Conditional Ror1 knockout reveals crucial involvement in lung adenocarcinoma development and identifies novel HIF-1α regulator.

We previously reported that ROR1 is a crucial downstream gene for the TTF-1/NKX2-1 lineage-survival oncogene in lung adenocarcinoma, while others have found altered expression of ROR1 in multiple cancer types. Accumulated evidence therefore indicates ROR1 as an attractive molecular target, though it has yet to be determined whether targeting Ror1 can inhibit tumor development and growth in vivo. To this end, genetically engineered mice carrying homozygously floxed Ror1 alleles and an SP-C promoter-driven human mutant EGFR transgene were generated. Ror1 ablation resulted in marked retardation of tumor development and progression in association with reduced malignant characteristics and significantly better survival. Interestingly, gene set enrichment analysis identified a hypoxia-induced gene set (HALLMARK_HYPOXIA) as most significantly downregulated by Ror1 ablation in vivo, which led to findings showing that ROR1 knockdown diminished HIF-1α expression under normoxia and clearly hampered HIF-1α induction in response to hypoxia in human lung adenocarcinoma cell lines. The present results directly demonstrate the importance of Ror1 for in vivo development and progression of lung adenocarcinoma, and also identify Ror1 as a novel regulator of Hif-1α. Thus, a future study aimed at the development of a novel therapeutic targeting ROR1 for treatment of solid tumors such as seen in lung cancer, which are frequently accompanied with a hypoxic tumor microenvironment, is warranted.

INTRODUCTION

Receptor tyrosine kinase‐like orphan receptor 1 (ROR1) is an evolutionarily conserved type‐I transmembrane receptor tyrosine kinase that plays an essential role in embryonic development. , Although ROR1 expression becomes markedly diminished after birth, , accumulating evidence indicates that it is highly upregulated and involved in various malignancies, including those of the lung, breast, pancreas, skin, ovary, and colon, and also hematopoietic cells. , , , , , , , , Four independent groups including our own have reported that the homeodomain transcription factor TTF‐1/NKX2‐1 plays a role as a lineage‐survival oncogene in lung adenocarcinoma. , , , However, that cannot be utilized as a molecular target for treatment, because of its essential roles for not only lung morphogenesis but also normal lung physiology, such as surfactant protein production and secretion. , Along this line, our subsequent identification of ROR1 as a direct transcriptional target of TTF‐1/NKX2‐1 in lung adenocarcinoma makes this molecule an attractive potential molecular target for treatment of lung adenocarcinoma. ROR1 sustains EGFR‐mediated PI3K‐AKT prosurvival signaling and inhibits ASK1‐p38 MAPK proapoptotic signaling in both kinase‐dependent and kinase‐independent manners, , while it also maintains caveolae structure and caveolae‐dependent endocytosis by functioning as a scaffold protein for caveolin‐1, cavin‐1, and cavin‐3 in a kinase‐independent fashion, and sustaining prosurvival signaling of other receptor tyrosine kinases (RTKs) including MET and IGF1R. ,

The present study was conducted to explore the importance of Ror1 in tumor development in vivo using an animal model. To this end, genetically engineered mice carrying both homozygous floxed Ror1 alleles and the SP‐C promoter–driven human mutant EGFR transgene were generated, which enabled the investigation of the roles of Ror1 in a lung adenocarcinoma model. The results are the first to show that Ror1 ablation inhibits tumor development in mice and significantly increases the survival of those bearing mutant EGFR‐elicited tumors. In addition, evidence indicating a novel function of ROR1 to sustain expression of HIF‐1α, a master transcriptional regulator of hypoxia‐induced genes, is presented.

MATERIALS AND METHODS

Generation of mouse models

All animal studies were performed after receiving approval for the protocols from the Institutional Animal Care and Use Committee of Nagoya University Graduate School of Medicine and Aichi Cancer Center. In brief, a neo‐loxP cassette carrying the FRT‐flanked PGK promoter–driven Neo gene was inserted into intron 4 of the Ror1 gene, while diphtheria toxin A‐fragment (DT‐A) driven by the PGK promoter was placed at the 5’ of the targeting genomic fragment (Figure 1A). Mouse C57BL/6 ES cells were transfected with the targeting vector by electroporation, and G418‐resistant clones were screened for homologous recombination by PCR. Positive clones were injected into Balb/c blastocysts to generate Ror1 +/NeoFRT mice, which were crossed with FLPe transgenic mice to remove the Neo cassette. Offspring were backcrossed to wild‐type C57BL/6J mice to remove the CAG‐FLPe transgene, resulting in generation of Ror1 +/ mice. Germline Ror1 knockout mice were generated by crossing Ror1 and CAG‐Cre mice, obtained from the Jackson Laboratory. Ror1 mice were also crossed with CAG‐CreER transgenic mice from the Jackson Laboratory and the resultant Ror1 was used to conditionally ablate the endogenous Ror1 gene upon induction of Cre recombinase in response to tamoxifen (TAM) administration.

