Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Oct 4.
Published in final edited form as: Mol Cancer Res. 2018 Jun 22;16(10):1458–1469. doi: 10.1158/1541-7786.MCR-17-0637

Hypoxia Promotes Resistance to EGFR Inhibition in NSCLC Cells via the Histone Demethylases, LSD1 and PLU-1

Yuhong Lu 1, Yanfeng Liu 1, Sebastian Oeck 1,3, Peter M Glazer 1,2,*
PMCID: PMC6170700  NIHMSID: NIHMS977875  PMID: 29934325

Abstract

The development of small-molecule tyrosine kinase inhibitors (TKIs) specific for epidermal growth factor receptors (EGFRs) with activating mutations has led to a new paradigm in the treatment of non-small cell lung cancer (NSCLC) patients. However, most patients eventually develop resistance. Hypoxia is a key micro-environmental stress in solid tumors that is associated with poor prognosis due, in part, to acquired resistance to conventional therapy. This study, documents that long-term, moderate hypoxia promotes resistance to the EGFR TKI, gefitinib, in the NSCLC cell line, HCC827, which harbors an activating EGFR mutation. Following hypoxic growth conditions, HCC827 cells treated with gefitinib upregulated N- cadherin, Fibronectin and Vimentin expression and downregulated E-cadherin, characteristic of an epithelial-mesenchymal transition (EMT) which prior studies have linked to EGFR TKI resistance. Mechanistically, knockdown of the histone demethylases, LSD1 and PLU-1, prevented and reversed hypoxia-induced gefitinib resistance, with inhibition of the associated EMT, suggesting that LSD1 and PLU-1 play key roles in hypoxia-induced gefitinib resistance and EMT. Moreover, hypoxia-treated HCC827 cells demonstrated more aggressive tumor growth in vivo compared to cells grown in normoxia, but inhibition of LSD1 function by shRNA- mediated knockdown or by the small-molecular inhibitor, SP2509, suppressed tumor growth and enhanced gefitinib response in vivo. These results suggest that hypoxia is a driving force for acquired resistance to EGFR TKIs through epigenetic change and coordination of EMT in NSCLC. This study suggests that combination of therapy with EGFR TKIs and LSD1 inhibitors may offer an attractive therapeutic strategy for NSCLCs.

Introduction

The epidermal growth factor receptor (EGFR) pathway plays a key role in cell proliferation and survival, and it is commonly dysregulated in many types of cancers (1). Activating mutations of this receptor have been identified in NSCLCs, leading to the clinical advancement of small molecule inhibitors targeting EGFRs with specific activating mutations (2,3). This new therapeutic approach has changed the clinical landscape for patients with advanced cancers of the lung, and EGFR TKIs have demonstrated efficacy in metastatic EGFR positive lung cancer patients (4,5).

However, while a recent study showed that first-generation EGFR TKIs significantly delayed disease progression, they had no effect on overall survival (6), as most patients eventually develop resistance (7,8). Recent studies have deepened our understanding of the molecular mechanisms underlying this acquired resistance. In more than 50% of resistant cases, the tumors have acquired secondary mutations in EGFR at exon 20 (T790M) (9). The amplification of other RTKs, like MET and HER2, or mutations in genes encoding downstream signaling components, like PIK3CA and BRAF, represent additional mechanisms of acquired resistance (10). Histologic transformation, particularly epithelial-to-mesenchymal transition (EMT), has also been reported in subsets of patients who have progressed on treatment with EGFR TKIs (11,12).

Hypoxia is a key feature in solid tumors that profoundly influences numerous aspects of tumor biology and is identified as an adverse prognostic factor (13,14). The negative impact of hypoxia on the efficacy of radio- and chemotherapy is well established (13,15,16). Hypoxia affects drug delivery, DNA repair, upregulation of resistance genes, and alters cell cycle and cell death pathways (13,17).

Here we show that long-term, moderate hypoxia promotes gefitinib resistance in the NSCLC cell line, HCC827, which harbors an activating EGFR mutation (18). In addition, after growth in hypoxia, gefitinib treatment of HCC827 cells induces N-cadherin expression, a mesenchymal marker, and down-regulates the epithelial marker, E-cadherin, with associated changes in cell motility reflective of EMT. Mechanistically, it is shown that knockdown of the histone demethylases, LSD1 and PLU-1, before hypoxia exposure prevents and knockdown after hypoxia exposure reverses the hypoxia-induced gefitinib resistance and EMT phenotype. Similarly, treatment of HCC827 cells that had acquired hypoxia-induced gefitinib resistance with the small molecule LSD1 inhibitor, SP2509, or the PLU-1 inhibitor, PBIT, re-sensitizes them to gefitinib.

In vivo, HCC827 cells grown in hypoxia prior to implantation in mice show more rapid tumor growth than cells grown in normoxia, but inhibition of LSD1 with SP2509 or shRNA knockdown of LSD1 suppresses this effect. In addition, treatment of mice with the combination of SP2509 and gefitinib further reduces tumor growth compared to either agent alone. These results identify hypoxia as a potential driving force for EGFR TKI resistance in NSCLC through epigenetic modification and EMT in a pathway that may be susceptible to reversal with LSD1 and/or PLU-1 inhibitors and other epigenetic agents.

Materials and Methods

Cells.

HCC827, and PC-9 cells were obtained from the ATCC (Manassas, VA) and grown in RPMI 1640 with 10% FBS. YLR086 is a patient-derived cell line, which was obtained from Dr. Katerina Politi (Yale University, Pathology Department), and the cells were cultured in RPMI with 10% FBS. All cell lines tested negative for mycoplasma.

Constructs.

Lentivirus shRNA vectors for LSD1 knock down were obtained from Sigma- Aldrich (LSD1–1: TRCN0000046068). The lentivirus shRNA vector for PLU-1 was obtained from Dr. Qin Yan (Yale University).

Hypoxia.

For moderate hypoxia (1% O2), the cells were cultured in a hypoxic incubator (Eppendorf incubator, Galaxy 170 R. New Brunswick), which is maintained to achieve a constant O2 concentration within the entire incubator for the indicated times. The CO2 level was maintained at 5% using an internal CO2 regulation system.

Western Blots.

Cells were lysed in RIPA buffer (25 mM Tris HCl PH 7.6, 150 mM NaCl, 1% Igepal CA-630, 1% sodium deoxycholate, 0.1% SDS) with protease inhibitor cocktail (Clontech). The primary antibodies used for Western blotting are listed in Supplemental Experimental Procedures.

Clonogenic survival assay.

After the cells (HCC827 or YLR086) were exposed to normoxia or hypoxia for 5 weeks, cells at 100,000 cells per 100-mm dish were then treated with gefitinib at 5 μΜ. Cells at 500 cells per 100-mm dish without gefitinib were used as controls for cloning efficiency. Resistance frequencies were calculated by dividing the number of clones growing under gefitinib selection by the effective number of cells plated (as determined by cloning efficiency in untreated cells).

Quantitative real-time PCR analysis.

For quantitative PCR analysis of endogenous N- cadherin mRNA expression, total RNA was prepared using the Absolutely RNA® Miniprep Kit (Agilent Technologies). A total of 2μg RNA was used to synthesize cDNA using the High Capacity cDNA Reverse Transcription kit (AB Applied Biosystems). The resulting cDNA was used in PCR reactions containing Taqman Universal PCR Master Mix (Applied Biosystems) and premixed Taqman probes and primers for Ν-Cadherin, and 18S (Applied Biosystems). The Mx3000p real time PCR system (Strata gene) was used to monitor the fluorescence intensity in real-time and calculated cycle thresholds.

Wound-healing assay.

Normoxia or hypoxia exposed HCC827 or YLR086 cells were plated to confluence in a 6-well plate. Then the cells were scraped and washed with PBS. The cells were then treated with DMSO or gefitinib and allowed to fill the wounded area for the indicated time. The plates were fixed by formalin and stained with hematoxylin for visualization by microscopy.

