ELSEVIER
Contents lists available at ScienceDirect
Computational and Structural Biotechnology Journal
journal homepage: www.elsevier.com/locate/csbj
COMPUTATIONAL ANDSTRUCTURAL BIOTECHNOLOGY JOURNAL
01010100100101
0010101
1010101011 001
0101001 01010
101010101110
101001 01010
010101001001
0101001 10101
010101001011 1010101011 10
0101001 10101
0101001 01010
101010101100
0101001 51010
010101001001010101001 10101
Research article
Identification of the H3K36me3 reader LEDGF/p75 in the pancancer landscape and functional exploration in clear cell renal cell carcinoma
Check for updates
Yuwei Zhang a,b,1, Wei Guo b,1, Yangkun Feng ª,1, Longfei Yang a,b, Hao Lin b, Pengcheng Zhou ®, Kejie Zhao , Lin Jiang ”,”, Bing Yao ”, "", Ninghan Feng a, b, c,
a Nantong University Medical School, Nantong, China
b Department of Urology, Jiangnan University Medical Center, Wuxi, China
” Wuxi School of Medicine, Jiangnan University, Wuxi, China
d Department of Endocrinology and Metabolism, First Affiliated Hospital of Nanjing Medical University, Nanjing, China
e Department of Medical Genetics, Nanjing Medical University, Nanjing, China
ARTICLE INFO
Keywords:
LEDGF/p75 H3K36me3 Pancancer
SETD2 Clear cell renal cell carcinoma
ABSTRACT
Lens epithelium-derived growth factor (LEDGF/p75) is a reader of epigenetic marks and a potential target for therapeutic intervention. Its involvement in human immunodeficiency virus (HIV) integration and the devel- opment of leukemia driven by MLL (also known as KMT2A) gene fusion make it an attractive candidate for drug development. However, exploration of LEDGF/p75 as an epigenetic reader of H3K36me3 in tumors is limited. Here, for the first time, we analyze the role of LEDGF/p75 in multiple cancers via multiple online databases and in vitro experiments. We used pancancer bulk sequencing data and online tools to analyze correlations of LEDGE/ p75 with prognosis, genomic instability, DNA damage repair, prognostic alternative splicing, protein in- teractions, and tumor immunity. In summary, the present study identified that LEDGF/p75 may serve as a prognostic predictor for tumors such as adrenocortical carcinoma, kidney chromophobe, liver hepatocellular carcinoma, pancreatic adenocarcinoma, skin cutaneous melanoma, and clear cell renal cell carcinoma (ccRCC). In addition, in vitro experiments and gene microarray sequencing were performed to explore the function of LEDGF/p75 in ccRCC, providing new insights into the pathogenesis of the nonmutated SETD2 ccRCC subtype.
1. Introduction
LEDGF/p75, encoded by PSIP1, was originally identified as a protein copurifying with the general transcriptional coactivator PC4 and described as a transcriptional coactivator related to stress and autoim- mune responses[1]. PSIP1 also codes for an alternative splicing isoform referred to as p52. Compared with p52, LEDGF/p75 has an integrase binding domain (IBD) in addition to the common PWWP domain[2]. LEDGF/p75 has been reported to play a key role in the development of human immunodeficiency virus (HIV) and MLL leukemia[3-5]. HIV integrase can recognize and bind to the IBD of LEDGF/p75 and hijack it to the transcriptionally active region of the genome, allowing the virus to replicate in large numbers[6,7]. Similarly, MLL/MENIN complex, an
important player in the development and progression of MLL leukemia, can bind to the IBD of LEDGF/p75 to promote development and pro- gression of the disease[8,9]. In fact, as a chromatin-binding protein, LEDGF/p75 mediates chromatin localization of several nuclear proteins [10-13].
Posttranslational modification of histones is an important branch of epigenetic inheritance that has been widely reported in various diseases, especially in cancer. H3K36me3 is a 3-methylated modification at the 36th K of histone H3 and mediates several key tumor processes, such as transcriptional elongation, DNA methylation, and DNA damage repair [14]. Studies have reported that LEDGF/p75 reads the H3K36me3 mark via its PWWP domain to recruit functional proteins to the region of actively transcribed genes[15,16]. However, its role in tumors is poorly
* Correspondence to: Department of Endocrinology and Metabolism, First Affiliated Hospital of Nanjing Medical University, 300 Guangzhou Road, Nanjing 210029, China.
** Correspondence to: Department of Medical Genetics, Nanjing Medical University, 101 Longmian Road, Nanjing 211166, China.
*** Correspondence to: Nantong University Medical School, 9 Qiangyuan Road, Nantong 226019, China. E-mail addresses: jlinna0000@163.com (L. Jiang), byao@njmu.edu.cn (B. Yao), n.feng@njmu.edu.cn (N. Feng).
1 These authors contributed equally to this work.
https://doi.org/10.1016/j.csbj.2023.08.023
understood. Therefore, the present study was performed to characterize the landscape of LEDGF/p75 across cancers for the first time.
Renal cell carcinoma includes more than 10 histological and mo- lecular subtypes, of which clear cell renal cell carcinoma (ccRCC) is the most common and accounts for the majority of deaths associated with kidney cancer[17]. Genetically, ccRCC results from high-frequency mutations or even deletion of multiple tumor-suppressor genes (VHL, 80%; PBRM1, 29-46%; BAP1, 6-19%; and SETD2, 8-30%), which leads to genomic instability and promotes defects in DNA repair pathways [18]. ccRCC can be classified into clinically and therapeutically relevant subtypes based on the molecular characteristics caused by these defects [19]. SETD2 is an RNA polymerase II-associated histone methyl- transferase that catalyzes H3K36me3, which is a transcriptional activity marker. Previous studies indicated that H3K36me3 is only added by SETD2. Although SMYD5 was recently reported to play a role in methylation, SETD2 is the most dominant specific methylase [20].
Research on the presence or absence of SETD2 is of key clinical significance for personalized treatment. Therefore, the present study was performed as preliminary functional exploration of LEDGF/p75, the main reader of H3K36me3, in ccRCC.
2. Materials and Methods
2.1. Acquisition of basic information about LEDGF/p75
The genomic view for the LEDGF/p75 gene was obtained from the GeneCards database[21]. The features for the domains and regions of LEDGF/p75 were obtained from the UniProt database[22]. A three-dimensional structure for LEDGF/p75 was constructed from AlphaFold[23]. The immunofluorescence graphs of the intracellular location of LEDGF/p75 in U-251MG cells (HPA019697) were obtained from the ATLAS database[24].
2.2. Interaction network of LEDGF/P75 and functional enrichment analyses
The LEDGF/p75 protein-protein interaction network with physical interactions was predicted via the GeneMANIA database[25], and the potential pathways are marked in colors. The LEDGF/p75 protein — protein interaction network with known experimental validations was also explored via the STRING database[26]. We further predicted scores in cancers and other diseases based on the LEDGF/p75 protein-protein interaction network via the canSAR database[27].
Gene Ontology (GO) analyses, including biological process, cellular component analyses and molecular function analyses, along with reac- tome pathways were predicted via the TISIDB database[28]. Gene set enrichment analysis (GSEA) was carried out via R software with ca- nonical pathway gene sets derived from the KEGG pathway database [29]. All R programs used in the present study were uploaded to GitHub (https://github.com/melondoctor/LEDGF/tree/master).
2.3. Analyses of LEDGF/p75’s correlation with histone modification, DNA mismatch repair, tumor environment, immune cell infiltration, and immune-related genes
We obtained pancancer expression profiles from the UCSC Xena database[30]. Associations between 5 histone modification genes, 5 DNA mismatch repair genes, tumor environment, immune cell infiltra- tion, immune-related genes and LEDGF/p75 expression in pancancer were visualized using R software.
2.4. Differential expression of LEDGF/p75 in normal and tumor groups
We processed expression data from the UCSC Xena database, deleted data with less than three normal samples, and analyzed the remaining data for 21 tumor types via R software. For the remaining 12 tumor
types, we added data from GTEx through GEPIA[31] for further analysis and identified differences in LEDGF/p75 expression between three tumor groups and corresponding normal groups. We further explored differential expression of LEDGF/p75 in different kinds of cells and obtained results from the ATLAS database.
For the LEDGF/p75 protein expression of tumor and normal groups, we searched the UALCAN database[32] and found 7 tumor types with different expression. Subsequently, we searched the ATLAS database and found corresponding immunohistochemical diagrams to show LEDGF/p75 protein expression.
2.5. Analyses between LEDGF/p75 expression and patient prognosis
Data on expression and survival in pancancer were obtained from the UCSC Xena database and analyzed via R software. Overall survival (OS), disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI) were analyzed to indicate patient prog- nosis. Cox regression analysis and Kaplan-Meier (K-M) analysis were used for forest plots and K-M plots, respectively. In addition, we analyzed LEDGF/p75 expression and clinical stages of patients across cancers.
2.6. Genomic alterations of LEDGF/p75 across cancers
We searched the cBioPortal database[33] for information about genomic alterations in LEDGF/p75 across cancers. We first identified the landscape of alteration frequency across cancers, including mutations, structural variants, amplifications, deep deletions and multiple alter- ations. Then, we searched the ratio of the alteration group in pancancer. We further obtained detailed information on copy-number alterations and mutations in LEDGF/p75. Finally, we analyzed the correlation be- tween genomic alterations of LEDGF/p75 and patient survival.
We processed the mutation data of pancancer from the UCSC Xena database and analyzed them via R software to identify information about the tumor mutational burden (TMB) and microsatellite instability (MSI).
2.7. Clinically relevant alternative splicing analyses of LEDGF/p75
To identify clinically relevant alternative splicing (AS) events, the OncoSplicing database[34] was searched for AS events for LEDGF/p75. We chose project 247053 for subsequent analyses, which was the only known splice type in SplAdder methodology according to the OncoS- plicing database. Pan plots indicate the reads in, reads out and percent spliced-in (PSI) values in pancancer and normal tissues. PanDiff plots compared the PSI differences of queried AS events (detected in more than 3 cancers) between cancers and adjacent or GTEx normal tissues. Finally, we explored the prognostic significance of LEDGF/p75 AS events across cancers via K-M plots.
