Zhou et al. Cancer Cell Int (2020) 20:134 https://doi.org/10.1186/s12935-020-01210-1 Cancer Cell International

PRIMARY RESEARCH Open Access Silencing of microRNA‑135b inhibits invasion, migration, and stemness of ­CD24+CD44+ pancreatic cancer stem cells through JADE‑1‑dependent AKT/mTOR pathway Jingyang Zhou1, Haihong Wang2, Jinhui Che2, Lu Xu2, Weizhong Yang2, Yunjiu Li2 and Wuyuan Zhou2*

Abstract Background: Recent studies have emphasized determining the ability of microRNAs (miRNAs) as crucial regulators in the occurrence and development of pancreatic cancer (PC), which continues to be one of the deadliest malignancies with few efective therapies. The study aimed to investigate the functional role of miR-135b and its associated mecha- nism to unravel the biological characteristics of tumor growth in pancreatic cancer stem cells (PCSCs). Methods: Microarray analyses were initially performed to identify the PC-related miRNAs and genes. The expression of miR-135b and PCSC markers in PC tissues and cells was determined by RT-qPCR and western blot analysis, respec- tively. The potential gene (JADE-1) that could bind to miR-135b was confrmed by the dual-luciferase reporter assay. To investigate the tumorigenicity, migration, invasion, and stemness of PC cells, several gain-of-function and loss-of- function genetic experiments were conducted. Finally, tumor formation in nude mice was conducted to confrm the results in vivo. Results: miR-135b was highly-expressed in PC tissues and PCSCs, which was identifed to specifcally target JADE-1. The overexpression of miR-135b promoted proliferation, migration, and invasion of PCSC, inhibited cell apoptosis and increased the expression of stemness-related factors (Sox-2, Oct-4, Nanog, Aldh1, and Slug). Moreover, miR-135b could promote the expression of phosphorylated AKT and phosphorylated mTOR in the AKT/mTOR pathway. Addi- tionally, miR-135b overexpression accelerated tumor growth in nude mice. Conclusions: Taken together, the silencing of miR-135b promotes the JADE-1 expression, which inactivates the AKT/mTOR pathway and ultimately results in inhibition of self-renewal and tumor growth of PCSCs. Hence, this study contributes to understanding the role of miR-135 in PCSCs and its underlying molecular mechanisms to aid in the development of efective PC therapeutics. Keywords: microRNA-135b, Jade family PHD fnger 1, Protein kinase B/mammalian target of rapamycin pathway, Pancreatic cancer, Stemness

Background Pancreatic cancer (PC) is a devastating malignancy

*Correspondence: [email protected] aficting the digestive system [1] with a 5-year survival 2 Department of Hepatopancreatobillary Surgery, Xuzhou City Cancer rate of less than 10% over the past decades [2]. According Hospital, No. 131 Huancheng Rd., Gulou District, Xuzhou 221000, Jiangsu, to GLOBOCAN 2018, PC remains one of the most lethal People’s Republic of China Full list of author information is available at the end of the article malignancies leading to 432,242 new deaths in 2018, as

© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://crea- tivecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdo- main/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhou et al. Cancer Cell Int (2020) 20:134 Page 2 of 16

patients rarely present symptoms before the advanced of Xuzhou City Cancer Hospital. Extensive eforts were stage of the disease [3, 4]. Currently, it was reported that made to minimize the pain and sufering of the experi- there is a subpopulation of cells in tumors, cancer stem mental animal. cells (CSCs), which are capable of self-renewal and gen- erating unique progeny that form the main tumor cell Bioinformatics prediction of PC related miRNA/gene component of tumors [5]. Tus, the possible reason for PC-related datasets were obtained from the Gene the high incidence of PC can be associated with the high Expression Omnibus (GEO) database (https​://www.ncbi. oncogenic ability of pancreatic CSCs (PCSCs) which can nlm.nih.gov/geo/) for diferential expression analysis. not only boast self-renewal ability but also produce later One miRNA expression dataset (GSE41369) and three generation with ability to diferentiate [6]. Intriguingly, coding gene expression datasets (GSE16515, GSE32676, the functional role of PSCSs in promoting the metastasis and GSE71989) were used to screen out diferentially of PC could be related to the resistance of PC to stand- expressed miRNAs/genes. Details of the datasets are ard chemotherapy and radiation treatment [7]. Tere is shown in Additional fle 1: Table S1. Te R-language afy a well-defned requirement to investigate more potential installation package (http://www.bioco​nduct​or.org/packa​ therapeutic approaches for the treatment of PC. ges/relea​se/bioc/html/afy.html) was applied to carry Located on 1q32.1, miR-135b is encoded in the frst out standardized pre-processing of datasets. Te limma intron of the LEMD1 gene, cancer/testis antigen [8]. package (http://maste​r.bioco​nduct​or.org/packa​ges/relea​ Specifcally, miR-135b is a well-documented oncogene, se/bioc/html/limma​.html) was used to screen out dif- possesses the ability to adjust the growth of various ferentially expressed miRNAs/genes between tumor tumors [9]. Moreover, several studies have illustrated the tissues and normal tissues and the corrected p-value aberrant expression of miR-135b in various malignant was expressed by adj. p. Val. Te miRNAs/genes with tumors, which highlights the crucial role of miR-135b in |log2FC| > 1.0 and adj. p. Val < 0.05 were considered multiple tumors [10, 11]. Of note, a Global microRNA signifcantly diferentially expressed whereas the heat expression profling of microdissected tissues has indi- map of diferentially expressed genes was drawn with cated the potential role of miR-135b might act as a novel heatmap package (https​://cran.r-proje​ct.org/web/packa​ biomarker for pancreatic ductal adenocarcinoma [12]. ges/pheat​map/index​.html). Te target genes of diferen- Additionally, another study by Jiang et al. has confrmed tially expressed miRNAs were predicted in the DIANA, the involvement of miR-135b in tumourigenesis and miRDB, miRWalk, and mirDIP. Ten, miRNA-gene pairs chemoresistance of PC [13]. diferentially expressed in PC and regulatory interac- Nevertheless, due to its high expression in renal proxi- tions were screened in connection with the Venn online mal tubules, JADE-1 has been suggested as a promising analysis tools-Calculate and custom Venn diagrams were candidate for renal tumor inhibitor [14] that accelerates drawn (http://bioin​forma​tics.psb.ugent​.be/webto​ols/ the apoptosis of renal cancer [15]. Importantly, the AKT/ Venn/). mammalian target of rapamycin (mTOR) is known as one of the most critical pathways in regulating cell develop- Cell culture ment, proliferation, and survival [16]. A further study has Human PCSCs (CD44­ +/CD24+/ESA+) were isolated indicated that the activation of the AKT/mTOR pathway from human PC cell line PANC-1 (cell bank of Chinese could enhance cell proliferation in PC cells [17]. Besides, Academy of Sciences, Shanghai, China) (source bank: another study suggested that the AKT/mTOR pathway https​://www.fengh​bio.cn) according to the method could also impact the survival ability and proliferation of described by Jun Sheng Fu et al. [20]. Te isolated PCSCs prostate cancer cells [18]. Additionally, the AKT pathway were cultured in the PCSC culture medium (Celprogen) could be regulated by miR-135b in colorectal cancer cells containing 1% N2, 2% B27, 100 ng/mL epidermal growth [19]. However, scarce data is determining the relationship factor, and 1% antifungal agent (Invitrogen, Shanghai, between miR-135b and JADE-1 and the efect of this cor- China), as well as 20 ng/mL human platelet, derived relation on PC progression. Considering these limitations growth factor (Sigma-Aldrich, Shanghai, China). Te here we aimed to investigate whether miR-135b target cells were cultured in a 37°C incubator with 5% ­CO2 and JADE-1 to regulate the AKT/mTOR pathway, thus modu- 95% saturated humidity [21]. lating the invasion, migration, and stemness of PCSCs. Cell grouping and transfection Materials and methods CD24+CD44+ESA+ cells were assigned into the ago- Ethics statement mir-negative control (NC) group, the agomir-miR-135b All the animal study was carried out by following the group, the antagomir-NC group, the antagomir-miR- Institutional Animal Ethics Care and Use Committee 135b group, the sh-NC group, the sh-JADE-1 group, the Zhou et al. Cancer Cell Int (2020) 20:134 Page 3 of 16

