MicroRNA expression and gene regulation drive breast cancer progression and metastasis in PyMT mice
Global miRNA expression profiles and potential targets in PyMT mice
There are 151 differentially expressed miRNAs at one or more time points during breast cancer progression in PyMT mice. They can be separated into two groups: 75 miRNAs over-expressed in tumor and the rest 76 under-expressed. 80 miRNAs are differentially expressed at two or more time points and, significantly, all of them show consistent directions of differential expression at different time points. This dichotomous differential expression pattern suggests that these miRNAs function as promoters or suppressors in breast cancer development. You can download the expression profiles of microRNAs and mRNAs here
We compared the expression profiles of 94 miRNAs differentially expressed in PyMT mice with their corresponding human orthologs in miRCancer and found a large degree of concordance in their expression patterns. When only breast cancer was considered, 77.14% of the intersecting miRNAs share the same expression direction (up- or down-regulation) in cancer between human and mice. When all cancer types were considered, the concordance rate increased to 89.06%.
Functional assessment of differential miRNAs regulation
Functional enrichment analysis was carried out via Genecodis tool including Gene Ontology Biological Process, KEGG Pathways and Panther Pathways annotations.
More regulated genes become involved in cancer-related biological processes and pathways, including cell adhesion, cell differentiation, multicellular organismal development, tight junction, and cell adhesion molecules along the cancer progression. You can download the functional analysis results here
By comparing the enriched terms with hallmarks of cancer propossed by Hanahan and Weinberg it can be seen that miRNA regulatory activity is focused on three different aspects of cancer and this regulation is more active during the tumor transitions from adenoma to carcinoma.
Transition patterns of miRNAs expression during cancer progression.
Transitions were calculated as the difference of expressions between two consecutive time points (t+1 and t). Based on their distribution and the magnitude and sense of their deviation from the mean, each transition was discretized as positive (+), negative (-) or flat (0). Because there are three classes for each of the three transitions between four consecutive time points, there are 27 different transition patterns: '+++', '++0', '++–', and so on. In our downstream analysis, we focused on miRNAs with a non-increase (involving only '0' or '+' transitions) or non-decrease (involving only '0' or '–' transitions) trend in their expression. The whole set of patterns and microRNAs can be downloaded here
| Pattern | microRNAs |
|---|---|
| [-00] | miR-669l, miR-669e, miR-669b, miR-3473c, miR-3964, miR-504, miR-201, miR-150, miR-708, miR-1958, miR-874 |
| [0-0] | miR-190b |
| [00-] | miR-137, miR-698, miR-133b, miR-133a, miR-203, miR-206 |
| [+0+] | miR-217 |
| [+00] | miR-669k, miR-3470b, miR-690, miR-196a, miR-1956, miR-543, miR-546, miR-1940 |
| [0++] | miR-325 |
| [0+0] | miR-694, miR-802, miR-3970, miR-672 |
| [00+] | miR-692, miR-376b, miR-204, miR-688, miR-1927, miR-326, miR-5107, miR-375, miR-1187, miR-1188, miR-541, miR-1945 |
Functional-Regulatory programs
The regulatory mechanisms of any biological processes involve more than one microRNA and its gene targets. MicroRNA's target prediction, gene expression profiles and functional annotations were integrated to obtain a global picture of miRNA gene regulatory programs behind breast cancer development. This data integration enabled us to identify regulatory modules or units from the mRNA-miRNA network at considering each studied cancer phase independently, but also microRNAs with positive or negative trends along cancer progression. The former can be downloaded here and the later here.