2.8. Exploration of the immunological roles of LEDGF/p75 across cancers
We first explored the association between LEDGF/p75 expression and immune subtypes across cancers via the TISIDB database. Subse- quently, we analyzed detailed subtypes, including C1 (wound healing), C2 (IFN-gamma dominant), C3 (inflammatory), C4 (lymphocyte depleted), C5 (immunologically quiet), and C6 (TGF-b dominant). We visualized the distribution across subtypes for the top six tumor types with the highest LEDGF/p75 expression. Heatmaps showed Spearman correlations between immunoinhibitors, chemokines, tumor infiltrating lymphocytes (TILs), and LEDGF/p75 expression across cancers.
We further searched the most confident results using gene expression data in the TIDE database[35], including parts of cancer, subtype, Pearson correlation with cytotoxic T lymphocyte level (CTL Cor), T-cell dysfunction score (T Dysfunction), survival risk score (Risk), survival risk score adjusted for the effect of cytotoxic T lymphocyte (Risk. adj),
and sample count in the dataset (count).
We then compared LEDGF/p75 expression levels across cell lines between pre- and postcytokine-treated samples with the TISMO data- base[36]; IFNy, IFNß, TNFa, and TGFb1 are included as cytokine treat- ments in the module.
Finally, we searched ROC Plotter[37] to explore the correlation be- tween LEDGF/p75 expression and immunotherapy. Receiver operating characteristic (ROC) curves indicated the high diagnostic value of LEDGF/p75 for assessing immunotherapy outcomes.
2.9. Cell culture
Human kidney cancer cell lines (i.e., Caki-1, 786-O, A498) and normal human kidney epithelial cells (HK-2) were acquired from Procell Life Science & Technology Company (Wuhan, China). All cell lines used in this study were tested and authenticated by DNA sequencing using the STR method (ABI 3730XL Genetic Analyzer) and tested for the absence of mycoplasma contamination (MycoAlert). All cell lines were cultured in commercial cell culture medium at 37 ℃ in a 5% CO2 atmosphere.
2.10. RNA isolation and quantitative real-time PCR (qRT-PCR)
TRIzol Reagent (Invitrogen, USA) was used to isolate total RNA. HiScript III SuperMix (Vazyme, China) was used to perform reverse transcription. qRT-PCR was used to analyze the expression level of mRNAs, as performed using a SYBR Green Kit (Yeasen, China) with the LightCycler® 96 SW 1.1 system (Roche, Switzerland).
2.11. Western blotting
RIPA buffer (Beyotime, China) mixed with protease inhibitor (Beyotime) was used to extract total cell protein. The proteins were separated and then transferred to a polyvinylidene difluoride mem- brane, which was incubated in 10% milk for 2 h at room temperature (RT). Subsequently, the membrane was incubated in the primary anti- body (1:1000, anti-LEDGF/p75: Abcam#ab177159; anti-H3K36me3: Cell Signaling Technology#4909 s; anti-H3: Proteintech#17168-1-AP; anti-ß-Actin: Proteintech#81115-1-RR) for 12 h at 4 ℃ and then treated with a matched secondary antibody (1:5000, Proteintech#SA00001-2) at RT for 2 h. Enhanced chemiluminescence (Tanon, China) was used for detection. ß-Actin and histone H3 were used as endogenous controls.
2.12. siRNA transfection
siRNA sequences, which were designed and synthesized by RiboBio (Guangzhou, China), used to target LEDGF/p75 were GGAAGA- TACCGACCATGAA (5’-3’, KD1) and GCAGCAACTAAACAATCAA (5’-3’, KD2). The siRNAs were transfected with Lipo3000 (Invitrogen) according to the instruction manual.
2.13. Cell counting kit-8 (CCK-8) and clone formation
A total of 2000 cells were seeded into 96-well plates and cultured for the indicated times. At each time point, 10 ul of CCK-8 reagent (Yeasen) was mixed with the cells for 1 h. Optical density (OD) values were measured at 450 nm.
For the clone formation experiment, 1000/well of the indicated cells were cultured in a 6-well plate for 10 days. Methanol was used to fix the cells for 30 min, followed by crystal violet staining for 30 min.
2.14. Transwell assay
Cell migration assays were performed with 24-well no-Matrigel Transwell chambers (Corning, USA). A total of 3 x 104 cells were cultured in the upper chamber suspended in 200 ul of medium without fetal bovine serum (FBS), and 600 ul of medium containing 10% FBS was
added to the bottom chamber. After overnight incubation, crystal violet was used to stain the cells on the lower surface of the chamber for 30 min. Images of three random fields were acquired using a fluorescence microscope, and the cells were counted.
2.15. Gene microarray analysis
786-O cells were selected for LEDGF/p75 knockdown treatment, and three biological replicates were prepared for gene microarray detection. Agilent SurePrint G3 Human Gene Expression v3 8×60K Microarray (DesignID:072363) chip experiments and data analysis of 6 samples were performed at Shanghai Ouyi Biomedical Technology Co., Ltd. China.
Feature Extraction software version 10.7.1.1 (Agilent Technologies) was used to process original images and extract original data. The original data were then standardized. Differential genes were screened according to fold change > 1.5 and P value < 0.05. Then, GO and KEGG enrichment analyses of differentially expressed genes were performed to determine biological functions and pathways.
2.16. Online databases
Information on all online databases used in the present study can be found in Table 2.
2.17. Statistical analyses
All bioinformatics analyses were conducted via R software (version 4.2.2), except for the results obtained from the online databases mentioned in this study. Independent t tests were used to compare normally distributed continuous variables and Mann-Whitney U tests to compare skewed continuous variables. All statistical tests were two- sided. P values less than 0.05 (* P < 0.05) were considered significant.
3. Results
3.1. Biological information of LEDGF/p75
PSIP1 is a protein-coding gene located in the short arm of the ninth chromosome (Supplementary Fig. 1 A). LEDGF/p75, encoded by PSIP1, has two functional domains: the PWWP domain (aa 1-91) and the IBD (aa 347-454) (Supplementary Fig. 1B-C)[38]. The PWWP domain reads the H3K36me3 mark, which is an important regulatory mode in epige- netics[39]. For the IBD, a protein binding hub, previous studies have reported several interacting proteins, such as HIV integrase 1 and 2, MENIN-MLL, CDCA7L, PogZ, CDC7-DBF4, and IWS1 [38].
We further explored localization of LEDGF/p75 in U-251MG cells and found that almost all of the protein in the nucleus, which was consistent with its chromatin identification function (Supplementary Fig. 1D).
3.2. LEDGF/p75 is involved in several diseases, especially cancer
We first focused on protein interactions of LEDGF/p75 and predicted proteins that physically bind to it via online tools. As displayed in Fig. 1A, LEDGF/p75 is mainly involved in viral infections and DNA repair, such as the viral life cycle, DNA repair complex and DNA binding. In addition, previous experiments confirmed that LEDGF/p75 binds to histones and histone-modification proteins (Fig. 1B). Finally, we pre- dicted the scores of LEDGF/p75 in cancer and other diseases based on protein-protein interactions, which highlighted the importance of LEDGF/p75 in cancer (Fig. 1C). As LEDGF/p75 was reported as a reader of histone modification marks such as H3K36me3, we further explored the correlation between LEDGF/p75 expression and several common histone modification-related genes. As shown in Fig. 1D, SETD2, a classic writer of H3K36me3, correlated highly positively with LEDGF/
A
KMT2B
CXCR4
CDCA7L
B
C
RBBPB
HSPB1
HIST2H3D
cancer score
HIST1H4B
HIST2H3A
PPP2R1A
CRYAB
HIST1H4C
APOBEC3G
DBF4
HIST1H2AJ
0
100
PSIP1
97%
KMT2A
HIMGA1
y
PSIP1
XRCCS
other disease score
HIST1H4H
HIST1H4F
KMT2A
HDGFL2
BANF1
XRCCB
0
100
D
44%
KPNA1
CAST
PPIA
XRCC4
Coexpression across cancer types
LIG4
PRMT5
P value
1
Physical Interactions
PRMT1
viral latency
0
DNA repair complex
non-recombinational repair
KMT2B
double-strand break repair
0.7
DNA secondary structure binding
KMT2A
viral life cycle
SETD2
Cor
ACC
BLCA
BRCA
CESC
CHOL
COAD
DLBC
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LAML
LGG
LIHC
LUAD
LUSC
MESO
PAAD
PCPG
PRAD
READ
SARC
SKCM
STAD
TGCT
THCA
THYM
UCEC
UVM
-0.5
OV
UCS
E
F
G
Running Enrichment Score
Running Enrichment Score
Running Enrichment Score
GBM
OV
PRAD
0.8
0.5
0.4
0.6
0.25
0.0
0.4
0.0
-0.
4
0.2
-0.25
0.0
-0.5
Ranked List Metric
OGOGO
Ranked List Metric
Ranked List Metric
10
00050
10
10
09090
-10
-10
-10
10000
20000
30000
40000
50000
10000
20000
30000
40000
50000
10000
20000
30000
40000
50000
Rank in Ordered Dataset
Rank in Ordered Dataset
Rank in Ordered Dataset
KEGG_CHEMOKINE_SIGNALING_PATHWAY
KEGG_ANTIGEN_PROCESSING_AND_PRESENTATION
KEGG_CALCIUM_SIGNALING_PATHWAY
KEGG_CYTOSOLIC_DNA_SENSING_PATHWAY
KEGG_AUTOIMMUNE_THYROID_DISEASE
KEGG_CELL_ADHESION_MOLECULES_CAMS
KEGG_NOD_LIKE_RECEPTOR_SIGNALING_PATHWAY
KEGG_CYTOSOLIC_DNA_SENSING_PATHWAY
KEGG_CYTOKINE_CYTOKINE_RECEPTOR_INTERACTION
KEGG_REGULATION_OF_AUTOPHAGY
KEGG_REGULATION_OF_AUTOPHAGY
KEGG_NEUROACTIVE_LIGAND_RECEPTOR_INTERACTION
KEGG_RIG ___ LIKE_RECEPTOR_SIGNALING_PATHWAY
KEGG_RIG_I_LIKE_RECEPTOR_SIGNALING_PATHWAY
KEGG_RIBOSOME
p75 expression in all 33 tumor types. Interestingly, apart from the pre- viously reported lysine methylation factors (SETD2, KMT2A, KMT2B), our results showed that LEDGF/p75 expression also correlated with arginine methylation factors (PRMT1, PRMT5) in some tumor types, suggesting a potential biological role.
