The mouse MLL/AF10(OM-LZ) leukemia cell line (12G) and MLL/AF10(OM-LZ) cells harboring wild-type or activating KRAS (KRAS^G12C) and wild-type or activating PTPN11 (PTPN11^G503A) were generated by the retroviral transduction of genes to 5-fluorouracil (5-FU)-enriched C57BL/6J (B6) mouse bone marrow (BM) cells. The mice were purchased from the National Laboratory Animal Center. These different cell types were either generated in the current study (AKw1G) or in previous studies (AK2G, AK3G, APw1 and APm1) (21,22,26) (Fig. 1A). All of these cell lines expressed the myelomonocytic markers. Mouse leukemia cells were cultured in the RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) with 20% fetal bovine serum (Hyclone; Cytiva), 2 mM L-glutamine, 100 µM 2-mercaptoethanol and 10 ng/ml IL-3 (R&D Systems, Inc.) for maintenance and proliferation analysis.

The total RNA of leukemia cells was prepared using TRIzol^® reagent (Gibco; Thermo Fisher Scientific, Inc.). The RNA was amplified, labeled and hybridized to the mouse genome 430A Array chip (12G vs. AK3G and 12G vs. AK2G), 430 2.0 Array chip (12G vs. AK2G and APw-1 vs. APm-1), or Clariom D Array chip (APw-1 vs. APm-1 and AKw1G vs. AK3G) (Affymetrix; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions (this procedure was performed by staff of the Genomic Medicine Research Core Laboratory at Chang Gung Memorial Hospital, Linkou, Taiwan). Microarray data are available at the NCBI Gene Expression Omnibus (GEO) website (https://www.ncbi.nlm.nih.gov/geo/; accession nos. GSE82156 and GSE134586) or can be downloaded from Chang Gung University website (21). Differential Expression Analysis was performed using Transcriptome Analysis Console software version 4.0 (Affymetrix; Thermo Fisher). A heat map was obtained using Cluster version 3.0 (http://bonsai.hgc.jp/~mdehoon/software/cluster/) and Java TreeView version 1.1.6r4 (http://jtreeview.sourceforge.net). DEGs with >1.5-fold-change in paired MLL/AF10 leukemia cells harboring wild-type and oncogenic KRAS or PTPN11 were then functionally annotated with GO enrichment analysis using online Database for Annotation, Visualization and Integrated Discovery (DAVID) version 6.8 annotation tools (https://david.ncifcrf.gov/). Statistical significance was evaluated using Fisher's exact test and corrected by Bonferroni correction for multiple testing. The GO ‘Molecular Function’ (MF) or ‘Biological Process’ (BP) categories with P<0.05 were considered statistically significant.

To validate the differential expression and drug responsiveness of HOXA11 in MLL-t AML, a meta-analysis was performed using the Oncomine™ database (https://www.oncomine.org/), including Valk leukemia, Wouters leukemia, Balgobind leukemia, and Haferlach leukemia (27–30).

To evaluate the expression levels of target genes, the total RNA of mouse leukemia cells or AML patient BM cells was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher). The total RNA was reverse transcribed into complementary DNA (cDNA) using random hexamers and SuperScript™ II reverse transcriptase (Thermo Fisher) according to the manufacturer's protocol. qPCR was performed using SYBR-Green PCR master mix (Applied Biosystems; Thermo Fisher Scientific, Inc.) and analyzed by ABI Prism 7900 system (Applied Biosystems; Thermo Fisher Scientific, Inc.). Primer sets were as follows: Mouse Hoxa11, 5′-GAAAACCTCGCTTCCTCCGA-3′ and 5′-ATAAGGGCAGCGCTTTTTGC-3′; mouse NF-κB inhibitor α (Nfkbia), 5′-GAGACCTGGCCTTCCTCAAC-3′ and 5′-TCTCGGAGCTCAGGATCACA-3′; mouse transcription factor p65 (Rela), 5′-TGGCTACTATGAGGCTGACCT-3′ and 5′-TGGTCTGGATTCGCTGGCTA-3′; mouse transformation-related protein p53 (Trp53), 5′-CCTCTCCCCCGCAAAAGAAA-3′ and 5′-GGCCCTTCTTGGTCTTCAGG-3′; mouse Gapdh, 5′-TTCACCACCATGGAGAAGGC-3′ and 5′-GGCATGGACTGTGGTCATGA-3′; human HOXA11, 5′-CGTCTTCCGGCCACACTGA-3′ and 5′-AGACGCTGAAGAAGAACTCCC-3′; and human ABL, 5′-TGGAGATAACACTCTAAGCATAACTAAAGG-3′ and 5′-GATGTAGTTGCTTGGGACCCA-3′. The thermocycling conditions were as follows: 95°C for 2 min; then 40 cycles of 95°C for 15 sec and 65°C for 30 sec. The gene expression levels of mouse genes and human HOXA11 were normalized against the housekeeping genes Gapdh and ABL, respectively. Fold-change was calculated using the 2^−∆∆Cq method (31). In the present study, leftover BM samples of clinical examination for initial diagnostic work-up of AML were used for gene expression analysis. Samples were obtained from patients admitted to Chang Gung Memorial Hospital (Taipei, Taiwan) between January 2002 and December 2010. The inclusion criteria were as follows: Adults and children with AML with MLL-t. The exclusion criteria were as follows: Non-MLL-t AML specimens. Among the 114 cases with MLL-t AML (56 men and 58 women, age range 0–84 years, median 27 years), it was determined that eight cases with MLL/AF10 or MLL/AF9 had sufficient remaining specimens available for HOXA11 expression level analysis. Of these eight cases, five had KRAS mutations and one had a PTPN11 mutation.