Chromatin immunoprecipitation (ChIP) assays.

ChIP assays were performed as described (19). The primer sequences for the E-Cadherin promoter were used as follows: 5’ - AGGCTAGAGGGTCACCGGTC (Forward), and 5’- ACAGCTGCAGGCTCGGACAGGTAA (Reverse). LSD1 antibody used for ChIP was purchased from Millipore (Cat#:17–10531).

Establishment of hypoxia-induced gefitinib resistant clones in HCC827 cells.

After HCC827 cells were exposed to 1%O2 for 35 days, hypoxic cells were selected with gefitinib at 5μm for 3 weeks, and the resistant clones were collected for further studies.

Xenograft studies.

Female athymic nu/nu mice (Envigo/Harlan) and NOD.CB17/Prkdcscid/NCrHsd (NSG) mice were used for in vivo xenograft studies. All studies were approved by the Yale University Institutional Animal Care and Use Committee (IACUC). Mice were quarantined for at least 1 week before experimental manipulation. For comparing tumor growth between the normoxic HCC827 cells and the hypoxic HCC827 cells in vivo, 5×106 normoxic or hypoxic HCC827 cells were implanted subcutaneously (5×106 cells in 0.1 cm3 PBS) in the right flank in athymic nu/nu mice. Mice were visually observed daily, and tumors were measured three times per week by calipers to determine the tumor volume using the formula: V=4/3π(length/2)(width/2)(depth/2).

For studying the effects of knockdown of PLU-1 or LSD1 in hypoxic HCC827 on the tumor growth, 1×105 HCC827 H-GFPsh, or HCC827 H-PLU-1sh cells, or HCC827 H-LSD1sh cells were implanted subcutaneously (1×105 cells in 0.1 ml PBS) in the right flank in NSG mice. Mice were visually observed daily, and tumors were measured three times per week by calipers to determine the tumor volume using the formula: V=4/3π(length/2)(width/2)(depth/2).

For studying the effects of LSD1 inhibitor, SP2509, on tumor initiation and growth by hypoxic HCC827 cells, 1×105 hypoxic HCC827 cells were implanted subcutaneously in the right flank in athymic nu/nu mice. On the same day, the mice were treated either with DMSO or SP2509 (Selleckchem) in DMSO at 25mg/Kg via intraperitoneal injection for 3 weeks. Mice were visually observed daily, and tumors were measured three times per week by calipers to determine the tumor volume using the formula above.

For studying the effects of combination therapy of gefitinib and LSD1 inhibitor, SP2509, on tumor growth, 5×106 hypoxic HCC827 cells were implanted subcutaneously in the right flank in athymic nu/nu mice. After the tumors reached 50 mm3, the mice were treated either with DMSO, Gefitinib (Selleckchem) alone, SP2509 alone, or gefitinib plus SP2509. Gefitinib was delivered three times/week via oral gavage (1mg/ml, gefitinib was stabilized in DMSO and diluted with PBS), and SP2509 was delivered via intraperitoneal injection (25 mg/kg) three times per week. Mice were visually observed daily, and tumors were measured as above.

Statistical analysis

Data are presented as means ± SEM and compared using Student’s t test or ANOVA with repeated measures when appropriate. All tests were two-sided. A P-value of less than 0.05 was considered statistically significant.

Results

Growth in moderate hypoxia promotes gefitinib resistance in HCC827 cells and primes cells for gefitinib-induced EMT.

As a key feature in solid tumors, hypoxia is a contributor to the development of therapy resistance (15,20). The main challenge in TKI targeted therapy, including EGFR TKIs, is the development of resistance in tumors. Previous studies have suggested that hypoxia may be one of the factors that promotes gefitinib resistance in NSCLC (21,22). We sought to test the hypothesis that hypoxia can promote such resistance and to investigate the underlying mechanism. To do so, HCC827 cells, a NSCLC cell line with an activating EGFR mutation (a small deletion in exon 19) (18), were exposed to normoxia or 1% oxygen conditions for 35 days. After this exposure, we placed both sets of cells in normoxia and measured their proliferation rates. We found that the hypoxia-exposed HCC827 cells had higher proliferation rates than the normoxic HCC827 cells (Fig. S1A). The normoxic and hypoxic cells were also tested for acquired resistance to gefitinib by clonogenic survival assay. We found that the gefitinib resistance rate was increased from 0.5% in normoxic HCC827 cells to close to 3% in the hypoxia exposed cells (Fig. 1A&1B), a 6-fold elevation. Similarly, serial cell counts to measure cell proliferation showed that after gefitinib treatment for two weeks, the number of HCC827 cells in the normoxic cell population was decreased from 0.6 million to 0.28 million; in contrast, the viable cell count in the hypoxia-exposed HCC827 cells increased from 0.6 million to 2 million after two weeks of gefitinib treatment (Fig. 1C). These data suggest that long term mild hypoxia exposure promotes gefitinib resistance in HCC827 cells.

Fig. 1. Hypoxia induces gefitinib resistance in HCC827 and primes cells for gefitinib- induced EMT.

Fig. 1.

Open in a new tab

HCC827 cells were exposed to normoxia (N) or to 1% O2 (H) for 5 weeks. Cells were collected for testing gefitinib sensitivity and presence of EMT. (A) Image of representative cell culture wells showing differential gefitinib-resistant colony formation following growth in normoxia or hypoxia and subsequent gefitinib selection. (B) Frequency of gefitinib-resistance in normoxic or hypoxic HCC827 cells. Error bars represent SEs from three replicates. (C) Cell growth inhibition by gefitinib in normoxic or hypoxic HCC827 cells measured by serial cell counts at indicated days of gefitinib treatment. (D) Western blot analyses to determine E- cadherin, N-cadherin, Fibronectin, and Vimentin expression levels in normoxic and hypoxic HCC827 cells with or without gefitinib treatment. (E) Quantitative real-time PCR analysis of N- cadherin mRNA levels in in normoxic and hypoxic HCC827 cells with or without gefitinib treatment. mRNA levels are expressed as the fold change relative to normoxic control HCC827 cells. (F) Wound-healing assay in normoxic and hypoxic HCC827 cells with or without gefitinib treatment. The cells were fixed after 6 days of gefitinib treatment.

During gefitinib treatment of the HCC827 cells, we observed morphologic changes on routine light microscopy in the previously hypoxic HCC827 cells that were characteristic of possible epithelial to mesenchymal transition (EMT), including losing regular cell shape and increasing cell motility (data not shown). These features were not seen in the cells that had been previously grown in normoxic conditions. Since EMT has been linked with EGFR TKIs resistance (12,23), we decided to interrogate EMT markers in both normoxic and hypoxic HCC827 cells. After normoxic and hypoxic HCC827 cells were treated with gefitinib for 6 days, western blot analyses showed decreased E-cadherin, an epithelial marker, and increased N- cadherin, Fibronectin, and Vimentin, as mesenchymal markers, in gefitinib treated hypoxic HCC827 cells but not in the normoxic cells (Fig. 1D). Interestingly, there were no differences seen in the expression of EMT markers between normoxic and hypoxic HCC827 cells in the absence of gefitinib treatment, suggesting that EMT specifically arises when hypoxic cells are challenged with gefitinib treatment. In keeping with this, N-cadherin expression was also increased in gefitinib-treated hypoxic HCC827 cells at the mRNA level (Fig. 1E). Next, a woundhealing assay was conducted to evaluate cell motility in both the normoxic and hypoxic HCC827 cells in the presence or absence of gefitinib. Compared to normoxic HCC827 cells, hypoxic HCC827 cells showed increased cell migration in both the DMSO treated control group and the gefitinib treated group (Fig. 1F). However, we did not observe any significant differences between normoxic and hypoxic HCC827 cells in a cell invasion assay (Fig. S1B). Taken together, our data suggest that long term hypoxia exposure promotes gefitinib resistance in NSCLC cells, and that this is accompanied by features of EMT.

