News Article | February 15, 2017
KrasG12DLSL/+, KrasG12DFSF/+, R26CreERT2 and R26Cag-LSL-Luc mice were generated by T. Jacks and obtained through the Jackson Laboratory30, 31, 32, 33. Tp53Frt/Frt mice were generated by D. Kirsch and obtained through the Jackson Laboratory34. The R26mTmG strain was generated by L. Luo and obtained through the Jackson Laboratory35. The Smarcb1Loxp/LoxP strain was provided by C. Roberts36. The Pdx1-Cre strain was obtained from A. M. Lowy through the Jackson Laboratory37. The Ptf1aCre/+ and Tp53LoxP/LoxP strains were provided by R.A.D.38, 39. The R26Cag-FlpoERT2 was generated by A. Joyner and obtained from the Jackson Laboratory40. The Cdh1Cfp strain was generated by H. Clevers and obtained through the Jackson Laboratory41. The KrasG12DFSF/+; Tp53Frt/Frt; R26CreERT2 were kept in a C57BL/6 background, the other strains were kept in a mixed C57BL/6 and 129Sv/Jae background. All animal studies and procedures were approved by the UTMDACC Institutional Animal Care and Use Committee. Animals were killed when sick or when they developed tumours larger than 15 mm in their greater diameter or ulcerated lesions. KC: KrasG12DLSL/+-Pdx1-Cre; KPC∆/∆: KrasG12DLSL/+-Tp53LoxP/LoxP-Pdx1-Cre; KSC∆/∆: KrasG12DLSL/+-Smarcb1LoxP/LoxP-Pdx1-Cre; KPSC∆/∆: KrasG12DLSL/+-Tp53LoxP/LoxP-Smarcb1LoxP/LoxP-Pdx1-Cre; CS∆/∆: Smarcb1LoxP/LoxP-Pdx1-Cre; KPC∆/∆-R26mTmG-Cdh1Cfp: KrasG12DLSL/+-Tp53LoxP/LoxP-Pdx1-Cre-R26mTmG/+-Cdh1Cfp/+; R26CreERT2/+-KPFrt/Frt: R26CreERT2/+-KrasG12DFSF/+-Tp53Frt/Frt; R26Cag-FlpoERT2/+-KP∆/∆: R26Cag-FlpoERT2/+-KrasG12DLSL/+-Tp53LoxP/LoxP; R26Cag-FlpoERT2/Cag-LSL-Luc/Cag-LSL-Luc-KP∆/∆: R26Cag-FlpoERT2/Cag-LSL-Luc/Cag-LSL-Luc-KrasG12DLSL/+-Tp53LoxP/LoxP; R26Cag-FlpoERT2/+-KPS∆/∆: R26Cag-FlpoERT2/+-KrasG12DLSL/+-Tp53LoxP/LoxP-Smarcb1LoxP/LoxP. Correct geneotype was determined by PCR analysis and gel electrophoresis at birth and at death. Males and females were equally represented in experimental cohorts. R26Cag-FlpoERT2/Cag-LSL-Luc-KP∆/∆: KrasG12DLSL and R26Cag-LSL-Luc are in cis. No sex bias was introduced during the generation of experimental cohorts. To generate pLSM5, a synthetic cassette (Geneart, Life Technologies) containing the U6 promoter and the Cre recombinase sequence under the human keratin 19 promoter (−1,114, +141) flanked by 2 TATA-Frt sites (XbaI-U6-TATA-Frt-EcoRI-hKrt19-NheI-Cre-TATA-Frt-HpaI) was cloned into the XbaI/HpaI site of the pSICO vector. A DNA fragment was liberated by XbaI/KpnI digestion and cloned into the XbaI/KpnI sites of the pLB vector42. The introduction of the TATA box into the Frt sites was designed as previously described43. To generate pLSM2, the human Keratin 19 promoter was cloned into the NotI /NheI sites of the pSICOR vector. The Flpo cassette was cloned into the NheI/EcoRI sites of the pSICOR-hKrt19 (pLSM1). A DNA fragment was liberated by KpnI/XbaI digestion and inserted into the KpnI/XbaI sites of the pLB vector to obtain the pLSM2 vector. The shRNA oligos were cloned into the HpaI/XhoI site as previously described43. All the constructs were verified by restriction digestion and sequencing. The pSICO, pSICOR, and pSICO-Flpo were made by T. Jacks31, 43. The pLB vector was created by S. Kissler. The pMSCV-LoxP-dsRed-Loxp-eGFP-Puro-WPRE vector was used for virus titration in HEK293 cells and provided by H. Clevers44. All plasmids were obtained through Addgene. The pMSCV-Neo vector was purchased from Clontech. shRNA sequences: Smarcb1-1 (5′-GGAAGAGGTGAATGATAAA-3′), Smarcb1-855 (5′-AGATAGGAACACAAGGCGAAT-3′), Smarcb1-857 (5′-GCCATCCGAAATACCGGAGAT-3′), Atf2 (5′-GAAGTTTCTAGAACGAAATAG-3′), c-Jun (5′-CAGTAACCCTAAGATCCTAAA-3′), Kras (5′-GGAAACAAGTAGTAATTGA-3′), Ern1 (5′-GCTGAACTACTTGAGGAATTA-3′), Mkk4 (5′-CCCATACATTGTTCAGTTCTA-3′), negative control (5′-GCAAGCTGACCCTGAAGTTCAT-3′). To amplify integrated vector from genomic DNA the following oligonucleotides were used: forward, 5′-CCCGGTTAATTTGCATATAATATTTC-3′; reverse, 5′-CATGATACAAAGGCATTAAAGCAG-3′. For constitutive knock-down experiments, the pLKO.1 system was used. Cells were briefly