dimethyl sulphoxide (DMSO) group (10 μmol/mL, St. synthesized by Shanghai GeneCore BioTechnologies Co., Louis, MO, USA), the perifosine (AKT/mTOR inhibitor) Ltd. (Shanghai, China) as illustrated in Table 1. Te fold group (10 μmol/mL, Selleck Chemicals, CA, USA), the changes were calculated using relative quantifcation -ΔΔCt agomir-miR-135b + DMSO (10 μmol/mL) group, and the ­(2 method). agomir-miR-135b + perifosine (10 μmol/mL) group. Te antagomirs and agomirs were designed and synthesized Western blot analysis by Dharmacon Inc. (Lafayette, CO, USA). Te total protein in cells was extracted by Radio Immu- CD24+CD44+ESA+ PCSCs were seeded into a 6-well noprecipitation Assay lysis bufer containing phenyl- plate and incubated overnight. Te cells were then methane sulfonyl fuoride (R0010, Beijing Solarbio treated with EntransterTM-R transfection solution com- Science & Technology Co., Ltd., Beijing, China). Te prised of agomir-miR-135b, antagomir-miR-135b, ago- protein concentration was determined according to the mir-NC, and antagomir-NC. instructions of the bicinchoninic acid kit (20201ES76, Lentiviral vector pLKO.1 (SCH002) was purchased Yeasen Company, Shanghai, China). Each well was loaded from Sigma-Aldrich (Shanghai, China). Te short hairpin with 30 μg sample and the protein was separated by the (sh)-JADE-1 and its corresponding control vector were polyacrylamide gel electrophoresis and transferred into a constructed by Wuhan Miaolingbio Inc. (Hubei, China) polyvinylidene fuoride membrane using a wet transfer- to establish a stable cell line. Te polybrene (6 mg/mL, ring method. Te membrane was blocked for 1 h with 5% Sigma-Aldrich, Shanghai, China) was added to the trans- bovine serum albumin at room temperature. Te primary fected cells with the aforementioned lentivirus. After antibodies to anti-human phosphorylated AKT (1:1000; 72 h of transduction, the cells were screened with 2 mg/ ab38449; Abcam Inc., Cambridge, UK), anti-human mL puromycin for 4 days. Table 1 RT-qPCR primer sequences Dual‑luciferase report gene assay Te 3′ untranslated region (3′UTR) fragment of JADE-1 miRNA/gene Primer sequence was amplifed by PCR and cloned into the pmirGLO vec- miR-135b F: 5′-CTG​TGG​CCT​ATG​GCT​TTT​CAT-3′ tor (Promega, Madison, WI, USA) by Shanghai Sangon R: 5′-GCT​CGC​CCC​TCA​CTG​TAG​-3′ Biotech Co., Ltd. (Shanghai, China) to construct wild- U6 F: 5′-CGC​TTC​GGC​AGC​ACA​TAT​AC-3′ type (WT) and mutant-type (MUT) recombinant dual- R: 5′-AAA​TAT​GGA​ACG​CTT​CAC​GA-3′ luciferase reporter plasmids. Te 293T cells (American JADE-1 F: 5′- AGG​GAT​TAA​AGT​GCT​CCC​CC-3′ Type Culture Collection [ATCC]) were assigned into 4 R: 5′-TGG​CGT​TTG​GCT​TGT​TAT​GG-3′ groups and co-transfected with miR-135b or NC and GAPDH F: 5′-AAT​GGG​CAG​CCG​TTA​GGA​AA-3′ pmirGLO-JADE-1-WT 3′UTR or pmirGLO-JADE- R: 5′-GCG​CCC​AAT​ACG​ACC​AAA​TC-3′ 1-MUT 3′UTR, respectively. After 30 h, the 293T cells CD24 F: 5′-CCC​CAC​CTT​GCC​TGCG-3′ were collected, followed by determining the luciferase R: 5′-AAA​TCT​GCG​TGG​GTA​GGA​GC-3′ activity using a fuorescent light detector (Promega Glo- CD44 F: 5′-TTA​CAG​CCT​CAG​CAG​AGC​AC-3′ max 20/20, Yuanpinghao Biotechnology Co. Ltd., Bei- R: 5′-TGA​CCT​AAG​ACG​GAG​GGA​GG-3′ jing, China) according to the provided instructions of ESA F: 5′-TGG​TTT​CAG​GGG​GCT​GTT​GT-3′ the dual-luciferase reporter gene detection kit (Promega, Madison, WI, USA). R: 5′-TCG​GCC​GTC​TCT​ACG​TCC​TC-3′ Sox-2 F: 5′-GCC​TGG​GCG​CCG​AGT​GGA​-3′ Reverse transcription‑quantitative polymerase chain R: 5′-GGG​CGA​GCC​GTT​CAT​GTA​GGT​CTG​-3′ reaction (RT‑qPCR) Oct-4 F: 5′-GCT​CGA​GAA​GGA​TGT​GGT​C-3′ Total RNA was extracted based on the Trizol method R: 5′-ATC​CTC​TCG​TTG​TGC​ATA​GTCG-3′ (Invitrogen, Carlsbad, California, USA). Total RNA was Nanog F: 5′-GCT​GAG​ATG​CCT​CAC​ACG​GAG-3′ then reversely transcribed into cDNA using the Rever- R: 5′-TCT​GTT​TCT​TGA​CTG​GGA​CCT​TGT​C-3′ tAIDTM First Strand cDNA Synthesis Kit (Fermentas, ALDH1 F: 5′-GCA​CGC​CAG​ACT​TAC​CTG​TC-3′ Termo Fisher, NY, USA). Te expression of miR-135b R: 5′-CCT​CCT​CAG​TTG​CAG​GAT​TAAAG-3′ was measured using a TaqMan miRNA assay (Ambion, Slug F: 5′-CGA​ACT​GGA​CAC​ACA​TAC​AGTG-3′ Austin, TX, USA) with U6 regarded as an internal refer- R: 5′-CTG​AGG​ATC​TCT​GGT​TGT​GGT-3′ ence. Te expression of JADE-1 was determined by the RT-qPCR reverse transcription quantitative polymerase chain reaction, microRNA- PrimeScript RT-PCR kits (TaKaRa, Shiga, Japan) and 135b miR-135b, JADE-1 jade family PHD fnger 1, GAPDH glyceraldehyde 3-phosphate dehydrogenase, ESA electrical stimulation for analgesia, F forward glyceraldehyde 3-phosphate dehydrogenase (GAPDH) primer, R reverse primer, Sox-2 SRY related HMG box-2, Oct-4 Octamer-4, ALDH1 was employed as an internal reference. All primers were aldehyde dehydrogenase isoform 1 Zhou et al. Cancer Cell Int (2020) 20:134 Page 4 of 16

AKT (1:1000; ab38449; Abcam Inc.), anti-mTOR (1:5000; Soft agar colony formation assay ab137133; Abcam Inc.), anti-phosphorylated mTOR Cells in the logarithmic growth phase in each group (1:1000; ab109268; Abcam Inc.), anti-human cleaved-cas- were detached using 0.25% trypsin and triturated pase-3 (1:1000; Cell Signaling Technology, Beverly, MA, into individual cells. Te cells were then suspended in USA), anti-human total caspase-3 (1:1000; #9662; Cell DMEM with 10% fetal bovine serum (FBS), seeded into Signaling Technology), anti-human cleaved-caspase-9 a 6-well plate, and subsequently cultured at 37 °C with antibody (1:1000; 20,750; Cell Signaling Technology), 5% ­CO2 with saturated humidity. Te culture medium anti-human total caspase-9 (1:1000; 9502; Cell Signal- was changed at regular intervals of 4 days. Te cells ing Technology), anti-SRY related HMG box-2 (Sox-2; were then fxed with a 4% polyformaldehyde solu- 1:1000; 2748; Cell Signaling Technology), anti-octamer- tion after 2 weeks, stained with 0.1% crystalized violet binding transcription factor 4 (Oct-4; 1:1000; 2750; Cell (Sigma-Aldrich, Shanghai, China), and counted manu- Signaling Technology), Nanog rabbit monoclonal anti- ally under a microscope. Te experiment was repeated body (1:5000; ab109250; Abcam Inc.), acetaldehyde dehy- 3 times. drogenases (Aldh1) rabbit monoclonal antibody (1:1000; ab52492; Abcam Inc.), anti-Slug antibody (5 µg/mL; Transwell assay ab51772; Abcam Inc.), JADE-1 rabbit polyclonal anti- Cell migration and invasion assessments were con- body (1:1000; ab155215; Abcam Inc.) and GAPDH rabbit ducted using the Transwell chamber (Boyden Cham- polyclonal antibody (1:2500; ab9485; Abcam Inc.) were bers; Corning, Cambridge, MA, USA). Te cells were cultured with the membrane overnight at 4 °C, followed 5 suspended in serum-free medium (5 × 10 cells/mL). by 1 h of incubation with horseradish peroxidase-labeled Cell migration assay was performed as follows: 200 µL corresponding secondary antibody at room tempera- cell suspension was added to the apical chamber of ture. By applying the electrochemiluminescence (ECL) Transwell. In the cell invasion experiment, 200 µL cells fuorescence detection kit (No. BB-3501, AmerSham GE were suspended in the apical chamber of Transwell Healthcare, Shanghai, China), the membrane was placed covered with Matrigel matrix glue (Corning, Cam- into a gel imager for exposure imaging. Te membranes bridge, MA, USA). All 600 µL of DMEM containing were then photographed using the Bio-Rad image analy- 10% FBS was added to the basolateral chamber of the sis system (Bio-Rad, Inc., Hercules, CA, USA) and subse- transwell. After the cells were incubated for 24–48 h, quently analyzed using the Quantity One v4.6.2 software. the cells in the apical chamber were erased using cot- Te relative protein content was expressed by the ratio ton swabs, fxed with 4% polyformaldehyde solution, of the gray value of the target protein band to that of the and stained with crystalized violet for 10 min (Sigma- GAPDH protein band. Aldrich, Shanghai, China). An inverted microscope (CX22; Olympus, Tokyo, Japan) was employed to ran- domly select 5 felds to observe, photograph, and sub- Cell counting kit‑8 (CCK‑8) sequently count the cell number. Te experiment was Te CCK-8 proliferation kit (Dojindo, Kumamoto, Japan) repeated 3 times [22]. was used for cell proliferation detection. Te cells were treated with 0.25% trypsin, followed by the addition of Flow cytometry Dulbecco’s modifed eagle’s medium (DMEM) to the trit- Flow cytometry was used to detect cell apoptosis. Te urate into individual cells. Following after, the cells were cells were seeded in a 6-well plate, and cultured for 48 h resuspended in stem cell medium for subsequent use. after treatment of virus plasmids. Te cells were treated After the cell density was adjusted, the cells were seeded with trypsin, washed twice with phosphate-bufered into a 96-well plate with 1000 cells per well in 100 μL cul- saline (PBS), and then resuspended in PBS (200 μL) ture medium, and the plate was gently rotated to achieve containing 10 μL RNAase (10 mg/mL) for incubation at uniform dispersion. Te cells were then cultured in an 37 °C for 30 min. Afterward, a 50 μL propidium iodide incubator under stable conditions of 37 °C with 5% ­CO2 (PI) solution was added and stored in an environment with saturated humidity. A total of 10 μL CCK-8 reagent void of light at room temperature for 20 min. Te was added at 0 h, 12 h, 24 h, 48 h and 72 h, respectively, whole experiment was conducted on the fow cytom- shaken gently, and mixed evenly. After incubation at eter (Accuri C6, BD Biosciences, San Jose, CA, USA) 37 °C for 2 h, the optical density (OD) of each well was [23]. Te upper right quadrant represented late apop- measured using an enzyme-linked immunosorbent assay totic cells, and the lower right quadrant represents (450 nm measuring wave). Each group experiment was early apoptotic cells. Apoptosis rate = late apoptosis performed in quintuplicate with a minimum of three rate early apoptosis rate. independent experiments. + Zhou et al. Cancer Cell Int (2020) 20:134 Page 5 of 16