GO analyses, including biological process, cellular component, and molecular function, indicated vital roles for LEDGF/p75 in both HIV infection and cancers (Supplementary Table 1). For instance, LEDGF/
p75 functions in viral latency, response to oxidative stress, and chro- matin binding. The reactome pathway of LEDGF/p75 illustrates the details of LEDGF/p75 in HIV integration and the viral life cycle (Sup- plementary Table 2). Then, we performed GSEA-KEGG analyses to further explore LEDGF/p75 function. LEDGF/p75 expression correlated with regulation of autophagy in GBM and OV and with cell adhesion molecules in PRAD, which indicates the potential function of LEDGE/ p75 in tumor metastasis and development (Fig. 1E-G). All TCGA
abbreviations are provided in Table 1.
3.3. LEDGF/p75 is differentially expressed across cancers
We analyzed LEDGF/p75 expression data from the UCSC Xena database and found LEDGF/p75 mRNA to be significantly differentially expressed in 15 of 21 tumor types. Among the 15 types, LEDGF/p75 was downregulated in 11 types (BLCA, BRCA, KICH, KIRC, KIRP, LUAD, LUSC, PRAD, READ, THCA, and UCEC) and highly expressed in another 4 (CHOL, HNSC, LIHC, and PCPG) (Fig. 2A). We supplemented data from GTEx normal tissues and performed the analyses for another 12 tumor types. We found LEDGF/p75 to be highly expressed in DLBC and THYM but reduced in OV (Fig. 2B-D).
Next, we explored LEDGF/p75 expression in various cell lines and found LEDGF/p75 to be differentially expressed in different cell lines (Fig. 2E). The top two cell lines with the highest LEDGF/p75 expression were HEL and NTERA-2, and the expression level of HEL cells was more than twice that of NTERA-2 cells. HEL, a cancer cell line derived from myeloid cells, is an erythroleukemia cell line (AML M6 in relapse after treatment for Hodgkin’s disease). Such high expression of LEDGF/p75 in HEL cells suggests that it may play a role in erythroleukemia. Previous studies have reported that LEDGF/p75 is essential for MLL-rearranged leukemogenesis[40]. Whether there is a deeper connection between the two diseases other than both being blood cancers and whether this connection is related to LEDGF/p75 remains unclear.
Then, we analyzed LEDGF/p75 protein expression across cancers via online tools. LEDGF/p75 was significantly reduced in 5 tumor types (BRCA, COAD, HNSC, LUAD, and UCEC) but markedly increased in another 2 (LIHC and OV) (Fig. 2F-L). As expected, immunohistochem- ical results from the ATLAS database showed trends similar to the above results.
These findings indicate significant differences in LEDGF/p75 expression across cancers. After integrated analysis of differences in
| Abbreviation | Full name |
|---|---|
| ACC | Adrenocortical Carcinoma |
| KIRC | Kidney Renal Clear Cell Carcinoma |
| PRAD | Prostate Adenocarcinoma |
| BLCA | Bladder Urothelial Carcinoma |
| KIRP | Kidney Renal Papillary Cell Carcinomal |
| READ | Rectum Adenocarcinoma |
| BRCA | Breast Invasive Carcinoma |
| LAML | Acute Myeloid Leukemia |
| SARC | Sarcoma |
| CESC | Cervical Squamous Cell Carcinoma |
| LGG | Lower Grade Glioma |
| SKCM | Skin Cutaneous Melanoma |
| CHOL | Cholangiocarcinoma |
| LIHC | Liver Hepatocellular Carcinoma |
| STAD | Stomach Adenocarcinoma |
| COAD | Colon Adenocarcinoma |
| LUAD | Lung Adenocarcinoma |
| TGCT | Testicular Germ Cell Tumors |
| DLBC | Diffuse Large B-cell Lymphoma |
| LUSC | Lung Squamous Cell Carcinoma |
| THCA | Thyroid Carcinoma |
| ESCA | Esophageal Carcinoma |
| MESO | Mesothelioma |
| THYM | Thymoma |
| GBM | Glioblastoma Multiforme |
| OV | Ovarian Serous Cystadenocarcinoma |
| UCEC | Uterine Corpus Endometrial Carcinoma |
| HNSC | Head and Neck Squamous Cell Carcinoma |
| PAAD | Pancreatic Adenocarcinoma |
| UCS | Uterine Carcinosarcoma |
| KICH | Kidney Chromophobe |
| PCPG | Pheochromocytoma and Paraganglioma |
| UVM | Uveal Melanoma |
| Database | Online link |
|---|---|
| GeneCards | https://www.genecards.org/ |
| Uniprot | https://www.uniprot.org/ |
| AlphaFold | https://alphafold.ebi.ac.uk/ |
| ATLAS | https://www.proteinatlas.org/ |
| GeneMANIA | http://genemania.org/ |
| STRING | https://cn.string-db.org/ |
| canSAR | https://cansar.ai/ |
| TISIDB | http://cis.hku.hk/TISIDB/index.php |
| KEGG pathway | http://www.gsea-msigdb.org/gsea/msigdb/human/collections. jsp |
| UCSC Xena | https://xena.ucsc.edu/ |
| GEPIA | http://gepia.cancer-pku.cn/index.html |
| UALCAN | http://ualcan.path.uab.edu/index.html |
| cBioPortal | https://www.cbioportal.org/ |
| OncoSplicing | http://www.oncosplicing.com/ |
| TIDE | http://tide.dfci.harvard.edu/ |
| TISMO | http://tismo.cistrome.org/ |
| ROC Plotter | https://www.rocplot.org/ |
| DepMap portal | https://depmap.org/portal/ |
LEDGF/p75 mRNA and protein expression, we found the same trend for LIHC, BRCA, LUAD and UCEC. Notably, LEDGF/p75 expression was opposite at the mRNA and protein levels in HNSC, which requires experimental verification.
3.4. LEDGF/p75 expression correlates with patient prognosis
To investigate the correlation between LEDGF/p75 expression and patient prognosis, we performed comprehensive analysis of expression and clinical data from the UCSC Xena database. Cox regression and K-M analyses were used to explore OS, DSS, DFI and PFI. As shown in the forest plot, LEDGF/p75 expression correlated negatively with OS in 4 tumor types (ACC, KICH, LIHC, and UCEC) but positively in 5 (CESC, KIRC, LGG, PAAD, and SKCM) (Fig. 3A). Regarding K-M plots, there was a negative correlation between LEDGF/p75 expression and OS in ACC but a positive correlation in 3 other tumor types (KIRC, LGG, and OV) (Fig. 3B-E). The results for DSS, DFI and PFI are shown in Supple- mentary Fig. 2 and Supplementary Fig. 3.
Furthermore, we analyzed the correlation between LEDGF/p75 expression and the clinical stage of patients. The results indicated that patients with advanced ACC generally expressed high levels of LEDGF/ p75; this phenomenon was reversed in patients with PAAD, SKCM, THCA, or KIRC (Fig. 3F-L). All these analyses indicate that LEDGF/p75 expression is closely associated with patient prognosis across cancers. Specifically, LEDGF/p75 is a potential oncogene in ACC, KICH and LIHC but a potential protective gene in PAAD and SKCM.
3.5. LEDGF/p75 genomic alterations in pancancer and correlation with patient prognosis
As genomic alterations may cause tumorigenesis, we explored LEDGF/p75 gene alterations in the cBioPortal database, which is a multidimensional cancer genomics dataset. We illustrate the landscape of alteration frequency in pancancer in Supplementary Fig. 4A, including mutation, structural variant, amplification, deep deletion, and multiple alterations. We then studied the percentage of altered groups in pancancer. Notably, the LEDGF/p75 gene was altered in nearly 50% of patients with 3 tumor types (COAD_POLE, ESCA POLE, and UCEC _- POLE) (Supplementary Fig. 4B). Putative copy number alterations from GISTIC showed that diploid and shallow deletions were the top two most common types (Fig. 4A). LEDGF/p75 mutation occurred at a low fre- quency (Fig. 4B), and specific mutation sites are shown in Fig. 4C.
Subsequently, we analyzed the TMB and MSI of LEDGF/p75 across cancers. The TMB results showed that LEDGF/p75 correlated negatively with 8 tumor types (KIRC, KIRP, LGG, PAAD, PRAD, THCA, UVM, and
A
Type Normal Tumor
8
**
LEDGF/p75 expression
0
…
—
A
2
0
BLCA
BRCA
CESC
CHOL
COAD-
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LIHC
LUAD
LUSC
PAAD
PCPG
PRAD
READ
STAD
THCA
UCEC
B
7
C
E
U-937
THP-1
8
NB-4
K-562
6
HMC-1
LEDGF/p75 expression
LEDGF/p75 expression
HL-60
HEL
HAP
5
0)
hTEC/SVTERT24-B
U-698
U-266/84
U-266/70
4
RPMI-8226
Myeloid
REH
4
MOLT-4
Karpas-707
Lymphoid
00
JURKAT
HDLM-2
Daudi
Mesenchymal
2
2
U-2197
U-2 OS
Muscle
RH-30
LHCN-M2
1
HHSteC
Endothelial
DLBC
OV
HBF TERT88
fHDF/TERT166-
Female reproductive
0
BJ hTERT+ SV40 Large T+ RasG12V-
BJ hTERT+ SV40 Large T+
system
T=47
N=337
T=426
N=88
BJ hTERT
Lung
ASC TERT
BJ
Proximal digestive
ASC diff
HSKMC TIME
tract
Eye
D
F
HUVEC TERT2
T-47d
SK-BR-3
Skin
SiHa
8
3
MCF7
Kidney & Urinary
2
hTERT-HME1
Hela
bladder
EFO-21
Male reproductive
LEDGF/p75 expression
Z-value
1.
