Elevated H3K27me3 levels sensitize osteosarcoma to cisplatin
- 274 Downloads
In osteosarcoma (OS), chemotherapy resistance has become one of the greatest issues leading to high mortality among patients. However, the mechanisms of drug resistance remain elusive, limiting therapeutic efficacy. Here, we set out to explore the relationship between dynamic histone changes and the efficacy of cisplatin against OS.
First, we found two histone demethylases associated with histone H3 lysine 27 trimethylation (H3K27me3) demethylation, KDM6A, and KDM6B that were upregulated after cisplatin treatment. Consistent with the clinical data, cisplatin-resistant OS specimens showed lower H3K27me3 levels than sensitive specimens. Then, we evaluated the effects of H3K27me3 alteration on OS chemosensitivity. In vitro inhibition of the histone methyltransferase EZH2 in OS cells decreased H3K27me3 levels and led to cisplatin resistance. Conversely, inhibition of the demethylases KDM6A and KDM6B increased H3K27me3 levels in OS and reversed cisplatin resistance in vitro and in vivo. Mechanistically, with the help of RNA sequencing (RNAseq), we found that PRKCA and MCL1 directly participated in the process by altering H3K27me3 on their gene loci, ultimately inactivating RAF/ERK/MAPK cascades and decreasing phosphorylation of BCL2.
Our study reveals a new epigenetic mechanism of OS resistance and indicates that elevated H3K27me3 levels can sensitize OS to cisplatin, suggesting a promising new strategy for the treatment of OS.
KeywordsOsteosarcoma H3 lysine 27 trimethylation (H3K27me3) KDM6A KDM6B Chemoresistance Apoptosis
Histone H3 lysine 27 trimethylation
Osteosarcoma (OS) is the most common primary malignant bone tumor worldwide, with an incidence of approximately one to three cases per million people per year; the male to female ratio is approximately 1.22:1 , and children/adolescents and those over the age of 75 with Paget’s disease or other bony lesions are the two susceptible populations [2, 3]. The femur (40%), tibia (20%), humerus (10%), and pelvis (8%) together account for almost 80% of the bones involved in OS based on reported cases; the other involved bones include the fibula, jaw, vertebrae, radius, ulna, and ribs . OS patients with metastasis and recurrence have a poor prognosis that is partly attributed to chemoresistance [4, 5]. Although several mechanisms have been reported to mediate chemoresistance and some chemoresistance-reversing agents have even been developed, the mechanisms remain unclear, and almost no substantial survival benefits are conferred by administration of the therapeutic agents . Above all, further exploration of the mechanism of chemoresistance and development of new strategies to reverse chemoresistance in OS are urgently needed.
Aberrant epigenetic reprogramming plays a pivotal role in tumorigenesis . Among epigenetic regulation factors, histone methyltransferases (HMTs) are frequently dysregulated in a spectrum of human tumors, which indicates that HMTs are potential therapeutic targets . Histone H3 lysine 27 trimethylation (H3K27me3) is frequently associated with gene repression and contributes considerably to tissue differentiation, stem cell properties, and cancer occurrence/progression . Increases and decreases in H3K27me3 levels are catalyzed by the histone methyltransferase enzyme enhancer of zeste homolog 2 (EZH2)  and two demethylases belonging to the Jumonji C (JmjC) domain-containing family: ubiquitously transcribed tetratricopeptide repeat on chromosome X (UTX, also called lysine-specific demethylase 6A [KDM6A]) and Jumonji D3 (JMJD3, also called KDM6B) . Thus far, increasing evidence has showed that dynamic methylation of H3K27 frequently occurs during the development of tumors, especially during the modulation of cancer stem cells (CSCs). H3K27me3 has been reported to be a negative modulator of breast CSCs (BCSCs), and upregulation of H3K27me3 content can suppress BCSCs ; the same phenomenon has also observed in ovarian CSCs  and in non-small cell lung cancer . However, it remains unclear whether alterations in histone modifications such as H3K27me3 foster tumorigenesis and chemoresistance in OS.
