Identification of key genes and miRNA-mRNA regulatory networks associated with bone marrow immune microenvironment regulations in multiple myeloma by integrative bioinformatics analysis

Figures & data

Table 1. Top 10 upregulated and top 10 downregulated DEmiRs in multiple myeloma.

ID	Adjusted P value	P value	LogFC	miRNA name
6193	1.52E-02	1.05E-04	2.536	hsa-miR-4301
4589	2.17E-02	1.89E-04	2.452	hsa-miR-548al
35638	3.73E-02	4.62E-04	2.443	hsa-miR-4693-3p
15327	4.67E-03	1.38E-05	2.424	hsa-miR-711
47809	4.51E-02	6.14E-04	2.403	hsa-miR-4799-3p
57759	3.70E-02	4.49E-04	2.331	hsa-miR-5583-5p
6709	1.52E-02	1.05E-04	2.222	hsa-miR-299-5p
50274	2.78E-02	2.82E-04	2.204	hsa-miR-6130
38405	1.83E-02	1.44E-04	2.196	hsa-miR-4481
50878	2.57E-02	2.46E-04	2.188	hsa-miR-758-3p
6629	2.05E-02	1.72E-04	−2.589	hsa-miR-4671-5p
3372	1.42E-02	9.12E-05	−2.607	hsa-miR-4681
15233	8.46E-03	3.76E-05	−2.638	hsa-miR-4755-5p
58185	1.16E-04	3.77E-08	−2.668	hsa-miR-1273a
60712	2.71E-02	2.74E-04	−2.699	hsa-miR-15b-3p
25588	1.16E-04	3.91E-08	−2.706	hsa-miR-3177-3p
25883	4.82E-02	6.78E-04	−2.762	hsa-miR-2355-3p
59150	9.95E-04	1.21E-06	−3.046	hsa-miR-4744
49869	3.20E-02	3.60E-04	−3.117	hsa-miR-378a-3p
49708	1.54E-02	1.09E-04	−3.296	hsa-miR-23a-5p

Table 2. Downregulated DemiRs are associated with the enrichment of KEGG pathways. MicroT web server v5.0 [22], and DIANA-TarBase v7.0 [23] are used for identifying common pathways.

KEGG pathways	P-value (MicroT-CDS v5.0)	P-value (TarBase v7.0)
Proteoglycans in cancer	3.22E-09	3.73E-14
Pathways in cancer	5.33E-09	3.69E-04
Hippo signaling pathway	7.22E-09	6.80E-07
Glioma	2.48E-07	8.62E-06
Phosphatidylinositol signaling system	2.48E-07	2.29E-02
TGF-beta signaling pathway	7.16E-07	1.92E-04
ErbB signaling pathway	1.21E-06	3.69E-04
Signaling pathways regulating pluripotency of stem cells	1.31E-06	4.57E-03
Fatty acid metabolism	3.02E-06	1.22E-11
Thyroid hormone signaling pathway	4.32E-05	1.92E-04
Wnt signaling pathway	5.35E-05	1.39E-02
Renal cell carcinoma	6.02E-05	8.33E-07
Fatty acid biosynthesis	6.74E-05	1.69E-03
Chronic myeloid leukemia	9.04E-05	8.62E-06
Focal adhesion	9.04E-05	3.62E-04
Endocytosis	1.99E-04	3.62E-04
Melanoma	2.48E-04	1.76E-02
Prostate cancer	2.50E-04	3.13E-05
Adherens junction	4.38E-04	1.30E-06
Colorectal cancer	6.91E-04	9.67E-04
Pancreatic cancer	9.28E-04	1.47E-03
Endometrial cancer	1.04E-03	5.51E-04
Neurotrophin signaling pathway	1.06E-03	5.67E-04
Acute myeloid leukemia	1.22E-03	6.69E-03
PI3K-Akt signaling pathway	1.22E-03	2.59E-02
Non-small cell lung cancer	1.38E-03	1.09E-04
FoxO signaling pathway	2.93E-03	1.26E-04
Oocyte meiosis	2.93E-03	3.32E-06
Ubiquitin mediated proteolysis	3.30E-03	3.32E-06
mTOR signaling pathway	3.79E-03	1.41E-04
Fatty acid degradation	5.89E-03	2.50E-05
p53 signaling pathway	6.28E-03	6.93E-04
Viral carcinogenesis	6.83E-03	1.06E-07
Prion diseases	7.62E-03	1.30E-06
Central carbon metabolism in cancer	8.51E-03	5.73E-04
Hepatitis B	1.02E-02	1.92E-04
Prolactin signaling pathway	1.14E-02	8.72E-03
Transcriptional misregulation in cancer	1.14E-02	4.52E-04
Regulation of actin cytoskeleton	1.17E-02	3.44E-02
MAPK signaling pathway	1.66E-02	3.35E-02
Bacterial invasion of epithelial cells	1.82E-02	2.50E-05
AMPK signaling pathway	2.05E-02	5.51E-04
Lysine degradation	2.12E-02	1.70E-06
Thyroid cancer	2.77E-02	1.44E-03
TNF signaling pathway	2.87E-02	7.92E-04
HIF-1 signaling pathway	3.06E-02	2.59E-02
Estrogen signaling pathway	4.05E-02	2.48E-07
N-Glycan biosynthesis	4.17E-02	1.25E-02

Figure 1. Construction of cluster-specific PPI of hub genes and cluster-specific pathway enrichment analysis. A. Hub genes of cluster 1 involved with PPI. B. KEGG pathways that are significantly associated with hub genes of cluster 1. C. Hub genes of cluster 2 involved with PPI. D. Three KEGG pathways are enriched in Cluster 2. E. PPI of hub genes in cluster 3. F. Three KEGG pathways are enriched associated with the hub genes of cluster 3. G-H. PPI of hub genes in cluster 4 and cluster 5.