BEWO
AN3-CA
system
6
0
SCLC-21H
Pancreas
HBEC3-KT
A549
-1
OE19
Gastrointestinal tract
hTERT-RPE1
-2
hTCEpi-
Liver & Gallbladder
4
BRCA
WM-115
SK-MEL-30
-3
HaCaT
Brain
N=18
T=125
A-431
RT4
RPTEC TERT1
NTERA-2
2
HEK 293
SuSa
PC-3
CAPAN-2
THYM
CACO-2
Hep G2
U-87 MG
U-251 MG
T=118
U-138 MG
N=339
SH-SY5Y
Normal
Tumor
GAMG
AF22
nTPM
0
200
400 800
900
G
H
3
I
3
3-
2
2
2
Z-value
1
Z-value
1
Z-value
1
0
0
0
-1
-1
-1
-2
-3
COAD
-2
HNSC
-2
LIHC
N=100
T=97
-3
N=71
T=108
-3
N=165
T=165
Normal
Tumor
Normal
Tumor
Normal
Tumor
C
K
3
3-
L
3
2
2
2
Z-value
1
Z-value
1
Z-value
1
0
0
0
-1
-1
-1
-2
LUAD
-2
-2
-3
-3
OV
UCEC
N=111
T=111
N=25
T=100
-3
N=31
T=100
Normal
Tumor
Normal
Tumor
Normal
Tumor
| pvalue | Hazard ratio | |
|---|---|---|
| ACC | 0.007 | 2.252(1.249-4.061) |
| BLCA | 0.728 | 0.972(0.828-1.141) |
| BRCA | 0.720 | 0.963(0.781-1.186) |
| CESC | 0.019 | 0.672(0.482-0.937) |
| CHOL | 0.739 | 0.863(0.362-2.055) |
| COAD | 0.891 | 0.980(0.729-1.316) |
| DLBC | 0.362 | 1.740(0.528-5.735) |
| ESCA | 0.400 | 1.183(0.800-1.751) |
| GBM | 0.530 | 0.923(0.718-1.186) |
| HNSC | 0.664 | 0.966(0.825-1.130) |
| KICH | 0.027 | 3.031(1.137-8.077) |
| KIRC | <0.001 | 0.601(0.453-0.798) |
| KIRP | 0.375 | 1.299(0.729-2.317) |
| LAML | 0.805 | 0.935(0.546-1.600) |
| LGG | <0.001 | 0.329(0.223-0.486) |
| LIHC | 0.034 | 1.259(1.017-1.559) |
| LUAD | 0.616 | 0.943(0.750-1.186) |
| LUSC | 0.285 | 0.906(0.756-1.086) |
| MESO | 0.293 | 1.297(0.799-2.105) |
| OV | 0.053 | 0.830(0.688-1.003) |
| PAAD | 0.019 | 0.662(0.469-0.934) |
| PCPG | 0.549 | 0.722(0.249-2.096) |
| PRAD | 0.073 | 3.835(0.883-16.650) |
| READ | 0.424 | 0.769(0.404-1.465) |
| SARC | 0.607 | 1.057(0.857-1.303) |
| SKCM | 0.023 | 0.826(0.700-0.974) |
| STAD | 0.356 | 0.875(0.658-1.162) |
| TGCT | 0.893 | 1.154(0.143-9.329) |
| THCA | 0.347 | 0.529(0.140-1.997) |
| THYM | 0.190 | 0.651(0.342-1.238) |
| UCEC | 0.049 | 1.303(1.001-1.697) |
| UCS | 0.535 | 1.190(0.687-2.061) |
| UVM | 0.873 | 1.043(0.624-1.742) |
A
Overall Survival
B
ACC
1.0
LEDGF/p75
Low
Survival probability
0.8
High
0.6
0.4
+
0.2
Overall Survival
HR = 2.39 (1.10-5.19)
0.0
P = 0.028
0
2.5
5
7.5
10
12.5
Time (years)
Low
39
29
16
6
2
High
40
21
8
4
2
1
C
KIRC
1.0
LEDGF/p75
Low
Survival probability
0.8
High
0.6
0.4
0.2
Overall Survival
HR = 0.61 (0.45-0.83)
0.0
P = 0.002
0
2.5
5
7.5
10
12.5
Time (years)
Low
269
165
71
23
6
0
High 270
166
82
32
7
0
D
LGG
1.0
LEDGF/p75
Low
Survival probability
0.8
High
0.6
0.4
0.2
Overall Survival
HR = 0.51 (0.36-0.72)
0.0
P < 0.001
0
5
10
15
0.12
0.50
2.0
8.0
Time (years)
Hazard ratio
Low
264
32
7
2
0
High
263
39
12
1
0
E
F
G
H
OV
ACC
BLCA
ESCA
1.0 -
LEDGF/p75
Low
Survival probability
0.8
High
71
0.00015
0.011
LEDGF/p75 expression
0.035
0.83
0.32
LEDGF/p75 expression
10.0
0.6
0.87
0.065
LEDGF/p75 expression
8
0.97
6
0.00029
0.52
0.12
0.22
0.4
5
0.3
0.45
7.5
0.45
0.56
6
0.031
0.0006
0.2
Overall Survival HR = 0.77 (0.59-0.99)
4
5.0
4
0.0
P = 0.043
3
0
5
10
15
2.5
Time (years)
2
2
Low
188
34
Stage | Stage II Stage III Stage IV
Stage | Stage II Stage III Stage IV
Stage | Stage II Stage III Stage IV
High
189
43
6
0
PAAD
J
SKCM
K
THCA
L
KIRC
0.21
0.79
0.71
0.86
LEDGF/p75 expression
0.8
LEDGF/p75 expression
0.21
LEDGF/p75 expression
6
0.62
0,72
0.6
LEDGF/p75 expression
8
0.16
0.99
0.17
0.14
7.5
0.13
0.0076
0.054
5
0.00023
0.25
0.035
0.02
6
0.0072
0.016
0.00052
0.81
5.0
4
4
E
4
B
2.5
3
2
2
2
Stage | Stage II Stage III Stage IV
Stage | Stage II Stage III Stage IV
Stage | Stage II Stage III Stage IV
Stage | Stage II Stage III Stage IV
A
B
Amplification
70
Not profiled
524
Gain
1158
No mutation
10317
5980
Multiple
3
Diploid
13
Shallow Deletion
Splice
3432
39
7
Deep Deletion
Truncating
Inframe
72
Missense
64
0
1k
2k
3k
4k
5k
6k
0
1k
2k
3k
4k
5k
6k
7k
8k
9k
10k
LEDGF/p75: Putative copy-number alterations from GISTIC
LEDGF/p75: Mutations
C
TMB
LEDGF/p75 Mutations
D
E278del
ACC
UCS
UVM
BLCA
5
UCEC
0.6
BRCA
CESC
THYM
4
CHOL
THCA
0 2
COAD
TGCT
0
DLBC
0
PWWP
LEDGF
STAD
-0.2
-0
0.4
ESCA
PTM (dbPTM)
0
100
200
300
400
530aa
SKCM
-0.6
GBM
Phosphorylation
SARC
HNSC
Acetylation
Methylation
READ
KICH
Glutathionylation
Malonylation
PRAD
KIRC
Sumoylation
Exon
PCPG
KIRP
2
3
4
5
6
8
9
10
11
12
13
14
15
16
PAAD
LÃML
OV
F
G
E
MESO
LUSC
LUAD
LIHC
LGG
MSI
MLH1
MSH2
MSH6
PMS2
EPCAM
pan-cancer
100%
UVM
ACC
BLCA
UCEC
UCS
0.4
BRCA
0.3.
CESC
ACC
:
Probability of Overall Survival
90%
Logrank Test P-Value: 4.676e-3
THYM
0.2
CHOL
BLCA
80%
Altered group
THCA
0.1
COAD
0
BRCA
E
70%
Unaltered group
TGCT
DLBC
CESC
:
#
#
60%
STAD
001
50%
-0.2
ESCA
CHOL
SKCM
-0.3
GBM
COAD
¿
:
40%
30%
SARC
HNSC
DLBC
E
:
20%
READ
KICH
ESCA
:
E
E
10%
PRAD
KIRC
GBM
E
E
E
#
0%
PCPG
KIRP
HNSC
:
E
:
0
40
80
Overall Survival (Months)
120
160
200
240
280
320
360
PAAD
LAML
OV
MESO
LIHC
LGG
KICH
:
:
#
LUSC
LUAD
KIRC
:
E
E
#
H
TGCT
BRCA
KIRP
:
E
:
100%
Logrank Test P-Value: 4.848e-3
LAML
:
E
E
:
90%
100%
Probability of Overall Survival
Probability of Overall Survival
Altered group
LGG
:
#
:
:
80%
90%
Unaltered group
70%
80%
LIHC
:
:
#
60%
70%
LUAD
:
:
60%
LUSC
:
E
:
#
50%
50%
MESO
:
:
40%
Logrank Test P-Value: 1.551e-4
40%
OV
:
:
E
30%
Altered group
30%
PAAD
E
¿
E
20%
Unaltered group
20%
PCPG
:
E
10%
10%
0%
PRAD
:
E
E
#
0
J
20 40 60 80 100 120 140 160 180 200 220 240 Overall Survival (Months)
0%
0
40
80
Overall Survival (Months)
120
160
200
240
280
READ
:
E
E
K
SARC
:
¿
E
HNSC
LIHC
¿
E
100%
Logrank Test P-Value: 8.918e-4
100%
SKCM
Logrank Test P-Value: 0.0368
Probability of Overall Survival
90%
Probability of Overall Survival
90%
STAD
Altered group
80%
Altered group
Unaltered group
80%
Unaltered group
TGCT
E
70%
70%
THCA
E
60%
60%
THYM
:
50%
50%
UCEC
40%
40%
UCS
30%
30%
UVM
20%
20%
Cor
P value
10%
10%
0%
0%
-0.4
0.8
0
1
20
40
60
80
100
120
140 1
160
Overall Survival (Months)
180 200
0
20
Overall Survival (Months)
40
60
80
100
120
THYM) but positively with 8 types (ACC, BLCA, LOAD, LAML, LUAD, READ, SKCM, and UCS) (Fig. 4D). The MSI results showed a positive correlation between 7 tumor types (ACC, BRCA, COAD, LUAD, READ, STAD, and UCEC) and LEDGF/p75 but a negative correlation only be- tween SKCM and LEDGF/p75 (Fig. 4E). Next, we investigated the po- tential function of LEDGF/p75 in DNA mismatch repair (MMR). As displayed in Fig. 4F, LEDGF/p75 expression correlated highly with MMR genes across cancers, which was consistent with a previously reported result that LEDGF/p75 is involved in DNA damage repair[41].
To further explore the clinical value of LEDGF/p75 gene alterations across cancers, we analyzed their association with patient OS. Our comprehensive analysis showed that patients in the altered group had shorter OS, which was significantly different (Fig. 4G). We then per- formed a separate analysis of 33 tumor types and found that LEDGF/p75 gene alterations predicted poor patient prognosis in TGCT, BRCA, HNSC, and LIHC (Fig. 4H-K). In conclusion, LEDGF/P75 genomic instability is widespread across cancers and suggests poor prognosis.