Here, we illustrate that alterations in H3K27me3 are highly correlated with cisplatin sensitivity in OS and that increased H3K27me3 levels can enhance cisplatin efficacy against OS. This effect may be mediated by direct downregulation of PRKCA and MCL1 expression and subsequent inactivation of RAF/ERK/MAPK cascades as well as decreased phosphorylation of BCL2.
Expression of KDM6A and KDM6B in OS cells and patient specimens
Downregulation of KDM6A and KDM6B sensitizes OS to cisplatin in vitro
Increasing H3K27me3 levels sensitizes OS to cisplatin in vitro
Upregulating H3K27me3 levels sensitizes OS to cisplatin in vivo
H3K27me3 modification is associated with CSC properties in OS
Alterations in H3K27me3 regulate cell apoptosis- and stemness-related genes involved in OS chemosensitivity to cisplatin
H3K27me3 alterations affect OS chemosensitivity through PRKCA via phosphorylation of BCL2 and modulation of RAF/ERK/MAPK cascades
OS is a rare bone malignancy that primarily affects children and adolescents. At present, patients are typically treated with surgery and intensive adjuvant chemotherapy. However, unfortunately, the 5-year survival outcome for recurrent or metastatic OS is poor. Since the survival time of OS has not changed over the past two decades , new strategies and agents urgently need to be introduced to provide greater benefit to patients.
Epigenetic alterations are known to contribute greatly to carcinogenesis in humans ; therefore, epigenetic modulation could be a novel and promising strategy for fighting cancers, especially OS. Based on this possibility, we screened a panel of DNA methyltransferases and demethylases that could be induced by cisplatin in OS cell lines. The results indicated that KDM6A and KDM6B, which are both demethylases of H3K27, were upregulated by cisplatin treatment. We then sought to explore the correlations between H3K27me3 levels and chemosensitivity in OS and the underlying mechanism. The results indicated that the H3K27 demethylases KDM6A and KDM6B were upregulated in OS and that higher H3K27me3 levels indicated better efficacy of cisplatin against OS. In addition, KDM6A/KDM6B knockdown or upregulation of H3K27me3 levels sensitized OS to cisplatin by enhancing apoptosis in tumor cells.
It is natural to associate the link between epigenetic alterations and cancer chemosensitivity with stem cell properties, since epigenetic modification events are pivotal for stem cell renewal and differentiation and since CSCs are recognized to be more resistant than normal stem cells to chemotherapeutic agents . Hence, in our study, we further tested CSC properties upon up- or downregulation of H3K27me3.
Our findings revealed that higher H3K27me3 levels were correlated with reduced CSC properties and reduced expression of CSC-related genes such as SOX9 and CD117. Although we did not observe a significant change in the expression of SOX2 (a master regulator of stemness phenotype), we found that SOX9, which was also a SOX family member, showed a significant downregulation upon GSK-J4 treatment. SOX9 (SRY (sex-determining region Y)-box 9) is a member of the SOX family of transcription factors that plays critical roles in embryonic development, lineage commitment, and stem cell maintenance . Likewise, CD117/c-kit is a tyrosine kinase receptor associated with cancer progression and normal stem cell maintenance , which also significantly downregulated upon GSK-J4 treatment.
In our study, RNAseq data indicated that PRKCA, MCL1, and other genes are also involved in this process. More importantly, through ChIP-qPCR analysis, we found that the expression of PRKCA and MCL1 was directly modulated by changes in H3K27me3 levels at the loci of these genes. The PKCs, which include PRKCA, are pivotal components of many signaling pathways that regulate diverse cellular functions including proliferation, angiogenesis, migration, invasion, metastasis, and resistance to apoptosis . PRKCA (the α-isozyme of PKC) is widely expressed in various tissues, and abnormal levels of PRKCA have been found in many transformed cell lines and in several human tumors . PRKCA has also been reported to be a dominant tumor progression factor that is associated with tumor proliferation and survival . Pharmacological inhibition of PRKCA can reduce the growth and survival of tumors, promote apoptosis, and sensitize tumor cells to chemotherapy . Mechanistically, the antiapoptotic action of PRKCA has been shown to be mediated by activation of RAF/ERK/MAPK cascades  and/or by phosphorylation of the antiapoptotic protein BCL2 . Moreover, MCL1, is a member of the antiapoptotic BCL2 family, which can suppress apoptosis . As the RAF/ERK/MAPK pathway is an important signaling pathway involved in cell survival, apoptosis inhibition, cell cycle progression, and proliferation, we reasonably associated PRKCA with the RAF/ERK/MAPK pathway and the phosphorylation of BCL2. Fortunately, in line with our hypothesis, RAF/ERK/MAPK pathway inactivation was found to participate in the process of OS sensitization to chemotherapy by increasing H3K27me3 levels on PRKCA and MCL1 gene loci.