FIGURE 1

Generation of Ror1 (RCE) mice. A, Schematic of wild‐type Ror1 allele, targeting vector, and targeted NeoFRT, floxed, and deleted allele. Recognition sites for KpnI, SacI, SpeI, and BamHI in the flanking region of exons 3 and 4 (filled boxes) are indicated. In the targeting vector, a 1.7‐kb DNA fragment carrying exons 3 and 4 and its flanking region was replaced by a 3.5‐kb fragment carrying 2 loxP sequences (red arrowheads) and PGK‐neo (NeoR) flanked by FRT sequences (open arrowheads). Diphtheria toxin A‐fragment (DT‐A; grey arrow) driven by the PGK promoter was inserted at 5' site of the vector. Scale bar represents 1.0 kb. B, Diagram of construct used to produce human mutant EGFR transgenic mice. C, Incidence of lung tumors in RCE mice determined by micro‐CT observation. D, Representative H&E staining of lung tumors from RCE mice at 16 weeks of age. Scale bars represent 200 μm (left) and 50 μm (right). E, Expression of Ror1 and Ttf1 mRNA in lung tumors in Ror1 (RE) mice as well as in normal lung tissues of Ror1 mice, both at 19 weeks of age. qRT‐PCR analysis was performed using representative lung tumors and normal lung tissues, one each from five RE and four Ror1 mice, respectively. Columns and bars indicate average and standard deviation, respectively. P‐values were calculated using a two‐tailed Student's t‐test. **P < .01. F, Expression of Ror1 and Ttf‐1 protein in lung tumors and normal lung tissues in RE and Ror1 mice, respectively, at 19 weeks of age. Fetal lung tissues were obtained from Ror1 mice on day E18.5. NCI‐H23 and PC‐9 cell lysates were used as negative and positive controls, respectively. β‐actin served as a loading control

To generate human mutant EGFR transgenic mice, a mutant EGFR construct lacking residues 746‐750 was generated by introducing a deletion into the wild‐type pcDNA3.1‐EGFR vector using a KOD ‐Plus‐ Mutagenesis Kit (TOYOBO, SMK‐101). The primers were F: 5′‐ACATCTCCGAAAGCCAACAAG‐3′ and R: 5′‐CTTGATAGCGACGGGAATTTT‐3′. The full‐length mutant EGFR cDNA and poly(A) addition signal sequence were amplified using the mutant pcDNA3.1‐EGFR vector as a template using PCR primers containing a SalI site for cloning. The primers were F: 5′‐ AAAGTCGACATGCGACCCTCCGGGACGGC ‐3′ and R: 5′‐AAAGTCGACTCATGCTCCAATAAATTCAC‐3′. After digestion with SalI, the amplified product was cloned downstream of the surfactant protein C (SP‐C) promoter in the pUC18 vector (a generous gift from Prof. Jeffery Whitsett). This mutant EGFR expression cassette was excised by NotI digestion and injected into fertilized C57BL/6J eggs.

Animal husbandry, genotyping, and treatment

All animals were maintained in specific pathogen‐free housing with adequate food and water according to guidelines approved by the Nagoya University Graduate School of Medicine and Aichi Cancer Center. Genomic DNA samples from mouse tail snips or ear punch samples were used for PCR genotyping. The primer sequences for PCR genotyping are listed in Table S1. For conditional knockout of Ror1, mice were injected intraperitoneally with TAM (100 mg/kg body weight) once every 5 days from 8 to 30 weeks of age. Cre‐mediated recombination of the floxed Ror1 alleles by TAM administration was confirmed by PCR using the following primer pair: Ror1 F: 5′‐CTATCAATAAGAAGGTTTAGGG‐3′ and Ror1 R: 5′‐CTCTCCCTCCTCTTTGTGAG‐3′. Gefitinib (10 mg/kg; SML1657, Sigma‐Aldrich) or the vehicle was administered by oral gavage once daily for 4 days.