3.6. Pancancer view of LEDGF/p75 alterative splicing and correlation with patient survival
Alternative splicing (AS) regulates the generation of multiple mRNA and protein products from a single gene. AS plays a crucial role in cancer progression, and cancer cells have general as well as cancer-type-specific and subtype-specific changes during splicing that may have prognostic value and contribute to cancer development and progression[42]. We chose the item (PSIP1_alt_3prime_247053) for subsequent analyses on the OncoSplicing database because it is the only known splice type in SplAdder, a bioinformatics tool for the analysis and quantification of alternative splicing events in RNA sequencing data. The read-in, read-out and PSI values are shown in Fig. 5A, and there were signifi- cant differences in LEDGF/p75 AS between tumor and normal tissues. We then visualized the PSI difference in tumor and adjacent normal tissues (Fig. 5B), and the result showed LUSC as the top result. However, the top three changed to LGG, GBM and TGCT when we compared the PSI difference between tumor and GTEx normal tissues (Fig. 5C).
We performed K-M analysis to explore the clinical value of LEDGE/ p75 AS. Consistent with our predictions, LEDGF/p75 AS suggested dif- ferences in patient prognosis in several tumor types. K-M curves of OS, DSS and PFI showed that a high LEDGF/p75 PSI indicated good prog- nosis in patients with SKCM (Fig. 5D-F). The same trend was also observed in patients with THCA and CESC (Fig. 5H-I). However, the opposite trend was observed in patients with COAD and LUAD (Fig. 5G, J-K). All of the above results imply the biological importance of LEDGF/ p75 AS events across cancers.
3.7. LEDGF/p75 is involved in cancer immune infiltration and immunotherapy
To further explore the immunomodulatory effects of LEDGF/p75, we analyzed correlations between immune cells, stromal cells, 22 immune cells, immune checkpoints and LEDGF/p75 expression across cancers (Supplementary Fig. 6); the results showed a tight association between LEDGF/p75 and tumor immunity. Subsequently, we explored whether LEDGF/p75 is differentially expressed in diverse cancer immune sub- types via the TISIDB database. Fig. 6A shows that LEDGF/p75 was significantly associated with immune subtypes in several tumors, and the top six are presented in Supplementary Fig. 5D-I. The detailed subtypes include C1 (wound healing), C2 (IFN-gamma dominant), C3 (inflammatory), C4 (lymphocyte depleted), C5 (immunologically quiet), and C6 (TGF-b dominant). In addition, we analyzed the association between immunoinhibitors, chemokines, TILs and LEDGF/p75 expres- sion (Supplementary Fig. 5A-C). As visualized in heatmaps, LEDGF/p75 expression correlated with several immunoinhibitors (CTLA4, IL10, PDCD1, etc.), chemokines (CXCL9, 10, 11, etc.) and TILs (activated CD4, Th2, CD56, etc.) in pancancer.
We searched the TIDE database to evaluate multiple published transcriptomic biomarkers to predict patient response. We list the most confident results about the correlation between LEDGF/p75 expression and CTLs, T dysfunction, and risks in Fig. 6C. The results showed a positive correlation between LEDGF/p75 expression and CTLs in breast cancer but a negative correlation in brain cancer. Moreover, high LEDGF/p75 expression indicated short overall survival for endometrial cancer patients.
We then explored the correlation between LEDGF/p75 and immu- notherapy. We compared LEDGF/p75 expression levels across cell lines between pre- and postcytokine-treated samples via the TISMO database, and the box plot is presented in Fig. 6B. Finally, we searched the ROC Plotter database to investigate the correlation between LEDGF/p75 expression and immunotherapy efficiency. Interestingly, high LEDGF/ p75 expression indicated effective immunotherapy results in ESCA PD- L1, STAD PD-1, SKCM CTLA-4, and SKCM PD-1 (Fig. 6D-G). Notably, the AUC was higher than 0.65 for ESCA PD-L1 and STAD PD-1, illus- trating the high value of the prediction model. In summary, LEDGF/p75 is involved in cancer immune infiltration and immunotherapy, which might guide personalized treatment of tumor patients.
3.8. LEDGF/p75 is highly expressed in ccRCC cells and significantly promotes proliferation and metastatic ability
Renal cell carcinoma includes more than 10 histological and mo- lecular subtypes, of which ccRCC is the most common and accounts for the majority of cancer-related deaths[17], and reduction or even dele- tion of H3K36me3 occurs due to the high mutation rate of SETD2 in patients with advanced ccRCC[43]. This phenomenon naturally strat- ifies ccRCC patients into subgroups; thus, the study of LEDGF/p75, the reader of H3K36me3, is of great significance. Although studies have reported the vital function of LEDGF/p75 in prostate cancer, leukemia and other kinds of tumors [44-46], there has been no study related to kidney cancer. Therefore, we are the first to conduct preliminary func- tional exploration of LEDGF/p75 in ccRCC.
Considering the presence of SETD2 mutations in ccRCC cell lines, we searched the DepMap portal database for relevant information. The re- sults revealed SETD2 mutations in A498 and Caki-1 cells: p.V2536fs and p.R400R, respectively. To determine expression of H3K36me3 in ccRCC cell lines, 786-O, A498 and Caki-1 cells were selected for western blot- ting. As shown in Fig. 7A, H3K36me3 was highly expressed in 786-O and Caki-1 cells but absent in A498 cells, as predicted by the database. For LEDGF/p75, all three ccRCC cell lines expressed higher levels than HK-2 cells. Therefore, we chose 786-O and Caki-1 cells for further studies.
To detect the impact of LEDGF/p75 on ccRCC cell characteristics, we first attempted to knock down LEDGF/p75. Both siRNAs tested achieved > 50% knockdown of the expression level of LEDGF/p75 (Fig. 7B-C). When LEDGF/p75 was knocked down, proliferation and migration ability of ccRCC cells were significantly reduced (Fig. 7D-G), which indicated that LEDGF/p75 is a potential oncogene in ccRCC.
3.9. LEDGF/p75 knockdown in ccRCC causes changes to cancer-related genes and pathways
To further explore the role of LEDGF/p75 in ccRCC, we knocked down LEDGF/p75 in 786-O cells and performed gene microarray anal- ysis (Fig. 8A). After LEDGF/p75 was knocked down, 655 genes were upregulated and 512 downregulated (Fig. 8B, Supplementary Table 3). Among them, the most down regulated protein-coding gene was ERO1L. We performed GO and KEGG analyses for all regulated genes. Consistent with our expectations, the transcriptional activity of cells was signifi- cantly changed after LEDGF/p75 knockdown (Figs. 8C, 8E). As H3K36me3 is a marker of active transcription[14], knockdown of LEDGF/p75, a reader of H3K36me3, is likely to cause transcriptional inhibition of some downstream genes. Therefore, we reperformed the GO and KEGG analyses of all downregulated genes.
A
| PSI | value | Reads-Out | Reads-In | |||||
|---|---|---|---|---|---|---|---|---|
| 0.00 0.25 | 0.75 0.50 | 1.00 | 500 0 | 1000 | 1500 0 | 200 400 | ||
| bo | ||||||||
| ACC-T (66) | 1 | |||||||
| BLCA-N (17) | 0 8 | P | 6 | |||||
| BLCA-T (333) | 00 | 0 | IDo | |||||
| BRCA-N (97) | 00 0 | P | P | |||||
| BRCA-T (927) | kTID 0 | Im | ||||||
| CESC-N (2) | V | |||||||
| CESC-T (264 | 1000 | 0 | 0 O | |||||
| 10 | · | - | A | |||||
| CHOL-N (4) | ||||||||
| CHOL-T (28 | Q | ₱ | 8 | |||||
| COAD-N (38 | O I 3 | A | ||||||
| COAD-T (269 | IL | p | ₾ | |||||
| DLBC-T (41 | ₱ | 1 | ||||||
| ESCA-N (8 | 4 0 | 0 | ||||||
| ESCA-T (136) | 10 | 0 | KDO | 10 | ||||
| GBM-T (165 | A | 9 | +0 | 1000 | ||||
| HNSC-N (32) | A T 0 T | lo | ||||||
| HNSC-T (401 | 1 | p | ||||||
| KICH-N (24 | . | . | ||||||
| KICH-T (5 (56 | 0 | P | ||||||
| KIRC-N (72) | 10 | p | ||||||
| KIRC-T (343 | 1 | - F | P | |||||
| KIRP-N (31 | . | P | ||||||
| KIRP-T (231) | A | 0 | 0 | ৳ | ||||
| LGG-T (520) LIHC-N (14) | . I - | OLD | HECIDO | TITIDD G | ||||
| ₱ | 0 | |||||||
| LIHC-T (211) | p | 00 | ||||||
| LUAD-N (39 | 6 | |||||||
| LUAD-T (428) | 0 0 | ACTO | 1 | |||||
| LUSC-N (47 | 0 | + | A | |||||
| LUSC-T (468) | AExo o | p | . | |||||
| MESO-T (75 | 0 | * | A | |||||
| OV-T (280) | 0 0 | I O | 4 | .. | ||||
| PAAD-N (3 | HI | A | ||||||
| PAAD-T (138) | . | · | ₱ | |||||
| PCPG-N (3 | VI | · | ||||||
| PCPG-T (160) | 00 | 10 | 100 | |||||
| PRAD-N (51 | I 0 | P | ||||||
| PRAD-T (432) | L | * | 0 | |||||
| READ-N (10 | b | |||||||
| READ-T (92) | 0 | |||||||
| SARC-N (1 | Y | |||||||
| SARC-T (231) | FOOD | |||||||
| C | SKCM-N (1 | I | I | |||||
| SKCM-T (404 | A | 100 | ||||||
| STAD-N (23 | 0 | . | b P | |||||
| STAD-T (290) | 100 | KID | 0 | |||||
| TGCT-T (142) | 0 | + | ||||||
| THCA-N (55) | . | + | ৳ | |||||
| THCA-T (470) | 100 | p | ||||||
| THYM-N | (2) + | + | U | |||||
| THYM-T (94) | P | |||||||
| UCEC-N (21 | C | · | A | |||||
| UCEC-T (145) | 40 | A | ||||||
| UCS-T (48) | 1 100 | V | ||||||
| UVM-T (35) | b | |||||||
| Adipose (149) | - | P | - | |||||
| Adrenal (56 | 00 | . | ||||||
| Artery (257 | 1000 | |||||||
| Bladder | (9 | 00 | ||||||
| Blood (114) | to | |||||||
| 0 | Ha | COLD COCO 0 | ||||||
| Brain (417) | 008 | |||||||
| Breast | (60) 0 | p | ||||||
| Cells (261 | 0 | ICD | 100 | |||||
| Cervix | (9) 4 | ¢ | ||||||
| Colon (73) | 0 | b | b | |||||
| Esophagus (181) | 0 | 8 | bo | ৳ | ||||
| Fallopian | (7) | |||||||
| Heart | (64) | D | 10 | |||||
| Intestine | (15) 0 V | O | ₱ | |||||
| Kidney | ( | 0 | ₱ | |||||
| Liver | (8) | Co | - | |||||
| Lung (128) | 0 | p | p | |||||
| (66) O | ৳ | p | ||||||
| Muscle | ||||||||
| Nerve (115) | p | Aco | ||||||
| Ovary | (39) | 0 | + | |||||
| Pancreas | 18) N | 0 | ₱ | . | ||||
| Pituitary | 24 0 | 0 | 0 | |||||
| Prostate | 38 | o | A | |||||
| Salivary | (4 | |||||||
| Skin (150) | 0 | b | p | |||||
| Spleen | (32 - | b | ||||||
| Stomach | 66 | D | p | |||||
| Testis | (67) 01- 0 | Ho | ol® | |||||
| O | ||||||||
| Thyroid | (120) 1 | lo | ||||||
| Uterus | (36) | .. | ||||||
| Vagina | (31) D | 4 | ₱ | 10 | ||||
Progression free interval
0.50
0.75
1.00-
H
0.00
Overall Survival
0.00
0.25
0.25
0.50
0.75
1.00-
D
-log10(FDR)
Log-rank p = 1.42e-02
THCA
Log-rank p = 2.16e-03
34.7
67.3
100
B
SKCM
0
2
0
0
-0.5
5
10
Years
Years
Median cutoff
PSI difference (Tumor-Normal)
-0.25
LIHC
10
PSI ⇐ 0.103(n=241)
Median cutoff
20
PSI>0.103(n=229)
PSI>0.105(n=196)
PSI ⇐ 0.105(n=199)
KIRP
0
15
30
LUSC
Progression free interval
Disease specific survival
0.00
0.25
0.50
0.75
1.00-
0.00
0.25
0.50
0.75
1.00
E
Log-rank p = 1.15e-02
CESC
Log-rank p = 9.62e-04
SKCM
0.25
0
0
5
10
Years
10
Years
0.5
20
15
PSI>0.096(n=132)
PSI ⇐ 0.096(n=132)
Median cutoff
PSI>0.105(n=195)
PSI ⇐ 0.105(n=194)
Median cutoff
0.24
0.19
0.15
0.10
Tumor PSI
20
30
U
Progression free interval
J
Progression free interval
F
-log10(FDR)
0.00
0.25
0.50
0.75
1.00-
0.00
0.25
0.50
0.75
1.00-
Log-rank p = 3.66e-02
LUAD
Log-rank p = 3.01e-02
SKCM
34.7
67.3
100
0
2
0
0
-0.5
GBM
5
LGG
10
Years
10
Median cutoff
Years
PSI difference (Tumor-GTEx)
-0.25
LIHC
TGCT
20
PSI>0.099(n=211)
PSI ⇐ 0.099(n=211)
PSI>0.105(n=196)
PSI ⇐ 0.105(n=199)
Median cutoff
BLCA
CESC
.