Onco‑spheroids formation assay immunoglobulin G antibody (1:1000, ab6721, Abcam) Te adherent cells were treated with 0.25% trypsin to was supplemented to culture the sections at 37 °C for prepare a single-cell suspension. Te cell density was 30 min. Afterward, the sections were developed by diam- 3 adjusted to 1 × 10 cells/mL, and then seeded into inobenzidine and counterstained with hematoxylin fol- growth medium containing stem cells (adding 1% N2, lowed by dehydration, clearing, sealing, and observation 2% B27, 100 ng/mL epidermal derived growth factor, under a microscope. PBS bufer instead of the primary and 1% antifungal agent [Invitrogen, Carlsbad, Cali- antibody was used as NC. Te proportion of positive fornia, USA]; 20 ng/mL human platelet growth factor, cells more than 10% was regarded as positive. Te stain- [Sigma-Aldrich, Shanghai, China]) in a low adsorption ing was mainly located in the cytoplasm or membrane or 6-well plate (Corning Inc., Corning, NY, USA) with cytoplasm, which was brownish-yellow. Five high power 2 mL/well. Te cells were then subsequently cultured visual felds were randomly selected to observe the posi- to obtain suspended cell spheres, and semi-quantitative tive expression rate of phosphorylated AKT, JADE-1, and liquid exchange was performed every 2 days. After such phosphorylated mTOR. a continuous culture for 10 days, the number of newly formed suspended cell spheres per well were recorded Statistical analysis and averaged under a microscope. Te experiment was All data were processed using the SPSS 21.0 statistical repeated 3 times [20, 21]. software (IBM Corp. Armonk, NY, USA). Measurement data were expressed as the mean ± standard deviation. In vivo study Normal distribution and homogeneity of variances were A total of 36 BALB/C male nude mice (aged 4–6 weeks; used to assess all data. A paired student’s t-test was weighing 18–20 g) were purchased from the experimen- used to compare paired data between the two groups. tal animal center of Southern Medical University (Guang- Whereas unpaired student’s t-test was used to compare dong, China) with all mice confrmed to have passed the unpaired data between two groups. One-way analysis of quarantine inspection. Te nude mice were randomly variance (ANOVA) was applied for comparison among divided into agomir-NC, agomir-miR-135b, antagomir- multiple groups followed by Tukey’s post hoc test. Two- NC, antagomir-miR-135b, agomir-miR-135b + DMSO, way ANOVA was performed to compare the data of cell and agomir-miR-135b + perifosine groups (n = 6 in each viability among multiple groups at diferent time points. group), raised in specifc pathogen-free (SPF) animal Repeated-measurement ANOVA was used to compare houses. Te successfully constructed and stably-trans- the data among multiple groups at diferent time points, 6 fected cells (1 × 10 cells) were mixed with 75 μL matrix followed by Bonferroni post hoc test. A p < 0.05 value was gel (50:50; Beckon Dickinson, Bedford, MA, USA). All considered to be indicative of statistical signifcance. 100 μL prepared tumor cell suspension was injected into the subcutaneous fanks of nude mice by a 1 mL-syringe. Results 2 miR‑135b is highly expressed in PCSCs Te tumor volume (V) = L (length) × W (width) × 0.5 was observed every 3 days and measured by a vernier cal- Bioinformatics was applied to predict the potential iper. After 30 days, the nude mice were euthanized and molecular targets of PC. R language was used to screen the subcutaneous tumors were completely stripped and diferentially expressed miRNA/genes from PC data- weighed. sets. Firstly, the diferentially expressed miRNAs were selected from miRNA dataset GSE41369. According to Immunohistochemistry the ascending order of adj. p. Val, we ranked the top 20 Te parafn-embedded sections were dewaxed and diferentially expressed miRNAs in GSE41369 (Fig. 1a). hydrated: sections were immersed in xylene I and II for Among these miRNAs, miR-135b was highly expressed 10 min, respectively, and in a diferent gradient of alco- in PC tumor tissues (log FC > 1 and adj. p. Val was the hol (100%, 95%, 80%, and 70%) for 2 min, respectively. smallest, Additional fle 1: Table S2). To study the difer- Te sections were immersed in 3% ­H2O2 for 10 min fol- ence of miR-135b expression in PC cells and PCSCs, the lowed by 10-min antigen retrieval. Te sections were fow cytometry was utilized to categorize the three-pos- then cooled down at room temperature, and supple- itive PCSCs and RT-qPCR was conducted to detect the mented with 5% bovine serum albumin, followed by the expression of miR-135b in both PC cells and PCSCs. Our 30-min culture at 37 °C. Te sections were then cultured results showed that (1.442 ± 0.152)% of the PANC-1 cells + + with 50 μL rabbit polyclonal antibodies (Abcam, Cam- were ­CD44 CD24 , while only (0.492 ± 0.149)% of the bridge, UK) to phosphorylated AKT (1:500, ab38449), cells were ­CD44+CD24+ESA+ (Fig. 1b). Te expression JADE-1 (1:500, ab155215), and phosphorylated mTOR of miR-135b in CD44­ +CD24+ and CD44­ +CD24+ESA+ (1:100, ab2732) overnight at 4 °C. Ten, goat anti-rabbit cells was signifcantly higher than that in PANC-1 cells, Zhou et al. Cancer Cell Int (2020) 20:134 Page 6 of 16

a b

C

Fig. 1 miR-135b was highly expressed in PCSCs. a The heat map of top 20 diferential miRNA expression in PC miRNA dataset GSE41369 (the abscissa indicated the sample number, the ordinate indicated the diferential gene, the histogram at the top right indicated the color scale and each rectangle in the map corresponded to a sample expression value); b The proportion of fow sorting positive cells in PANC-1 cells; C, RT-qPCR was used to detect miR-135b expression in cells; *p < 0.05 vs. the PANC-1 group, #p < 0.05 vs. the ­CD44+CD24+ group. The measurement data were expressed as mean standard deviation. Unpaired student’s t-test was used to compare unpaired data between two groups. One-way ANOVA was ± applied for comparison among multiple groups, followed by Tukey’s post hoc test. The experiment was repeated 3 times

but the expression of miR-135b in ­CD44+CD24+ESA+ cytometry whereas the expression of apoptosis-related cells was signifcantly higher than that in ­CD44+CD24+ proteins (total caspase-3, total caspase-9, cleaved-cas- cells (Fig. 1c, p < 0.05). Terefore, these results suggested pase-3, and cleaved-caspase-9) was determined by west- that miR-135b exhibited high expression in PCSCs. ern blot analysis. Our results showed that miR-135b signifcantly inhibited the apoptosis in PCSC and the miR‑135b inhibitor suppresses viability while enhances expressions of cleaved-caspase-3 and cleaved-caspase-9 apoptosis of PCSCs were decreased accordingly, but no signifcant difer- To study the efect of miR-135b on the biological charac- ence in expression of total caspase-3 and total cas- teristics of PCSCs, the viability of PCSCs was measured pase-9 after diferent transfections were found (Fig. 2b, using the CCK-8 method. Our results showed that ago- c, p < 0.05). Tereafter, the cloning ability of PCSCs was mir-miR-135b promoted cell viability (Fig. 2a, p < 0.05). tested and we found that miR-135b elevated the colony Meanwhile, the apoptosis of PCSCs was detected by fow formation of cells (Fig. 2d, p < 0.05). Te above-reported

(See fgure on next page.) Fig. 2 miR-135b inhibition could suppress viability, invasion, and migration and promote apoptosis in PCSCs. a CCK-8 assay was used to test the viability of PCSCs. b Flow cytometry was applied to detect apoptosis of PCSCs. c Western blot analysis was utilized to detect the expression of apoptotic proteins. d The cloning ability of PCSCs was tested by soft agar colony forming experiment. e onco-spheroids formation assay was used to detect the sphere-forming ability of PCSCs ( 400). f Transwell assay was applied to examine the migration and invasion of PCSCs ( 200). g × × Western blot analysis and RT-qPCR were used to detect expression of tumor stem cell markers. *p < 0.05 vs. the agomir-NC group; #p < 0.05 vs. the antagomir-NC group. The measurement data were expressed as the mean standard deviation. One-way ANOVA was applied for comparison ± among multiple groups, followed by Tukey’s post hoc test. Two-way ANOVA was performed to compare the data of cell viability among multiple groups at diferent time points. The experiment was repeated 3 times Zhou et al. Cancer Cell Int (2020) 20:134 Page 7 of 16

a agomir-NC b agomir-NC agomir-miR-135b agomir-miR-135b antagomir-NC antagomir-NC 4 4 4 agomir-NC 4 agomir-miR-135b antagomir-NCantagomir-miR-135b antagomir-miR-135b antagomir-miR-135b # 10 10 10 1.5 10.03% 10 5.04% 12.12% 21.15% 25 3 3 3 * 3 10 10 10 10 20 * %) 2 2 2 1.0 2 * 10 10 10 10 at e( 450nm) PI PI PI * PI 15 sr e( 1 1 1 1 si 10 10 10 * 10 to 10 valu 0.5 * 0 0 0 # 0