It remains unclear whether histone modification dynamics are the driving force of tumorigenesis and chemoresistance. PRKCA may also perform proapoptotic functions ; therefore, further investigation into the mechanism that influences the PRKCA functional mode is needed in the future. In addition, increasing numbers of studies have also demonstrated that epigenetic therapies enhance the efficacy of adoptive immunotherapies  and immune checkpoint inhibitors . Thus, whether the immune system is involved in the process of chemosensitization also needs further exploration.
In summary, we have shown, for the first time, that increasing H3K27me3 levels can promote the efficacy of cisplatin against OS by upregulating BMF and downregulating PRKCA, MCL1, SOX9, CD117, and KLF4, which subsequently dephosphorylates BCL2 and deactivates RAF/ERK/MAPK cascades. These findings may provide a promising strategy for enhancing chemotherapeutic efficacy against OS and improving prognoses of patients.
Patients with histologically confirmed OS who were involved in this study provided their informed consent, and all studies were performed following the guidelines of the ethics committee of Shanghai Ninth People’s Hospital. Twenty patients (six females and 14 males; there was no significant difference in composition regarding age or sex) were enrolled in this study (the clinical information is summarized in Additional file 6: Tables S1 and S2) and were divided into two groups based on their cisplatin efficacy outcomes: a cisplatin-sensitive group (cisplatin sensitivity 50–100%) and a cisplatin-resistant group (cisplatin sensitivity 0–50%).
After deparaffinization, rehydration, antigen retrieval, and blockade of endogenous peroxidase, tissues were incubated with anti-TUNEL (Roche), anti-cleaved Caspase 3 (Servicebio), anti-Ki67 (Servicebio), and anti-CD31 (Abways) primary antibodies at 4 °C overnight. Next, secondary antibodies were applied, and diaminobenzidine (DAB) (Dako) solution was used as a chromogen. Finally, the sections were counterstained with hematoxylin (Sigma-Aldrich) to identify nuclei.
Cell lines and cell culture
The human OS cell lines 143B and HOS were purchased from the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China). Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) (HyClone) supplemented with 10% fetal bovine serum (FBS; Gibco) and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin) at 37 °C in a humidified atmosphere with 5% CO2.
Cell viability assay
A cell viability assay was performed as previously described . Briefly, cells were added into 96-well plates at a density of 5 × 103 cells/well. The cells were incubated with or without the indicated agents for 72 h, and then CCK-8 reagent was added into each well at a volume of 1:10 of that of the culture medium. The cells were then cultured for 2 h while protected from light. Data were collected by reading the optical density (OD) at 450 nm.
RNA extraction and RT-qPCR
RNA extraction and RT-qPCR were performed as previously reported . In brief, RNA was isolated from cell lines using TRIzol Reagent (Invitrogen), and RNA samples (1 μg) were subjected to RT-qPCR using an RT-PCR kit (Takara) according to the manufacturer’s protocols. Relative quantification (RQ) was calculated as RQ = 2−ΔΔCt. The sequences of the primers are listed in Additional file 6: Table S3.
Protein extraction and western blot analysis
Cells were dissolved in lysis buffer containing protease inhibitors. Then, the proteins were subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to 0.22 μm polyvinylidene fluoride (PVDF) membranes. The membranes were blocked with 5% nonfat milk at room temperature for 2 h and then incubated with primary antibodies, including antibodies against GAPDH (Cell Signaling Technology), H3K27me3 (Active Motif), H3 (Active Motif), PRKCA (Abways), BCL2 (Cell Signaling Technology), pBCL2 (Cell Signaling Technology), ERK (Cell Signaling Technology), pERK (Cell Signaling Technology), RAF (Abways), pRAF (Abways), γH2AX (Abways), SOX9 (Abways), CD117 (Abways), and cleaved Caspase3 (Abways), overnight at 4 °C. After three washes in Tris-buffered saline with Tween 20 (TBST), the membranes were incubated with corresponding secondary antibodies for 2 h at room temperature.