15
OV
LUSC
THCA
20
30
0
ACC
PAAD
Disease free interval
Progression free interval
0.00
0.25
0.50
0.75
1.00
K
0.00
0.25
0.50
0.75
1.00-
G
PRAD
Log-rank p = 5.77e-03
LUAD
Log-rank p = 4.32e-02
COAD
0
0.25
0
5
3
Years
0.5
10
Years
6
.
PSI ⇐ 0.099(n=127)
Median cutoff
PSI>0.053(n=133)
PSI ⇐ 0.053(n=134)
Median cutoff
15
PSI>0.099(n=130)
9
0.24
0.19
0.15
0.10
Tumor PSI
20
12
Fig. 5. LEDGF/p75 alternative splicing correlates with patient prognosis. (A) Read-in, read-out, and percent spliced in (PSI) values of LEDGF/p75 in pancancer
and normal tissues. The red and gray labels represent cancers and adjacent normal tissues, respectively; black labels represent normal tissues. The parts labeled with
“Reads-In” and “Reads-Out” on the Y-axis represent read count values, indicating exon splicing in or splicing out, respectively. (B-C) PSI differences between tumor
and adjacent normal tissues and between tumor and GTEx normal tissues. The red line refers to 0.05, the dot size represents the tumor PSI value, and different cancers
are marked in different colors. (D-K) The PSI value of LEDGF/p75 correlates with patient prognosis in several kinds of cancer.
A
Kruskal-Wallis Test (-log10pv)
12
10
8
6
4
2
0
ACC BLCA
BRCA
CESC
CHOL
COAD
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LGG
LIHC
LUAD
LUSC
MESO
OV
PAAD
PCPG
PRAD
READ
SARC
SKCM
STAD
TGCT
THCA
UCEC
UCS
UVM
B
E Baseline IFNb
# IFNg TGFb1 TNFa
IFNb vs. Baseline
* IFNg vs. Baseline TGFb1 vs. Baseline TNFa vs. Baseline
Ę
4T1_GSE110912(n=6)
4T1_XW33589424(n=15)
4T1_RTM28723893(n=12)
B16_GSE149824(n=8)
B16_SSG33589424(n=16)
B16_GSE110708(n=6)
B16_GSE107670(n=6)
B16_GSE106390(n=6)
B16_GSE85535(n=7)
B16_RTM28723893(n=8) CT26_RTM28723893(n=12)
E0771_XW33589424(n=4)
EMT6_XW33589424(n=6)
KPC_RTM28723893(n=12) LLC_RTM28723893(n=44)
MC38_GSE112251(n=12)
MC38_RTM28723893(n=48)
MOC1_RU31562203_LZ5733(n=18)
MOC2_RU31562203(n=7)
MOC22_RU31562203(n=4)
Panc02_RTM28723893(n=12)
Renca_RTM28723893(n=11)
C
-
*
*
1
*
**
**
**
*
*
**
5
6
7
8
LEDGF/p75 log(TPM)
| . Cancer | . Subtype | CTL Cor | T . Dysfunction | ¢ Risk | + Risk.adj | + Count |
|---|---|---|---|---|---|---|
| Breast | TN | 0.245 | 0.387 | 0.380 | 0.962 | 233 |
| Melanoma | Metastatic | -0.004 | -1.091 | -0.456 | -0.471 | 317 |
| Endometrial | -0.076 | 1.663 | 2.213 | 2.118 | 541 | |
| Leukemia | AML | -0.128 | -0.2350 | 0.691 | 0.477 | 79 |
| Brain | Neuroblastoma | -0.202 | -0.856 | 0.957 | 0.697 | 389 |
0.3
r= 0.245 , p= 0.000159
r= - 0.202 , p=6.2e-05
1.0
Continuous z= 2.21 , p= 0.0269
0.2
LEDGF/p75
5000
0.8
0.1
LEDGF/p75
Survival Fraction
0.0
0.6
-0.1
3000
0.4
0.2
1000
LEDGF/p75 Top (n=11)
-0.3
0.0
LEDGF/p75 Bottom (n=530)
-0.1
0.1
0.2
0.3
0.4
0.5
0
4000
8000
12000
0
50
100
150
200
CTL
CTL
OS (month)
D
ESCA PD-L1
E
STAD PD-1 LEDGF/p75
F
SKCM CTLA-4
G
SKCM PD-1
4000
LEDGF/p75
5000
LEDGF/p75
LEDGF/p75
2500
Gene expression
3000
Gene expression
2000
Gene expression
4000
Gene expression
3000
2000
1500
3000
2000
2000
1000
1000
1000
500 1000
500
Non-responder
Responder
Non-responder
Responder
Non-responder
Responder
Non-responder
Responder
1.0
1.0
1.0
1.0
True positive rate
0.8
True positive rate
0.8
True positive rate
0.8
True positive rate
0.8
0.6
0.6
0.6
0.6
0.4
0.4
0.4
0.4
AUC: 0.68
AUC:0.755
AUC: 0.596
AUC:0.582
0.2
p-value: 3.8e-04
strongest cutoff: 1210
0.2
p-value: 1.1e-03
p-value: 3.3e-02
strongest cutoff: 1653
0.2
strongest cutoff: 1637
0.2
p-value:3.2e-03
TPR: 0.71
strongest cutoff: 1436
0.0
TNR: 0.61
0.0
TPR: 0.92
TNR: 0.67
0.0
TPR: 0.67
TNR: 0.57
0.0
TPR:0.59
TNR:0.53
1.0
0.8
0.6
0.4
1.0
0.8
0.6
0.4
False positive rate
0.2
0.0
False positive rate
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
False positive rate
False positive rate
0.4
0.2
0.0
A
B
C
HK-2
786-O
A498
Caki-1
Relative LEDGF/p75 expression
LEDGF/p75
1.5
NC
KD1
KD2
1.0
p52
LEDGF/p75
0.5
H3K36me3
**
0.0
ß-Actin
H3
NC
KD1
KD2
D
F
NC
KD1
KD2
400-
786-O
Colony number
300-
**
OD value (450nm)
2.0
200-
**
- NC
Caki-1
100-
1.5
+ KD1
+ KD2
**
0
NC
KD1
KD2
1.0
500-
0.5
786-O
Colony number
400-
300-
*
0.0
**
0
24
48
72
96
200-
100-
Time (h)
0
NC
KD1
KD2
E
Caki-1
G
NC
KD1
KD2
OD value (450nm)
Migration cell number
1500-
3
1000-
NC
KD1
A
Caki-1
500
2.
KD2
*
0
NC
KD1
KD2
1
1500-
786-O
Migration cell number
1000-
0
0
24
48
72
96
500
**
Time (h)
0
NC
KD1
KD2
Interestingly, the results of GO analysis showed that after LEDGE/ p75 knockdown, the activity of several functional proteins, such as the Wnt protein, were changed (Fig. 8D). The p53 signaling pathway was also affected by changes in LEDGF/p75 (Fig. 8F).
4. Discussion
Studies to date on LEDGF/p75 have mainly focused on HIV and MLL diseases, but there are few studies on its role in tumors. In fact, as a reader of histone modification marks, LEDGF/p75 mediates chromatin binding of many nuclear proteins, thus playing an important biological function in tumors[14]. As a main reader of H3K36me3, its potential role in cancer is worthy of further study. Here, for the first time, we comprehensively analyze the LEDGF/p75 landscape across cancers
using multiple online databases, in vitro experiments and gene micro- array sequencing. The present study detailed the basic information, clinical significance, genomic instability, alternative splicing, and can- cer immunity related to LEDGF/p75.