OD # 10 Ap op 10 10 10 5 # 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 # 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 Annexin V FITC Annexin V FITC Annexin V FITC Annexin V FITC 0.0 # 0 01224364872 cdTime (h) agomir-NC cleaved-Caspase3 agomir-NC agomir-miR-135b agomir-miR-135b antagomir-NC antagomir-NC antagomir-miR-135b cleaved-Caspase9 4

n antagomir-miR-135b

io 150 # # total-Caspase3 * 3 agomir-NCagomir-miR-135b n xpress 100 ne ei total-Caspase9 2 * omatio # ot

* yf pr

GAPDH ve 1 50 Colon

C b 5 Relati -N 35b -NC ir 1 ir 0 3 9 m e 0 om s se-3 se-9 go pase a ag -miR-13 pa pa p antagomir-NC antagomir-miR-135b ir-miR- ta as as s s C a an -C - l c ca agomir-NC om d a al ve ed t t ag av to to tagomir e agomir-miR-135b an clea cl e antagomir-NC antagomir-miR-135b

150 s ell c * 00

r5 100

pe # ere

ph 50

25 μm 25 μm 25 μm 25 μm rs mo

agomir-NC agomir-miR-135b antagomir-NCantagomir-miR-135b Tu 0

f Invasion Migration

agomir-NC agomir-NC agomir-miR-135b agomir-miR-135b antagomir-NC antagomir-NC antagomir-miR-135b antagomir-miR-135b

50 * 150 50 μm 50 μm 50 μm 50 μm * ells

agomir-NC agomir-miR-135b agomir-NC agomir-miR-135b 40 cells 100 # 30

# grated nvasiv ec mi fi 20 of 50 er be ro 10 mb Num

50 μm 50 μm 50 μm 50 μm Nu 0 0 antagomir-NC antagomir-miR-135b antagomir-NCantagomir-miR-135b

g agomir-NC agomir-NC Slug agomir-miR-135b agomir-miR-135b antagomir-NC antagomir-NC Nanog antagomir-miR-135b antagomir-miR-135b * 2.5 * n 3 * * * * Aldh1 * * 2.0 xpression xpressio 2 1.5 # *

Sox-2 ne # # # * # # ## RN Ae 1.0

protei # 1 Oct-4 #

tive 0.5 la Re Relativ em GAPDH 0 0.0 g g 1 4 g 1 -2 -4 u o -2 - h x t dh t Slug d C b C b Sl an l ox Al Oc 5 A S Oc Nano So -N -N N ir -13 r mi om iR o g m -miR-135 a ir- tag ir m om an g a tago n a Zhou et al. Cancer Cell Int (2020) 20:134 Page 8 of 16

results demonstrated that overexpression of miR-135b transfected with the agomir-miR-135b, the agomir-NC, could promote the proliferation of PCSCs and inhibit cell the antagomir-miR-135b, and the antagomir-NC respec- apoptosis. tively. Te growth of tumor in nude mice was analyzed Onco-spheroids formation assay was used to test the by recording the growth, volume, and weight of tumor- sphere-forming ability of PCSCs. Our results exhib- bearing mice subcutaneously. Te tumor weight was ited that compared with the agomir-NC group, the measured by peeling of the tumor tissue on the 30th sphere-forming ability of the agomir-miR-135b group day. Our results manifested that compared with the was signifcantly enhanced. However, compared to the agomir-NC group, the tumor growth rate of PCSCs in antagomir-NC group, the ability of stem cells to form nude mice of the agomir-miR-135b group, the PCSCs in spheres in the antagomir-miR-135b group was mark- nude mice were signifcantly accelerated, the tumor vol- edly decreased (Fig. 2e, p < 0.05). Transwell assay was ume was obviously increased, and the tumor weight was employed to examine the migration and invasion of markedly raised. However, compared with the antago- PCSCs and results indicated that the migration and mir-NC group, the subcutaneous tumors of nude mice invasion of PCSCs in the agomir-miR-135b group were in the antagomir-miR-135b group grew slowly whereas signifcantly increased compared with the agomir-NC the tumor volume and weight were remarkably reduced group. Moreover, compared to the antagomir-NC group, (Fig. 3a–c, p < 0.05). Tese results indicated that the low the migration and invasion of stem cells in the antago- expression of miR-135b in PCSCs inhibited tumorigen- mir-miR-135b group were signifcantly reduced (Fig. 2f, esis in vivo. p < 0.05). Western blot analysis and RT-qPCR analyses were used to detect the mRNA and protein expression miR‑135b is predicted to target JADE‑1 to regulate PC of tumor stem cell markers (Sox-2, Oct-4, Nanog, Aldh1 progression and Slug). Intriguingly, our results indicated that the To further investigate the molecular mechanism that overexpression of miR-135b could promote the mRNA miR-135b may participate in PC progression, the target and protein expression of these tumor stem cell markers genes of miR-135b were predicted from the DIANA, (Fig. 2g, p < 0.05). Taken together, our results indicated miRDB, miRWalk, and mirDIP, the results of which that the downregulated miR-135b could afect the cellu- demonstrated that 1116, 472, 10,132, and 1159 target lar physiology of PCSCs. genes were obtained respectively. Venn map (Fig. 4a) provided the data indicating 192 intersected genes, and Low expression of miR‑135b inhibits tumorigenesis these intersected genes were taken as the target genes and tumor growth in vivo of miR-135b for subsequent analyses. In the mean- To identify the tumorigenicity and tumor growth of time, the diferentially expressed genes were screened PCSCs with low expression of miR-135b, tumor forma- out from PC gene datasets i.e., GSE16515, GSE32676, tion assay in nude mice was performed using PCSCs and GSE71989, of which 475, 213, and 646 genes were

a b agomir-NC c agomir-NC agomir-miR-135b agomir-miR-135b antagomir-NC antagomir-NC 1500 antagomir-miR-135b 2.0 antagomir-miR-135b )

3 * agomir-NC ) 1.5 mm 1000 * *

e( * # agomir-miR-135b *

um * * 1.0 ol * rv 500 antagomir-NC # * # # 0.5 # # Tumor weight (g Tu mo # # antagomir-miR-135b # 0 0.0 36912151821242730 Time (d) Fig. 3 Inhibited miR-135b suppressed tumorigenesis and tumor growth in vivo. a Macroscopic observation of tumor size on 30th day. b measurement of tumor volume in nude mice with vernier caliper. c Determination of tumor mass in nude mice by weighing method; The tumorigenicity and growth of PCSCs were detected by tumor formation in nude mice. n 6, *p < 0.05 vs. the agomir-NC group, #p < 0.05 vs. the = antagomir-NC group; The measurement data were expressed as mean standard deviation. One-way ANOVA was applied for comparison among ± multiple groups, followed by Tukey’s post hoc test. Repeated measurement ANOVA was used to compare the data among multiple groups at diferent time points, followed by the Bonferroni post hoc test Zhou et al. Cancer Cell Int (2020) 20:134 Page 9 of 16

a b c

d ef

Fig. 4 JADE-1 was predicted to be a target gene of miR-135b. a Comparison among the results of predicting miR-135b target genes from DIANA, miRDB, miRWalk and mirDIP. b The similarities and diferences of down-regulated expressed genes in pancreatic cancer gene datasets GSE16515, GSE32676 and GSE71989. c JADE-1 was both the target gene of miR-135b and the down-regulated expressed gene of pancreatic cancer. d, e The heat map of GSE16515 and GSE71989 gene expression datasets for the top 60 down-regulated diferential genes (the abscissa represents the sample number, the ordinate represents the diferentially expressed genes, the histogram at the upper right represents the color scale, and each rectangle in the map corresponds to a sample expression value). f The expression of JADE-1 in pancreatic cancer gene dataset GSE32676

down-regulated in each dataset respectively. Compari- miR‑135b inhibits the expression of JADE‑1 sons of these diferentially expressed genes from the Te target gene JADE-1 binding sites of miR-135b 3 datasets revealed there were 15 intersected genes were predicted through the bioinformatics online tool (Fig. 4b). Tereafter, these 15 intersected down-reg- Targetscan (Fig. 5a). Te dual-luciferase reporter gene ulated genes in PC tissues of the 3 datasets were used assay illustrated that compared to the other 3 groups, as candidate genes which were further aligned with the the luciferase activity in the JADE-1-WT 3’UTR and target genes of miR-135b (Fig. 4c). Only one intersected miR-135b mimic co-transfected group was signifcantly gene, namely, JADE-1, was identifed, which suggested decreased (Fig. 5b). RT-qPCR and western blot analysis that the low expression of JADE-1 in PC might be regu- demonstrated that the JADE-1 expression in the ago- lated by miR-135b. Te gene expression heat maps of mir-miR-135b group was signifcantly lower than that GSE16515 and GSE71989 datasets are shown in Fig. 4d, in the agomir-NC group. Moreover, compared to the e respectively while the expression changes of JADE-1 antagomir-NC group, JADE-1 expression in PCSCs of in GSE32676 are shown in Fig. 4f, which indicated that the antagomir-miR-135b group was markedly increased JADE-1 was poorly expressed in PC tissues. Neverthe- (p < 0.05; Fig. 5c, d). Tus, it was suggested that miR- less, a previous study has revealed that JADE-1 is an 135b suppressed the expression of JADE-1. anti-oncogene with the ability to inhibit the AKT path- way [15], which poses signifcant importance in PC [23]. Terefore, we speculated that miR-135b might tar- get JADE-1 and regulates the growth of PC tissues. Zhou et al. Cancer Cell Int (2020) 20:134 Page 10 of 16

a Predicted consequential pairing of target region (top) Site Context++ Context++ score Weighted Conserved branch P and miRNA (bottom) type score percentile context++ score length CT