Table 3. MCODE identified hub gene-associated five significant clusters in MM.

Cluster	Score	Nodes	Edges	Hub genes
1	13.2	31	198	UBA52, TOP1, CENPE, ANAPC7, RBL1, CDKN1A, TFDP1, RAD50, SMC6, CDKN2A, ORC3, TOP2A, BORA, SKP2, RRM1, MCM8, PRIM1, ANAPC4, RBBP4, RFC5, NCAPD2, WRN, MSH2, CDC16, CCNB1, RB1, CCNE1, CDC23, PCNA, MDC1, RAD51C
2	10.32	26	129	BTRC, SPDL1, KIF20A, CENPF, SLBP, FBXW8, NUP37, SMC1A, WDHD1, HUWE1, CDC20, NUP107, FBXO5, UBE2O, NCAPG, XPO1, RNF7, ESCO2, FBXL20, PSMD1, FBXO7, NUP160, PSMB8, DSN1, MKI67, CKAP5
3	5.091	12	28	BPTF, BRCA2, GTF2H1, CDC6, POLR2B, RFC1, ATF2, ORC4, ORC2, BRCA1, TP53, TOP2B
4	4.667	7	14	HDAC2, CTCF, TBP, SMARCC2, KDM1A, SUPT16H, SMARCC1
5	2.727	12	15	CXCL12, NOC3L, CAT, CEP290, WDR36, NAT10, PRKAR2B, BBS10, IGF1, NEDD1, STAT5B, GART

Figure 2. Differentially expressed hub genes were targeted by DEmiRs. A. Identification of upregulated and downregulated hub genes. B. The upregulated 9 hub genes interact with downregulated DEmiRs. C. The downregulated 52 hub genes interact with upregulated DEmiRs.

Figure 3. DEmiRs targeting hubs genes are associated with survival prognosis in multiple myeloma patients. Kaplan-Meier curves show that the expression of two hub genes (CDKN2A and TP53) correlates with survival prognosis (Disease-specific survival) in MM patients (log-rank test, P < 0.05).

Figure 4. The expression level of CDKN2A is associated with the regulation of the tumor microenvironment in MM. A. Immune scores and microenvironment scores are negatively correlated with the expression levels of CDKN2A. B. Immune signatures are significantly correlated with the expression levels of CDKN2A.

Figure 5. Positive correlation of CDKN2A with immunosuppressive markers in MM microenvironment. Immunosuppressive markers PDCD1 (PD-1), CTLA4, FOXP3, HHLA2, CD70, TMIGD2, TNFSF9, and IL5 are positively correlated with the expression levels of CDKN2A in the microenvironment of MM.

Figure 6. The expression validation of CDKN2A and its co-regulatory DEmiRs in the independent MM cohorts. A. The expression level of CDKN2A co-regulatory miRNAs: hsa-mir-492, hsa-mir-24, hsa-mir-192, and hsa-mir-10b are consistently downregulated in the GSE16558. B. The expression level of CDKN2A is consistently upregulated in the GSE16558 [17] and GSE13591[18].

Figure 7. MiRNA-CDKN2A interactions in MM. The downregulation of miRNAs interacts with the upregulation of CDKN2A. This co-regulatory network is associated with the immunosuppressive TME of MM.

Paraskevopoulou MD, Georgakilas G, Kostoulas N, et al. DIANA-MicroT Web server v5.0: service integration into MiRNA Functional analysis workflows. Nucleic Acids Res. 2013;41:W169–W173. doi:10.1093/nar/gkt393.

 PubMed Web of Science ®Google Scholar

Vlachos IS, Paraskevopoulou MD, Karagkouni D, et al. DIANA-TarBase v7.0: indexing more than half a million experimentally supported MiRNA:MRNA interactions. Nucleic Acids Res. 2015;43:D153–D159. doi:10.1093/nar/gku1215.

 PubMed Web of Science ®Google Scholar

Gutiérrez NC, Sarasquete ME, Misiewicz-Krzeminska I, et al. Deregulation of MicroRNA expression in the different genetic subtypes of Multiple Myeloma and correlation with gene expression profiling. Leukemia. 2010;24:629–637. doi:10.1038/leu.2009.274.

 PubMed Web of Science ®Google Scholar

Agnelli L, Mosca L, Fabris S, et al. A SNP Microarray and FISH-based procedure to detect allelic imbalances in Multiple myeloma: An Integrated Genomics Approach Reveals a Wide gene dosage effect. Genes. Chromosomes Cancer. 2009;48:603–614. doi:10.1002/gcc.20668.

 PubMed Web of Science ®Google Scholar

Data availability

The GEO datasets GSE125363, GSE125361, GSE47552, GSE2658, GSE136324, GSE16558, and GSE13591 used in this study are available in The National Biotechnology Information Center Gene (NCBI-Gene) database (https://www.ncbi.nlm.nih.gov/gene).