The PWWP and IBD of LEDGF/p75 perform the functions of chro- matin recognition and protein binding, respectively; specifically, pro- cesses of chromatin and DNA binding, transcription regulation, protein-protein interactions, epitope recognition, HIV integration, and stress survival are involved, and homology to the hepatoma-derived growth factor protein family is notable[38]. Accordingly, LEDGF/p75 plays an important role in a variety of biological processes, consistent with the results of our GO analysis, protein interaction prediction, and GSEA across cancers.
LEDGF/p75 expression varied widely among normal groups, tumor
A
B
· Up regulation: 655
· Down regulation: 512
· No significance: 57033
8
Fold change = - 1.50
Fold change = 1.50
6
-log10(P value)
ERO1L
4
2
0.8
Z score expression
0.4
pValue = 0.05
0.0
-0.4 5
0
-0.8N
-4
-2
0
2
4
NC1
NC2
NC3
KD1
KD2
KD3
log2 Fold change (LEDGF/p75-KD vs. NC)
C
All regulated genes
D
Down regulated genes
molecular function
molecular function
cellular component
7 -
cellular component 4
biological process
biological process
pValue=0.05
6
-log10(P value)
pValue=0.05
5
3
-log10(P value)
4 -
2.
3 -
2.
1 .
1
DNA binding transcription factor selvity
protein phosphalas - phosphata ~ actraty AS Misphosphate 5-post fatorescurity RNA polymerase Il proximal promotor sentence
0
motal lof Winding
Mene Inhibitor activity
a vararansterase activity sont
histone methy transtornoverseny
double-stranded DNA binding chromo shadow comail binding
pericentric heterocimamatin synaptic vesicle miaan
nuclear heterogene nucce
integral component of organelle membrane
nuclear pericentric heterdone maan
exon-exon junction complex
autophagosome men wane
auto cription, DNA-fengtated
regulation of transcription, DNAMONOTHed
negative regulation of endothelial ced migration
tary acid tigation
synaptic vesicle docking
response so
negative regulation of JUN kinh cold
regulation of nitric oxide These cavity
Iron-sulfur cluster aunembly
0
double-stranded DNA binding
Wal-protein birto
acyl-CoA dehydrogenase acity flavin adenine dinucleotide biro
Wat-activated receptor ach metalloendopeptidase acht
ARF guanyi-nucleotide exchange factor acuity
gamma-tubran
gamma- kibusin Ce metallopeutdash
metalloperadase om
integral component of membrane
pericentric heterochronntin
Cytoplasmic stress grande
chromosome, centromencourt
reticulum
endoplasmic rescu
negative regulation of endothelial cell migririnn engottendi coppiason
positive regulation of NF-kappab transcription
interferon
response to ne ral process testcase whereren delen a
cotranslational protein targeting to membran
embryonic pattern specific
cellular protein complex localizauy
nitric oxide mediated signal transduuion
drug transmembrane transport
inositol-
E
F
All regulated genes
Down regulated genes
pValue
pValue
Transcriptional misregulation in cancer
p53 signaling pathway
Other types of O-glycan biosynthesis
0.12
Endocytosis
0.15
Lysine degradation
0.1
Cellular senescence
Mannose type O-glycan biosynthesis
Other types of O-glycan biosynthesis
Gap junction
0.075
Vascular smooth muscle contraction
0.1
N-Glycan biosynthesis
0.05
alpha-Linolenic acid metabolism
p53 signaling pathway
Circadian entrainment
Adherens junction
0.025
Herpes simplex infection
0.05
Inositol phosphate metabolism
Glycerophospholipid metabolism
Phosphatidylinositol signaling system
Cell cycle
ListHit
Transcriptional misregulation in cancer
Choline metabolism in cancer
ListHit
MicroRNAs in cancer
Inflammatory bowel disease (IBD)
Cushing syndrome
2
Leukocyte transendothelial migration
1
Vascular smooth muscle contraction
Adherens junction
Regulation of actin cytoskeleton
Cell cycle
Long-term depression
6
Autophagy - animal
4
Fluid shear stress and atherosclerosis
Rheumatoid arthritis
Leukocyte transendothelial migration
Fatty acid degradation
Cholinergic synapse
Inflammatory bowel disease (IBD)
12
Gap junction
Chemokine signaling pathway
8
Cell adhesion molecules (CAMs)
Th1 and Th2 cell differentiation
Glycosaminoglycan biosynthesis
Ubiquinone and other terpenoid-quinone biosynthesis
Homologous recombination
Endocrine and other factor-regulated calcium regulation
Axon guidance
Notch signaling pathway
Amoebiasis
Valine, leucine and isoleucine degradation
Circadian entrainment
Prostate cancer
NF-kappa B signaling pathway
Endocrine and other factor-regulated calcium regulation
Human T-cell leukemia virus 1 infection
Choline metabolism in cancer
Breast cancer
AGE-RAGE signaling pathway in diabetic complication
Gastric cancer
Staphylococcus aureus infection
1.0
1.8
2.6
3.4
4.2
5.0
0
2
4
6
8
Enrichment score
10
Enrichment score
groups and cell lines, suggesting a disease-specific nature of LEDGF/p75. It is worth noting that LEDGF/p75 expression in HEL cells was much higher than that in dozens of other cells. As a cancer cell line derived from myeloid cells, HEL is an erythroleukemia cell line (AML M6 in relapse after treatment for Hodgkin’s disease). Whether there is a deeper connection between erythroleukemia and MLL (known to be dependent on LEDGF/p75) other than both being blood cancers and whether that connection is related to LEDGF/p75 is a question that has not yet been answered.
Compared with normal tissues, LEDGF/p75’s transcription and protein expression levels were lower in BRCA, LUAD, and UCEC and higher levels in LIHC. However, LEDGF/p75 transcriptional and protein levels in COAD, HNSC, OV, and KIRC did not seem to be uniform. There are many reasons for this. TCGA is a database based on tumor data, and the number of normal tissues is far less than the number of tumor tissues, which may cause a degree of error. In addition, posttranscriptional regulation and posttranslational modification have a great impact on transcription and translation level. Therefore, experiments need to be carried out to verify the findings.
Furthermore, we comprehensively analyzed the OS, DSS, DFI, PFI and clinical stages of patients. High expression of LEDGF/p75 in ACC, KICH, and LIHC suggested poor prognosis, with the opposite in PAAD and SKCM.
Genomic instability leads to the development of tumors, including mutation, structural variant, amplification, deep deletion, and multiple alterations. Our study indicates that genomic alterations in LEDGF/p75 occur in multiple cancers and are associated with poor prognosis in patients with TGCT, BRCA, HNSC, and LIHC. We also analyzed the relationship between LEDGF/p75 and tumor immunity. Our study re- ports the immune subtypes of LEDGF/p75, the relationship between LEDGF/p75 and immunoinhibitors, chemokines and TILs, and the effi- cacy of immunotherapy. Our analysis may provide additional treatment options for patients with cancer.
Previous studies have reported that LEDGF/p75 plays different key roles in cancers such as cervical cancer[47], breast cancer[48], ovarian cancer[49], and prostate cancer[50]. However, there is still no report about the function of LEDGF/p75 as a reader of H3K36me3 in kidney cancer.
A high proportion of SETD2 mutations in patients with ccRCC resulted in substantial reduction or even deletion of H3K36me3, natu- rally grouping patients into clinically and therapeutically relevant sub- types. Therefore, studying LEDGF/p75, a key reader of H3K36me3, is of great significance for clinical diagnosis and treatment of ccRCC patients. We selected 786-O and Caki-1 cells with high H3K36me3 expression to perform functional experiments after LEDGF/p75 was knocked down. The results suggested that LEDGF/p75 is a potential oncogene in ccRCC. Interestingly, LEDGF/p75 is significantly protected from mutations in the ccRCC cohort, which is consistent with its role as an oncogene. Therefore, targeting LEDGF/p75 interference may be a feasible personalized therapy for ccRCC patients without SETD2 mutation. However, the current experimental results are only a preliminary exploration of this hypothesis, and experiments are still needed for rigorous verification in the future.
To further explore the role of LEDGF/p75 in ccRCC, we knocked down LEDGF/p75 and performed gene microarray analysis. Considering that H3K36me3 is a transcriptional activation mark, we focused on the genetic functions that were lowered after LEDGF/p75 knockdown. Interestingly, we found that the p53 signaling pathway changed, which is worthy of further research. After LEDGF/p75 was knocked down, ERO1L was the most significantly decreased protein-coding gene among 512 downregulated genes. Overexpression of ERO1L, an endoplasmic reticulum oxidase, is related to the development and progression of many cancers, such as lung adenocarcinoma, glioblastoma and low- grade glioma, pancreatic ductal adenocarcinoma, and kidney renal papillary cell carcinoma[51]. Whether there is a regulatory axis of LEDGF/p75-ERO1L in SETD2 nonmutant ccRCC is worth exploring.
In summary, we performed multidirectional analysis of LEDGF/p75 across cancers and identified it as a prognostic biomarker. The present study preliminarily explored its potential function in kidney cancer, especially for personalized treatment of patients with SETD2 nonmutant ccRCC.
Funding
This study was funded by Wuxi Taihu Lake Talent Plan, Leading Talents in Medical and Health Profession Project: Research and appli- cation of early screening and accurate diagnosis and treatment of prostate cancer (THRCJH20200104).
CRediT authorship contribution statement
Ninghan Feng, Bing Yao, and Lin Jiang designed this study and provided clinical guidance as well as data interpretation. Yuwei Zhang, Wei Guo, and Yangkun Feng performed the analyses and experiments. Yuwei Zhang and Wei Guo prepared the figures for this study. Longfei Yang, Hao Lin, Pengcheng Zhou and Kejie Zhao checked the data. Yuwei Zhang drafted the article. All authors reviewed the manuscript, provided comments and approved the final version.
Declaration of Competing Interest
The authors declare that they have no conflicts of interest.
Data Availability
The data of this study are available from the corresponding author on reasonable request.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.csbj.2023.08.023.
References
[1] Ge H, Si Y, Roeder RG. Isolation of cDNAs encoding novel transcription coactivators p52 and p75 reveals an alternate regulatory mechanism of transcriptional activation. EMBO J 1998;17:6723-9.