PLU-1 and LSD1, the histone lysine demethylases, are required for hypoxia induced gefitinib resistance.

Histone demethylases and their associated chromatin modulations have been linked to EGFR TKIs resistance in NSCLC cells (24,25). In addition, PLU-1 has been reported to promote drug resistance in melanomas (26) and to drive metastasis and EMT in hepatocellular cancer cells (27), while LSD1 has been shown to promote invasion and metastasis through facilitating EMT in many cancer cell types (28–30). These studies provided a further rationale for the investigation of whether PLU-1 and/or LSD1 play a role in hypoxia induced gefitinib resistance in HCC827 cells. Hence, we examined global H3K4 methylation levels in the normoxic and hypoxic HCC827 cell with or without gefitinib treatment. We observed decreased H3K4 dimethylation in the gefitinib treated hypoxic HCC827 cells but not in the normoxic cells (Fig. S2A). In prior work, we had observed that the H3K4 demethylase, PLU-1, is upregulated in hypoxic conditions (19). Here, we confirmed that PLU-1 expression is increased in hypoxic HCC827 cells, both in presence or absence of gefitinib (Fig.S4A). To further probe the roles of PLU-1 and LSD1, we established stable small hairpin RNA (shRNA)-mediated knockdown of PLU-1 or LSD1 in HCC827 cells, along with a control line containing shRNA targeting GFP (Fig. 2B). The PLU-1 knockdown HCC827 cells, designated as HCC827 PLU-1sh (Fig. 2B), and the LSD1 knockdown HCC827 cells, designated as HCC827 LSD1sh (Fig. 2B), along with control HCC827 GFPsh cells, were exposed to normoxia or 1% oxygen for 35 days, and clonogenic survival assays were performed to test gefitinib sensitivity (Fig. 2A). Consistent with the results above, in the control HCC827 GFPsh cells, gefitinib resistance was increased about 6-fold after hypoxia exposure. However, knockdown of PLU-1 or LSD1 prevented the hypoxia-induced gefitinib resistance (Fig. 2C).

Fig. 2. Knockdown of PLU-1 or LSD1 prevents hypoxia induced gefitinib resistance, and blocks EMT in HCC827 cells.

Fig. 2.

Open in a new tab

After HCC827 cells with PLU-1 knockdown (HCC827 PLU- 1sh) or LSD1 knockdown (HCC827 LSD1sh) and control cells transduced with GFP shRNA (HCC827 GFPsh) were exposed to normoxia (N) or to hypoxia (H; 1% O2), cells were collected for testing gefitinib sensitivity and presence of EMT. (A) Schematic of experimental timeline in HCC827 GFPsh, HCC827 PLU-1sh and HCC827 LSD1sh cells. (B) Western blot analyses to determine PLU-1 and LSD1 knockdown in HCC827 cells. (C) Frequency of gefitinib resistance following exposure of HCC827 GFPsh and HCC827 PLU-1sh (Left), or HCC827 GFPsh and HCC827 LSD1 (Right) cells to normoxia or hypoxia (1% O2). Error bars represent SEs from three replicates. (D) Western blot analyses to determine N-cadherin expression levels in normoxic or hypoxic HCC827 GFPsh and HCC827 PLU-1sh with or without gefitinib treatment (Left), and N-cadherin expression levels in normoxic or hypoxic HCC827 GFPsh and HCC827 LSD1sh (Right) with or without gefitinib treatment (E) Wound-healing assay in normoxic and hypoxic HCC827 GFPsh and HCC827 PLU-1sh cells with or without gefitinib treatment (Left), and in normoxic and hypoxic HCC827 GFPsh and HCC827 LSD1sh cells with or without gefitinib treatment (Right). The cells were analyzed after 7 days of gefitinib treatment.

Similarly, cell proliferation assays showed that, upon gefitinib treatment for 10 days, hypoxic HCC827 GFPsh cells showed increased growth compared to the normoxic HCC827 GFPsh cells (p<0.05), but shRNA knockdown of LSD1, abrogated this difference in cell growth in the presence of gefitinib treatment between cells that had been exposed to hypoxia or normoxia (p<0.05 (Fig. S3 A). For comparison, we also evaluated cell proliferation rates in the absence of gefitinib exposure in the normoxic and hypoxic HCC827GFP cells, the normoxic and hypoxic HCC827 LSD1sh cells, and the normoxic and hypoxic HCC827 PLU-1sh cells. We found that the hypoxic HCC827 GFPsh showed higher proliferation than normoxic HCC827 GFPsh cells, but hypoxic HCC827 LSD1sh cells and hypoxic HCC827 PLU-1sh cells did not show higher proliferation compared to their corresponding normoxic cell lines (Fig. S4B).

To test if PLU-1 or LSD1 is associated with hypoxia-induced EMT in HCC827 cells, N- Cadherin expression was measured in HCC827 GFPsh, HCC827 PLU-1sh and HCC827 LSD1sh cells after exposure to normoxia or hypoxia with or without gefitinib treatment. N- cadherin expression was elevated in hypoxic HCC827 GFPsh cells after gefitinib treatment (Fig. 2C); however, this N-cadherin induction was blocked in the PLU-1 knockdown cells or LSD1 knockdown cells (Fig. 2D). Next, the wound healing assay was conducted to test the impact of PLU-1 or LSD1 on cell motility in the HCC827 cells. Without gefitinib exposure (DMSO treated control groups), hypoxic HCC827 GFPsh, hypoxic HCC827 PLU-1sh and hypoxic LSD1sh cells showed more motility than the normoxic cells. However, in the gefitinib treated groups, active wound repair was only observed in hypoxic HCC827 GFPsh cells; knockdown of PLU-1or LSD1 reduced the migratory capacity of the cells (Fig. 2E). Together, these findings suggest that PLU- 1 and LSD1 are required for hypoxia induced gefitinib resistance in the HCC827 cells, and knockdown of PLU-1 or LSD1 prior to hypoxic exposure prevents acquired gefitinib resistance induced by hypoxia.

Knockdown of PLU-1 or LSD1 reverses hypoxia induced gefitinib resistance.

We next asked whether inhibition of PLU-1 or LSD1 function would reverse hypoxia-induced gefitinib resistance in HCC827 cells. HCC827 cells were exposed to normoxia or hypoxia at 1% O2 for 5 weeks as before, and then we knocked down PLU-1 or LSD1 by shRNA in the HCC827 cells that had been exposed to hypoxia (Fig. 3A&B), named as HCC827 H-PLU-1sh and HCC827 H- LSD1sh. For comparison, we knocked down GFP using GFP shRNA in both normoxic HCC827 and hypoxic HCC827 cells, named as HCC827 N-GFPsh and HCC827 H-GFPsh (Fig. 3A&B). The four cell lines, HCC827 N-GFPsh, HCC827 H-GFPsh, HCC827 H-PLU-1sh and HCC827 H- LSD1sh cells, were assayed for gefitinib resistance (Fig. 3A). Again, compared to HCC827 N- GFPsh cells, HCC827 H-GFPsh cells showed significantly increased gefitinib resistance (about 7-fold) by clonogenic assay; however, neither the HCC827 H-PLU-1sh nor the HCC827 H- LSD1sh cells showed increased gefitinib resistance compared to the HCC827 N-GFPsh cells (Fig. 3C). Again, cell proliferation assay showed that after gefitinib treatment for 10 days, in HCC827 H-GFPsh cells, the cell growth in gefitinib was around 6% of non-gefitinib-treated control, which is significantly higher than in the HCC827 N-GFPsh cells (Fig. S3B) (p<0.05), but in HCC827 H-PLU-1 or HCC827 H-LSD1sh, there were no significant differences in cell growth compared to HCC827 N-GFPsh with gefitinib treatment (p<0.05 (Fig. S3B). These data suggest that targeting PLU-1 or LSD1 after hypoxia exposure can reverse gefitinib resistance induced by hypoxia. Functionally, we confirmed that knockdown of LSD1 or PLU-1 yielded elevated H3K4 dimethylation levels (Fig. S2B).