selected in puromycin before experiments. The murine Myc open reading frame was purchased from Genecopoeia and subcloned into the EcoRI/BglII sites of the pMSCV-Neo vector. The pLenti-PKG Gfp-Puro was obtained from Addgene45. In the pLSM2-shRNA system/mouse strain, we crossed a latent allele of oncogenic KrasG12DFSF/+ that can be activated by Flpo-mediated recombination with a conditional Tp53Frt/Frt allele that, similarly, can be ablated in a time-restricted, tissue-specific manner by expressing a codon-optimized Flpo recombinase (provided by lentiviral delivery and under a tissue specific promoter)31, 34, 46. In addition, we introduced a tamoxifen-inducible Cre recombinase (CreERT2) that is expressed in virtually all tissues30. The pLSM2-shRNA system/vector was designed as follows. The lentiviral vector expresses the codon-optimized Flpo recombinase under the human KRT19 promoter and a constitutive shRNA under the U6 promoter. The entire cassette is flanked by LoxP sites and can be removed by Cre-mediated recombination in a time-restricted manner. The orthotopic injection of the virus results in the activation of oncogenic Kras and inactivation of Tp53 along with the RNAi-mediated depletion of Smarcb1 in the pancreatic epithelial compartment. The treatment with caerulein (performed according to the staggered protocol described previously6), starts 1 week after the viral injection and results in robust activation of a ductal differentiation program in the acinar compartment (acinar ductal metaplasia) and in a proliferative response6. Tamoxifen treatment results in Cre-mediated recombination at the LoxP sites in the genome of the integrated provirus, deletion of the shRNA cassette and restoration of expression of the gene target. In the pLSM5-shRNA system/mouse strain, we crossed mouse strains carrying a latent oncogenic KrasG12DLSL/+ allele (activated by Cre-mediated recombination) with a conditional Tp53LoxP/LoxP allele (along with a conditional Smarcb1LoxP/LoxP allele in some experiments) that, similarly, can be ablated in a time-restricted, tissue-specific manner by expressing a Cre recombinase (provided by lentiviral delivery)32, 36, 38. In addition, we introduced a tamoxifen-inducible Flpo recombinase (FlpoERT2) which is expressed in virtually all tissues under a strong promoter (CAG)40. The pLSM5-shRNA lentiviral vector expresses a codon-optimized Cre recombinase under the human KRT19 promoter and a latent shRNA that can be activated by Flpo-mediated recombination and the deletion of a Frt-Stop-Frt cassette. A TATA-box cassette was introduced into the Frt sites to increase shRNA expression upon Flpo-mediated recombination. The system allows the generation of primary tumours and the depletion of a gene of interest at a desired time. Infectious viral particles were produced using psPAX2 and pMD2G helper plasmids. For transfection, 293T cells were cultured in DMEM containing 10% FBS (Gibco), 100 IU ml−1 penicillin (Gibco), 100 μg ml−1 streptomycin (Gibco) and 4 mM caffeine (Sigma-Aldrich) and transfected using the polyethylenimine method. Virus-containing supernatant was collected 48–72 h after transfection, spun at 3,000 r.p.m. for 10 min and filtered through 0.45-μm low-protein-binding filters (Corning)47. High-titre preparations were obtained by multiple rounds of ultracentrifugation at 23,000 r.p.m. for 2 h each. Viral titre was quantified in HEK293T cells stably transduced with a Cre-inducible GFP reporter44. For orthotopic injections, a previously described protocol was partially modified13. In brief, virus was resuspended in a solution of OPTI-MEM and polybrene (8 μg ml−1). Mice were anaesthetized using a ketamine/xylazine solution (150 mg kg−1 and 10 mg kg−1, respectively). Shaved skin was disinfected with betadine and ethanol and 1-cm incisions were performed through the skin/subcutaneous and muscular/peritoneal layers. The spleen and tail of the pancreas were identified and exposed and multiple injections were performed in the pancreatic tail and body (2 × 108–5 × 108 IU per mouse). The