen‐dependent transcription but also interacts with other transcription factors, including those of the TEAD family. TEADs are well known for recruiting the Hippo pathway co‐activator YAP to enhancers, which promotes expression of genes that enhance proliferation, survival, stemness, and invasion 5. YAP is overexpressed in a wide variety of human cancers and promotes breast cancer metastasis. However, YAP‐TEAD links to AIB1 have not been described so far.

Kushner and colleagues now show that AIB1 physically interacts with YAP and TEAD4 proteins, and that AIB1 is required for a subset of YAP‐TEAD‐induced transcripts in normal mammary epithelial cells and early‐stage breast cancer cells 1. This AIB‐YAP‐TEAD4 interaction is required for cell invasiveness in mammospheres and correlates with tumorigenicity of cell‐line derived xenografts. Kushner et al then set out to define AIB1‐YAP co‐regulated genes. Somewhat unexpectedly, genome‐ and transcriptome‐wide analyses done in MCF10A breast epithelial cells expressing combinations of AIB1, YAP, and TEAD identify four subgroups of targets: YAP activated, AIB1‐YAP co‐activated, YAP repressed, and AIB1‐YAP co‐repressed (Fig 1). The activity of AIB1 is found to be required for a subset of YAP‐activated genes (e.g., ANKRD1, CTGF, and CYR61), but also for a significant fraction of YAP‐repressed genes (e.g., S100A9, RAB7B, and TNFSF10). Importantly, known YAP1 target genes are found in both AIB1‐dependent groups.

Figure 1. Interactions between AIB1‐ER and YAP‐TEAD signaling pathways in breast cancer progression.

(A) The nuclear receptor co‐activator AIB1 and the Hippo pathway co‐activator YAP converge upon TEAD‐binding sites for transcription activation and repression. A subset of AIB1/YAP co‐repressed genes at the 1q21.3 locus is regulated by AIB1 recruitment of the tumor suppressor ANCO1. (B) YAP‐TEAD complexes are recruited to a subset of estrogen response elements (EREs) via an interaction with the ER and are required for estrogen‐induced transcription and breast cancer progression. The right panel (bounded by a dotted line) provides a hypothetical model of AIB1‐ER and YAP‐TEAD interactions within an enhancer with adjacent ERE and Hippo response elements (HRE).

The cooperative effect of AIB1 and YAP on shared enhancers matches their known roles as co‐activators. However, the need for AIB1 for repression of YAP targets implies that another factor must be involved, likely a transcriptional repressor. Using ChIP, Hi‐C and RT–qPCR techniques, the authors find that the ANCO1 transcriptional repressor (ankyrin repeat‐containing cofactor 1; also Ankrd11), which binds AIB1 and represses ER target genes 2, 6, is also recruited by AIB1 into YAP‐TEAD4 complexes to repress YAP targets, particularly those in the 1q21.3 region associated with breast cancer aggressiveness 7. ANCO1 was originally recovered from a yeast two‐hybrid screen for proteins that interact with p160 nuclear receptor co‐activators (including AIB1) and recruits histone deacetylases to inhibit ER transactivation 6. Intriguingly, mutations in ANCO1 are tightly linked to KBG syndrome, an autosomal dominant disorder characterized by distinctive facial features, developmental delay, short stature, and skeletal anomalies 8. Whether AIB1 or YAP‐TEAD contribute to KBG phenotypes is not known. However, evidence from cancer databases indicates that ANCO is mutated or deleted in a range of human cancer samples (The Cancer Genome Atlas Program: https://www.cancer.gov/tcga), suggesting that it might have tumor suppressive properties. Consistent with this idea, Kushner et al find that high ANCO1 correlates with enhanced survival in breast cancer patients and that ANCO1 expression is low in ductal carcinoma in situ (DCIS). These data suggest that ANCO1 loss and de‐repression of AIB1‐YAP‐TEAD promote breast cancer progression, and that ANCO1 status could serve as marker of cancer stage.

In addition to introducing ANCO1 as a YAP repressor, the study also contributes to an emerging body of data pointing to crosstalk between YAP and pathways that involve AIB1. AIB1 and the related p160 co‐activators NCOA‐1 and NCOA‐2 interact with nuclear hormone receptors (e.g., estrogen and androgen receptors—ER and AR) that would, in turn, seem to be candidate Hippo‐interacting factors. Indeed, a recent paper 9 supports a model of YAP‐ER crosstalk by a mechanism that resembles that proposed by Kushner et al (Fig 1B, left panel). This study found YAP‐TEAD bind directly to ER and are required for estrogen‐induced gene transcription and breast cancer growth. AIB1 co‐purifies with ER from breast cancer cells but its presence in the ER‐YAP‐TEAD complex was not tested. The TEAD component of the ER‐YAP‐TEAD trimer may not contact DNA, but it remains undetermined if a physical interaction between AIB1‐ER and YAP‐TEAD could occur between two adjacent DNA‐bound complexes (Fig 1B, right panel). A model based on enhancer co‐occupancy has been proposed to explain physical and functional interactions between the Drosophila melanogaster homologs of YAP and AIB1 10, respectively, Yorkie (Yki) and Taiman (Tai). The Tai‐Yki interaction is based on two PPxY (proline‐proline‐x‐tyrosine) motifs in the Tai transactivation domain (TAD) that interact with the two conserved WW domains in Yki and result in co‐activation of shared transcriptional targets. Interestingly, human AIB1 lacks PPxY motifs, indicating that the AIB‐YAP complex documented by Kushner et al may be conserved from flies to man, but mediated by a different interaction module in each species. Intriguingly, the AIB1 paralogs NCOA1 and NCOA2 each contain a single PPxY in their TADs. However, to date, physical interactions between YAP and NCOA1/2 have not been reported.

In summary, the study by Kushner et al presents evidence that the loss of ANCO1 repression on shared AIB1‐YAP targets at the 1q21.3 locus plays a critical role during breast cancer progression. Additional work will be required to determine whether different proteins can recruit ANCO1 to other transcriptional complexes, and to identify factors responsible for ANCO1‐dependent and independent repression of AIB1‐YAP target genes. As studies of ANCO1 have been performed largely in cultured cells, a logical next step would be to assess in vivo function of ANCO1 or its homologs in YAP/AIB1‐regulated pathways in whole organisms or tissues. These combined approaches may ultimately shed light on conserved roles of ANCO1 and AIB‐YAP in developmental and pathologic tissue growth, and in turn speed the development of targeted therapeutic interventions.

EMBO Reports (2020) 21: e49647