Isolation of mouse embryonic fibroblasts (MEFs)

Mouse embryonic fibroblasts were prepared from day‐13.5 embryos of Ror1 and Ror1 mice, as described in Supplementary Methods. MEFs after four passages were treated for 5 days with 4‐hydroxytamoxifen (4‐OHT) to ablate the Ror1 floxed alleles, followed by harvesting for DNA and protein extraction.

Micro‐CT analysis

From 8 weeks of age, each mouse was anesthetized with isoflurane and scanned weekly using a SkyScan 1176 micro‐CT analyzer. For acquisition of micro‐CT images, 282 projection data were collected in one full rotation of the gantry at X‐ray settings of 80 kV and 313 μA with aluminum and copper filters. All images were reconstructed using the N‐Recon program according to the manufacturer's protocol. Tumor volume in CT images was estimated by subtracting volume that exhibited air from total lung volume using Image J software (https://imagej.nih.gov/ij/). Normal pulmonary blood vessels were difficult to distinguish from similarly sized small tumors; thus they were included together with tumors in the calculation. Total tumor volume was calculated as the sum of tumors in two axial planes: one encompassing the seventh vertebra and the space between the third and fourth sternebrae, and the other the ninth vertebra and the space between the fourth and fifth sternebrae (Figure S1).

Western blotting and quantitative real‐time (qRT)‐PCR analyses

Western blotting analyses were performed according to standard procedures using Immobilon‐P filters (Millipore), an enhanced chemiluminescence system (GE Healthcare), and antibodies listed in Table S2. qRT‐PCR analysis was also performed according to standard procedures, as described in Supplementary Methods, with the primers listed in Table S3. A TaqMan gene expression assay (Applied Biosystems) was also used to determine Ror1 (Mm00443463_m1), Ttf‐1(Mm00447558_m1), and 18s (Hs99999901_s1) mRNA expression.

Immunohistochemistry

Mouse tissues were collected, then fixed with 10% neutral buffered formalin for 48 hours at room temperature before embedding in paraffin. Hematoxylin and eosin (H&E) and immunohistochemical staining was performed using standard methods, as described in detail in Supplementary Methods. Two investigators (NA and HI) determined the percentage of MCM2‐positive cells in 24 tumors from three TAM‐treated mice and 27 tumors from three control mice under 10× magnification.

Microarray analysis

To obtain gene expression data using a dye‐swap strategy (GSE161182), 50 ng of total RNA from tumor tissues of TAM‐treated (n = 3) and control (n = 3) Ror1 mice was used for fluorescent (Cy‐5 or Cy‐3) generation of labeled cDNA. Microarray analysis was done using SurePrint G3 Mouse 8x60K Microarray G4851B, ver. 2 (Agilent Technologies) and GeneSpring, ver. 13.1 (Agilent), essentially as previously described. Gene set enrichment analysis (GSEA) was performed as detailed in Supplementary Methods.

Cell cultures

NCI‐H23 and NCI‐H1975 cells were purchased from the American Type Culture Collection, and PC‐9 cells from RIKEN Cell Bank. NCI‐H1975 cells overexpressing siRNA‐resistant wild‐type (ROR1‐WTm) or kinase‐dead (ROR1‐KDm) ROR1 used in the reconstitution experiments were previously reported. NCI‐H1975 and PC‐9 cells were free from mycoplasma contamination and authenticated by short tandem repeat DNA profiling performed in February 2019 at the Japanese Collection of Research Bioresources, National Institute of Biomedical Innovation of Japan. NCI‐H1975 and PC‐9 cells were grown under a normoxia (21% O2) or hypoxia (1% O2) condition for 24 hours at 37ºC/5% CO2 and then harvested for Western blotting and qRT‐PCR analyses.

siRNA and transfection

NCI‐H1975 and PC‐9 cells were transfected with 20 nmol/L siRNA using Lipofectamine RNAiMAX (Life Technologies), according to the manufacturer's instructions, using the following siRNAs: sicontrol (AllStars Negative Control siRNA, Qiagen) and siROR1 (Hs_ROR1_3, Qiagen). Cells were cultured for 24 hours and then placed under hypoxia or normoxia for 24 hours, with and without 5 μmol/L MG132 or DMSO for the final 6 hours prior to harvesting.