muscular/peritoneal planes were closed using continuous absorbable sutures. The skin/subcutaneous planes were closed using interrupted absorbable sutures. Analgesia was achieved with buprenorphine (0.1 mg kg-1 BID). At 7 days after surgery, mice were treated with caerulein as previously described6. Mice were monitored for tumour formation twice per week. For tamoxifen treatment, after tumours were detected, mice were treated with tamoxifen (Sigma) by intraperitoneal injection. A total of 100 μl of tamoxifen solution (15 mg ml−1 in corn oil) was injected every other day, giving five injections in total. Treatment cycles were repeated every 2 weeks if appropriate. In orthotopic secondary transplantation studies, tamoxifen treatment was started 5 days after surgery. For orthotopic transplantations experiments, 2 × 105 cells were resuspended in a 2:1 solution of OPTI-MEM (Gibco) and Matrigel (BD Biosciences, 356231) and transplanted into the tail of the pancreas of 6–9-week-old mice in a single injection (25 μl). For subcutaneous transplantation studies, tumour cells were resuspended in OPTI-MEM (Gibco) and Matrigel (BD Biosciences, 356231) (2:1 dilution) and injected subcutaneously into the flank of 6–9-week-old NCr Nude female mice (Taconic). Liver-seeding experiments were performed as described previously48. Liver weight was measured fresh at necropsy. For transplantation in a limiting dilution, 1 × 103, 2 × 102 or 2 × 10 tumour cells were resuspended in a 2:1 solution of OPTI-MEM (Gibco) and Matrigel (BD Biosciences, 356231) and injected into the flank of 6–9-week-old NCr Nude female mice (Taconic). Mice were observed for 24–34 weeks. The TIC frequency was calculated using L-Calc Limiting Dilutions Software (Stem Cell Technologies) and expressed as proportion of TIC ± s.e.m. The following primary antibodies were used for immunofluorescence, immunohistochemistry and immunoblotting: phospho-p44/42 MAPK (Erk1/2, Thr202/Tyr204) (D13.14.4E, Cell Signaling Technologies #4370); phospho-MEK1/2 (Ser221, 166F8) (Cell Signaling Technologies #2338), SMARCB1 (Sigma Aldrich # HPA018248); SMARCB1 (BD Transduction Laboratories #612111); Vinculin (E1E9V, Cell Signaling Technologies #13901); vimentin (D21H3, Cell Signaling Technologies #5741); CDH1 (4A2, Cell Signaling Technologies #14472); nestin (rat-401 Millipore #Mab 353); Ki67 (Thermo Scientific #RM9106); Sox9 (Millipore #AB-5535); Pdx1 (Millipore # 06-1385); cleaved caspase 3 (A175, Cell Signaling Technologies #9661); phospho-p38 (D3F9) (Cell Signaling Technologies #4511); p38α (Cell Signaling Technologies #9218); JNK (Cell Signaling Technologies #9252); phospho-JNK (Thr183/Tyr185, 81E11, Cell Signaling Technologies #4668); ATF2 (20F1 Cell Signaling Technologies #9226), phospho-ATF2 (Thr69/71, Cell Signaling Technologies #9225); c-Jun (60A8, Cell Signaling Technologies #9165); phospho-c-Jun (Ser73, D47G9, Cell Signaling Technologies #3270); ubiquitin (Cell Signaling Technologies #3933); IRE1-α (14C10, Cell Signaling Technologies #3294); PERK (D11A8, Cell Signaling Technologies #5683); XBP-1 s (D2C1F, Cell signaling Technologies #12782); ATF6 (70B1413, Abcam #11909); ATF6 (NovusBio # NBP1-77251); SEK1/MKK4 (Cell Signaling Technologies #9152); phospho-SEK1/MKK4 (Ser257, C36C11, Cell Signaling Technologies #4514); c-Myc (D3N8F, Cell Signaling Technologies #13987). The following chemical reagents were used: gemcitabine (LC Laboratories), bortezomib (LC Laboratories), carfilzomib (LC Laboratories), NVP-AUY-922 (LC Laboratories), ganetespib (Selleck Chemicals) SP600125 (LC Labs), BIRB796 (LC Labs), tunicamycin (Selleck Chemicals). Senescence-associated β-galactosidase staining was performed with a senescence β-Galactosidase Staining Kit (Cell Signaling Technologies) according to the manufacturer’s instructions. Tumour-derived cells and primary lines were cultured in vitro for <5 passages prior to experimentation. Aldefluor-based cell sorting (Stem Cell Technologies) was performed according to the manufacturer’s instructions. Cells showing a fluorescence signal