Fig. 3. Knockdown of PLU-1 or LSD1 reverses hypoxia induced gefitinib resistance.

Fig. 3.

Open in a new tab

HCC cells were exposed to normoxia (N) or to 1% O2 (H) for 5 weeks. Hypoxic HCC827 cells with LSD1 knockdown or with PLU-1 knockdown, named as HCC827 H-LSD1sh and HCC827 H-PLU-1sh respectively, were established. Control cells transduced with a GFP shRNA in normoxic or hypoxic HCC827, named HCC827 N-GFPsh or HCC827 H-GFPsh, were also established. Then all the cell lines were tested for gefitinib sensitivity and EMT. (A) Schematic of experimental timeline. (B) Western blot analyses to determine PLU-1 and LSD1 knockdown in HCC827 H-LSD1sh and HCC827 H-PLU-1sh cells. (C) Frequency of gefitinib resistance in HCC827 N-GFPsh, HCC827 H-GFPsh, HCC827 H-LSD1sh and HCC827 H-PLU- 1sh cells. Error bars represent SEs from three replicates. (D) Western blot analyses to determine N-cadherin expression levels in HCC827 N-GFPsh, HCC827 H-GFPsh, HCC827 H- LSD1sh and HCC827 H-PLU-1sh cells with or without gefitinib treatment. (E) Wound-healing assay in HCC827 N-GFPsh, HCC827 H-GFPsh, HCC827 H-LSD1sh and HCC827 H-PLU-1sh cells with or without gefitinib treatment. The cells were analyzed after 7 days of gefitinib treatment.

To examine whether knockdown of PLU-1 or LSD1 after hypoxia exposure affects EMT during gefitinib treatment, the N-cadherin expression in all four cell lines was tested with or without gefitinib treatment. It was found that N-cadherin expression was only induced in gefitinib treated HCC827 H-GFPsh cells (Fig. 3D). Knockdown of PLU-1 or LSD1 after hypoxia exposure attenuated N-cadherin expression in HCC827 cells (Fig. 3D). In keeping with the N-cadherin expression data, the wound healing assay showed that wound repair was only observed in gefitinib treated HCC827 H-GFPsh cells, as knockdown of PLU-1 or LSD1 in hypoxic HCC827 cells diminished cell motility during gefitinib treatment (Fig. 3E). These data suggest that even though hypoxia potentiates EMT in HCC827 cells, knockdown of PLU-1 or LSD1 blocks this process. Taken together, our data indicate that targeting PLU-1 or LSD1 can effectively reverse hypoxia induced gefitinib resistance in HCC827 cells and also block the associated EMT.

The LSD1 inhibitor, SP2509, and PLU-1 inhibitor, PBIT, attenuate hypoxia-induced gefitinib resistance in HCC827 cells.

LSD1 over expression has been reported in a variety of tumors and has been associated with cancer cell growth and metastasis (32–34). Numerous small molecule inhibitors of LSD1 have been developed and are in preclinical testing in several cancer types, including acute myeloid leukemia, breast cancer and neuroblastoma (35,36). Therefore, we sought to test whether LSD1 inhibition is able to attenuate gefitinib resistance induced by hypoxia. We used SP2509, a selective LSD1 inhibitor that has been shown to have antitumor effects both in vitro and in vivo (35). We first exposed HCC827 cells to normoxia or 1% O2 for 5 weeks, and we performed clonogenic survival assays to test the effect of SP2509 on gefitinib sensitivity in hypoxic HCC827 cells. We found that SP2509 sensitized the hypoxic HCC827 to gefitinib treatment (Fig. 4A). We also tested the effect of SP2509 in the wound healing assay. Similar to LSD1 knockdown data shown above, SP2509 also blocked hypoxia induced cell migration during gefitinib treatment (Fig. 4B).

Fig. 4. LSD1 inhibitor, SP2509, and PLU-1 inhibitor, PBIT, attenuate gefitinib resistance in HCC827 cells.

Fig. 4.

Open in a new tab

After HCC827 cells were exposed to normoxia (N) or to 1% O2 (H) for 5 weeks, cells were treated with gefitinib alone or combination of LSD1 inhibitor, SP2509, or combination of with PLU-1 inhibitor, PBIT. (A) Frequency of gefitinib resistance in normoxic or hypoxic HCC827 cells in the presence of the LSD1 inhibitor, SP2509. Cell survival was tested in normoxic HCC827 and hypoxic HCC827 when exposed to 0.1 μΜ SP2509 (Left). Selection in the presence of gefitinib at 5 μΜ, or gefitinib combined with SP2509 at 0.1 μΜ was performed (Right). Errors represent SEs from three replicates. (B) Wound-healing assay in normoxic and hypoxic HCC827 cells treated with gefitinib alone or combination of gefitinib with SP2509. The cells were analyzed after 6 days of gefitinib treatment. (C) Frequency of gefitinib resistance in normoxic or hypoxic HCC827 cells in the presence of the PLU-1 inhibitor, PBIT. Cell survival was tested in normoxic HCC827 and hypoxic HCC827 when exposed to 0.5 μΜ PBIT (Left). Selection in the presence of gefitinib at 5 μΜ, or gefitinib combined with PBIT at 0.5 μΜ was performed (Right). Errors represent SEs from three replicates. (D) Wound-healing assay in normoxic and hypoxic HCC827 cells treated with gefitinib alone or with a combination of gefitinib and PBIT. The cells were analyzed after 6 days of treatment.

Similarly, we also tested the PLU-1 inhibitor, PBIT, for its effect on gefitinib resistance induced by hypoxia. Again, we used normoxic and hypoxic HCC827 cells, and performed clonogenic survival assays to test the effect of PBIT on gefitinib sensitivity in hypoxic HCC827 cells. As we found in SP2509, PBIT sensitized the hypoxic HCC827 to gefitinib treatment (Fig. 4C). In the wound healing assay, similar to the PLU-1 knockdown data shown above, PBIT also blocked the hypoxia-induced increase in cell migration during gefitinib treatment (Fig. 4D).

Hypoxia-induced gefitinib resistant HCC827 sub-clones, C2–5 and C2–8, show change in gene expression consistent with EMT, mediated by LSD1.

To further explore the mechanisms by which hypoxia induces gefitinib resistance through EMT, we established hypoxia-induced, gefitinib resistant sub-clones by gefitinib selection in hypoxic HCC827 cells. Clones C2–5 and C2–8 were selected at random for further study. First, we performed western blot analysis, revealing decreased E-cadherin expression along with increased N-cadherin, Fibronectin, and Vimentin expression in the hypoxia-induced gefitinib resistant clones, C2–5 and C2–8, compared to the parental HCC827 cells (Fig. 5A). By immunofluorescence assay, N- cadherin expression was observed only in a gefitinib resistant clone, C2–8 (Fig. 5B). Next, we examined EMT transcription factor expression at the protein level in both the gefitinib resistant clones and the parental HCC827 cells. We found no differences in Snail expression between parental HCC827 cells and gefitinib resistant clones, C2–5 and C2–8, but we did observe increased expression of both SLUG and Twist 1 in both C2–5 and C2–8 (Fig. 5C). By coimmunoprecipitation assay, we observed SLUG and LSD1 association in the gefitinib resistant clones, C2–5 and C2–8, but not in the parental cells (Fig. 5D). In keeping with this, chromatin immune precipitation (ChIP) assays revealed increased LSD1 binding to E-cadherin promoter in the C2–8 compared to the parental cells. Taken together, the data suggest that hypoxia induces gefitinib resistance through EMT via SLUG/LSD1 interaction and LSD1 activity at the E- cadherin promoter.