Statistical analysis

Continuous variables were compared using a two‐tailed unpaired Student's t‐test. Survival analysis was performed using Kaplan‐Meier survival curves. A log‐rank test was used to evaluate the statistical significance of differences of survival curves. Statistical analysis was done using the Prism 8.0 application (GraphPad).

RESULTS

To determine the roles of Ror1 in lung adenocarcinoma development in vivo in an animal model, we generated Ror1 conditional knockout mice by placing loxP sites at the 5' and 3' positions of exons 3 and 4, respectively, of the Ror1 gene (Ror1; Figure 1A), in which all major functional domains were deleted upon Cre induction. To confirm the occurrence of intended loss of functional Ror1 by recombination, germline Ror1 knockout mice were first generated by crossing Ror1 with CAG‐Cre transgenic mice. While Ror1‐heterozygous mice did not exhibit any obvious phenotypes, intercrosses of Ror1‐heterozygous mice resulted in neonatal lethality in germline homozygous Ror1 knockout mice (Figure S2A), a finding consistent with a previous study. Next, Ror1 mice were crossed with CAG‐CreER transgenic mice, which made it possible to ablate the endogenous Ror1 gene upon induction of Cre recombinase in response to TAM administration. To validate generation of Ror1‐null alleles by recombination of the floxed Ror1 alleles, MEFs isolated from E13.5 embryos of Ror1 and Ror1 mice were treated in vitro with 4‐OHT for 5 days. We observed efficient recombination of the Ror1 gene, leading to nearly complete depletion of Ror1 expression in Ror1 MEFs, but not in Ror1 MEFs (Figure S2B,C).

The in vivo effects of Ror1 depletion in adulthood were first investigated using conditional knockout mice. Ror1 mice were injected intraperitoneally with TAM (100 mg/kg body weight) or oil once every 5 days from 8 to 30 weeks of age. While recombination of Ror1 was confirmed in normal organs, including lung, stomach, small intestine, colon, liver, kidney, and spleen tissues (Figure S3A), all Ror1 mice with TAM administration appeared healthy, and no significant differences in body weight or survival were observed between Ror1 mice administered with TAM and those that received oil (data not shown). In addition, Western blot analysis showed a low but appreciable level of Ror1 mRNA expression in mice without Cre induction (Figure S3B), whereas no appreciable pathologic changes were observed in the lung, stomach, small intestine, colon, liver, kidney, or spleen tissues of Ror1 mice with TAM administration (Figure S3C), suggesting that Ror1 is not essential in adulthood.

A lung adenocarcinoma model was also established by generating transgenic mice expressing human mutant EGFR driven by the pulmonary SP‐C promoter, which is active only in type II alveolar epithelial cells (SPC‐hEGFR; Figure 1B). Twenty‐seven pups were obtained following injection of the transgene into fertilized eggs. A line positive for the transgene‐produced offspring that developed lung adenocarcinoma at 8 weeks of age or older was then selected for further analysis. Ror1 and Ror1 mice were crossed with SPC‐hEGFR mice, which generated Ror1 (hereafter RCE) and Ror1 (hereafter RE) mice, respectively. qRT‐PCR assays using human EGFR‐specific primers and mRNAs extracted from various tissues confirmed lung‐specific expression of human EGFR (Figure S4A). RCE mice developed multifocal lung adenocarcinomas at 8 weeks of age or older, with 100% incidence by 16 weeks of age, the same as observed in SPC‐hEGFR mice (Figure 1C,D). Treatment with gefitinib led to rapid tumor regression, suggesting that lung tumors driven by human mutant EGFR are highly dependent on EGFR signaling (Figure S4B,C). Furthermore, increased levels of expression of Ror1 and Ttf‐1/Nkx2‐1 at both the mRNA and protein levels were noted in the lung tumors as compared with normal lung tissues (Figure 1E,F), which are findings similar to those previously reported for human lung adenocarcinoma cases with increased TTF‐1 and ROR1 expression. , , , ,

To determine the role of Ror1 in lung tumorigenesis in vivo in an animal model, RCE mice without any detectable lung tumors seen in a micro‐CT examination were administered an intraperitoneal (i.p.) injection of 100 mg/kg of TAM once every 5 days from 8 weeks of age. Ablation of Ror1 significantly delayed tumor development, allowing all mice to survive until 30 weeks of age, whereas control RCE mice administered with oil alone (n = 6) showed marked disease progression and a median survival time of 23 weeks (P = .0006, log‐rank test; Figure 2A‐C). Tumors arising in RCE mice conditionally knocked out for Ror1 with TAM treatment had a considerably preserved existing lung parenchyma architecture as compared with RCE mice treated with the solvent (Figure 2D). Survival curves derived from RE mice treated with TAM (n = 5) were not significantly different from those of RCE mice treated with oil alone (Figure S5), indicating that TAM administration did not have a significant effect on survival.