above the average of the diethylaminobenzaldehyde-treated negative controls were considered positive. Protein synthetic rate was assessed using the Click-iT Plus OPP Alexa Fluor 594 Protein Synthesis Assay Kit (Life Technologies) according to the manufacturer’s instructions. The rate of incorporation of OPP was assessed by FACS analysis. Cells cultured in the presence of 20 μM Cycloheximide (Sigma Aldrich) were used as negative technical controls. After staining, samples were acquired using a BD FACS Canto II flow cytometer. Cell sorting experiments were performed using BD Influx cell sorter. For details see ref. 49. Data were analysed by FlowJo (Tree Star). Patient-derived samples were obtained from patients who had given informed consent under Institutional Review Board (IRB)-approved protocols LAB07-0854 chaired by J.B.F. (UTMDACC) and IRB00044588 chaired by L. D. Wood (JHMI). The establishment of human PDX lines was described in detail previously50, 51. Passage-1 PDXs were dissociated using collagenase and dispase (collagenase IV–dispase 4 mg ml−1; Invitrogen) at 37 °C for 1 h and single-cell suspensions were then transduced with a lentiviral GFP reporter (pLenti-PKG GFP-Puro) and transplanted into NOD SCID immunocompromised mice. Experimental cohorts were generated by serial transplantations in vivo. Cells were isolated from primary pancreatic tumours as previously described52. Cells derived from primary mouse tumours were kept in culture as spheres in semi-solid media for <5 passages. After explant, tumours were digested at 37 °C for 1 h (collagenase IV-dispase 4 mg ml−1; Invitrogen). Single-cell suspensions were plated in DMEM (Lonza) supplemented with 2 mM glutamine (Invitrogen), 10% FBS (Lonza), 40 ng ml−1 hEGF (PeproTech), 20 ng ml−1 hFGF (PeproTech), 5 μg ml−1 h-insulin (Roche), 0.5 μM hydrocortisone (Sigma), 100 μM β-mercaptoethanol (Sigma), 4 μg ml−1 heparin (Sigma), penicillin (Gibco) 100 IU ml−1 and streptomycin (Gibco) 100 μg ml−1. Methocult M3134 (StemCell Technologies) was added to the culture medium to a final concentration of 0.8% (v/v)) to keep tumour cells growing as clonal spheres and not aggregates. Spheres were collected and digested with 0.25% trypsin (Gibco) to single cells and re-plated. For 2D tumour cultures, cells were kept in DMEM containing 10% FBS (Gibco), 100 IU ml−1, penicillin (Gibco) and 100 μg ml−1 streptomycin (Gibco). For in vivo transplantation studies, low-passage (<5) tumour cells were used. For GDAs, single-cell suspensions were generated using collagenase and dispase (collagenase IV–dispase 4 mg ml−1, Invitrogen) and transplanted immediately into recipient mice. Experimental cohorts were generated by serial transplantations in vivo. The following was performed as previously described53 with modifications. Pancreata were harvested and digested at 37 °C for 45 min (collagenase IV, 4 mg ml−1) and passed through a 100-μm nylon cell strainer to separate the acinar fraction from larger ducts. The ductal fraction underwent additional digestion with 0.25% trypsin (Gibco) for 5 min at 37 °C and mechanical disruption. The two fractions were combined and plated on collagen IV-coated plates (Corning) in modified PDEC medium: DMEM/F12 (Gibco), 15 mM HEPES (Invitrogen), 5 mg ml−1 d-glucose (Sigma Aldrich), 1.22 mg ml−1 nicotinamide (Sigma Aldrich), 5 nM 3,3,5-tri-iodo-l-thyronine (Sigma Aldrich), 1 μM dexamethasone (Sigma Aldrich), 100 ng ml−1 cholera toxin (Sigma Aldrich), 5 ml l−1 insulin-transferrin-selenium (BD), penicillin (Gibco), 100 μg ml−1 streptomycin (Gibco), 0.1 mg ml−1 soybean trypsin inhibitor (Sigma Aldrich), 40 ng ml−1 EGF (Sigma Aldrich), 25 μg ml−1 bovine pituitary extract (Invitrogen), 100 μM β-mercaptoethanol (Sigma) and 10% FBS (Gibco). Cells were passaged at low confluency until exhaustion or escaper clones were established. For drug treatments, spheres were collected, washed, digested with trypsin and repeatedly counted (Countless, Invitrogen). Equal numbers of live cells were incubated with bortezomib (5 nM), carfilzomib (5 nM), NVP-AUY-922 (50 nM), ganetespib (50 nM), 4-hydroxy-tamoxifen (250 nM) and tunicamycin (200 nM). Spheroids were manually counted under a Nikon Eclipse Ti microscope using a click-counter. Experiments were repeated at least three times and error bars represent the s.d. of technical replicates from a representative experiment. For orthotopic end point survival studies, 6–9-week-old female mice were transplanted orthotopically with 2 × 105 cells resuspended in a 2:1 solution of OPTI-MEM (Gibco) and Matrigel (BD Biosciences, 356231). GEMM-derived-allografts and PDXs were briefly dissociated and passaged in vivo in NCr Nude and NOD SCID female mice, respectively, to limit the phenotypic changes associated with 2D cultures. Tumour volumes were measured according to the formula l × w2/2, where w represents tumour width. Clinical response was determined as the ratio of tumour volume at the end of the treatment to the volume at the beginning of the treatment. Gemcitabine was administered intraperitoneally at 100 mg kg−1 every 4 days for 16 days; NVP-AUY-922 was administered intraperitoneally at 75 mg kg−1 every other day for 16 days; BIRB796 was administered by oral gavage every second day at 40 mg kg−1 for 16 days; SP600125 was administered intraperitoneally at 40 mg kg−1 every day for 16 days. Gemcitabine was dissolved in phosphate buffer saline, AUY922 was dissolved 10% DMSO/25% water/65% PEG 400, SP600125 was resuspended in PBS and DMSO and BIRB796 was prepared as previously reported54. Animals were imaged on a 4.7T Bruker Biospec (Bruker BioSpin) equipped with 6-cm inner-diameter gradients and a 35-mm inner-diameter volume coil. Multi-slice T2-weighted images were acquired in coronal and axial geometries using a rapid acquisition with relaxation enhancement (RARE) sequence with TR/TE of 2,000/38 ms, matrix size 256 × 192, 0.75-mm slice thickness, 0.25-mm slice gap, 4 × 3-cm FOV, 101-kHz bandwidth, 3 NEX. Axial scan sequences were gated to reduce respiratory motion. Detection of luciferase activity was performed in an IVIS-100 imaging system. Five minutes before the procedure, mice were injected intraperitoneally with d-luciferin, bioluminescence substrate (Perkin Elmer) according to the manufacturer’s instructions. Living Image 4.3 software (Perkin Elmer) was used for analysis of the images after acquisition. Tumour samples were fixed in 4% formaldehyde for 24 h at room temperature, moved into 70% ethanol for 12 h, and then embedded in paraffin (Leica ASP300S). After cutting (Leica RM2235) and baking at 60 °C for 20 min for de-paraffinization, slides were treated with Citra-Plus Solution (BioGenex) for antigen unmasking according to the manufacturer’s instructions. For immunohistochemical staining, endogenous peroxidases were inactivated by 3% hydrogen peroxide at room temperature for 15 min. Non-specific signals were blocked using 5% BSA and 5% goat serum for 1 h. Tumour samples were stained with primary antibodies for 12 h at 4 °C and the Mouse on Mouse Kit (Vector Laboratory) was used when appropriate according to the manufacturer’s instructions. For immunostaining, ImmPress (Vector Laboratory) were used as secondary antibodies and Nova RED (Vector Laboratory) was used for detection. Images were captured with a Nikon DS-Fi1 digital camera using a wide-field Nikon Eclipse Ci microscope. For immunofluorescence, secondary antibodies conjugated to Alexa488, Alexa647 and Alexa555 (Molecular Probes) were used. Images were captured with a Hamamatsu C11440 digital camera, using a wide-field Nikon EclipseNi microscope and a Nikon high-speed multi-photon confocal microscope A1 R MP. Total staining score was weighed to the intensity and prevalence (percentage of positive tumour cells and intensity score of 0 to 3) in random fields at 20× magnification. Quantitative analysis was performed using Image J and Immunoratio programs according to the providers’ instructions at 20× original magnification. TEM was performed at the UTMDACC High Resolution Electron Microscopy Facility. Samples were fixed with a solution containing 