Fig. 5. Hypoxia-induced gefitinib resistant HCC827 sub-clones, C2–5 and C2–8, show changes in gene expression consistent with EMT, mediated by LSD1.

Fig. 5.

Open in a new tab

After exposure to hypoxia for 35 days, HCC827 cells were grown in the presence gefitinib at 5 μΜ, and individual clones were collected for further study. (A) Western blot analyses to determine E-cadherin, N- cadherin, Fibronectin, and Vimentin expression levels in parental HCC827 cells and hypoxia- induced gefitinib resistant HCC827 sub-clones, C2–5 and C2–8. (B). Immunofluorescence assay to examine N-cadherin expression in hypoxia-induced HCC827 sub-clone, C2–8. (C). Western blot analyses to determine expression levels of EMT-associated transcription factors, Snail, SLUG, and Twistl, in parental HCC827 cells compared to hypoxia-induced gefitinib resistant HCC827 sub-clones, C2–5 and C2–8. (D). Co-Immunoprecipitation (co-IP) assay indicates increased SLUG and LSD1 association in hypoxia-induced gefitinib resistant HCC827 sub- clones, C2–5 and C2–8. (E). Chromatin immunoprecipitation (ChIP) assay to examine LSD1 occupancy at the Ε-Cadherin promoter by qPCR in HCC827 parental cells versus hypoxia- induced gefitinib resistant clone, C2–8, (Left). Representative agarose gels containing E- Cadherin promoter region PCR amplification products are shown (Right).

To ask whether hypoxia-induced gefitinib resistance might be accompanied by an EGFR second mutation at T790M, which is frequently associated with EGFR TKI resistance, we performed droplet digital PCR (ddPCR) to detect this mutation in genomic DNA. We found that EGFR T790M mutation was not detected above background in either HCC827 hypoxic cells or hypoxia-induced gefitinib resistant sub-clones (Fig. S9), suggesting that the EGFR T790M mutation is not associated with hypoxia-induced gefitinib resistance in HCC827 cells.31

Moderate hypoxia restores gefitinib resistance in patient-derived NSCLC cells, and LSD1 and PLU-1 are required for regaining resistance.

The YLR086 cell line was obtained from a patient with NSCLC with an activating EGFR mutation (exon 19 deletion), who had developed resistance to EGFR TKIs treatment. This cell line does not have the common second site mutation, T790M. Interestingly, when we cultured the cell line without any EGFR TKIs in vitro for about two months, the cell line became sensitive to gefitinib treatment. Because the response of these cells to gefitinib seemed to be modulated by growth conditions, we tested whether hypoxia could restore resistance to gefitinib in this cell line. We exposed the YLR086 to either normoxia or 1% O2, and after two weeks, we found that gefitinib resistance rate was increased from 0.29% in normoxic YLR086 cells to 1.29% after hypoxia exposure (Fig. S5A). After 5 weeks of hypoxic exposure, one third of the hypoxic YLR086 cells became resistant to gefitinib, which is 9 times higher than the normoxic YLR086 cells (which showed 3.35% resistance to gefitinib) (Fig. S5A). The cell proliferation assay also showed that the hypoxia induced more cell survival during gefitinib treatment of the YLR086 cells (Fig. S5B). We also performed the wound-healing assay to test cell motility in both the normoxic and hypoxic YLR086 cells. Compared to the normoxic YLR086 cells, the hypoxic YLR086 cells exhibited strong cell migration in both the DMSO treated control group and the gefitinib treated group (Fig. S5C), which is consistent with our findings in the HCC827 cells above. We also tested global H3K4 methylation levels in both the normoxic and hypoxic YLR086 cells with or without gefitinib treatment. By western blot, we found decreased H3K4 dimethylation only in the gefitinib treated hypoxic YLR086 cells (Fig. S5D). These results indicate that hypoxic exposure can restore the gefitinib resistance in these patient derived NSCLC cells, and that histone modifications occurred during this process.

To determine if PLU-1 or LSD1 is also engaged in the restoration of gefitinib resistance by hypoxia in YLR086 cells, we knocked down PLU-1 or LSD1 by shRNA, along with control GFP knockdown (Fig. S6B), similar to what we did in HCC827 cells. Then we exposed YLR086 GFPsh, YLR086 PLU-1sh, and YLR086 LSD1sh cells either to normoxia or 1% O2 for 5 weeks (Fig. S6A). As expected, by clonogenic survival assays, in control YLR086 GFPsh cells, the gefitinib resistance was increased by about 10-fold after hypoxia exposure (Fig. S6C). However, knockdown of PLU-1 or LSD1 prevented the hypoxia induced gefitinib resistance in YLR086 cells (Fig. S6C). In the wound healing assay, we also found robust wound repair only in the hypoxic YLR086 GFPsh cells, as knockdown of PLU-1 or LSD1 resulted in a reduction of cell migratory capacity compared to the control cells, especially with gefitinib treatment (Fig. S6D). Taken together, our results suggest that LSD1 and PLU-1 are required for hypoxia mediated restoration of gefitinib resistance in the YLR086 cells.

Lastly, we tested whether knockdown of PLU-1 or LSD1 could reverse hypoxia induced restoration of gefitinib resistance in the YLR086 cells. As we did in HCC827 cells (Figs. 2A & 3A), we knocked down GFP in both the normoxic YLR086 cell and the hypoxic YLR086 cells, named as YLR086 N-GFPsh and YLR086 H-GFPsh (Fig. S7A). We also knocked down PLU-1 or LSD1 in the hypoxia exposed YLR086 cells, named as YLR086 H-PLU-1sh or YLR086 H- LSD1sh cells (Fig. S7A & B). We performed clonogenic survival assays, and again, found that, compared to the YLR086 N-GFPsh cells, the YLR086 H-GFPsh cells showed significantly increased gefitinib resistance of about 7-fold. In comparison, the YLR086 H-LSD1sh cells did not show any increased resistance (Fig. S7C), showing that LSD1 knockdown can reverse the acquired resistance. Somewhat surprisingly, PLU-1 knockdown did not reverse the hypoxia induced gefitinib resistance in the YLR086 cells (Fig. S7C).

In the wound healing assay, we observed the expected increased wound repair in the gefitinib treated HCC827 H-GFPsh cells, but knockdown of LSD1 in the hypoxic YLR086 cells diminished wound repair during gefitinib treatment (Fig. S7D). In keeping with the gefitinib resistance results above, knockdown of the PLU-1 in hypoxic YLR086 cells showed no effects on the cell motility in theYLR086 cells. Taken all together, the above results indicate that hypoxia exposure re-establishes EGFR TKIs resistance in patient derived NSCLC cells, and this can be revered by LSD1 knockdown but not by PLU-1 knockdown.

Chemical inhibition or shRNA knockdown of LSD1 suppresses hypoxic HCC827 cell growth in vivo in tumor xenografts.

To test the implications of the above findings in vivo, we established tumor xenografts using normoxic HCC827 (HCC827 N) and hypoxic HCC827 (HCC827 H) cells which, as described above, represent HCC827 cells exposed either to normoxia or 1% O2 for 5 weeks in vitro, respectively. We found that the hypoxic HCC827 cells formed tumors much faster than the normoxic HCC827 cells in vivo (Fig. 6A), in keeping with the above results showing that hypoxia exposure transforms HCC827 cells into a more aggressive phenotype. Next, we wanted to understand the extent to which LSD1 or PLU-1 might play a role in the more aggressive growth of the hypoxic HCC827 cells in vivo. To do so, we established stable PLU-1 or LSD1 knockdown sub-populations in hypoxic HCC827 cells, designated HCC827 H-PLU-1sh and HCC827 H-LSD1sh, similar to the scheme shown in Fig. 3A, which are pooled populations. We also established a control with shRNA targeting GFP, HCC827 H-GFPsh.