Position 2033-2040 of PHF17 3’ UTR5’ GAGGGAUAAUUUGUCAAGCCAUA 8mer -0.22 90 -0.213.748 0.75 hsa-miR-135b-5p 3’ AGUGUAUCCUUACUUUUCGGUAU

agomir-NC agomir-NC b mimic-NC c agomir-miR-135b d agomir-miR-135b

miR-135b mimic antagomir-NC n antagomir-NC antagomir-miR-135b # 1.5 4 antagomir-miR-135b JADE-1 rotei 1.0 # 0.8 3

1.0 GAPDH ADE- 1p *

fJ 0.6

2 b C b no * 0.4 -135 ir-N -135 0.5 * mir-NC m o R o 1 -mi g -miR ag ir ta ir

xpressio 0.2 m om an o ee Relative luciferase activity ag tag 0.0 0 an 0.0 JADE-1-WT3'UTR JADE-1-MUT3'UTR Relative expression of JADE-1 mRNA Relativ Fig. 5 miR-135b suppressed JADE-1 expression specifcally. a The binding site of miR-135b to JADE-1 (aliasing PHF-17) 3′UTR was analyzed by Targetscan. b dual-luciferase reporter gene assay was used to determine the targeting relationship between miR-135b and JADE-1 (*p < 0.05 vs. the agomir-NC group); c RT-qPCR was applied to measure JADE-1 mRNA expression in PCSCs of each group. d western blot analysis was used to examine the expression of JADE-1 protein in PCSCs of each group. *p < 0.05 vs. The mimic-NC or antagomir-NC group, #p < 0.05 vs. the agomir-NC group. The measurement data were expressed as mean standard deviation. One-way ANOVA was applied for comparison among multiple groups, ± followed by Tukey’s post hoc test. The experiment was repeated 3 times

agomir-NC p < 0.05). Hence, overexpression of miR-135b activated agomir-miR-135b p-AKT the AKT/mTOR pathway through inhibition of JADE-1. antagomir-NC antagomir-miR-135b AKT 1.5 sh-NC p-mTOR sh-JADE-1 & miR‑135b activates the AKT/mTOR pathway to promote * * & mTOR 1.0 cell proliferation and restrict apoptosis of PCSCs GAPDH # # C C 0.5 To investigate whether miR-135b-inhibited JADE-1 pro- -N NC ir r- DE-1 mi sh-N om iR-135b motes the AKT/mTOR pathway to afect the charac- m go miR-135b ag r- ta r- sh-JA Relative protein expression omi an 0.0 g teristics of PCSCs, CCK-8 assay was frst employed to a tagomi p-AKT/AKT p-mTOR/mTOR an examine the viability of PCSCs. Our results showed that Fig. 6 miR-135b promoted activation of AKT/mTOR pathway via cell viability was enhanced after silencing the JADE-1, inhibition of JADE-1 determined by western blot analysis; *p < 0.05 vs. the agomir-NC group; #p < 0.05 vs. the antagomir-NC group; but perifosine could reverse the promotion of miR-135b &p < 0.05 vs. the sh-NC group; The measurement data were expressed on cell viability (Fig. 7a, p < 0.05). Tereafter, fow cytom- as mean standard deviation. One-way ANOVA was applied for etry was used to examine the apoptosis rate of PCSCs. ± comparison among multiple groups, followed by Tukey’s post hoc Our results showed that perifosine enhanced whereas the test. The experiment was repeated 3 times silencing of JADE-1 signifcantly reduced the apoptotic rate of the cells. Meanwhile, western blot analysis mani- fested that the expression of apoptosis-related proteins (cleaved-caspase-3 and cleaved-caspase-9) was decreased miR‑135b promotes activation of the AKT/mTOR signaling accordingly by JADE-1 silencing but enhanced by peri- pathway via inhibition of JADE‑1 fosine, while expression of total caspase-3 and total Te AKT/mTOR signaling pathway plays a crucial role in caspase-9 showed no signifcant change after diferent the development of PC [24]. Hence, we set out to inves- transfections (Fig. 7b, c, p < 0.05). Next, the cloning ability tigate the efects conferred by miR-135b and JADE-1 of PCSCs was examined, we found that perifosine could on the AKT/mTOR signaling-pathway related proteins. reverse the advance efect of miR-135b on PCSC cloning Western blot analysis documented that overexpressed ability. Te cloning ability of cells was strengthened after miR-135b or silencing JADE-1 accelerated the phospho- JADE-1 silencing (Fig. 7d, p < 0.05). Te above results rylation of AKT and mTOR. However, silencing miR- showed that overexpression of miR-135b decreased 135b led to a drop in the expression of phosphorylated JADE-1 expression and activated the AKT/mTOR path- AKT and phosphorylated mTOR, but no signifcant way to promote proliferation and inhibit apoptosis of efect on AKT and mTOR expression was found (Fig. 6, PCSCs. Zhou et al. Cancer Cell Int (2020) 20:134 Page 11 of 16

sh-NC sh-JADE-1 absh-NC DMSO sh-JADE-1 Perifosine DMSO agomir-miR-135b + DMSO Perifosine agomir-miR-135b + Perifosine agomir-miR-135b + DMSO agomir-miR-135b + # 1.5 agomir-miR-135b + Perifosine sh-NC sh-JADE-1 DMSO Perifosine agomir-miR-135b + DMSO Perifosine 30 4 4 4 4 4 4 ) ) 10 10 10 10 * 10.16% 10 4.56% 11.12% 10 24.15% 5.03% 13.65% (% 3 * 3 3 3 3 1.0 3 20 10 10 10 10 10 10 & * * & & rate 2 2 2 2 2 & 2 * 10 10 10 10 10 10 PI PI PI PI PI & PI * #

1 10 1 1 1 1 0.5 1 & # 10 10 10 10 10 10 Apoptos is OD value (450nm * # 0 0 0 0 # 0 0 10 10 10 # 10 10 10 0 1 2 3 4 0 1 2 3 4 0.0 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 10 10 10 10 10 10 10 10 10 10 0 Annexin V FITC Annexin V FITC Annexin V FITC Annexin V PE Annexin V FITC Annexin V PE 01224364872 Time (h) sh-NC c sh-NC d sh-JADE-1 cleaved-Caspase3 sh-JADE-1 DMSO DMSO Perifosine Perifosine cleaved-Caspase9 agomir-miR-135b + DMSO agomir-miR-135b + DMSO agomir-miR-135b + Perifosine agomir-miR-135b + Perifosine total-Caspase3 4 150 # sh-NCsh-JADE-1 DMSO * & 3

# tion total-Caspase9 100 expression ma

2 fo # * & GAPDH rotein * & 50 ep

1 Colony C -1 O e E in ine MS s D o Relativ sh-N D rif 0 + DMSO e 0 Perifos b 3 9 9 Perifosineagomir-miR-135b + DMSO agomir-miR-135b + Perifosine sh-JA + P e e -3 - s s se se 135 b a a a a 35 sp sp sh-NC miR- Casp Casp a ca d- d- l c l e e e ta ta mir- o o sh-JADE-1 r-miR-1 av av t t go i e e a cl cl DMSO gom a Perifosine agomir-miR-135b + DMSO agomir-miR-135b + Perifosine 150

ells &

0c * 50 100 er

25 μm 25 μm 25 μm 25 μm 25 μm 25 μm ep #

pher 50

sh-NC sh-JADE-1 DMSO Perifosine agomir-miR-135b + DMSO agomir-miR-135b rs + Perifosine Tu mo 0 f sh-NC sh-NC sh-JADE-1 sh-JADE-1 DMSO DMSO Perifosine Perifosine agomir-miR-135b + DMSO agomir-miR-135b + DMSO agomir-miR-135b + Perifosine agomir-miR-135b + Perifosine 50 μm 50 μm 50 μm 50 μm 50 μm 50 μm 50 & 200

s & s * *

sh-NC sh-JADE-1 DMSO sh-NC sh-JADE-1 DMSO cell cell 40 150 # 30 grated vasive

in 100 # mi of

20 of er ber 50 10 mb 50 μm 50 μm 50 μm 50 μm 50 μm 50 μm Num Nu Perifosineagomir-miR-135b + DMSO agomir-miR-135b + Perifosine Perifosineagomir-miR-135b + DMSO agomir-miR-135b + Perifosine 0 0

g sh-NC Slug sh-NC sh-JADE-1 sh-JADE-1 DMSO DMSO Perifosine Nanog Perifosine agomir-miR-135b + DMSO agomir-miR-135b + DMSO agomir-miR-135b + Perifosine 4 agomir-miR-135b + Perifosine 4 Aldh1 n