Colony formation assay
Cells were seeded into six-well plates at a density of 1000 cells/well. After cell attachment, different agents were added. On day 7, the colonies were fixed and stained with 0.1% crystal violet. The number of colonies was counted manually.
Flow cytometry analysis
A flow cytometry assay was performed as described previously . In brief, the percentage of cells undergoing apoptosis was evaluated by flow cytometry (Beckman Gallios). Cells were incubated with different treatments for 48 h, harvested and resuspended in 500 μl of staining buffer with Annexin V-allophycocyanin (APC) and 7-AAD. The data were analyzed using FlowJo software.
Tumorsphere formation assay
Single-cell suspensions were added to ultra-low-attachment multiwell plates (Costar) in serum-free chemically defined medium  with different treatments. Half of the volume was replaced with fresh medium every second day. Images were obtained on day 10.
shRNA transfection and lentivirus packaging
KDM6A and KDM6B shRNA and the full protein-coding PRKCA open reading frame were synthesized by Sangon Biotech (Shanghai, China) and cloned into the pLVX plasmid (Open Biosystems) (sequences listed in Additional file 6: Table S4). Empty vectors were used as a negative control. Then, lentivirus was packaged using psPAX2 and pMD2G vectors. To obtain stable cell lines, lentivirus supernatant was added to the cells, and the cells were cultured with 2 μg/ml puromycin for 2 weeks.
Xenograft experiments and histologic analysis
All animals were housed and maintained in specific pathogen-free (SPF) conditions. All animal experimental protocols were approved by the ethics committee of Shanghai Ninth People’s Hospital. Twenty-four athymic nude mice (6–8 weeks old) each were injected subcutaneously with 1 × 106 143B cells. On day 5 after injection, the mice were randomly divided into four groups (n = 6): the Ctrl group, the GSK group, the Cis group, and the Cis+GSK group. Beginning on day 5, the GSK group and the Cis+GSK group received ip injections of GSK-J4 (50 mg/kg) every day. Beginning on day 7, the Cis group and the Cis+GSK group received ip injections of cisplatin (8 mg/kg biw). The volume of each tumor was measured as (length × width2/2) weekly after implantation until the mice were sacrificed.
Cells were fixed and stained with antibodies against γH2AX (Abways) or H3K27me3 (Active Motif) after being treated with different agents for 24 h. After being washed in phosphate-buffered saline (PBS) three times, the cells were stained with the antibodies Alexa Fluor 555 (Invitrogen) and Alexa Fluor 488 (Invitrogen) for 1 h at room temperature. Images were obtained with a Cell Observer (ZEISS).
RNAseq and KEGG analysis
RNA samples were obtained from control (Ctrl), cisplatin-treated (Cis) (cisplatin 1 μM, 24 h), cisplatin- and EPZ-6438-cotreated (Cis+EPZ) (cisplatin 1 μM, 24 h and EPZ-6438 10 μM, 24 h), and cisplatin- and GSK-J4-cotreated (Cis+GSK) (cisplatin 1 μM, 24 h and GSK-J4 15 μM, 24 h) 143B cells. After the proper pretreatments, the samples were subjected to RNAseq on a HiSeq 2500 system by Oebiotech Co. Ltd. Gene expression levels were analyzed with the programs htseq-count  and Cufflinks . All differentially expressed gene lists were generated with DESeq software .
Control and EPZ-6438 or GSK-J4 treated 143B cells (2 × 107) were crosslinked, lysed, and sheared using UCD-300 (Bioruptor) to ~ 200–700 base pairs in length. Then, ChIP was performed using an EZ ChIP kit (Millipore; 17–371) according to the manufacturer’s instructions, followed by quantification of ChIP-enriched DNA by RT-qPCR. The primer sequences are listed in Additional file 6: Table S3.