FIGURE 2

Ror1 required at early stage of lung tumor development. A, Kaplan‐Meier survival curves comparing Ror1 (RCE) mice administered with tamoxifen (TAM) (n = 9) or oil (n = 6). P‐values were calculated using a log‐rank test. B, Representative axial images of RCE mice administered with TAM or oil at 0, 9, and 18 weeks after treatment. Scale bars represent 5 mm. C, Tumor volume/lung volume ratios calculated using Image J at 0, 9, and 18 weeks after treatment. In bottom panels, columns indicate average and bars standard deviation. P‐values were calculated using a two‐tailed Student's t‐test. NS, not significant. *P < .05. D, Representative H&E staining and Victoria blue staining, and counterstaining with nuclear fast red (NFR) of lung tumors from RCE mice administered with TAM (n = 3) or oil (n = 3). Scale bars represent 200 μm

We also examined whether Ror1 knockout inhibits growth of already detectable tumors because ablation of Ror1 at an early time point without a detectable tumor may not reflect the therapeutic impact of ROR1 inhibition in future clinical applications. Weekly micro‐CT examinations of RCE mice (n = 9) from 11 to 16 weeks of age were performed, and i.p. injections of 100 mg/kg of TAM or solvent oil alone once every 5 days were initiated when a tumor larger than 3 mm2 was detected. Ror1 ablation resulted in marked retardation of tumor progression, with these RCE mice showing significantly better survival as compared with those treated with the solvent alone (P = .0031, log‐rank test; Figure 3A‐C).

FIGURE 3

Ror1 ablation delays lung tumor growth. A, Kaplan‐Meier survival curves comparing Ror1 (RCE) mice administered with tamoxifen (TAM) (n = 9) or oil (n = 7). P‐values were calculated using a log‐rank test. B, Representative axial images of RCE mice administered with TAM or oil at 0, 4, and 8 weeks after treatment. Scale bars represent 5 mm. C, Tumor volume/total lung volume ratios were calculated using Image J every 4 weeks after treatment. In bottom panels, columns indicate average and bars standard deviation. P‐values were calculated using a two‐tailed Student's t‐test. NS, not significant. *P < .05

We previously observed that ROR1 knockdown affected RTK signaling towards AKT and p38 in TTF‐1/NKX2‐1‐positive human lung adenocarcinoma cell lines in vitro. However, it was interesting that no readily noticeable changes in regard to activation of Akt, p38 Mapk, Erk, or tyrosine kinase receptors, including EGFR, Met, Her2, and IGF‐1R, were observed in tumors arising in RCE mice conditionally knocked out for Ror1 in the present study (Figure 4A). In this regard, the human mutant EGFR transgene was found to be expressed at a very high level (Figure S6), which possibly results in maintaining the activities of these signaling pathways through heterodimerization and transactivation of RTKs, , , , , thus conferring insensitivity to Ror1 ablation. Nevertheless, Ror1 ablation significantly suppressed tumor growth, which was accompanied by a marked decrease in expression of the proliferation marker MCM2 (Figure 4B). We speculated that Ror1 may also regulate other crucial molecules in lung adenocarcinoma cells. To gain molecular insight into pathways perturbed by Ror1 knockout in mice, GSEA based on the mRNA expression profiles of lung tumors arising in RCE mice with and without TAM administration was performed. As a result, a gene set of hypoxia‐induced genes (HALLMARK_HYPOXIA) was found to be the most significantly downregulated by Ror1 ablation (P < .0001, permutation test; NES = −1.82; FDR = 0.0083; Figure 4C), suggesting a potential regulatory role of Ror1 in regard to expression of hypoxia‐induced genes.

