CC BY
17
ISOFORMS OF MIR-148A AND MIR-203A ARE PUTATIVE SUPPRESSORS OF COLORECTAL CANCER
Nersisyan SA1-2E3
1 National Research University Higher School of Economics (HSE), Moscow, Russia
2 Institute of Molecular Biology (IMB) of the National Academy of Sciences of the Republic of Armenia, Yerevan, Armenia
MicroRNAs are short non-coding molecules which regulate translation in a gene-specific manner. MicroRNA isoforms that differ by few extra or missing nucleotides at the 5'-terminus (5'-isomiR) show strikingly different target specificity. This study aimed to identify functional roles of 5'-isomiR in colorectal cancers. Transcriptomic targets of microRNA isoforms were predicted using bioinformatics tools miRDB and TargetScan. The sets of putative targets identified for 5'-isomiR were integrated with mRNA and microRNA sequencing data for primary colorectal tumors retrieved from The Cancer Genome Atlas Colon Adenocarcinoma (TCGA-COAD) database. The network of interactions among miRNA, their targets and transcription factors was built using the miRGTF-net algorithm. The results indicate that microRNA isoforms highly expressed in colorectal cancer and differing by a single nucleotide position at the 5'-terminus have < 30% common targets. The regulatory network of interactions enables identification of the most engaged microRNA isoforms. Anti-correlated expression levels of canonical microRNA hsa-miR-148a-3p and its putative targets including CSF1, ETS1, FLT1, ITGA5, MEIS1, MITF and RUNX2 proliferation regulators suggest an anti-tumor role for this molecule. The canonical microRNA hsa-miR-203a-3p|0 and its 5'-isoform bind different sets of anti-correlated putative targets, although both of them interact with genes involved in the epithelial-mesenchymal transition: SNA12 and TNC.
Keywords: isomiR, colorectal cancer, regulatory networks, miRGTF-net, TCGA
Funding: the study was supported by HSE Basic Research Program.
Acknowledgement: the author thanks Aleksey Galatenko of the HSE Laboratory of Molecular Physiology for the fruitful critique and valuable comments.
Compliance with ethical standards: the study complies with the ethical principles of the World Medical Association Declaration of Helsinki.
Correspondence should be addressed: Stepan A. Nersisyan Vavilova, 7, Moscow, 117312, Russia; [email protected]
Received: 29.04.2022 Accepted: 22.05.2022 Published online: 30.05.2022
DOI: 10.24075/brsmu.2022.028
ИЗОФОРМЫ микроРНК MIR-148A И MIR-203A ПРЕДПОЛОЖИТЕЛЬНО ИГРАЮТ РОЛЬ СУПРЕССОРОВ КОЛОРЕКТАЛЬНОГО РАКА
С. А. Нерсисян1'2^
1 Национальный исследовательский университет «Высшая школа экономики», Москва, Россия
2 Институт молекулярной биологии Национальной академии наук Республики Армения, Ереван, Армения
Изоформы микроРНК — класс коротких некодирующих РНК, осуществляющих регуляцию экспрессии генов. Изоформы микроРНК отличаются от канонических микроРНК несколькими нуклеотидами на концах молекулы, причем вариации с 5'-концов микроРНК изменяют множество генов-мишеней. Целью работы было провести анализ функциональной активности 5'-изоформ микроРНК в тканях колоректального рака. Мишени 5'-изоформ микроРНК были предсказаны с помощью биоинформатических программ miRDB и TargetScan. Полученные данные о мишенях 5'-изоформ микроРНК были интегрированы с данными секвенирования мРНК и изоформ микроРНК образцов первичных колоректальных опухолей проекта The Cancer Genome Atlas Colon Adenocarcinoma. Для построения сети взаимодействий изоформ микроРНК, их мишеней и транскрипционных факторов по интегрированным данным использовали алгоритм miRGTF-net. Показано, что высокоэкспрессированные при колоректальном раке изоформы микроРНК, различающиеся одним нуклеотидом на 5'-конце молекулы, имеют не более 30% общих мишеней. В регуляторной сети взаимодействий выявлены наиболее активные изоформы микроРНК. Уровни экспрессий канонической микроРНК hsa-miR-148a-3p и ее предсказанных мишеней, являющихся регуляторами клеточной пролиферации (CSF1, ETS1, FLT1, ITGA5, MEIS1, MITF, RUNX2), были значимо отрицательно коррелированы, откуда может следовать противоопухолевая роль данной молекулы. Каноническая микроРНК hsa-miR-203a-3p|0 и ее 5'-изоформа были антикоррелированы с различными генами-мишенями, но при этом обе потенциально подавляли экспрессию генов, вовлеченных в эпителиально-мезенхимный переход: SNAI2 и TNC.