& & & * 3 * 3 * &

* Sox-2 xpressio & * & expression & & * * & * * ne 2 2 # #

Oct-4 protei #

mRNA # # # # # # * & 1 #

1 tive GAPDH la Re Relative 0 0 1 SlugNanog Aldh1Sox-2Oct-4 Slug Nanog Aldh1 Sox-2 Oct-4 O ne ne -NC S SO si DE- M fo sh A D DM ri + Perifosi b sh-J 5 + Pe 3 b 1 5 R- 13 -mi ir om ir-miR- ag m go a Fig. 7 miR-135b activates AKT/mTOR pathway to promote cell proliferation and restrict apoptosis of PCSCs. a CCK-8 was used to test the proliferation of PCSCs. b Flow cytometry was applied to detect apoptosis. c Western blot analysis was utilized to examine the expression of apoptotic protein. d Clone forming assay was employed to determine the activity of PCSCs. e Onco-spheroids formation assay was used to test the sphere forming ability of PCSCs ( 400). f Transwell was applied to detect the migration and invasion of PCSCs ( 200). g Western blot analysis and × × RT-qPCR were carried out to detect expression of tumor stem cell markers. *p < 0.05 vs. the sh-NC group; #p < 0.05 vs. the DMSO group; &p < 0.05 vs. the agomir-miR-135b DMSO group. The measurement data were expressed as the mean standard deviation. One-way ANOVA was applied for + ± comparison among multiple groups, followed by Tukey’s post hoc test. Two-way ANOVA was performed to compare the data of cell viability among multiple groups at diferent time points. Repeated measurement ANOVA was used to compare the data among multiple groups at diferent time points, followed by the Bonferroni post hoc test. The experiment was repeated 3 times Zhou et al. Cancer Cell Int (2020) 20:134 Page 12 of 16

Te stem cell onco-spheroids formation assay described enhanced positive expression rates of phosphorylated that compared with the sh-NC group, the rate of PCSC AKT and phosphorylated mTOR while the antagomir- sphere-forming in the sh-JADE-1 group was signifcantly miR-135b exhibited the opposite trends as compared higher. Compared to the DMSO group, the sphere-form- to the antagomir-NC (p < 0.05). However, JADE-1 posi- ing rate of PCSCs in the perifosine group was decreased. tive expression rate did not potently change and positive Compared with the agomir-miR-135b + DMSO group, expression rates of phosphorylated AKT and phospho- the sphere-forming rate in the agomir-miR-135b + peri- rylated mTOR were diminished (p < 0.05; Fig. 8d). In fosine group was signifcantly decreased (Fig. 7e, summary, miR-135b activates the AKT/mTOR pathway, p < 0.05). Transwell assay was employed to investigate the thus enhancing the tumorigenicity of PCSCs in vivo. migration and invasion ability of PCSCs (Fig. 7f, p < 0.05). Our results exhibited that compared with the sh-NC Discussion group, the migration and invasion abilities of PCSCs in PC is considered as the most devastating malignancies the sh-JADE-1 group were signifcantly enhanced. Com- known to man, characterized by rarely exhibiting symp- pared with the DMSO group, the migration and inva- toms at the early stage, poor prognosis, rapid growth, sion of PCSCs in the perifosine group were decreased. and expansion [25]. Hence, a comprehensive understand- Compared with the agomir-miR-135b + DMSO group, ing of the molecular mechanisms associated with the the migration and invasion of PCSCs in the agomir- physiology of PC is urgently required. A recent study has miR-135b + perifosine group were remarkably weak- highlighted the critical role of miRNAs in prognostic and ened. Western blot analysis and RT-qPCR were used to therapeutic perspectives for various cancers including PC detect the mRNA and protein expression of some repre- [26]. Te purpose of the present study was to investigate sentative tumor stem cell markers (Sox-2, Oct-4, Nanog, the roles of miR-135b in biological characteristics and Aldh1, Slug,) and our results showed that the mRNA and tumor growth of PCSCs. Intriguingly, Our data from can- protein expression of tumor stem cell markers in the sh- cer cell lines and animal experiments demonstrated that JADE-1 group was signifcantly higher than that in the miR-135b down-regulated the JADE-1 to promote pro- sh-NC group. Compared to the DMSO group, the mRNA liferation and suppress apoptosis of PCSCs via the AKT/ and protein expression of tumor stem cell markers in the mTOR pathway. perifosine group was declined. As compared to the ago- Being a member of miRNAs, miR-135b has been mir-miR-135b + DMSO group, the mRNA and protein reported to be involved in growth in many malignancies. expression of tumor stem cell markers in the agomir- Particularly, a study by Lin et al. has indicated the upreg- miR-135b + perifosine group was signifcantly declined ulation of miR-135b in highly invasive non-small-cell (Fig. 7g, p < 0.05). Tese results suggested that miR-135b lung cancer cells, which ultimately enhances the cancer might afect the biological characteristics of PCSCs by cell migratory and invasive abilities in vitro [27]. Moreo- targeting JADE-1 to activate the AKT/mTOR pathway. ver, Arigoni et al. have denoted the upregulation of miR- 135b in basal or normal-like human breast cancers and miR‑135b activates AKT/mTOR pathway to elevate that it was strongly correlated with patient survival and the tumorigenicity of PCSCs in vivo early metastasis [28]. In gastric cancer cells, miR-135b To elucidate the efect of miR-135b activating the AKT/ has also been reported to accelerate cell proliferation mTOR pathway on the tumorigenicity of PCSCs in vivo, and promote other cellular function, including cell inva- tumor xenografts in nude mice were conducted with sion, colony formation, migration, and sphere formation PCSCs transfected with agomir-miR-135b + DMSO in gastric cancer cells [29], which was consistent with the and the agomir-miR-135b + perifosine. By recording the results of our study. growth of tumors in subcutaneous tumor-bearing mice, Moreover, here we also attempted to unravel the the growth trend of tumors in nude mice was statistically mechanism by which miR-135b activate the AKT/mTOR analyzed. Te tumor weight was measured by peeling of pathway to regulate the stem cells by inhibiting JADE-1 the tumor mass on the 30th day. Our results indicated expression, which subsequently afects the stemness of that compared with the agomir-miR-135b + perifosine PCSCs and promotes tumor growth. Our study identifed group, PCSCs grew rapidly in the subcutaneous part of JADE-1 as a novel gene specifcally regulated by miR-135b nude mice, whereas in the agomir-miR-135b + DMSO in PCSCs. As a transcription factor, JADE-1 represents a group, the tumor volume and weight exhibited signifcant component of the HBO1 complex characterized by acet- increase (Fig. 8a–c). Meanwhile, our results of immuno- ylation of histone protein [30]. Interestingly, JADE-1 has histochemistry documented that in contrast to the ago- been identifed as the frst member of the JADE protein mir-NC group, tumor of mice in the agomir-miR-135b family containing plant homology domain with the func- group had reduced JADE-1 positive expression rate and tion of suppressing proliferation and advancing apoptosis Zhou et al. Cancer Cell Int (2020) 20:134 Page 13 of 16

a b agomir-miR-135b + Perifosine c agomir-miR-135b + Perifosine agomir-miR-135b + DMSO agomir-miR-135b + DMSO 1500 1.5 * ) ) agomir-miR-135b + Perifosine 3

mm * 1.0 1000 * * e( *

um *

ol *

rv * 0.5 agomir-miR-135b + DMSO 500 * Tumor weight (g Tu mo

0 0.0 36912151821242730 Time (d)

d agomir-NC agomir-miR-135b antagomir-NC antagomir-miR-135b JADE-1 25 μm 25 μm 25 μm 25 μm 25 μm 25 μm agomir-miR-135b + DMSO

80 agomir-miR-135b + Perifosine * & &

) *

p-AK T 60 (% 25 μm 25 μm 25 μm 25 μm 25 μm 25 μm # rate 40 # # R * 20 Positive p-mTO 25 μm 25 μm 25 μm 25 μm 25 μm 25 μm

agomir-NC agomir-miR-135b antagomir-NC antagomir-miR-135b agomir-miR-135b + DMSO agomir-miR-135b + Perifosine 0 JADE-1 p-AKTp-mTOR Fig. 8 miR-135b activates AKT/mTOR pathway to elevate the tumorigenicity of PCSCs. a Macroscopic observation of tumor size. b Measurement of tumor volume in nude mice with vernier caliper. c Determination of tumor mass in nude mice by weighing method. d Immunohistochemistry was performed to detect positive expression of JADE-1, phosphorylated AKT and phosphorylated mTOR ( 400); n 6, *p < 0.05 vs. the × = agomir-miR-135b perifosine group; #p < 0.05 vs. the antagomir-NC group; &p < 0.05 vs. the agomir-miR-135b DMSO group. n 6. The + + = measurement data were expressed as mean standard deviation. Unpaired student’s t-test was used to compare unpaired data between the two ± groups. Repeated measurement ANOVA was used to compare the data among multiple groups at diferent time points, followed by the Bonferroni post hoc test