FIGURE 4

Molecular pathways perturbated by Ror1 ablation in vivo. A, Western blotting analysis of Ror1, p‐AKT (T308), AKT, p‐p38 (T180/Y182), p38, p‐Erk1/2 (T203/Y205), Erk1/2, p‐EGFR (Y845), p‐EGFR (Y992), p‐EGFR (Y1068), EGFR, p‐Met (Y1234/Y1235), Met, p‐Her2 (Y877), Her2, p‐IGF1R (Y1135/Y1136), IGF1R, and β‐actin in normal lung tissues from Ror1 mice (n = 2), and lung tumors from Ror1 (RCE) mice administered with tamoxifen (TAM) (n = 4) or oil (n = 2). B, Representative images showing immunohistochemical analysis of MCM2 and percentages of MCM2‐positive cells in 24 tumors from RCE mice administered with TAM (n = 3) and 27 tumors from RCE mice administered with oil (n = 3) Scale bars represent 200 μm. Columns indicate average and bars standard deviation. **P < .01, two‐tailed Student's t‐test. C, Hallmark gene sets were significantly decreased in lung tumors in RCE mice administered with TAM as compared to those in RCE mice administered oil alone. P < .01 and false discovery rate (FDR) q < 0.05 were considered to indicate significance. Result for the “HALLMARK_HYPOXIA” gene set was shown on the right. P‐values were calculated based on gene permutation. NES, normalized enrichment score

Hypoxia is a commonly observed phenomenon in solid tumors because of an associated rapid increase in tumor mass and disproportional and inadequate tumor angiogenesis. Under a hypoxic condition, hypoxia‐inducible factor 1 alpha (HIF‐1α) has a pivotal role to mediate adaptive cellular responses to hypoxia. Accordingly, we examined the expression of Hif‐1α in lung tumors that developed in RCE mice. Expression levels of Hif‐1α were greater in lung tumors as compared with normal lung tissues and significantly decreased by Ror1 ablation (Figure S7A). Furthermore, the expression levels of Hif‐1α target genes associated with glucose metabolism, including Hk2, Ldha, and Aldoa, were significantly decreased in accordance with Ror1 ablation–mediated suppression of Hif‐1α. On the other hand, expression levels of Hif‐1α–regulated growth factors, such as Vegfa and Tgfa, were not affected (Figure S7B), suggesting the possibility of maintenance of Vegfa and Tgfa expression due to EGFR transgene overexpression. ,

Next, whether ROR1 is involved in regulation of HIF‐1α in human lung adenocarcinoma cell lines was examined. ROR1 knockdown using siRNA diminished HIF‐1α expression under normoxia (21% O2) and clearly hampered induction of HIF‐1α under hypoxia (1% O2) in NCI‐H1975 (Figure 5A), findings that were substantiated by decreased expression of the hypoxia‐responsive, HIF‐1α–regulated gene LOX (Figure 5B). While ROR1 knockdown modestly affected expression levels of HIF1A mRNA under normoxia, it did not elicit any changes under hypoxia (Figure 5C). HIF‐1α is known to be primarily controlled by polyubiquitination and proteasomal degradation under normoxia. In this regard, treatment with MG132, a proteasome inhibitor, did not restore HIF‐1α protein expression under either a normoxic or hypoxic condition in cells treated with siROR1 (Figure 5D), suggesting involvement of another mechanism in ROR1‐mediated regulation of HIF‐1α protein. Similar findings were obtained with PC‐9 cells (Figure S8A‐D). Finally, findings obtained with siROR1‐treated NCI‐H1975 cells reconstituted with wild‐type or kinase‐dead ROR1 showed that wild‐type but not kinase‐dead ROR1 with silent mutations at the siRNA binding site recovered hypoxia‐mediated induction of HIF‐1α (Figure 6), suggesting that ROR1 regulates HIF‐1α expression in a kinase‐dependent manner.