Ключевые слова: изоформы микроРНК, колоректальный рак, регуляторные сети, miRGTF-net, TCGA Финансирование: исследование осуществлено в рамках Программы фундаментальных исследований НИУ ВШЭ.
Благодарности: Алексею Галатенко из лаборатории молекулярной физиологии НИУ ВШЭ за критику авторских идей и ценные замечания.
Соблюдение этических стандартов: исследование проведено с соблюдением этических принципов Хельсинкской декларации Всемирной медицинской ассоциации.
РК1 Для корреспонденции: Степан Ашотович Нерсисян
ул. Вавилова, д. 7, г Москва, 117312, Россия; [email protected]
Статья получена: 29.04.2022 Статья принята к печати: 22.05.2022 Опубликована онлайн: 30.05.2022 DOI: 10.24075/vrgmu.2022.028
MicroRNAs are short non-coding RNA molecules which regulate gene expression at the post-transcriptional level [1]. Effective binding of microRNA to their target transcripts depends on perfect match between seed-region (nucleotides 2-7 counting from the 5'-terminus of microRNA) and its complementary site in mRNA [2]. The microRNA-mRNA binding inhibits translation and may also promote mRNA degradation provided a sufficient
number of complementary bonds outside the seed-region [2]. Various microRNAs act as tumor suppressors and oncogenes in many cancers [3-5].
MicroRNA processing endonucleases Drosha and Dicer may cleave the long precursor molecules with certain imprecision, giving rise to microRNA isoforms with few extra or missing nucleotides at the termini [6]. Meanwhile, the exact positioning of
5'-terminus is pivotal for target specificity, as it defines the seed-region. MicroRNA 5'-isoforms (5'-isomiR) differing by a single extra or missing terminal nucleotide may target quite different sets of transcripts compared with the prototype canonical microRNA.
Colorectal cancer (CRC) is the third most common and the second most fatal cancer globally [7]. MicroRNAs have been widely implicated in the mechanisms of CRC progression and metastasis. For instance, miR-200 family has been shown to inhibit ZEB1 and ZEB2 genes that encode key transcription factors (TF) of the epithelial-mesenchymal transition (EMT) characteristic of CRC [8]. Accordingly, suppression of miR-200 at the level of transcription and/or processing facilitates EMT and metastasis [8]. MicroRNA expression profiles are actively employed as a source of putative CRC markers, both diagnostic and prognostic [9]. Contemporary research on the role of microRNA isoforms in CRC is focused on comparative evaluation of their expression levels in tumors and matching healthy tissues [10]. To the best of our knowledge, functional activity of 5'-isomiR in CRC has remained outside the focus.
A great diversity of microRNAs and at least an order of magnitude higher number of regulatory interactions they participate in (with about 200 targets per microRNA on average [1]) invoke the use of bioinformatic approaches. One of them is building and analysis of regulatory networks, with microRNA molecules and genes represented by graph's vertices and the interactions represented by edges between particular microRNAs and their targets [11]. The construction of regulatory networks is traditionally based on two sources: (1) knowledge databases of interactions between different types of molecules and (2) correlation analysis of mRNA and microRNA expression levels in clinical samples. We have previously developed an algorithm that allows integration of these two sources in a single network with the addition of TF as another class of major regulatory molecules [12]. The miRGTF-net algorithm affords complete and reliable description of intracellular interaction landscapes in cell/tissue types of interest.
This study aimed to identify functional roles of 5'-isomiR in CRC by integration of the gene expression profiles, sequence-based target prediction and TF activity data using the miRGTF-net algorithm.
METHODS
Target prediction for microRNA isoforms
Pri-microRNA hairpin sequences and canonical Drosha and Dicer cleavage sites were retrieved from miRBase version 21 (https://www. mirbase.org). Designation of 5'-isomiR uses standard nomenclature: a digit after the vertical slash indicates a 5' to 3' shift in the cleavage position with regard to the canonical variant. For example, hsa-miR-10a-5p|+1 corresponds to the canonical hsa-miR-10a-5p devoid of the first nucleotide counting from the 5'-terminus.
