ORIGINAL ARTICLE LMTK3 is implicated in endocrine resistance via multiple signaling pathways J Stebbing 1 , A Filipovic 1 , LC Lit 1 , K Blighe 2 , A Grothey 1 , Y Xu 1 , Y Miki 3 , LW Chow 3,4 , RC Coombes 1 , H Sasano 3 , JA Shaw 2 and G Giamas 1 Resistance to endocrine therapy in breast cancer is common. With the aim of discovering new molecular targets for breast cancer therapy, we have recently identified LMTK3 as a regulator of the estrogen receptor-alpha (ERa) and wished to understand its role in endocrine resistance. We find that inhibition of LMTK3 in a xenograft tamoxifen (Tam)-resistant (BT474) breast cancer mouse model results in re-sensitization to Tam as demonstrated by a reduction in tumor volume. A whole genome microarray analysis, using a BT474 cell line, reveals genes significantly modulated (positively or negatively) after LMTK3 silencing, including some that are known to be implicated in Tam resistance, notably c-MYC, HSPB8 and SIAH2. We show that LMTK3 is able to increase the levels of HSPB8 at a transcriptional and translational level thereby protecting MCF7 cells from Tam-induced cell death, by reducing autophagy. Finally, high LMTK3 levels at baseline in tumors are predictive for endocrine resistance; therapy does not lead to alteration in levels, whereas in patient’s plasma samples, acquired LMTK3 gene amplification (copy number variation) was associated with relapse while receiving Tam. In aggregate, these data support a role for LMTK3 in both innate (intrinsic) and acquired (adaptive) endocrine resistance in breast cancer. Oncogene (2013) 32, 3371–3380; doi:10.1038/onc.2012.343; published online 6 August 2012 Keywords: LMTK3; breast cancer; estrogen receptor; endocrine resistance; HSPB8; oncogene INTRODUCTION With an annual incidence of 1.3 million cases worldwide and 465,000 deaths, breast cancer remains the most frequently diagnosed type of cancer and the leading cause of cancer mortality in females. 1 Because ERa is expressed in over 70% of breast cancers, it is not surprising that many treatments thus far have focused on targeting ERa, directly or indirectly, by inhibiting its activity, stability and/or biosynthesis. 2–5 Nevertheless, many patients become resistant to these treatments and relapse at a rate of 1–1.5% of those diagnosed per annum. Although mutations in the ERa are rarely found, 6 other mechanisms have been associated with tamoxifen (Tam) resistance, namely phosphorylation of ERa. 7 Thus, deciphering the role of kinases in modulating ERa activity may derive new druggable targets with the potential to reverse this resistance, including re-sensitizing cells to Tam. 8,9 Using a high throughput screen with activity of estrogen-responsive genes as a quantitative read-out, we have recently identified Lemur Tyrosine Kinase 3 (LMTK3), for which a role had not previously been assigned, as a ‘master’ oncogenic regulator of ERa with transcriptional and translational effects. 10 To establish a role of LMTK3 in reversing endocrine resistance, we examined the effects of its inhibition in a Tam-resistant xenograft model and subsequent sensitivity to Tam. To under- stand the signaling pathways implicated, we performed a genome-wide gene expression analysis using the Tam-resistant cell line (BT474) with small interfering RNA (siRNA) knockdown of LMTK3, which revealed that LMTK3 is able to regulate the transcription of several genes, some of which have been previously described to have a role in ERa-positive breast cancer. We demonstrate that inhibition of LMTK3 decreased the total protein levels of HSBP8 (HSP22), a member of the heat-shock protein family, recently identified as a potential target for Tam resistance via a high-throughput ectopic expression screen. 11 Overexpression of LMTK3 promoted cell survival in MCF7 cells (Tam sensitive) after Tam treatment by reducing autophagy, suggesting a potential mechanism by which LMTK3 acts in the development on Tam resistance. Resistance is typically innate (intrinsic) or acquired (adaptive), both factors that limit the success of therapies in the clinic. Whereas the latter is more common, we have previously shown in a large cohort that LMTK3 levels are predictive, as well as prognostic, in terms of response to Tam. Now, we have compared LMTK3 levels in responding breast cancers treated with endocrine therapy versus non-responding breast cancers during neo- adjuvant treatment. 12,13 Higher LMTK3 levels at baseline were associated with tumors that were less likely to respond to therapy. We have recently used SNP6.0 arrays to profile the ‘circulating cancer genome’ comparing circulating free DNA (cfDNA) samples isolated from plasma with the primary tumor in 50 breast cancer patients during follow-up. Specific copy number variations (CNVs) were detected in cfDNA, mirroring the primary tumor, up to twelve years after diagnosis despite no other evidence of disease, suggesting dormancy in the majority of patients. 14 Because of the high density of probes, 906 600 for SNPs and 946 000 for CNVs, it is possible to interrogate SNP/CNV data for most genomic intervals. We therefore assayed CNV data in the LMTK3 gene in these 1 Department of Surgery and Cancer, Division of Cancer, Imperial College London, Hammersmith Hospital Campus, Du Cane Road, London, UK; 2 Department of Cancer Studies and Molecular Medicine, University of Leicester, Leicester, UK; 3 Department of Pathology, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan and 4 UNIMED Medical Institute, Comprehensive Centre for Breast Diseases, Hong Kong. Correspondence: Dr G Giamas, Department of Surgery and Cancer, Hammersmith Hospital Campus, Imperial College London, Du Cane Road, London W12 0NN, UK. Email: g.giamas@imperial.ac.uk Received 4 June 2012; revised 20 June 2012; accepted 21 June 2012; published online 6 August 2012 Oncogene (2013) 32, 3371–3380 & 2013 Macmillan Publishers Limited All rights reserved 0950-9232/13 www.nature.com/onc