FIGURE 5

Involvement of ROR1 in regulation of HIF‐1α protein expression. A, Western blot analysis of ROR1 and HIF‐1α under normoxia (21% O2) or hypoxia (1% O2). B and C, Quantitative RT‐PCR of LOX (B) and HIF1A (C) in NCI‐H1975 cells treated with sicontrol or siROR1 under normoxia (21% O2) or hypoxia (1% O2). P‐values were calculated using a two‐tailed Student's t‐test. NS, not significant. **P < .01. D, Western blot analysis of ROR1 and HIF‐1α in NCI‐H1975 cells treated with sicontrol or siROR1 under normoxia (21% O2) or hypoxia (1% O2), with 5 μmol/L MG132 or DMSO for 6 h prior to harvesting. β‐actin served as a loading control in experiments shown in (A) and (D)

FIGURE 6

Regulation of HIF‐1α expression by ROR1 in kinase‐dependent manner. Western blot analysis of ROR1 and HIF‐1α in presence or absence of siROR1 treatment in NCI‐H1975 cells overexpressing siRNA‐resistant wild‐type (ROR1‐WTm) or kinase‐dead (ROR1‐KDm) ROR1. ROR1‐WTm and ROR1‐KDm carry multiple silent mutations at the binding site of siROR1. VC, vector control

DISCUSSION

We and others have previously identified TTF‐1/NKX2‐1 as a lineage‐survival oncogene of lung adenocarcinoma, , , , , after which our subsequent identification of ROR1 as a crucial downstream factor of TTF‐1/NKX2‐1–mediated signaling strongly supported the notion of ROR1 as an attractive molecular target for developing novel therapeutics for lung cancers, , , , as TTF‐1/NKX2‐1 cannot be targeted due to its essential role in maintaining physiological functions of the lung. , In addition, accumulated evidence suggests crucial functions of ROR1 in regulation of cell proliferation, apoptosis, self‐renewal of cancer stem cells, epithelial‐mesenchymal transition, and metastasis in multiple types of cancer, , , , while a significant association of ROR1 expression with worse overall survival has also been reported. , These findings have led to exploration of ROR1‐targeted therapies, including monoclonal antibodies, , CAR‐T cells, , and small molecule inhibitors. , However, to the best of our knowledge, the importance of Ror1 in development and progression of any cancer type has never been proven in vivo in an animal model. In the present study, examinations of conditional Ror1 knockout mice carrying the human mutant EGFR transgene demonstrated that conditional Ror1 ablation strongly inhibits lung tumor development and growth in association with markedly better survival. Interestingly, the lung parenchymal architecture was found to be better preserved in tumor tissues arising in TAM‐treated RCE mice as compared with those treated with oil alone, a finding consistent with lower level of malignancy of tumors lacking Ror1.

The present results revealed that ROR1 is involved in regulation of HIF‐1α expression in lung adenocarcinoma cells under both normoxic and hypoxic conditions. HIF‐1α is a master regulator of adaptive responses to hypoxia and has profound effects on various aspects of cancer development, including proliferation, angiogenesis, metabolism, migration and invasion, metastasis, resistance to therapy, and poor prognosis. , , , Therefore, direct targeting of HIF‐1α or inhibition of downstream effector pathways is considered to have great promise for treatment of cancer. In addition, targeting RTKs has been suggested as an alternative means to suppress HIF‐1α–mediated signaling. , , , In this regard, HER2 and EGFR have been shown to induce HIF‐1α protein via the PI3K/AKT/MTOR signaling pathway in breast cancer cell lines, , while VEGF/VEGFR1‐mediated MAPK signaling was found to promote HIF‐1α transcriptional activity in neuroblastoma cells by increasing phosphorylation of HIF‐1α. It was also reported that various RTKs, including VEGFR1, PDGFR‐β, RET, and EGFR, regulate HIF‐1α mRNA expression. Notably, inhibition of a single RTK was shown to have only a minimal effect on HIF‐1α expression, perhaps because of redundant regulation of HIF‐1α by multiple RTKs. With this in mind, it is interesting to note that the present results clearly demonstrated that ROR1 inhibition alone was sufficient to suppress HIF‐1α expression under both normoxic and hypoxic conditions despite sustainment of RTK signaling.

In conclusion, conditional Ror1 knockout mice carrying a human mutant EGFR transgene were established in the present study. Experiments with that animal model clearly demonstrated for the first time that Ror1 has crucial involvement in lung adenocarcinoma development and growth in vivo. In addition, the importance of ROR1 in regulation of HIF‐1α expression was also shown. ROR1 thus appears to be a promising molecular target for the development of a novel therapeutic to treat solid tumors such as lung cancer, which are frequently accompanied with a hypoxic microenvironment.

DISCLOSURE

The authors have no competing interests to declare.

Fig S1‐S8

Click here for additional data file.

Table S1‐S3

Click here for additional data file.

Supplementary Material

Click here for additional data file.