via the Wnt pathway [31, 32]. Te AKT is a well-known play a critical role in accelerating the proliferation and protein kinase, encoded by homologs of intracellular ret- survival of PC cells [37]. Peculiarly, miR-135b has been roviruses [33]. Besides, mTOR is widely understood to be illustrated as a promoter of cisplatin resistance in gastric an evolutionarily conservative nutrition-sensitive pro- cancer cells by activating the MAPK pathway [38]. More- tein kinase capable of adjusting the growth and metabo- over, miR-135b has also been reported to activate the lism of eukaryocytes [34]. With the indispensable role in PI3K/AKT pathway to promote chemoresistance of colo- cell metabolism and other critical functions, the AKT/ rectal cancer [39]. Considering these above-described mTOR pathway has been regarded as an important target fndings we may speculate that miR-135b may possess an for treating malignancies [35]. A previous study has sug- oncogenic role in PC to regulate these pathways, but fur- gested that the AKT pathway could be activated by up- ther studies are required to confrm these fndings. regulating miR-135b, thus suppressing the development Moreover, the present study has also indicated that of colorectal cancer cells [19]. Additionally, overexpres- miR-135b-mediated activation of the AKT/mTOR path- sion of miR-135b elevated the expression of phosphoryl- way via down-regulation of JADE-1, further enhancing ated PI3K, AKT, and mTOR in hypoxia-treated human the expression of the stemness-related genes, such as umbilical vein endothelial cells, thus activating the PI3K/ Sox2, Nanog, and Oct-4. However, scarce studies are AKT/mTOR pathway [36], which was concurred with determining the relationship between miR-135b and our results. Notably, decreased expression of JADE-1 (a the stemness-linked genes, except for certain miRNAs. major regulator of the AKT pathway) has been attrib- For example, miR-200c has been reported to negatively uted to the poor prognosis and activation of AKT in the regulate the expression of Sox-2 to suppress the AKT renal cancer cell, thus, increasing the cell invasion [15]. pathway [40]. Te expression of miR-148a can sup- Of note, the AKT/mTOR pathway has been indicated to press the Oct-4/Sox-2-stimulated stem-cell like tumor Zhou et al. Cancer Cell Int (2020) 20:134 Page 14 of 16

miR-135b

JJADE-1

JADE-1 Caspase Cascade

P P Akt Caspa Proliferation Self renewal Invasion and migration Survival Tumorigenesis P mTOROORR Expression

Oct-4 Sox-2 Slug Aldh-1 Nanog

Fig. 9 miR-135b-regulation of JADE-1 inhibited the growth of cancer stem cells and tumors in PC via the AKT/mTOR pathway. miR-135b could promote sphere-forming, cloning, invasion, migration, tumor formation of PCSCs in vivo. The mechanism may be performed through inhibiting JADE-1 and activating AKT/mTOR pathway, thus afecting the stemness of PCSCs and promoting tumor growth

cell proliferation in glioblastoma based on the suppres- Conclusion sion of stem cell markers [41]. Oct-4 has been reported Taken together, the critical fndings of our study sug- as a major transcription factor in regulating the self- gest that miR-135b negatively regulated the JADE-1 and renewal and versatility in embryonic stem cells [42]. controlled the stem cell activity in PC via the Akt/mTOR Importantly, Sox2 belongs to the SRY-related high pathway (Fig. 9). Notably, these fndings highlight the mobility group box (SOX) gene family that encodes potential of miR-135b as a promising therapeutic target transcription factors and possesses a single HMG for PC. However, due to the lack of comprehensive data DNA-binding feld [43]. Furthermore, removal of Sox2 about the correlation between miR-135b with JADE-1 in pancreatic ductal glandular cancer cells enhanced and the AKT/mTOR pathway in PC progression, further the inhibition activity of small-molecule aiming at AKT investigations are indispensable to reveal the underlying [44]. Furthermore, Nanog is viewed as one of the key mechanism by which miR-135b infuences the PCSCs. transcription factors in preserving self-renewal and the undiferentiated condition of pluripotent stem cells Supplementary information [45]. Tus, Sox2, Nanog, and Oct-4 commonly preserve Supplementary information accompanies this paper at https​://doi. the regulatory system and maintain the versatility and org/10.1186/s1293​5-020-01210​-1. self-renewal of stem cells [46]. Terefore, we concluded that inhibition of JADE-1 on the AKT/mTOR pathway Additional fle 1: Table S1. Specifc information of pancreatic cancer datasets. Table S2. LogFC and p values of top 20 diferentiated miRNA afects the stemness of PCSCs and suppresses tumor expression in pancreatic cancer in GSE41369 dataset. growth. Zhou et al. Cancer Cell Int (2020) 20:134 Page 15 of 16

Abbreviations 8. Khatri R, Subramanian S. MicroRNA-135b and its circuitry networks as miRNAs: MicroRNAs; PC: Pancreatic cancer; PCSCs: Pancreatic cancer stem potential therapeutic targets in colon cancer. Front Oncol. 2013;3:268. cells; GEO: Gene Expression Omnibus; DMSO: Dimethyl sulphoxide; WT: 9. Wang R, Xu B, Xu H. TGF-beta1 promoted chondrocyte proliferation by Wild-type; MUT: Mutant-type; ATCC​: American Type Culture Collection; regulating Sp1 through MSC-exosomes derived miR-135b. Cell Cycle. Aldh1: Acetaldehyde dehydrogenases; TBST: Tris-bufered saline tween; ECL: 2018;17(24):2756–65. Electrochemiluminescence. 10. Ozcan O, Kara M, Yumrutas O, Bozgeyik E, Bozgeyik I, Celik OI. MTUS1 and its targeting miRNAs in colorectal carcinoma: signifcant associations. Acknowledgements Tumour Biol. 2016;37(5):6637–45. We would like to acknowledge the helpful comments on this paper received 11. Lauvrak SU, Munthe E, Kresse SH, Stratford EW, Namlos HM, Meza-Zepeda from our reviewers. LA, Myklebost O. Functional characterisation of osteosarcoma cell lines and identifcation of mRNAs and miRNAs associated with aggressive Authors’ contributions cancer phenotypes. Br J Cancer. 2013;109(8):2228–36. JZ and HW designed the study. JC and LX collated the data, carried out data 12. Munding JB, Adai AT, Maghnouj A, Urbanik A, Zollner H, Lifers ST, analyses. WY and YL produced the initial draft of the manuscript. WZh and JZ Chromik AM, Uhl W, Szafranska-Schwarzbach AE, Tannapfel A, Hahn SA. contributed to drafting the manuscript. All authors read and approved the Global microRNA expression profling of microdissected tissues identifes fnal manuscript. miR-135b as a novel biomarker for pancreatic ductal adenocarcinoma. Int J Cancer. 2012;131(2):E86–95. Funding 13. Jiang W, Zhao S, Shen J, Guo L, Sun Y, Zhu Y, Ma Z, Zhang X, Hu Y, Xiao This study was supported by The plan to the introduction of clinical medical W, Li K, Li S, Zhou L, Huang L, Lu Z, Feng Y, Xiao J, Zhang EE, Yang L, Wan teams of Xuzhou city (2018TD011). R. The MiR-135b-BMAL1-YY1 loop disturbs pancreatic clockwork to pro- mote tumourigenesis and chemoresistance. Cell Death Dis. 2018;9(2):149. Availability of data and materials 14. Xiao-Fen W, Ting C, Jie L, Deng-Yang M, Qing-Feng Z, Xin L. Correlation The datasets generated/analyzed during the current study are available. analysis of VHL and Jade-1 gene expression in human renal cell carci- noma. Open Med. 2016;11(1):226–30. Ethics approval and consent to participate 15. Zeng L, Bai M, Mittal AK, El-Jouni W, Zhou J, Cohen DM, Zhou MI, Cohen All the animal study was carried out by following the Institutional Animal Eth- HT. Candidate tumor suppressor and pVHL partner Jade-1 binds and ics Care and Use Committee of Xuzhou City Cancer Hospital. Extensive eforts inhibits AKT in renal cell carcinoma. Cancer Res. 2013;73(17):5371–80. were made to minimize the pain and sufering of the experimental animal. 16. Lunardi A, Webster KA, Papa A, Padmani B, Clohessy JG, Clohessy JG, Bronson RT, Pandolf PP. Role of aberrant PI3K pathway activation in Consent for publication gallbladder tumorigenesis. Oncotarget. 2014;5(4):894–900. Not applicable. 17. Suo L, Chang X, Zhao Y. The orexin-A-regulated Akt/mTOR pathway pro- motes cell proliferation through inhibiting apoptosis in pancreatic cancer Competing interests cells. Front Endocrinol. 2018;9:647. The authors declare that they have no competing interests. 18. Gao S, Zhao Z, Wu R, Wu L, Tian X, Zhang Z. MiR-1 inhibits prostate cancer PC3 cells proliferation through the Akt/mTOR signaling pathway by bind- Author details ing to c-Met. Biomed Pharmacother. 2019;109:1406–10. 1 Class 182, Queen Mary School, Medical Department, Nanchang University, 19. Guo J, Hao J, Jiang H, Jin J, Wu H, Jin Z, Li Z. Proteasome activator subunit Nanchang 330031, People’s Republic of China. 2 Department of Hepatopan- 3 promotes pancreatic cancer growth via c-Myc-glycolysis signaling axis. creatobillary Surgery, Xuzhou City Cancer Hospital, No. 131 Huancheng Rd., Cancer Lett. 2017;386:161–7. Gulou District, Xuzhou 221000, Jiangsu, People’s Republic of China. 20. Fu J, Rodova M, Roy SK, Sharma J, Singh KP, Srivastava RK, Shankar S. GANT-61 inhibits pancreatic cancer stem cell growth in vitro and in NOD/ Received: 7 November 2019 Accepted: 10 April 2020 SCID/IL2R gamma null mice xenograft. Cancer Lett. 2013;330(1):22–32. 21. Shankar S, Nall D, Tang SN, Meeker D, Passarini J, Sharma J, Srivastava RK. Resveratrol inhibits pancreatic cancer stem cell characteristics in human and KrasG12D transgenic mice by inhibiting pluripotency maintaining factors and epithelial-mesenchymal transition. PLoS ONE. 2011;6(1):e16530. References 22. Huang M, Tang SN, Upadhyay G, Marsh JL, Jackman CP, Shankar S, Srivas- 1. Yang Y, Sun Y, Wang H, Li H, Zhang M, Zhou L, Meng X, Wu Y, Liu P, Liu X, tava RK. Embelin suppresses growth of human pancreatic cancer xeno- Zhang J, Tan X. MicroRNA-221 induces autophagy through suppressing grafts, and pancreatic cancer cells isolated from KrasG12D mice by inhib- HDAC6 expression and promoting apoptosis in pancreatic cancer. Oncol iting Akt and Sonic hedgehog pathways. PLoS ONE. 2014;9(4):e92161. Lett. 2018;16(6):7295–301. 23. Sharma N, Nanta R, Sharma J, Gunewardena S, Singh KP, Shankar S, 2. Zhang Z, Han H, Rong Y, Zhu K, Zhu Z, Tang Z, Xiong C, Tao J. Hypoxia Srivastava RK. PI3K/AKT/mTOR and sonic hedgehog pathways cooperate potentiates gemcitabine-induced stemness in pancreatic cancer cells together to inhibit human pancreatic cancer stem cell characteristics and through AKT/Notch1 signaling. J Exp Clin Cancer Res. 2018;37(1):291. tumor growth. Oncotarget. 2015;6(31):32039–60. 3. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global 24. Xu Z, Zhang Y, Jiang J, Yang Y, Shi R, Hao B, Zhang Z, Huang Z, Kim JW, cancer statistics 2018: gLOBOCAN estimates of incidence and mor- Zhang G. Epidermal growth factor induces HCCR expression via PI3K/ tality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. Akt/mTOR signaling in PANC-1 pancreatic cancer cells. BMC Cancer. 2018;68(6):394–424. 2010;10:161. 4. Ryan DP, Hong TS, Bardeesy N. Pancreatic adenocarcinoma. N Engl J Med. 25. Chen W, Zhou Y, Zhi X, Ma T, Liu H, Chen BW, Zheng X, Xie S, Zhao B, Feng 2014;371(11):1039–49. X, Dang X, Liang T. Delivery of miR-212 by chimeric peptide-condensed 5. Perusina Lanfranca M, Thompson JK, Bednar F, Halbrook C, Lyssiotis C, supramolecular nanoparticles enhances the sensitivity of pancreatic Levi B, Frankel TL. Metabolism and epigenetics of pancreatic cancer stem ductal adenocarcinoma to doxorubicin. Biomaterials. 2019;192:590–600. cells. Semin Cancer Biol. 2019;57:19–26. 26. Qadir MI, Faheem A. miRNA: a diagnostic and therapeutic tool for pancre- 6. Herreros-Villanueva M, Bujanda L, Billadeau DD, Zhang JS. Embry- atic cancer. Crit Rev Eukaryot Gene Expr. 2017;27(3):197–204. onic stem cell factors and pancreatic cancer. World J Gastroenterol. 27. Lin CW, Chang YL, Chang YC, Lin JC, Chen CC, Pan SH, Wu CT, Chen HY, 2014;20(9):2247–54. Yang SC, Hong TM, Yang PC. MicroRNA-135b promotes lung cancer 7. Zhou P, Li B, Liu F, Zhang M, Wang Q, Liu Y, Yao Y, Li D. The epithelial to metastasis by regulating multiple targets in the Hippo pathway and mesenchymal transition (EMT) and cancer stem cells: implication for LZTS1. Nat Commun. 2013;4:1877. treatment resistance in pancreatic cancer. Mol Cancer. 2017;16(1):52. 28. Arigoni M, Barutello G, Riccardo F, Ercole E, Cantarella D, Orso F, Conti L, Lanzardo S, Taverna D, Merighi I, Calogero RA, Cavallo F, Quaglino Zhou et al. Cancer Cell Int (2020) 20:134 Page 16 of 16