Each sample was analyzed in triplicate, and the experiments were repeated three times. The mean, standard error of the mean (SEM), and P values based on two-tailed t tests were calculated with Excel (Microsoft). Differences were considered significant at P < 0.05 (*P < 0.05, **P < 0.01).
We thank Oebiotech Co. Ltd. for their help.
This study was supported by grants from the National Key R&D Program of China (2016YFC1100600); the Shanghai Ninth People’s Hospital Lin+ Program (JYLJ012); and the National High Technology R&D Program (863 Program) (2015AA033603).
Availability of data and materials
All original data are available upon request.
YH, YJ, and CL conceived the project and designed the experiments; CH, YJ, and JS performed the experiments; and YH, YJ, and CL revised the manuscript and submitted the final version. All authors read and approved the final manuscript.
Ethics approval and consent to participate
This study was approved by the ethics committee of Shanghai Ninth People’s Hospital and was performed in accordance with the relevant guidelines.
Consent for publication
No parts of this manuscript are being considered for publication elsewhere.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 3.Moore DD, Luu HH. Osteosarcoma. In: Peabody TD, Attar S, editors. Orthop Oncol Prim Metastatic Tumors Skelet Syst [Internet]. Cham: Springer International Publishing; 2014. p. 65–92. https://doi.org/10.1007/978-3-319-07323-1_4.
- 4.Yang J, Zhang W. New molecular insights into osteosarcoma targeted therapy. Curr Opin Oncol. 2013;25:398–406 Available from: http://content.wkhealth.com/linkback/openurl?sid=WKPTLP:landingpage&an=00001622-201307000-00011. Cited 30 Apr 2018.CrossRefGoogle Scholar
- 5.Dieudonné F-X, Marion A, Haÿ E, Marie PJ, Modrowski D. High Wnt signaling represses the proapoptotic proteoglycan syndecan-2 in osteosarcoma cells. Cancer Res. 2010;70:5399–408 American Association for Cancer Research. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20530678. Cited 30 Apr 2018.CrossRefGoogle Scholar
- 13.Sakaki H, Okada M, Kuramoto K, Takeda H, Watarai H, Suzuki S, et al. GSKJ4, a selective jumonji H3K27 demethylase inhibitor, effectively targets ovarian cancer stem cells. Anticancer Res. 2015;35:6607–14 International Institute of Anticancer Research. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26637876. Cited 19 Nov 2018.PubMedGoogle Scholar
- 14.Watarai H, Okada M, Kuramoto K, Takeda H, Sakaki H, Suzuki S, et al. Impact of H3K27 demethylase inhibitor GSKJ4 on NSCLC cells alone and in combination with metformin. Anticancer Res. 2016;36:6083–92 International Institute of Anticancer Research. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27793936. Cited 19 Nov 2018.CrossRefGoogle Scholar
- 17.De Santa F, Totaro MG, Prosperini E, Notarbartolo S, Testa G, Natoli G. The histone H3 Lysine-27 demethylase Jmjd3 links inflammation to inhibition of polycomb-mediated gene silencing. Cell. 2007;130:1083–94 Available from: http://www.ncbi.nlm.nih.gov/pubmed/17825402. Cited 29 Apr 2018.CrossRefGoogle Scholar
- 19.Knutson SK, Warholic NM, Wigle TJ, Klaus CR, Allain CJ, Raimondi A, et al. Durable tumor regression in genetically altered malignant rhabdoid tumors by inhibition of methyltransferase EZH2. Proc Natl Acad Sci. 2013;110:7922–7 Available from: http://www.ncbi.nlm.nih.gov/pubmed/23620515. Cited 29 Apr 2018.CrossRefGoogle Scholar
- 20.Reinhardt HC, Aslanian AS, Lees JA, Yaffe MB. p53-deficient cells rely on ATM- and ATR-mediated checkpoint signaling through the p38MAPK/MK2 pathway for survival after DNA damage. Cancer Cell. 2007;11:175–89 NIH Public Access. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17292828. Cited 30 Apr 2018.CrossRefGoogle Scholar
- 21.Sobczyńska-Rak A, Silmanowicz P, Polkowska I. Tumor angiogenesis – factors influencing the development of a tumor vascular network and assessment of neoangiogenesis in histopathological samples. Med Weter. 2016;72:542–8 Available from: http://www.medycynawet.edu.pl/index.php/archives/382/5563-summary-med-weter-72-9-542-548-2016.Google Scholar
- 23.Beck B, Blanpain C. Unravelling cancer stem cell potential. Nat Rev Cancer. 2013:727–38 Available from: http://www.ncbi.nlm.nih.gov/pubmed/24060864. Cited 29 Apr 2018.