[2] Nishizawa Y, Usukura J, Singh DP, Chylack Jr LT, Shinohara T. Spatial and temporal dynamics of two alternatively spliced regulatory factors, lens epithelium- derived growth factor (ledgf/p75) and p52, in the nucleus. Cell Tissue Res 2001; 305:107-14.
[3] Llano M, Saenz DT, Meehan A, Wongthida P, Peretz M, Walker WH, et al. An essential role for LEDGF/p75 in HIV integration. Science 2006;314:461-4.
[4] Okuda H, Kanai A, Ito S, Matsui H, Yokoyama A. AF4 uses the SL1 components of RNAP1 machinery to initiate MLL fusion- and AEP-dependent transcription. Nat Commun 2015;6:8869.
[5] Schroder AR, Shinn P, Chen H, Berry C, Ecker JR, Bushman F. HIV-1 integration in the human genome favors active genes and local hotspots. Cell 2002;110:521-9.
[6] Ciuffi A, Bushman FD. Retroviral DNA integration: HIV and the role of LEDGF/p75. Trends Genet 2006;22:388-95.
[7] Wang GP, Ciuffi A, Leipzig J, Berry CC, Bushman FD. HIV integration site selection: analysis by massively parallel pyrosequencing reveals association with epigenetic modifications. Genome Res 2007;17:1186-94.
[8] Canella A, Van Belle S, Brouns T, Nigita G, Carlon MS, Christ F, et al. LEDGF/p75- mediated chemoresistance of mixed-lineage leukemia involves cell survival pathways and super enhancer activators. Cancer Gene Ther 2022;29:133-40.
[9] Yokoyama A, Cleary ML. Menin critically links MLL proteins with LEDGF on cancer-associated target genes. Cancer Cell 2008;14:36-46.
[10] De Rijck J, Bartholomeeusen K, Ceulemans H, Debyser Z, Gijsbers R. High- resolution profiling of the LEDGF/p75 chromatin interaction in the ENCODE region. Nucleic Acids Res 2010;38:6135-47.
[11] Sharma S, Cermakova K, De Rijck J, Demeulemeester J, Fabry M, El Ashkar S, et al. Affinity switching of the LEDGF/p75 IBD interactome is governed by kinase- dependent phosphorylation. Proc Natl Acad Sci USA 2018;115:E7053-62.
[12] Huang J, Gurung B, Wan B, Matkar S, Veniaminova NA, Wan K, et al. The same pocket in menin binds both MLL and JUND but has opposite effects on transcription. Nature 2012;482:542-6.
[13] Cermakova K, Demeulemeester J, Lux V, Nedomova M, Goldman SR, Smith EA, et al. A ubiquitous disordered protein interaction module orchestrates transcription elongation. Science 2021;374:1113-21.
[14] Xiao C, Fan T, Tian H, Zheng Y, Zhou Z, Li S, et al. H3K36 trimethylation-mediated biological functions in cancer. Clin Epigenetics 2021;13:199.
[15] Pradeepa MM, Sutherland HG, Ule J, Grimes GR, Bickmore WA. Psip1/Ledgf p52 binds methylated histone H3K36 and splicing factors and contributes to the regulation of alternative splicing. PLoS Genet 2012;8:e1002717.
[16] Turlure F, Maertens G, Rahman S, Cherepanov P, Engelman A. A tripartite DNA- binding element, comprised of the nuclear localization signal and two AT-hook motifs, mediates the association of LEDGF/p75 with chromatin in vivo. Nucleic Acids Res 2006;34:1653-65.
[17] Hsieh JJ, Purdue MP, Signoretti S, Swanton C, Albiges L, Schmidinger M, et al. Renal cell carcinoma. Nat Rev Dis Prim 2017;3:17009.
[18] Hsieh JJ, Le VH, Oyama T, Ricketts CJ, Ho TH, Cheng EH. Chromosome 3p Loss- Orchestrated VHL, HIF, and Epigenetic Deregulation in Clear Cell Renal Cell Carcinoma. J Clin Oncol 2018;36. JCO2018792549.
[19] Jonasch E, Walker CL, Rathmell WK. Clear cell renal cell carcinoma ontogeny and mechanisms of lethality. Nat Rev Nephrol 2021;17:245-61.
[20] Zhang Y, Fang Y, Tang Y, Han S, Jia J, Wan X, et al. SMYD5 catalyzes histone H3 lysine 36 trimethylation at promoters. Nat Commun 2022;13:3190.
[21] Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, et al. The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr Protoc Bioinforma 2016;54. 1 30 31-31 30 33.
[22] C. UniProt, UniProt: the Universal Protein Knowledgebase in 2023, Nucleic Acids Res, (2022).
[23] Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature 2021;596:583-9.
[24] Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oksvold P, Mardinoglu A, et al. Proteomics. Tissue-based map of the human proteome. Science 2015;347:1260419.
[25] Warde-Farley D, Donaldson SL, Comes O, Zuberi K, Badrawi R, Chao P, et al. The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res 2010;38. W214-220.
[26] Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, et al. The STRING database in 2021: customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res 2021; 49:D605-12.
[27] Mitsopoulos C, Di Micco P, Fernandez EV, Dolciami D, Holt E, Mica IL, et al. canSAR: update to the cancer translational research and drug discovery knowledgebase. Nucleic Acids Res 2021;49:D1074-82.
[28] Ru B, Wong CN, Tong Y, Zhong JY, Zhong SSW, Wu WC, et al. TISIDB: an integrated repository portal for tumor-immune system interactions. Bioinformatics 2019;35:4200-2.
[29] Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 2005;102:15545-50.
[30] Goldman MJ, Craft B, Hastie M, Repecka K, McDade F, Kamath A, et al. Visualizing and interpreting cancer genomics data via the Xena platform. Nat Biotechnol 2020; 38:675-8.
[31] Tang Z, Li C, Kang B, Gao G, Li C, Zhang Z. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res 2017; 45:W98-102.
[32] Chandrashekar DS, Karthikeyan SK, Korla PK, Patel H, Shovon AR, Athar M, et al. UALCAN: an update to the integrated cancer data analysis platform. Neoplasia 2022;25:18-27.
[33] Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401-4.
[34] Zhang Y, Yao X, Zhou H, Wu X, Tian J, Zeng J, et al. OncoSplicing: an updated database for clinically relevant alternative splicing in 33 human cancers. Nucleic Acids Res 2022;50:D1340-7.
[35] Fu J, Li K, Zhang W, Wan C, Zhang J, Jiang P, et al. Large-scale public data reuse to model immunotherapy response and resistance. Genome Med 2020;12:21.
[36] Zeng Z, Wong CJ, Yang L, Ouardaoui N, Li D, Zhang W, et al. TISMO: syngeneic mouse tumor database to model tumor immunity and immunotherapy response. Nucleic Acids Res 2022;50:D1391-7.
[37] 1 Fekete JT, Gyorffy B. ROCplot.org: Validating predictive biomarkers of chemotherapy/hormonal therapy/anti-HER2 therapy using transcriptomic data of 3,104 breast cancer patients. Int J Cancer 2019;145:3140-51.
[38] Cermakova K, Weydert C, Christ F, De Rijck J, Debyser Z. Lessons Learned: HIV Points the Way Towards Precision Treatment of Mixed-Lineage Leukemia. Trends Pharm Sci 2016;37:660-71.
[39] Wang H, Farnung L, Dienemann C, Cramer P. Structure of H3K36-methylated nucleosome-PWWP complex reveals multivalent cross-gyre binding. Nat Struct Mol Biol 2020;27:8-13.
[40] El Ashkar S, Schwaller J, Pieters T, Goossens S, Demeulemeester J, Christ F, et al. LEDGF/p75 is dispensable for hematopoiesis but essential for MLL-rearranged leukemogenesis. Blood 2018;131:95-107.
[41] Ui A, Chiba N, Yasui A. Relationship among DNA double-strand break (DSB), DSB repair, and transcription prevents genome instability and cancer. Cancer Sci 2020; 111:1443-51.
[42] Bonnal SC, Lopez-Oreja I, Valcarcel J. Roles and mechanisms of alternative splicing in cancer - implications for care. Nat Rev Clin Oncol 2020;17:457-74.
[43] Xie Y, Sahin M, Sinha S, Wang Y, Nargund AM, Lyu Y, et al. SETD2 loss perturbs the kidney cancer epigenetic landscape to promote metastasis and engenders actionable dependencies on histone chaperone complexes. Nat Cancer 2022;3: 188-202.
[44] Ortiz-Hernandez GL, Sanchez-Hernandez ES, Ochoa PT, Elix CC, Alkashgari HR, McMullen JRW, et al. The LEDGF/p75 Integrase Binding Domain Interactome Contributes to the Survival, Clonogenicity, and Tumorsphere Formation of Docetaxel-Resistant Prostate Cancer Cells. Cells 2021;10.
[45] Roudaia L, Speck NA. A MENage a Trois in leukemia. Cancer Cell 2008;14:3-5.
[46] Daugaard M, Kirkegaard-Sorensen T, Ostenfeld MS, Aaboe M, Hoyer-Hansen M, Orntoft TF, et al. Lens epithelium-derived growth factor is an Hsp70-2 regulated guardian of lysosomal stability in human cancer. Cancer Res 2007;67:2559-67.
[47] Leitz J, Reuschenbach M, Lohrey C, Honegger A, Accardi R, Tommasino M, et al. Oncogenic human papillomaviruses activate the tumor-associated lens epithelial- derived growth factor (LEDGF) gene. PLoS Pathog 2014;10:e1003957.
[48] Singh DK, Gholamalamdari O, Jadaliha M, Ling X, Lin Li, YC, Zhang Y, et al. PSIP1/ p75 promotes tumorigenicity in breast cancer cells by promoting the transcription of cell cycle genes. Carcinogenesis 2017;38:966-75.
[49] Sapoznik S, Cohen B, Tzuman Y, Meir G, Ben-Dor S, Harmelin A, et al. Gonadotropin-regulated lymphangiogenesis in ovarian cancer is mediated by LEDGF-induced expression of VEGF-C. Cancer Res 2009;69:9306-14.
[50] Liedtke V, Rose L, Hiemann R, Nasser A, Rodiger S, Bonaventura A, et al. Over- expression of LEDGF/p75 in HEp-2 cells enhances qutoimmune IgG response in patients with benign prostatic hyperplasia-a novel diagnostic approach with therapeutic consequence? Int J Mol Sci 2023;24.
[51] Shergalis AG, Hu S, Bankhead 3rd A, Neamati N. Role of the ERO1-PDI interaction in oxidative protein folding and disease. Pharm Ther 2020;210:107525.