E. miR-135b coordinates progression of ErbB2-driven mammary 39. Liu B, Liu Y, Zhao L, Pan Y, Shan Y, Li Y, Jia L. Upregulation of microRNA- carcinomas through suppression of MID1 and MTCH2. Am J Pathol. 135b and microRNA-182 promotes chemoresistance of colorectal 2013;182(6):2058–70. cancer by targeting ST6GALNAC2 via PI3K/AKT pathway. Mol Carcinog. 29. Han TS, Voon DC, Oshima H, Nakayama M, Echizen K, Sakai E, Yong 2017;56(12):2669–80. ZWE, Murakami K, Yu L, Minamoto T, Ock CY, Jenkins BJ, Kim SJ, Yang 40. Lu YX, Yuan L, Xue XL, Zhou M, Liu Y, Zhang C, Li JP, Zheng L, Hong M, Li HK, Oshima M. Interleukin 1 up-regulates MicroRNA 135b to promote XN. Regulation of colorectal carcinoma stemness, growth, and metastasis infammation-associated gastric carcinogenesis in mice. Gastroenterol- by an miR-200c-Sox2-negative feedback loop mechanism. Clin Cancer ogy. 2019;156(4):1140–55. Res. 2014;20(10):2631–42. 30. Panchenko MV, Zhou MI, Cohen HT. von Hippel-Lindau partner Jade-1 is 41. Lopez-Bertoni H, Lal B, Li A, Caplan M, Guerrero-Cazares H, Eberhart CG, a transcriptional co-activator associated with histone acetyltransferase Quinones-Hinojosa A, Glas M, Schefer B, Laterra J, Li Y. DNMT-dependent activity. J Biol Chem. 2004;279(53):56032–41. suppression of microRNA regulates the induction of GBM tumor-propa- 31. Foy RL, Chitalia VC, Panchenko MV, Zeng L, Lopez D, Lee JW, Rana SV, gating phenotype by Oct4 and Sox2. Oncogene. 2015;34(30):3994–4004. Boletta A, Qian F, Tsiokas L, Piontek KB, Germino GG, Zhou MI, Cohen HT. 42. Chang JH, Au HK, Lee WC, Chi CC, Ling TY, Wang LM, Kao SH, Huang Polycystin-1 regulates the stability and ubiquitination of transcription YH, Tzeng CR. Expression of the pluripotent transcription factor OCT4 factor Jade-1. Hum Mol Genet. 2012;21(26):5456–71. promotes cell migration in endometriosis. Fertil Steril. 2013;99(5):1332–9. 32. Borgal L, Habbig S, Hatzold J, Liebau MC, Dafnger C, Sacarea I, Ham- 43. Song Y, Xiao L, Fu J, Huang W, Wang Q, Zhang X, Yang S. Increased merschmidt M, Benzing T, Schermer B. The ciliary protein nephro- expression of the pluripotency markers sex-determining region Y-box 2 cystin-4 translocates the canonical Wnt regulator Jade-1 to the and Nanog homeobox in ovarian endometriosis. Reprod Biol Endocrinol. nucleus to negatively regulate beta-catenin signaling. J Biol Chem. 2014;12:42. 2012;287(30):25370–80. 44. Chen J, Tian W, He H, Chen F, Huang J, Wang X, Chen Z. Downregula- 33. Xu W, Yang Z, Lu N. A new role for the PI3K/Akt signaling pathway in the tion of miR200c3p contributes to the resistance of breast cancer cells to epithelial-mesenchymal transition. Cell Adhes Migr. 2015;9(4):317–24. paclitaxel by targeting SOX2. Oncol Rep. 2018;40(6):3821–9. 34. Kapahi P, Chen D, Rogers AN, Katewa SD, Li PW, Thomas EL, Kockel L. 45. Siu MK, Wong ES, Kong DS, Chan HY, Jiang L, Wong OG, Lam EW, Chan With TOR, less is more: a key role for the conserved nutrient-sensing TOR KK, Ngan HY, Le XF, Cheung AN. Stem cell transcription factor NANOG pathway in aging. Cell Metab. 2010;11(6):453–65. controls cell migration and invasion via dysregulation of E-cadherin and 35. Beauchamp EM, Platanias LC. The evolution of the TOR pathway and its FoxJ1 and contributes to adverse clinical outcome in ovarian cancers. role in cancer. Oncogene. 2013;32(34):3923–32. Oncogene. 2013;32(30):3500–9. 36. Wei S, Wang H. Ligustrazine promoted hypoxia-treated cell growth by 46. Xiao L, Song Y, Huang W, Yang S, Fu J, Feng X, Zhou M. Expression of upregulation of miR-135b in human umbilical vein endothelial cells. Exp SOX2, NANOG and OCT4 in a mouse model of lipopolysaccharide- Mol Pathol. 2019;106:102–8. induced acute uterine injury and intrauterine adhesions. Reprod Biol 37. Rozengurt E. Mechanistic target of rapamycin (mTOR): a point of con- Endocrinol. 2017;15(1):14. vergence in the action of insulin/IGF-1 and G protein-coupled receptor agonists in pancreatic cancer cells. Front Physiol. 2014;5:357. 38. Zhou J, Chen Q. Poor expression of microRNA-135b results in the inhibi- Publisher’s Note tion of cisplatin resistance and proliferation and induces the apoptosis Springer Nature remains neutral with regard to jurisdictional claims in pub- of gastric cancer cells through MST1-mediated MAPK signaling pathway. lished maps and institutional afliations. FASEB J. 2019;33(3):3420–36.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions