The T-ALL cell lines Jurkat, CEM and MOLT4 were cultured in RPMI1640 (Gibco, NY, USA) with 10% fetal bovine serum (FBS). They were maintained at 37°C in a humidified incubator with 5% CO2.

The peripheral blood were collected from healthy human and red blood cell lysis was conducted with lysis buffer (Solarbio, Beijing, China). The CD45⁺ T cells were isolated with magnetic beads (Stemcell, Vancouver, Canada) according to the manufacturer’s instructions.

Total RNA of T-ALL cells was extracted with TRIzol Reagent (Thermo Fisher Scientific, MA, USA) and 1ug total RNA was reverse-transcribed into complementary DNA(cDNA) using the RevertAid First Strand cDNA Synthesis Kit (Thermo). Quantitative real-time PCR was performed to determine the RNA expression using SYBR Green master mix (DBI^®Bioscience, Ludwigshafen, German) with specific primers listed below. All of the reactions were run in triplicate and the relative levels of lncRNA were normalized to 18S rRNA. The sequences of the primers were as follows: hGAPDH Forward: 5’- CTTTTGCGTCGCCAGCCGAG -3’; hGAPDH Reverse: 5’- CCAGGCGCCCAATACGACCA -3’; PPM1A-AS Forward: 5’- AGTCCTGGACAGTCTTTAGGC -3’; PPM1A-AS Reverse: 5’- AGGTGTGTGCTGGGAAATGT -3’.

Cytoplasmic and nuclear RNA were isolated and purified using the PARIS™ Kit (ThermoFisher, #AM1921) according to the manufacturer’s instructions.

For knockdown assay, oligos encoding shRNA specific for PPM1A-AS were ligated into pSUPER.retro.puro, and the fragment containing the H1 promoter and hairpin sequences was subcloned into the lentiviral shuttle pCCL.PPT.hPGK.GFP.Wpre. The shRNA target sequences were as follows: shPPM1A-AS-1, GCATCAAGAAGAACAGCTA; shPPM1A-AS-2, GGTTGATCTGTGCGGCAAA. For overexpression assay, lncRNA PPM1A-AS sequence was ligated into the lentiviral shuttle pCCL.PPT.hPGK.IRES.GFP/pre. These plasmids were used to produce lentivirus in HEK293T cells with the packaging plasmids pMD2.BSBG, pMDLg/pRRE and pRSV-REV. Cells were infected with lentivirus and sorted by GFP signal to generate a stable cell line.

To assess cell proliferation ability, Jurkat, CEM and MOTL4 cells infected with control lentivirus, lncRNA PPM1A-AS-knockdown lentivirus or lncRNA PPM1A-AS-overexprssion lentivirus were seeded into 96-well plates at a density of 2000 cells per well. At 0, 24, 48, 72, 96, 120 hours after the cells were seeded, CCK-8 reagent (Dojingdo, Japan) was mixed with the cells for 1h incubation at 37°C. The absorbance value was measured at 450nm with Microplate reader.

The EdU assay was conducted with Cell-Light EdU Apollo567 In Vitro Kit (RiboBio Co., Guangzhou, China) according to the manufacturer’s instructions of suspension cell. Briefly, Jurkat, CEM and MOTL4 cells were infected with control, lncRNA PPM1A-AS-knockdown or lncRNA PPM1A-AS-overexprssion lentivirus and then 5000 cells of each group were planted in 24-well plates. 24 hours later, the cells were remarked with EdU and made smears. After cell fixation and Apollo staining, the slides were observed and took photos under a microscope at 100×magnification.

Jurkat and CEM cells infected with control lentivirus, lncRNA PPM1A-AS-knockdown lentivirus or lncRNA PPM1A-AS-overexprssion lentivirus were collected and incubated with annexin V and PIfor 15min. The apoptotic cells were detected by BD flow cytometer. The annexin positive but PI negative staining indicated the early apoptotic cells. The annexin V and PI both positive staining indicated cells in necrosis (post-apoptotic necrosis or late apoptosis). The proportion of total apoptotic cells are the sum of these two parts of cells.