- 24.Bongiovanni L, Caposano F, Romanucci M, Grieco V, Malatesta D, Brachelente C, et al. Survivin and Sox9: potential Stem cell markers in canine Normal, hyperplastic, and neoplastic canine prostate. Vet Pathol. 2018. https://doi.org/10.1177/0300985818794161.
- 26.Szklarczyk D, Morris JH, Cook H, Kuhn M, Wyder S, Simonovic M, et al. The STRING database in 2017: quality-controlled protein–protein association networks, made broadly accessible. Nucleic Acids Res. 2017;45:D362–8 Oxford University Press. Available from: https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkw937. Cited 10 Aug 2018.CrossRefGoogle Scholar
- 27.Nakashima S. JB minireview-protein kinase C isotypes and their specific function protein kinase ca (PKCa): regulation and biological function. J Biockem. 2002; Available from: https://academic.oup.com/jb/article-abstract/132/5/669/919783.
- 34.Budihardjo I, Oliver H, Lutter M, Luo X, Wang X. Biochemical pathways of caspase activation during apoptosis. Annu Rev Cell Dev Biol. 1999;15:269–90 Annual Reviews 4139 El Camino Way, P.O. Box 10139, Palo Alto, CA 94303–0139, USA. Available from: http://www.annualreviews.org/doi/10.1146/annurev.cellbio.15.1.269. Cited 18 Aug 2018.CrossRefGoogle Scholar
- 35.Gill J, Ahluwalia MK, Geller D, Gorlick R. New targets and approaches in osteosarcoma. Pharmacol Ther. 2013;137:89–99 Pergamon. Available from: https://www.sciencedirect.com/science/article/pii/S0163725812002057?via%3Dihub. Cited 12 Aug 2018.CrossRefGoogle Scholar
- 41.Carter C. Protein kinase C as a drug target implications for drug or diet prevention and treatment of cancer. Curr Drug Targets. 2000;1:163–83 Available from: http://www.eurekaselect.com/openurl/content.php?genre=article&issn=1389-4501&volume=1&issue=2&spage=163. Cited 9 Aug 2018.CrossRefGoogle Scholar
- 46.He C, Wu T, Hao Y. Anlotinib induces hepatocellular carcinoma apoptosis and inhibits proliferation via Erk and Akt pathway. Biochem Biophys Res Commun. 2018;503:3093–9 Available from: https://linkinghub.elsevier.com/retrieve/pii/S0006291X18317820. Elsevier Ltd.CrossRefGoogle Scholar
- 47.Jiang C, He C, Wu Z, Li F, Xiao J. Histone methyltransferase SETD2 regulates osteosarcoma cell growth and chemosensitivity by suppressing Wnt/β-catenin signaling. Biochem Biophys Res Commun. 2018;502:382–8 Academic Press. Available from: https://www.sciencedirect.com/science/article/pii/S0006291X18312531?via%3Dihub. Cited 29 July 2018.CrossRefGoogle Scholar
- 49.Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31:166–9 Oxford University Press. Available from: https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btu638. Cited 25 May 2018.CrossRefGoogle Scholar
- 50.Roberts A, Trapnell C, Donaghey J, Rinn JL, Pachter L. Improving RNA-Seq expression estimates by correcting for fragment bias. Genome Biol. 2011;12:R22 BioMed Central. Available from: http://genomebiology.biomedcentral.com/articles/10.1186/gb-2011-12-3-r22. Cited 25 May 2018.CrossRefGoogle Scholar
- 51.Anders S, Huber W. Differential expression of RNA-Seq data at the gene level – the DESeq package. [cited 25 May 2018]; Available from: http://bioconductor.org/packages/release/bioc/vignettes/DESeq/inst/doc/DESeq.pdf
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.