Some of these regulatory modules appear concurrently along the whole progression or in some sequential phases. Interestingly, most of them are specially related with several transmembrane circulations:
| Module | #genes | #microRNAs | Recurrent regulated genes |
|---|---|---|---|
| ABC transporters | 22 | 69 | DNAH8 ABCC5 ABCA8A ABCB4 ABCA8B |
| GPCR + Cell surface receptor | 18 | 63 | ADGRB2 ADGRE5 ADGRG3 ADGRL1 ADGRL4 FZD9 L1CAM |
| Fatty acid metabolism | 16 | 57 | ACSL1 ACADL ECH1 ECHS1 HPGD PPARA SNCA |
| Citokine-citokine receptor + Inmune response | 50 | 100 | IL1B CX3CR1 PDGFRA GDF5 PDGFRB CXCL15 CXCL1 TNFRSF11A IL7R SBSPON BMP6 |
| Epidermal growth factor | 32 | 87 | C1RA CCBE1 BMP1 NPNT NRG4 CD93 DLK1 ADGRE5 FAT4 WIF1 LAMA4 ADGRL4 TLL1 ADAM19 |
| Cell adhesion molecules | 41 | 86 | JAM3 MPZ PTPRM MPZL1 CADM3 NLGN2 |
| Calcium ion transport | 16 | 56 | ATP2C2 GJA4 CACHD1 TRPM6 CACNA1H RYR2 CYP27B1 CACNA1A RAMP2 JPH2 |
| Calcium signaling + GPCR, rodopsion-like domain | 40 | 80 | MYLK NOS1 PDGFRA AGTR1A PTGER3 HRH1 PDGFRB HTR7 PTGER1 HTR4 CACNA1H ADRB3 RYR2 SLC8A2 ADCY3 CACNA1A HTR2A LPAR2 S1PR1 TSPO |
| Fibronectin domain | 49 | 101 | CHL1 PHYHIP SPEG EPHA2 MYLK PTPRM MYOM3 |
| GTP-binding + RAS GTPase | 22 | 68 | RAB36 RAB3D RAB15 |
| Angiogenesis | 18 | 52 | EPHA2 ANGPT2 JAM3 TNFAIP2 CCBE1 ENPEP FGF1 CYP1B1 EFNB2 S1PR1 TGFA RAMP2 MMP19 |
| Axon guidance + Semaphorin domain | 23 | 76 | SRGAP3 SEMA4G EPHA2 SEMA3A PLXNA1 EFNB1 SEMA3C DPYSL2 NFATC2 RGS3 L1CAM ABLIM2 LRRC4C SLIT3 PLXND1 ITGB3 |
On example is the regulatory module involved in the fatty acid metabolism. There are in total 31 genes and 67 miRNAs participating in this module, and only 7 of them appear to be simultaneously regulated by microRNAs the 3 different time points (ACSL1, ACADL, ECH1, ECHS1, HPGD, PPARA and SNCA). After studying their expression profiles along all breast cancer subtypes in TCGA dataset, we found OLR1, FAAH consistently overexpressed and SNCA, ME1, PPARA and PPARG under-expressed mostly in Her2, Luminal A and B subtypes. These genes show accordingly expressed in PyMT model at different phases of cancer progression or along the whole tumor development. MicroRNAs persistently regulating these genes in our model are miR-143, miR-27b, miR-141, miR-200a and miR-148a
Suplementary material
Detailed results of this study can be found in the following links:
- Expression profiles of differentially expressed microRNAs [Download]
- Expression profiles of mRNAs [Download]
- Hallmarks of cancer and Enriched biological annotations correspondence [Download]
- Novel microRNAs and potentially regulated targets [Download]
- Functional terms enriched in regulated genes in week 6, 8, 10 and 12 of cancer progression [Download]
- Regulatory modules considering each cancer phase independently [Download]
- Regulatory modules considering miRNAs with positive or negative trends along the cancer progression [Download]
Contact
Albert Einstein College of Medicine
Michael F. Price Center
1301 Morris Park Avenue , Room 353A
Bronx, NY 10461
Tel: 718.678.1139
zhengdong.zhang | ruben.nogales at einstein.yu.edu
You can also visit the ZDZ Laboratory Web Site