Total cell lysate from 5×10⁶ leukemia cells was prepared by direct lysis of cells with RIPA buffer [20 mM Tris-Cl (pH 7.5) 150 mM NaCl, 1% Triton X-100, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 0.5 mM PMSF]. The amount of total protein was assayed using a Pierce^® BCA Protein Assay (Pierce; Thermo Fisher Scientific, Inc.). The lysates (20 µg/lane) were electrophoresed on 10% polyacrylamide gel, and subsequently transferred to an Immobilon membrane (EMD Millipore). The membrane was blocked in 5% bovine serum albumin at 4°C (Sigma-Aldrich; Merck KGaA) for 1 h, and then incubated with primary antibodies at 4°C overnight against mouse Hoxa11 (1:5,000; cat. no. NBP1-80228; Novus Biologicals, Ltd.), β-actin or Gapdh (1:10,000; cat. nos. sc-47778 or sc-32233; Santa Cruz Biotechnology, Inc.), followed by incubation for 2 h at room temperature with horseradish peroxidase-conjugated anti-mouse or anti-rabbit IgG secondary antibodies (1:5,000; cat. nos. C04001 or C04003; Croyez Bioscience Co., Ltd.). Western blots were developed with a Western Lightning Plus ECL kit (PerkinElmer, Inc.) and the images were visualized by Analytik Jena™ UVP ChemStudio PLUS and VisionWorks™ software (version 9.0; Analytik Jena US LLC).

The in vivo leukemogenic potential of the leukemia cell lines was determined by BM transplantation assay using male B6 mice (n=45, age, 6–8 weeks; weight, 20–24 g). Mice were maintained in pathogen-free devices under a controlled animal housing conditions (temperature 20±3°C, humidity 60–70% with a 12-h light /dark cycle and access to food and water ad libitum). Briefly, leukemia cells (APm1-shV, APm1-shH11-2, 12G-V, and 12G-H11) were intraperitoneally (i.p.) injected into mice (1×10⁶ cells/100 µl 1X PBS/mouse; n=10 mice for each leukemia cell line) that had received a sublethal dose of γ-irradiation (5.25 Gy) unless otherwise stated. Mice i.p. injected with normal saline (n=5) served as controls. To monitor leukemia development, peripheral blood (100 µl) was collected weekly for cytologic analysis and complete blood count measurement using Hemavet 950 (Drew Scientific; Erba Diagnostics, Inc.) or BC-5000 (Mindray Medical International Limited) hemocytometers. Mice were sacrificed when moribund (7–15 weeks post-injection). The moribund state was defined as mice displaying leukocytosis, with hunched posture, weakness, shortness of breath and 20% weight loss. Mice were euthanized by an i.p. injection of Zoletil (50 mg/kg) and Rompun (xylazine, 10 mg/kg) or by inhalation of isoflurane (3–5%), followed by cervical dislocation (32). BM, peripheral blood, ascites, organs and tumor masses were collected and weighed.

To generate stable gene knockdown cell lines, AK3G or APm-1 cells were infected with lentivirus expressing shRNA against Hoxa11 [The RNAi Consortium (cat. nos. TRCN0000413738 and TRCN0000417739)] at a multiplicity of infection of 1 and selected in RPMI-1640 medium containing puromycin (2.5 µg/ml) for a total of 2 weeks. A third generation system was used. Cells stably transfected with blank lentiviral vector pLKO_025 were used as negative controls. All lentiviruses were obtained from the National RNAi Core Facility at the Institute of Molecular Biology/Genomic Research Center, Academia Sinica (Taiwan).

The full-length Hoxa11 gene (~1 Kb) was amplified from cDNA of AK3G cells by PCR using the following primers: 5′-GAAGATCTCCCAAGGTAGCCCAATGATG-3′ and 5′-CCGCTCGAGCCAGTAGGCTGGAGCCTTAG-3′. The PCR product was digested with restriction enzymes BglII and XhoI, and was subsequently cloned into the BglII and XhoI sites of the retroviral vector pMSCVpuro (Clontech Laboratories, Inc.). The fidelity of nucleotide sequences of the Hoxa11 gene was confirmed by Sanger sequencing. Plasmid DNAs of pMSCV-Hoxa11 and pMSCVpuro were transfected into EcoPack2-293 cells (Invitrogen; Thermo Fisher Scientific, Inc.) to package retroviruses. Viral titer was determined by infection of NIH/3T3 (American Type Culture Collection; ATCC^® CRL-1658™), a murine fibroblast cell line, with serial diluted supernatant to generate puromycin-resistant colonies. To generate 12G cell lines with ectopic expression of Hoxa11 (12G-H11-1 and 12G-H11-2), 12G cells were mixed with retroviruses at a 1:1 ratio and selected in RPMI complete medium containing puromycin (2.5 µg/ml) for a total of 2 weeks. The cells transduced with MSCVpuro retroviruses were used as negative controls (12G-V1 and 12G-V3). The presence of Hoxa11 gene was confirmed by PCR amplification of the 1-Kb product from the genomic DNA of 12G-H11 cell lines using cloning primers. The fidelity of nucleotide sequences of Hoxa11 was confirmed by Sanger sequencing.

For cytologic analysis, cells were cytospinned at 700 × g for 3 min or smeared, air-dried, and stained with Liu reagents (Tonyar Biotech, Inc.) at room temperature. For immunophenotypic analysis, cells were stained at 4°C for 15 min with phycoerythrin-macrophage-1 antigen (Mac-1), phycoerythrin-CD115 and allophycocyanine-Ki-67 antibodies (cat. nos. RM2804-3, 12-1152-82 and 17-5698-82; eBioscience; Thermo Fisher) followed by flow cytometric analysis using FACSCanto II Cell Analyzer and FACSDiva software version 5.0 (BD Biosciences). For cell proliferation analysis, cells were assessed at indicated time points using Cell Counting Kit-8 (CCK-8/WST-8; Dojindo Molecular Technologies, Inc.) according to the manufacturer's instructions. To determine Ara-C resistance, cells were cultured in RPMI-1640 complete medium and a gradient concentration of Ara-C (0, 128, 320, 640, 1,600, 3,200, 16,000 or 40,000 ng/ml) for 24 h. Cell viability was measured using CCK-8 (incubation time 2 h). To determine apoptotic cell rate, cells were treated with Ara-C (0 or 160 ng/ml) for 24 h, followed by Annexin-V/propidium iodide (Sigma-Aldrich; Merck KGaA) staining in the dark and flow cytometric analysis using a FACSCanto II Cell Analyzer and FACSDiva software version 5.0 (BD Biosciences).

The competitive engraftment and clonal expansion assay was described previously (22,33). Briefly, paired cells (AK3G-shV vs. AK3G-shH11 and 12G-V vs. 12G-H11) were mixed at a ratio of 1:1 and then i.p. injected into B6 mice (1×10⁶ cells/mouse). The mice were sacrificed at days 43 and 57. Mice BMs and spleens were collected and used to extract genomic DNA. To amplify 300-bp DNA fragments of the pMSCVpuro vector (from 12G-V) or the region spanning the pMSCVpuro-Hoxa11 junction (from 12G-H11) from genomic DNA, PCR was performed using the following primers: MSCV 5′ primer 5′-CCCTTGAACCTCCTCGTTCAGCC-3′ in combination with MSCV 3′ primer 5′-GAGACGTGCTACTTCCATTTGTC-3′ or Hoxa11 primer 5′-GAGTAGCAGTGGGCCAGATTGC-3′. The PCR products were sequenced using MSCV 5′ primer, and the peak height of the 62th nucleotide (C for 12G-H11 and T for 12G-V) was measured. The (C/C+T) peak height ratio was converted to the cell ratio (12G-H11/12G-H11+12G-V) by aligning to the standard curve. The standard curve was generated by assessing the relationship between nucleotide peak height ratios and cell ratios from cell mixtures with mixed paired cells in ratios of 10:0, 9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, 1:9 and 0:10.