The control or infected cells were rinsed with PBS and lysed in RIPA Lysisbuffer (Beyotime, China) supplemented with Protease and Phosphatase Inhibitor (Cell Signaling Technology Inc., USA) on ice for 30min. The cell lysates were centrifuged for 10min (12000 g, 4°C) and the supernatant was collected. The protein concentration was calculated with Pierce BCA protein assay kit (Thermo) and equivalent quantities of protein were separated on 10% SDS-PAGE gels. Then the proteins were transferred onto a nitrocellulose membrane (Bio-Rad, CA, USA). After blocking with 5% non-fat milk at room temperature for 1h, the membranes were immunostained with primary antibodies at 4°C overnight, washed three times in TBST, and then incubated with secondary antibody at room temperature for 1h. Finally, the protein bands on membranes were detected with an enhanced chemiluminescence reagent (Millipore) and captured using a luminescence instrument (Tanon, Shanghai, China). The gray density of protein bands was determined by Image J software. The primary antibodies were listed below: Akt (#9272, Cell Signaling Technology); T308-pAkt (#13038, Cell Signaling Technology), S473-pAkt (#4060, Cell Signaling Technology); STAT-3(#9132, Cell Signaling Technology), T705-pSTAT3 (#9131, Cell Signaling Technology), S727-pSTAT3 (#9134, Cell Signaling Technology); Notch4 (ab184742, Abcam).

Female NOD-SCID mice (4-6 weeks old) were purchased from the Model Animal Research Center of Nanjing University (Nanjing, China) and housed under pathogen-free conditions in Tianjin Medical University. Jurkat cells were infected with control lentivirus or lncRNA PPM1A-AS-knockdown lentivirus (2×10⁶), suspended in 200ul PBS and then injected into the mice by tail intravenous injection. After 45 days later, the peripheral blood was taken from mice, the mice were sacrificed, the spleens were photographed and weighed and the bone marrow were dissected. The animal study was reviewed and approved by Tianjin Medical University Animal Care and Use Committee.

RNA was isolated from Jurkat cells injected with control lentivirus or lncRNAPPM1A-AS-knockdownlentivirus using Trizol (Invitrogen) following manufacture instructions. Biological samples in duplicate were submitted to Novogene Co., Ltd for RNA sequencing. Barcoded sequence libraries were constructed using TruSeq RNA Sample Prep kit (Illumina), and sequenced on a HiSeq 2000 instrument.

Data analyses were undertaken with GraphPad PRISM 8.0. The results were shown as mean ± standard deviation (SD), from three independent experiments and analyzed via the Student’s t-test. The value p <0.05 was considered statistically significant.

To identify lncRNAs differentially expressed in T-ALL, we collected and analyzed RNA sequencing (RNA-seq) data of 10 T-ALL patients and 2 human whole thymus extracts from public database (GSE110636 (13), GSE57982 (10), the sample numbers were listed in Supplementary Table 1 ). The 10 T-ALL patients were belonging to different genetic subgroups. The heatmap was showed in Figure 1A . Among these lncRNAs, we found that CTD-2184C24.2 (ENST00000553775) was obviously increased in T-ALL patients ( Figure 1B ). CTD-2184C24.2 is a transcript antisense to PPM1A. So, we named this lncRNA as PPM1A-AS after this gene. Moreover, we tested the expression level of PPM1A-AS in T-ALL cell lines Jurkat, CCRF-CEM and MOLT4. The T cells extracted from healthy human blood with CD45⁺ magnetic beads were used as control. As a result, PPM1A-AS was significantly overexpressed in all three T-ALL cell lines than T cells extracted from healthy persons ( Figure 1C ). Finally, we verified the distribution of PPM1A-AS in T cells. No matter in normal or cancerous T cells, PPM1A-AS was existing in both nucleus and cytoplasm, mainly in nucleus ( Figure 1D ). These results also indicated that the PPM1A-AS had no shuttle between nucleus and cytoplasm during T-ALL formation.