3% glutaraldehyde plus 2% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.3, for 1 h. After fixation, the samples were washed and treated with 0.1% Millipore-filtered cacodylate-buffered tannic acid, post-fixed with 1% buffered osmium tetroxide for 30 min, and stained en bloc with 1% Millipore-filtered uranyl acetate. The samples were dehydrated in increasing concentrations of ethanol, infiltrated and embedded in LX-112 medium. The samples were polymerized at 60 °C for 2 days. Ultra-thin sections were cut using a Leica Ultracut microtome, stained with uranyl acetate and lead citrate in a Leica EM Stainer and examined using a JEM 1010 transmission electron microscope (JEOL) at an accelerating voltage of 80 kV. Digital images were obtained using an AMT Imaging System (Advanced Microscopy Techniques Corp). Protein lysates were resolved on 5–15% gradient polyacrylamide SDS gels (Bio-Rad) and transferred onto PVDF membranes (Bio-Rad) according to the manufacturer’s instructions. Membranes were incubated with the indicated primary antibodies, washed in TBST buffer and probed with HRP-conjugated secondary antibodies. The detection of bands was carried out upon chemi-luminescence reaction followed by film exposure (Denville Scientific). In vitro and in vivo data are presented as the mean ± s.d. Results from limiting dilutions analysis (LDA) were expressed as the proportion of TIC ± s.e.m. Differences in stem-cell frequencies between groups were determined using a chi-squared test (2-tailed)55, 56. Comparisons between biological replicates were performed using a two-tailed Student’s t-test. Results from survival and incidence experiments were analysed with a log-rank (Mantel–Cox) test and expressed as Kaplan–Meier survival curves. Results from contingency tables were analysed using the two-tailed Fisher’s exact test (GraphPad software). Group size was determined on the basis of the results of preliminary experiments. No statistical methods were used to determine sample size. Group allocation and analysis of outcome were not performed in a blinded manner. Samples that did not meet proper experimental conditions were excluded from the analysis. DNA and RNA were isolated using DNeasy Blood and Tissue Kit (Qiagen) and RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. Gene expression profiling was performed at the UTMDACC Microarray Core Facility on a Gene Chip Mouse Genome 430 2.0 Array (Affymetrix). The robust multi-array average method was used with default options (with background correction, quantile normalization, and log transformation) to normalize raw data from batches using R/Bioconductor’s affy package (12925520) and analysed with GSEA c3.tft.v4.0 (TFT) and c6.all.v4.0. (Oncogenic Signatures); HOMER (20513432) was also used to identify significantly enriched biological pathways or processes for the differentially expressed genes57, 58. Subgroup information (Classical, QM-PDA, Exocrine-like) for each gene was provided to a heuristic optimization method (stochastic gradient descent) to minimize objective function. The objective function output was used to calculate decision boundaries with a support vector machine approach to optimize the partitioning of subtypes. The obtained microarray signal values for each probe were used for proper classification. The decision surface for multi-class datasets was plotted with Python package matplotlib. To control for random occurrence, we permutated the classification subtypes provided to the stochastic gradient descent function and randomized trainings yield ambiguous classifications, suggesting that gene expression signatures in our model is overlapping with the previous pancreas cancer subgroups. Clinical and pathological data for patient samples are provided in Supplementary Table 1. Microarray data supporting the findings of this study have been deposited in the Gene Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/) under accession number GSE83754. All other data are available from the corresponding author (G.G.) upon reasonable request.
Seviour E.G.,University of Texas M. D. Anderson Cancer Center |
Sehgal V.,University of Texas M. D. Anderson Cancer Center |
Lu Y.,University of Texas M. D. Anderson Cancer Center |
Luo Z.,University of Texas M. D. Anderson Cancer Center |
And 17 more authors.
Oncogene | Year: 2016
The myc oncogene is overexpressed in almost half of all breast and ovarian cancers, but attempts at therapeutic interventions against myc have proven to be challenging. Myc regulates multiple biological processes, including the cell cycle, and as such is associated with cell proliferation and tumor progression. We identified a protein signature of high myc, low p27 and high phospho-Rb significantly correlated with poor patient survival in breast and ovarian cancers. Screening of a miRNA library by functional proteomics in multiple cell lines and integration of data from patient tumors revealed a panel of five microRNAs (miRNAs) (miR-124, miR-365, miR-34b∗, miR-18a and miR-506) as potential tumor suppressors capable of reversing the p27/myc/phospho-Rb protein signature. Mechanistic studies revealed an RNA-activation function of miR-124 resulting in direct induction of p27 protein levels by binding to and inducing transcription on the p27 promoter region leading to a subsequent G1 arrest. Additionally, in vivo studies utilizing a xenograft model demonstrated that nanoparticle-mediated delivery of miR-124 could reduce tumor growth and sensitize cells to etoposide, suggesting a clinical application of miRNAs as therapeutics to target the functional effect of myc on tumor growth. © 2016 Macmillan Publishers Limited All rights reserved.
PubMed | University of Houston, University of Texas M. D. Anderson Cancer Center, UTMDACC, Cambridge Institute of Public Health and Oregon Health And Science University
Type: Journal Article | Journal: Oncogene | Year: 2016
The myc oncogene is overexpressed in almost half of all breast and ovarian cancers, but attempts at therapeutic interventions against myc have proven to be challenging. Myc regulates multiple biological processes, including the cell cycle, and as such is associated with cell proliferation and tumor progression. We identified a protein signature of high myc, low p27 and high phospho-Rb significantly correlated with poor patient survival in breast and ovarian cancers. Screening of a miRNA library by functional proteomics in multiple cell lines and integration of data from patient tumors revealed a panel of five microRNAs (miRNAs) (miR-124, miR-365, miR-34b*, miR-18a and miR-506) as potential tumor suppressors capable of reversing the p27/myc/phospho-Rb protein signature. Mechanistic studies revealed an RNA-activation function of miR-124 resulting in direct induction of p27 protein levels by binding to and inducing transcription on the p27 promoter region leading to a subsequent G1 arrest. Additionally, in vivo studies utilizing a xenograft model demonstrated that nanoparticle-mediated delivery of miR-124 could reduce tumor growth and sensitize cells to etoposide, suggesting a clinical application of miRNAs as therapeutics to target the functional effect of myc on tumor growth.
Ayala-Ramirez M.,University of Houston |
Jasim S.,Washington University in St. Louis |
Feng L.,UTMDACC |
Ejaz S.,University of Houston |
And 10 more authors.