All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by National Institutes of Health (publication no. 85–23, revised 1996) and were carried out according to the protocol approved by the Animal Research Committee of Chang Gung Memorial Hospital (IACUC No. 2014092403; Taoyuan, Taiwan). Human sample collection was conducted in accordance with the Declaration of Helsinki and was approved by the Chang Gung Memorial Hospital Research Ethics Committee (IRB No. 96-1748B). Written informed consent was obtained from all patients. For patients under the age of 18, consent/permission was obtained from the parents/guardians.

Statistical analyses were performed using SPSS software version 20.0 (SPSS, Inc). The statistical significance of differences in gene expression levels of the two groups was compared using a Mann-Whitney test. Survival analysis was conducted according to the Kaplan-Meier method, and differences in survival were assessed using the log-rank test. Drug sensitivity was compared using an unpaired two-sample Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Our previous studies demonstrated that mice transplanted with AK2G, AK3G or APm-1 had shorter survival than those transplanted with the 12G and APw-1, respectively (21,22,26). Moreover, APm-1 cells have been reported to be more resistant to daunorubicin, but more sensitive to Ara-C than APw-1 cells (22). To identify genes involved in the acceleration of leukemia development and drug resistance by cooperation of MLL/AF10 with RAS signaling mutations, transcriptomic profiling were compared between 6 paired MLL/AF10(OM-LZ) leukemia cells harboring wild-type and oncogenic KRAS (12G vs. AK3G, 12G vs. AK2G and AKw1G vs. AK3G) or PTPN11 (APw-1 vs. APm-1) using cDNA microarray data. A total of 23 DEGs (seven upregulated and 16 downregulated) with >1.5-fold-change were identified in AK3G and APm-1 cell lines compared with 12G/AKw1G and APw-1 cells, respectively (Fig. 1B). The GO terms ‘BP’ and ‘MF’ enriched in the 23 DEGs included ‘innate immune response’, ‘immune system process’, ‘actin filament binding’, ‘cellular response to interferon-alpha’, ‘sequence-specific DNA’ and ‘regulation of gene expression’ (Fig. 1C). Among these enriched GO terms, three terms (‘innate immune response’, ‘immune system process’ and ‘cellular response to interferon-alpha’) are related to the immune response; ‘actin filament binding’ is related to cell protrusion and migration, while the last two terms (‘sequence-specific DNA’ and ‘regulation of gene expression’) are related to the regulation of gene expression by transcription factors.

A total of four genes, Hoxa11, PR domain zinc finger protein 5 (Prdm5), Iroquois-class homeodomain protein IRX-5 (Irx5) and homeobox protein PKNOX2 (Pknox2), were mapped to the GO ‘BP’ term ‘sequence-specific DNA’; other than Pknox2, these genes were upregulated in the AK3G and APm-1 cell lines according to cDNA microarray data (Fig. 1B). To investigate whether these genes were upregulated in the patients with AML with MLL-t, KRAS or PTPN11 mutations, meta-analysis was performed using datasets deposited by Valk (285 AML cases), Wouters (503 AML cases), Haferlach (542 AML cases), and Balgobind (237 childhood AML cases) in the Oncomine™ clinical research data repository for gene expression change. The results revealed that two (Balgobind and Haferlach) of the four AML series showed significant differences (P<0.01) in HOXA11 expression for patients with AML with or without MLL-t (11q23-r) using reporter ID 208493 (Fig. 2A). One (Valk) of the two AML series showed significant differences (P<0.05) in HOXA11 expression for patients with AML with or without KRAS activating mutations (KRASm) (Fig. 2B). In the Balgobind series, there was a trend of higher HOXA11 expression in patients with AML with PTPN11 mutations than those without PTPN11 mutations, but the difference was not statistically significant (Fig. 2C). This is probably due to the low case numbers (n=5). On the other hand, there were no significant differences in IRX5 and PRDM5 expression in leukemia cells of patients with AML with or without MLL-t, KRAS or PTPN11 mutations (Fig. S1A-F).

The results showed that MLL/AF10⁺ patients with AML with KRAS or PTPN11 mutations had higher levels of HOXA11 expression than patients with wild-type KRAS and PTPN11 genes (Fig. 2D). Similarly, MLL/AF9⁺ patients with AML with KRAS mutations expressed higher levels of HOXA11 compared with the patient with wild-type KRAS (Fig. 2E). These results supported that cooperation of MLL-t with oncogenic KRAS or PTPN11 mutation induces HOXA11 expression in patients with AML.

Based on the cDNA microarray data, Hoxa10 and Hoxa11 were upregulated (>1.5-fold) in AK3G compared with 12G cells (Fig. 3A). Compared with AKw1G, AK3G cells had higher expression levels (>1.5-fold) of Hoxa5, Hoxa6, Hoxa10 and Hoxa11 (Fig. 3B). Compared with APw-1, only Hoxa11 was upregulated in APm-1 cells (Fig. 3C). These data suggested that KRAS and PTPN11 mutations had overlapping, but not exactly the same effects on Hoxa gene expression in MLL/AF10 leukemia cells. Conversely, no significant differences in the expression levels of Hoxb, Hoxc, Hoxd and Meis1 genes were observed in these three pairs of mouse leukemia cell lines (Fig. 3A-C).

RT-qPCR and western blotting confirmed that the transcriptional and translational levels of Hoxa11 were increased in MLL/AF10(OM-LZ) mouse leukemia cell lines harboring KRAS mutation (AK2G, AK3G) and PTPN11 mutation (APm-1) compared with cells harboring wild-type KRAS (12G, AKw1G) or wild-type PTPN11 (APw-1) (Fig. 3D and E).

To determine the role of Hoxa11 in the leukemogenesis of MLL/AF10 leukemia cells harboring RAS pathway mutations, two lentivirus-based shRNAs that targeted mouse Hoxa11 gene (shH11-1 and shH11-2) were stably transduced into APm-1 cells. RT-qPCR analysis of the Hoxa11 knockdown APm-1 (APm-1-shH11-1 and APm-1-shH11-2) cell lines showed that the expression levels of Hoxa11 were reduced to 86 and 46%, respectively, compared with the control cell line (APm-1-shV) (Fig. 4A). By combining the fold changes in Hoxa11 between APw-1 and APm-1 and between APm-1-shV and APm-1-shH11-1, the expression level of Hoxa11 in APm-1-shH11-2 cells was estimated to be 2.4-fold higher than in APw-1 cells (Figs. 3D and 4A). Western blot analysis further confirmed that the protein levels of Hoxa11 were decreased in APm-1-shH11 cell lines compared with the APm-1-shV control cell line (Fig. 3E). Mice i.p. injected with APm-1-shH11-2 cells, which had lower Hoxa11 expression, had significantly longer survival than those injected with APm-1-shV cells (median 64 days vs. 50 days; P<0.01; Fig. 4B and Table I). To further support the role of Hoxa11 in the survival of leukemic mice, Hoxa11 overexpressed 12G cells (12G-H11-1 and 12G-H11-2) were generated by transduction of retrovirus-based full-length Hoxa11 gene into 12G cells. The 12G cells transduced with empty retroviruses (12G-V1 and 12G-V3) served as controls. Overexpression of Hoxa11 in the 12G-H11-1 and 12G-H11-2 cell lines compared with the control 12G-V1 and 12G-V3 cell lines was confirmed by western blotting and RT-qPCR (Figs. 3E and 4C). BM transplantation assay data showed that 12G-H11-1 mice had significantly shorter survival than control 12G-V1 mice (median 76 days vs. 91 days; P<0.001; Fig. 4D and Table I). These results indicated that Hoxa11 plays a critical role in the survival of leukemic mice induced by MLL/AF10 leukemia cells and MLL/AF10 cells harboring activating PTPN11 mutation.

Compared with the control cell lines, Hoxa11-knockdown APm-1 cells and Hoxa11-overexpression 12G cells had similar blast-like morphology (Fig. S2A and C, left column), similar percentages of CD115⁺ cells (Fig. S2A and C, right column), and similar cell proliferation rates in vitro (Fig. S2B and D). These results suggested that the role of Hoxa11 in the acceleration of disease progression in the leukemia mice was not caused by changes in differentiation potential or proliferation rate of the MLL/AF10 leukemia cells. Flow cytometry analysis of Ki-67, which detects proliferating cells, was performed on the leukemia cells (Mac1⁺) obtained from BM of 12G-V1 and 12G-H11-1 mice. The results showed that 12G-H11-1 cells were more actively proliferating in vivo (Fig. 4E). To investigate whether Hoxa11 enhanced growth advantage of MLL/AF10 leukemia cells in vivo, competitive engraftment and clonal expansion assays were performed. Compared with the control 12G-V1, 12G-H11-1 cells were more competitive to engraft to and expand in BM and spleen (Figs. 4F and S3A-C). These results suggested that Hoxa11 promotes disease progression, at least partly, by promoting leukemia cell recruitment and proliferation in their niche.

On the other hand, AK3G cells were also stably transduced with the two lentivirus-based shRNAs. Although the knockdown efficiencies of the shRNAs were significant (Fig. S4), the Hoxa11 expression levels in AK3G-shH11-2 cells were estimated to be 16.9-fold and 21.7-fold higher than that of 12G and AKw1G cells, respectively, based on the combined RT-qPCR results (Figs. 3D and S4). Moreover, AK3G cells had higher expression levels of Hoxa5, Hoxa6 and Hoxa10 compared with AKw1G (Fig. 3B). Due to the high Hoxa11 expression levels and the possible complementary effect of the Hoxa5-a10 genes (11), no further assays were performed to assess survival and drug sensitivity of AK3G-shV and AK3G-shH11 cells in this study.

To determine whether Hoxa11 affects Ara-C resistance of MLL/AF10 leukemia cells harboring PTPN11^G503A, APm-1-shV and APm-1-shH11-2 cells were treated with a gradient concentration of Ara-C for 24 h. Compared with APm-1-shV cells, APm-1-shH11-2 cells exhibited significantly higher viability at Ara-C concentrations between 128–1,600 ng/ml (P<0.01; Fig. 5A). The median lethal dose (LD50) of Ara-C for APm-1-shV and APm-1-shH11-2 cells was 235 and 315 ng/ml, respectively. By contrast, Hoxa11-overexpressing 12G-H11-1 cells showed significantly lower cell viabilities at all tested concentrations of Ara-C (P<0.005; Fig. 5B). The LD50 of Ara-C for 12G-V and 12G-H11 cells was 459 and 224 ng/ml, respectively. Further studies revealed that the apoptosis rate of APm-1-shH11-2 cells was lower than that of APm-1-shV cells before (17.6 vs. 25.4%, respectively) and after (41.3 vs. 45.2%, respectively) Ara-C treatment (160 ng/ml) for 24 h (Fig. 5C). The apoptosis rate of 12G-H11-1 cells was higher than that of 12G-V1 cells before (30.5 vs. 14.9%, respectively) and after Ara-C treatment (57.8 vs. 37.2%, respectively) (Fig. 5D). These results indicated that cooperation of MLL/AF10 with PTPN11^G503A upregulated Hoxa11, which in turn increased cell apoptosis and rendered cells more sensitive to Ara-C.

As Hoxa11 encodes a DNA-binding transcription factor and affects apoptosis of leukemia cells, the present study next determined the expression levels of three apoptosis-related genes in Hoxa11-knockdown APm-1 and Hoxa11-overexpression 12G cells using RT-qPCR analysis. It was found that the silencing of Hoxa11 significantly increased the expression levels of Nfkbia, Rela and Trp53, whereas the overexpression of Hoxa11 significantly reduced the expression level of these genes (Fig. 5E). These results suggested that Hoxa11 induces cell apoptosis, at least partly, via regulation of apoptosis-related gene expression.

To determine whether the expression of HOXA11 in leukemia cells of patients with AML was associated with chemotherapy drug sensitivity, a meta-analysis was performed using a data set deposited by Heuser consisting of 33 AML cases enrolled in the AML-SHG 01/99 trial, in the Oncomine™ clinical research data repository for gene expression change (34). The results of this analysis showed that responders of chemotherapy or AML induction/consolidation had higher HOXA11 expression levels than non-responders using reporter AA598674 (Fig. 5F-a and F-b). Collectively, these findings indicated that patients with AML with higher HOXA11 expression are associated with an improved response to chemotherapy with Ara-C.