In order to explore the function of PPM1A-AS, stable clones with knockdown and overexpression of PPM1A-AS were generated with shRNA and plasmid via lentivirus technology in T-ALL cell lines, Jurkat, CEM and MOTL4. Firstly, we evaluated the efficiency of PPM1A-AS knockdown ( Figure 2A and Supplementary Figure 1A ) and overexpression by RT-qPCR ( Figure 3A and Supplementary Figure 1D ). The PPM1A-AS RNA level was about 25% lower by two different shRNAs and hundreds higher by overexpression vector. Then we performed CCK-8 and EdU assays to test the cell proliferation ability. Compared to the control group, cells infected with PPM1A-AS shRNA lentivirus showed significantly slower proliferation rates ( Figures 2B, C and Supplementary Figures 1B, C ) while overexpression of PPM1A-AS enhanced cell proliferation ( Figures 3B, C and Supplementary Figures 1E, F ). Besides, we examined cell apoptosis by flow cytometry. We found that knockdown of PPM1A-AS could induce T-ALL cell death ( Figure 2D ). Taken together, we conclude that lncRNA PPM1A-AS may have the potential to serve as an oncogene in T-ALL.

To further determine the role of lncRNA PPM1A-AS in T-ALL, we established tumor xenograft model in NOD-SCID mice. We cultured Jurkat cells infected with blank lentivirus or lncRNA PPM1A-AS-konckdown lentivirus and transplanted mice by tail intravenous injection. About 45 days later, we sacrificed the mice, photographed and measured the spleens, detected the proportion of tumor cells in peripheral blood and bone marrow. As the pictures showed in Figure 4A , the spleens from mice of the control group were much enlarged than the PPM1A-AS knockdown group. The statistical data were in Figure 4B and the spleens’ average weight of control mice was obviously larger. Moreover, the proportion of tumor cells in peripheral blood ( Figure 4C ) or bone marrow ( Figure 4D ) from control mice were much more than the mice of PPM1A-AS knockdown group, which means PPM1A-AS knockdown can repress T-ALL tumor development in vivo.

To further investigate the mechanism by which PPM1A-AS promotes T-ALL progression, we performed RNA-seq in Jurkat cells stably transduced with shRNAs targeting PPM1A-AS or a non-targeting control. Two biological replicates were included in each group and high correlations were observed between the replicates ( Figure 5A ). We then performed differential gene expression analysis, and identified 288 and 313 genes significantly upregulated and downregulated after knockdown of PPM1A-AS ( Figure 5B and Supplementary Table 2 ). The major signal pathways that the differently expressed genes involved in were analyzed by KEGG database ( Figure 5C ). Among these pathways, Notch signaling pathway and PI3K-Akt signaling pathway are the major pathways participated in T-ALL tumorigenesis and development (14). We then tested if PPM1A-AS has any role in the well-known oncogenic pathways involved in T-ALL. AKT, NOTCH and STAT are core genes of PI3K-Akt signaling pathway, Notch signaling pathway and JAK-STAT3 signaling pathway respectively (15). We chose proteins of these genes and examined their expression after PPM1A-AS knockdown. The whole protein levels of STAT3 and Akt were not changed but the phosphorylated protein (T308, S473 of Akt and T705, S727 of STAT3) were significantly downregulated by PPM1A-AS knockdown ( Figure 5D ). Besides, the Notch4 ( Figure 5D ) but not Notch1 (data were not showed) protein level was also consistent with PPM1A-AS RNA level. Therefore, these data showed that PPM1A-AS could regulate multiple proteins and thus have roles in several T-ALL related pathways.