European Journal of Endocrinology | Year: 2013
Objective: Adrenocortical carcinoma (ACC) is a rare malignancy with a poor prognosis. Herein, we describe the clinical features and outcomes for a large series of ACC patients. Design and methods: Retrospective review of ACC patients seen at The University of Texas MDAnderson Cancer Center from 1998 through 2011. Results: A total of 330 patients with median age at diagnosis of 48.5 years; 12 (3.6%) patients were under 18 years. Hormonally functioning tumors represented 41.8% (n=138) of all cases. Surgical resection for the primary tumor was done in 275 (83.3%) patients (45 at MDAnderson (16.4%)). For those who had surgical resection, the median local-recurrence-free time was 1.04 years. Factors associated with local recurrence included positive surgical margins (P=0.007) and advanced disease stage (P=0.026). Median overall survival time for all patients was 3.21 years. Median survival times were 24.1, 6.08, 3.47, and 0.89 years for stages I, II, III, and IV respectively. In multivariable analysis, older age, functioning tumors, and higher disease stage remained significant prognostic factors associated with poor survival. Conclusion: ACC prognosis remains poor with the use of currently available treatments. Older age, functioning tumors, and incomplete resections are clinical factors associated with poor survival. Surgical expertise is important to achieve complete resections and to improve outcome. © 2013 European Society of Endocrinology.
PubMed | UTMDACC
Type: Journal Article | Journal: Journal of clinical oncology : official journal of the American Society of Clinical Oncology | Year: 2016
3593 Background: Preoperative radiotherapy with continuous infusional 5-fluorouracil (CI-5FU) improves sphincter preservation and produces pathological complete remissions (pCR) in 10-30% of patients (pts) with LARC. Capecitabine (Xeloda), an oral fluoropyrimidine, is preferentially converted to 5-FU by high intra-tumor thymidine phosphorylase that can be further activated by radiotherapy.The primary endpoint of this phase II study is to evaluate the rate of pathological response with secondary endpoints on survival, recurrence, quality of life, and tumor markers. Capecitabine (825 mg/m2 PO bid/7days during XRT was administered with concomitant boost XRT (52.5 Gy/30 fractions to the primary/perirectal nodes, and 45Gy to the pelvis) in 54 pts with LARC ( T3N0 or any T/N1 staged by EUS). Eligible patients recieved capecitabine 1250 mg/m2 PO bid day 1-14 every 21 for four cycles as adjuvant treatment after surgery.Pts characteristics included median age: 56 (36-78), male:female ratio: 1.7; ECOG performance status 0-1 100%; 42 pts (92%) had T3N0 or T3N1 disease. The pathological response of the 44 patients who underwent resection to date includes 7 pts (16%) who achieved pCR and 28 pts (64%) had a partial response resulting in a 79% overall response rate. Stable disease was evident in 5 pts (11%) and 2 pts (5%) had progressive disease. Two pts (5%) who achieve complete clinical remissions refused surgery. Most toxicities were Grade 1/2: hand foot syndrome (27%), diarrhea (25%), fatigue (21%), leukopenia (15%), anemia (15%), and proctitis (10%). Gr 3 toxicities were limited: diarrhea (10%) and radiation dermatitis (6%), bilirubinemia (4%) and hand and foot syndrome (2%). Gr 4 diarrhea occurred only in one pt.Concomitant boost radiotherapy plus capecitabine is very well tolerated and produces pathological response rates that are comparable to CI-5FU for LARC. Capecitabine also avoids the discomfort and potential complications, like infection and phlebitis, associated with CI-5FU. Full report will be presented at the meeting. *Study supported by Roche [Table: see text].
PubMed | UTMDACC, University of Texas M. D. Anderson Cancer Center and Methodist Hospital Research Institute
Type: | Journal: Oncogene | Year: 2016
KRas is mutated in a significant number of human cancers and so there is an urgent therapeutic need to target KRas signalling. To target KRas in lung cancers we used a systems approach of integrating a genome-wide miRNA screen with patient-derived phospho-proteomic signatures of the KRas downstream pathway, and identified miR-193a-3p, which directly targets KRas. Unique aspects of miR-193a-3p biology include two functionally independent target sites in the KRas 3UTR and clinically significant correlation between miR-193a-3p and KRas expression in patients. Rescue experiments with mutated KRas 3UTR showed very significantly that the anti-tumour effect of miR-193a-3p is via specific direct targeting of KRas and not due to other targets. Ex vivo and in vivo studies utilizing nanoliposome packaged miR-193a-3p demonstrated significant inhibition of tumour growth, circulating tumour cell viability and decreased metastasis. These studies show the broader applicability of using miR-193a-3p as a therapeutic agent to target KRas-mutant cancer.Oncogene advance online publication, 26 September 2016; doi:10.1038/onc.2016.308.