Fig. 6. Inhibition of LSD1 function by SP2905 or by shRNA knockdown suppresses hypoxic HCC827 cell growth in vivo.

Fig. 6.

Open in a new tab

(A) Tumor volume growth curves for normoxic and hypoxic HCC827 cell xenografts. (B) Tumor volume growth curves for HCC827 H-GFPsh (Hypoxic HCC827 with GFP shRNA) and HCC827 H-PLU-1sh cells (Hypoxic HCC827 with PLU-1 knockdown) xenografts. 1×105 HCC827 H-GFPsh and HCC827 H-PLU-1sh cells were implanted subcutaneously; (two-tailed Student’s t-test, p<0.001 for all time points). (C) Tumor volume growth curves for HCC827 H-GFPsh (same as B) and HCC827 H-LSD1sh cells (Hypoxic HCC827 with LSD1 knockdown) xenografts. 1×105 HCC827 H-GFPsh or HCC827 H- LSD-1sh cells were implanted subcutaneously. (two-tailed Student’s t-test, p<0.001 for all time points). (D) Tumor volume growth curves for hypoxic HCC827 xenografts treated with either SP2509 (25mg/kg) or vehicle control. 1×105 hypoxic HCC827 cells were implanted subcutaneously and SP2509 was given on the same day as implantation. (E) Tumor volume growth curves for hypoxic HCC827 xenografts treated with either vehicle control, gefitinib alone (1mg/kg), SP2509 (25mg/kg) alone, or a combination of gefitinib and SP2509, beginning when the tumor volumes reached 50 mm3 (Anova test, p<0.05). (F) Tumor Kaplan-Meier curves after treatment as described in (E). Significantly higher “survival” (defined as tumor volume smaller than 5X of tumor volume at first treatment) was seen in mice injected with gefitinib and SP2509 versus gefitinib alone (***p<0.05) and versus vehicle control. There is no significant difference between vehicle versus gefitinib alone, or vehicle versus SP2509 alone.

To compare these cells with respect to tumor formation and growth, we implanted 1×105 cells into the flanks of immune-deficient NSG mice by subcutaneous injection. We used NSG instead of nude mice because pilot studies showed better overall tumor take rates compared to nude mice for the HCC827 cells (data not shown). We found tumor take rates were 100% among all three cell lines: HCC827 H-GFPsh, HCC827 H-PLU-1sh and HCC827 H-LSD1sh, as determined by emergence of a palpable tumor in the mice. However, the HCC827 H-PLU-1sh and the HCC827 H-LSD1sh tumors grew much more slowly than the control HCC827 H-GFPsh tumors (Figs. 6B and C). These data indicate that HCC827 tumor growth is disrupted without PLU-1 or LSD1 function.

Based on the above results suggesting that LSD1 function plays a role in the aggressive growth of the hypoxic HCC827 cells in vivo, we tested whether the small molecule LSD1 inhibitor, SP2509, could affect hypoxic HCC827 tumor initiation and/or growth. We implanted 1×105 HCC827 H cells (which had been exposed to 1% O2 for 5 weeks in vitro before implantation, as above), in mice by subcutaneous injection. The SP2509 treatment started at the same day as the tumor cell implantation and continued three times per week for 3 weeks. We found that the tumor initiation rate was much lower in the SP2509 treated group than the control group (30% versus 63.5%; Fig. S8A). Further, SP2509 almost completely suppressed tumor growth (Fig. 6D). We then repeated this experiment except that we implanted 10-fold more HCC827 H cells (5×106), with SP2509 treatment as above. Again, we observed that SP2509 significantly suppressed HCC827 H tumor growth (Fig. S8B).

Prompted by these findings, we asked whether LSD1 inhibition can potentiate gefitinib antitumor activity in vivo, especially for the more aggressive hypoxic HCC827 cells. We implanted 5×106 HCC827 H cells in mice by subcutaneous injection and started SP2509 and/or gefitinib treatment when the tumors reached a size of approximately 50 mm3. The treatment continued 3 times per week for 3 weeks. We found that, by themselves, gefitinib and SP2509 each partially inhibited HCC827 H cell tumor growth (Fig. 6E); however, the combination of SP2509 and gefitinib further inhibited tumor growth (Fig. 6E, p<0.05). Moreover, the combination also significantly prolonged the time to reach 5X the starting volume compared to both control and gefitinib alone group (Fig. 6F, p<0.05).

Taken together, the functional blockage of LSD1 by knockdown or inhibitor hinders tumor initiation and growth. Combination of gefitinib and LSD1 inhibitor, SP2509, significantly inhibits hypoxic HCC827 tumor growth and improves survival.

Discussion.

Here we show that hypoxia, a key feature of the tumor microenvironment, induces gefitinib resistance in HCC827 cells, a NSCLC cell line that harbors an activating EGFR mutation, and restores gefitinib resistance in YLR086, a patient derived NSCLC cell line. We also find that hypoxia induced gefitinib resistance is accompanied by features of EMT. Moreover, we have found that the histone demethylases, LSD1 and PLU-1, are required for developing and maintaining gefitinib resistance and the accompanying EMT induced by hypoxia. In an in vivo tumor study, we demonstrate that disruption of LSD1 function by a small molecule inhibitor (SP2509) or by shRNA knockdown inhibits tumor growth of hypoxia-exposed HCC827 cells. Notably, a combination of SP2509 and gefitinib significantly reduced tumor growth compared to single agent treatment.

Several studies have previously suggested that cancer cell lines undergoing EMT develop resistance to EGFR inhibitors (12,37,38). In a tumor biopsy from NSCLC patients with EGFR mutation, a subpopulation of mesenchymal tumor cells was identified, and they appeared to mediate resistance to EGFR inhibitor therapy (5); however, inducing a mesenchymal-to- epithelial transition (MET), which is a reversion of EMT, reverses EGFR inhibitor resistance (39). Our work here suggests that prolonged hypoxia exposure induces EGFR TKIs resistance in a manner that is accompanied by EMT and that depends on epigenetic regulation. In keeping with this, there is substantial evidence that chromatin modifications also have a key role during EMT (40).

The development of drug resistance during cancer therapy can be due to newly acquired mutations occurring in genes in the direct pathway or in complementary pathways (41,42). However, epigenetic alternations are increasingly being recognized as important mechanisms of acquired drug resistance (43,44). Experimentally, instances of reversible drug resistance have been documented with drug tolerance reversible by epigenetic therapy, such as histone deacetylase (HDAC) inhibitors and/or DNA demethylating agent (25). Interestingly, decreased levels of H3K4 methylation and upregulated levels of JARID1A/RBP2/KDM5A are key findings in resistant sub-populations of cancer cells (25). In addition, recent studies have shown that BET bromodomain inhibition can re-sensitize resistant breast cancer cells to ERBB2 or PI3K inhibitors (45,46). These findings all indicate that epigenetic change can be a common driving force for acquired drug resistance in cancer therapy.

Tumor hypoxia is an important mediator of resistance to cancer therapy through multiple mechanisms (13,15,47). Tumor hypoxia also has significant effects on epigenetic regulation. In previous work, we found that hypoxia leads to reduction of H3K4 methylation levels through regulation of the histone demethylases, LSD1 and PLU-1/KDM5B, mediating silencing of the DNA repair genes, BRCA1 and MLH (19,48). In the present study, we found that LSD1 and PLU-1/KDM5B are also required for developing EGFR TKI resistance induced by hypoxia in the HCC827 NSCLC cell line. Notably, we found that interruption of LSD1 or PLU-1/KDM5B function not only blocked but also reversed the gefitinib resistance induced by hypoxia. In addition, in the patient derived NSCLC cell line, YLR086, (which was initially resistant to gefitinib but became sensitive after 2 months in culture without drug,), we found that hypoxia exposure converted the cells back to the drug resistant state. LSD1 and PLU-1 were required for this regained resistance under hypoxic conditions. Lastly, in hypoxia-induced gefitinib resistant HCC827 sub-clones, there is increased occupancy of LSD1 at the E-Cadherin promoter compared to parental HCC827 cells.

Currently, a number of targeted epigenetic therapies are being pursued in clinical trials, including LSD1 inhibitors (49,50). In our study, we found not only that the LSD1 inhibitor, SP2509, is able to re-sensitize hypoxia-induced gefitinib resistant cells to the drug in vitro but also that the combination of gefitinib and SP2509 significantly inhibited HCC827 tumor growth in vivo compared to gefitinib alone. These results suggest that agents targeting LSD1 could be utilized in combination with EGFR TKIs to possibly prevent (or even reverse) the acquisition of resistance by that may occur through epigenetic mechanisms.

Supplementary Material

Supplement Information

Acknowledgments

We thank Dr. Katerina Politi for providing us the YLR086. a patient derived NSCLC cell line.

This work was supported by NIH grant (R01ES005775) to PMG.

Footnotes

Conflict of Interest Statement:

The authors declare no potential conflicts of interest.

References

  • 1.Normanno N, De Luca A, Bianco C, Strizzi L, Mancino M, Maiello MR, et al. Epidermal growth factor receptor (EGFR) signaling in cancer. Gene 2006;366(1):2–16. [DOI] [PubMed] [Google Scholar]
  • 2.Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, et al. EGFR Mutations in Lung Cancer: Correlation with Clinical Response to Gefitinib Therapy. Science 2004;304(5676):1497–500. [DOI] [PubMed] [Google Scholar]
  • 3.Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating Mutations in the Epidermal Growth Factor Receptor Underlying Responsiveness of Non-Small-Cell Lung Cancer to Gefitinib. New England Journal of Medicine 2004;350(21):2129–39. [DOI] [PubMed] [Google Scholar]
  • 4.Mok TS, Wu Y-L, Thongprasert S, Yang C-H, Chu D-T, Saijo N, et al. Gefitinib or Carboplatin- Paclitaxel in Pulmonary Adenocarcinoma. New England Journal of Medicine 2009;361(10):947–57. [DOI] [PubMed] [Google Scholar]
  • 5.Sequist LV, Waltman BA, Dias-Santagata D, Digumarthy S, Turke AB, Fidias P, et al. Genotypic and Histological Evolution of Lung Cancers Acquiring Resistance to EGFR Inhibitors. Science Translational Medicine 2011;3(75): 75ra26–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lee CK, Brown C, Gralla RJ, Hirsh V, Thongprasert S, Tsai C-M, et al. Impact of EGFR Inhibitor in Non-Small Cell Lung Cancer on Progression-Free and Overall Survival: A Meta-Analysis. JNCI: Journal of the National Cancer Institute 2013;105(9):595–605. [DOI] [PubMed] [Google Scholar]
  • 7.Pakkala S, Ramalingam SS. Epidermal Growth Factor Receptor Mutated Advanced Non-Small Cell Lung Cancer. Hematology/Oncology Clinics of North America 2017;31(1):83–99. [DOI] [PubMed] [Google Scholar]
  • 8.Lin Y, Wang X, Jin H. EGFR-TKI resistance in NSCLC patients: mechanisms and strategies. American Journal of Cancer Research 2014;4(5):411–35. [PMC free article] [PubMed] [Google Scholar]
  • 9.Pao W, Chmielecki J. Rational, biologically based treatment of EGFR-mutant non-small-cell lung cancer. Nat Rev Cancer 2010;10(11):760–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ohashi K, Maruvka YE, Michor F, Pao W. Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor-Resistant Disease. Journal of Clinical Oncology 2013;31(8):1070–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Jakobsen KR, Demuth C, Sorensen BS, Nielsen AL. The role of epithelial to mesenchymal transition in resistance to epidermal growth factor receptor tyrosine kinase inhibitors in nonsmall cell lung cancer. Translational Lung Cancer Research 2016;5(2):172–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Byers LA, Diao L, Wang J, Saintigny P, Girard L, Peyton M, et al. An Epithelial-Mesenchymal Transition Gene Signature Predicts Resistance to EGFR and PI3K Inhibitors and Identifies Axl as a Therapeutic Target for Overcoming EGFR Inhibitor Resistance. Clinical Cancer Research 2013;19(1):279–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Scanlon SE, Glazer PM. Multifaceted control of DNA repair pathways by the hypoxic tumor microenvironment. DNA Repair 2015;32:180–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Vaupel P The Role of Hypoxia-Induced Factors in Tumor Progression. The Oncologist 2004;9(suppl 5):10–7. [DOI] [PubMed] [Google Scholar]
  • 15.Tan Q, Saggar JK, Yu M, Wang M, Tannock IF. Mechanisms of Drug Resistance Related to the Microenvironment of Solid Tumors and Possible Strategies to Inhibit Them. The Cancer Journal 2015;21(4):254–62. [DOI] [PubMed] [Google Scholar]
  • 16.Gray LH, Conger AD, Ebert M, Hornsey S, Scott OCA. The Concentration of Oxygen Dissolved in Tissues at the Time of Irradiation as a Factor in Radiotherapy. The British Journal of Radiology 1953;26(312):638–48. [DOI] [PubMed] [Google Scholar]
  • 17.Masoud GN, Li W. HIF-1α pathway: role, regulation and intervention for cancer therapy. Acta Pharmaceutica Sinica B 2015;5(5):378–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Cragg MS, Kuroda J, Puthalakath H, Huang DC, Strasser A. Gefitinib-induced killing of NSCLC cell lines expressing mutant EGFR requires BIM and can be enhanced by BH3 mimetics. PLoS medicine 2007;4(10):1681–89; discussion 90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lu Y, Chu A, Turker MS, Glazer PM. Hypoxia-Induced Epigenetic Regulation and Silencing of the BRCA1 Promoter. Molecular and Cellular Biology 2011;31(16):3339–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Khaled G, Rana A, Marwan E-S, Hala G-M. Targeting Hypoxia for Sensitization of Tumors to Radio- and Chemotherapy. Current Cancer Drug Targets 2013;13(6):670–85. [DOI] [PubMed] [Google Scholar]
  • 21.An S-M, Lei H-M, Ding X-P, Sun F, Zhang C, Tang Y-B, et al. Interleukin-6 identified as an important factor in hypoxia- and aldehyde dehydrogenase-based gefitinib adaptive resistance in non-small cell lung cancer cells. Oncology Letters 2017;14(3):3445–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Murakami A, Takahashi F, Nurwidya F, Kobayashi I, Minakata K, Hashimoto M, et al. Hypoxia Increases Gefitinib-Resistant Lung Cancer Stem Cells through the Activation of Insulin-Like Growth Factor 1 Receptor. PLoS ONE 2014;9(1):e86459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Suda K, Tomizawa K, Fujii M, Murakami H, Osada H, Maehara Y, et al. Epithelial to Mesenchymal Transition in an Epidermal Growth Factor Receptor-Mutant Lung Cancer Cell Line with Acquired Resistance to Erlotinib. Journal of Thoracic Oncology 2011;6(7):1152–61. [DOI] [PubMed] [Google Scholar]
  • 24.Gale M, Sayegh J, Cao J, Norcia M, Gareiss P, Hoyer D, et al. Screen-identified selective inhibitor of lysine demethylase 5A blocks cancer cell growth and drug resistance. Oncotarget 2016;7(26):39931–44 doi 10.18632/oncotarget.9539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sharma SV, Lee DY, Li B, Quinlan MP, Takahashi F, Maheswaran S, et al. A Chromatin-Mediated Reversible Drug-Tolerant State in Cancer Cell Subpopulations. Cell 2010;141(1):69–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Roesch A, Vultur A, Bogeski I, Wang H, Zimmermann Katharina M, Speicher D, et al. Overcoming Intrinsic Multidrug Resistance in Melanoma by Blocking the Mitochondrial Respiratory Chain of Slow-Cycling JARID1B high Cells. Cancer Cell;23(6):811–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Tang B, Qi G, Tang F, Yuan S, Wang Z, Liang X, et al. JARID1B promotes metastasis and epithelial- mesenchymal transition via PTEN/AKT signaling in hepatocellular carcinoma cells. 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Feng J, Xu G, Liu J, Zhang N, Li L, Ji J, et al. Phosphorylation of LSD1 at Ser112 is crucial for its function in induction of EMT and metastasis in breast cancer. Breast Cancer Research and Treatment 2016;159(3):443–56 doi 10.1007/s10549-016-3959-9. [DOI] [PubMed] [Google Scholar]
  • 29.Ambrosio S, Amente S, Sacca CD, Capasso M, Calogero RA, Lania L, et al. LSD1 mediates MYCN control of epithelial-mesenchymal transition through silencing of metastatic suppressor NDRG1 gene. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Luo H, Shenoy Anitha K, Li X, Jin Y, Jin L, Cai Q, et al. MOF Acetylates the Histone Demethylase LSD1 to Suppress Epithelial-to-Mesenchymal Transition. Cell Reports;15(12):2665–78. [DOI] [PubMed] [Google Scholar]
  • 31. Cai C, He Housheng H, Chen S, Coleman I, Wang H, Fang Z, et al. Androgen Receptor Gene Expression in Prostate Cancer Is Directly Suppressed by the Androgen Receptor Through Recruitment of Lysine-Specific Demethylase 1. Cancer Cell;20(4):457–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters AHFM, et al. LSD1 demethylates repressive histone marks to promote androgen-receptor-dependent transcription. Nature 2005;437(7057):436–9. [DOI] [PubMed] [Google Scholar]
  • 33.Chen C, Zhao M, Yin N, He B, Wang B, Yuan Y, et al. Abnormal Histone Acetylation and Methylation Levels in Esophageal Squamous Cell Carcinomas. Cancer Investigation 2011;29(8):548–56. [DOI] [PubMed] [Google Scholar]
  • 34.Hayami S, Kelly JD, Cho H-S, Yoshimatsu M, Unoki M, Tsunoda T, et al. Overexpression of LSD1 contributes to human carcinogenesis through chromatin regulation in various cancers. International Journal of Cancer 2011;128(3):574–86 doi 10.1002/ijc.25349. [DOI] [PubMed] [Google Scholar]
  • 35.Fiskus W, Sharma S, Shah B, Portier BP, Devaraj SGT, Liu K, et al. Highly effective combination of LSD1 (KDM1A) antagonist and pan-histone deacetylase inhibitor against human AML cells. Leukemia 2014;28(11):2155–64 doi 10.1038/leu.2014.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Mould DP, McGonagle AE, Wiseman DH, Williams EL, Jordan AM. Reversible Inhibitors of LSD1 as Therapeutic Agents in Acute Myeloid Leukemia: Clinical Significance and Progress to Date. Medicinal Research Reviews 2015;35(3):586–618. [DOI] [PubMed] [Google Scholar]
  • 37.Frederick BA, Helfrich BA, Coldren CD, Zheng D, Chan D, Bunn PA, et al. Epithelial to mesenchymal transition predicts gefitinib resistance in cell lines of head and neck squamous cell carcinoma and non-small cell lung carcinoma. Molecular Cancer Therapeutics 2007;6(6):1683–91. [DOI] [PubMed] [Google Scholar]
  • 38.Fuchs BC, Fujii T, Dorfman JD, Goodwin JM, Zhu AX, Lanuti M, et al. Epithelial-to-Mesenchymal Transition and Integrin-Linked Kinase Mediate Sensitivity to Epidermal Growth Factor Receptor Inhibition in Human Hepatoma Cells. Cancer Research 2008;68(7):2391–9. [DOI] [PubMed] [Google Scholar]
  • 39.Lee CG, McCarthy S, Gruidl M, Timme C, Yeatman TJ. MicroRNA-147 Induces a Mesenchymal-To- Epithelial Transition (MET) and Reverses EGFR Inhibitor Resistance. PLOS ONE 2014;9(1):e84597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.McDonald OG, Wu H, Timp W, Doi A, Feinberg AP. Genome-scale epigenetic reprogramming during epithelial-to-mesenchymal transition. Nat Struct Mol Biol 2011;18(8):867–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Levy BP, Rao P, Becker DJ, Becker K. Attacking a Moving Target: Understanding Resistance and Managing Progression in EGFR-Positive Lung Cancer Patients Treated With Tyrosine Kinase Inhibitors. Oncology (Williston Park) 2016;30(7):601–12. [PubMed] [Google Scholar]
  • 42.Tetsu O, Phuchareon J, Eisele DW, Hangauer MJ, McCormick F. AKT inactivation causes persistent drug tolerance to EGFR inhibitors. Pharmacological Research 2015;102:132–7. [DOI] [PubMed] [Google Scholar]
  • 43.Brien Gerard L, Valerio Daria G, Armstrong Scott A. Exploiting the Epigenome to Control Cancer- Promoting Gene-Expression Programs. Cancer Cell 2016;29(4):464–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Easwaran H, Tsai H-C, Baylin Stephen B. Cancer Epigenetics: Tumor Heterogeneity, Plasticity of Stem-like States, and Drug Resistance. Molecular Cell 2014;54(5):716–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Stratikopoulos Elias E, Dendy M, Szabolcs M, Khaykin Alan J, Lefebvre C, Zhou M-M, et al. Kinase and BET Inhibitors Together Clamp Inhibition of PI3K Signaling and Overcome Resistance to Therapy. Cancer Cell 2015;27(6):837–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Stuhlmiller Timothy J, Miller Samantha M, Zawistowski Jon S, Nakamura K, Beltran Adriana S, Duncan James S, et al. Inhibition of Lapatinib-Induced Kinome Reprogramming in ERBB2-Positive Breast Cancer by Targeting BET Family Bromodomains. Cell Reports 2015;11(3):390–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Erler JT, Cawthorne CJ, Williams KJ, Koritzinsky M, Wouters BG, Wilson C, et al. Hypoxia- Mediated Down-Regulation of Bid and Bax in Tumors Occurs via Hypoxia-Inducible Factor 1- Dependent and -Independent Mechanisms and Contributes to Drug Resistance. Molecular and Cellular Biology 2004;24(7):2875–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Lu Y, Wajapeyee N, Turker Mitchell S, Glazer Peter M. Silencing of the DNA Mismatch Repair Gene MLH1 Induced by Hypoxic Stress in a Pathway Dependent on the Histone Demethylase LSD1. Cell Reports 2014;8(2):501–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Mohammad Helai P, Smitheman Kimberly N, Kamat Chandrashekhar D, Soong D, Federowicz Kelly E, Van Aller Glenn S, et al. A DNA Hypomethylation Signature Predicts Antitumor Activity of LSD1 Inhibitors in SCLC. Cancer Cell 2015;28(1):57–69. [DOI] [PubMed] [Google Scholar]
  • 50.Harris William J, Huang X, Lynch James T, Spencer Gary J, Hitchin James R, Li Y, et al. The Histone Demethylase KDM1A Sustains the Oncogenic Potential of MLL-AF9 Leukemia Stem Cells. Cancer Cell 2012;21(4):473–87. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement Information
NIHMS977875-supplement-Supplement_Information.pdf (530.3KB, pdf)

RESOURCES