For the target prediction, canonical microRNA and isomiR sequences were loaded in miRDB version 6.0 [13] and TargetScan version 7.2 [2]. For miRDB, putative targets with prediction scores > 80 were qualified as valid in accordance with the developer's recommendations. The TargetScan predictions were tailored to the number of miRDB predictions by choosing the appropriate number of the strongest interactions for each microRNA isoform.
Collection and analysis of sequencing data for mRNA and microRNA isoforms
The publicly available raw sequencing data for mRNA and miRNA isoforms were retrieved from The Cancer Genome
Atlas Colon Adenocarcinoma (TCGA-COAD) project available at the Genomic Data Commons Data Portal (https://portal.gdc. cancer.gov/). The data were normalized in the edgeR package version 3.30.0 [14] using the Trimmed Mean of M-values (TMM) normalization algorithm yielding TMM-normalized Reads Per Kilobase of transcript per Million mapped reads (TMM-RPKM) matrices for mRNA expression and TMM-normalized Reads Per Million mapped reads (TMM-RPM) matrices for microRNA expression.
The 5'-isomiR expression matrices were sorted by total expression in the studied samples and the cumulative distribution functions were calculated. The minimal set of 5'-isomiR covering 95% of all sequencing reads were qualified as highly expressed and used in further analysis.
Construction of a regulatory network of interactions among microRNA isoforms, their targets and transcription factors
The miRGTF-net algorithm [12] was applied to build a regulatory network of interactions among microRNA isoforms, their targets and transcription factors. A major advantage of the algorithm is the ability to integrate mRNA and miRNA isoform expression data retrieved from TCGA-COAD with biological knowledge databases including:
- TRRUST version 2 (https://www.grnpedia.org/trrust/): interactions between TF and genes;
- TransmiR version 2 (http://www.cuilab.cn/transmir): interactions between TF and microRNA;
- miRDB, TargetScan (see previous sections): interactions between microRNA isoforms and their targets;
- miRIAD (https://www.miriad-database.org): coexpression of host genes and their intronic microRNA.
The analysis involved a canonical sequence of steps of the miRGTF-net algorithm briefly described as follows. At first we built a network based on the interactions retrieved from databases. Then we calculated Spearman correlation coefficients based on the corresponding expression levels retrieved from TCGA-COAD. Edges connecting molecules with weakly correlated expression levels were eliminated (cutoff for the absolute value of Spearman's correlation was determined as the 0.9 quantile of the correlation distribution). Edges between microRNA isoforms and their putative targets showing positive correlations and edges between host genes and their intronic microRNA showing negative correlations were removed as well.
Next, we evaluated the strength of the linear relationships between expression of each vertex and its direct regulators. The corresponding linear models were built with the ridge regression method. Model quality was evaluated by the coefficient of determination (R2) and strength and direction of regulation were evaluated using standardized regression coefficients (^-coefficients). The vertices and the edges were filtered using the default threshold values of miRGTF-net: ^-coefficient absolute value > 0.3 and 90% of the highest R2. Accordingly, the final network contained vertices corresponding to regulators of expression along with regulated entities (genes and microRNA isoforms).
The search of strongly connected components was carried out using the NetworkX package version 2.8 (https://networkx. org). The regulatory networks were visualized in Gephi (https:// gephi.org) and yED Graph Editor (yWorks GmbH; Germany).
Functional enrichment analysis
Functional annotation of gene sets (targets of microRNA isoforms) was carried out using DAVID web service version
Fig. 1. Expression distributions for 55 most highly expressed microRNA 5'-isoforms in a sample of colorectal cancer tissues. Horizontal lines In the boxes are median values, box limits correspond to lower and upper quartiles, whisker termini correspond to minimal and maximal values
12/2021 [15] and Gene Ontology (GO) biological pathway descriptions [16].
RESULTS
Expression of microRNA isoforms in CRC samples
The analyzed TCGA-COAD dataset contained expression profiles of mRNA and 5'-isoforms of microRNA in a sample of 426 primary CRC tumor specimens. We identified a total of 55 highly expressed microRNA isoforms, 10 of them non-canonical (Fig. 1). Two of the identified non-canonical 5'-isomiR species accounted for over 1% of total microRNA expression each: hsa-miR-192-5p|+1 (2.4%) and hsa-miR-10a-5p|+1 (1.3%).
A shift of the 5'-terminus in microRNA results in altered seed-region of the mature molecule with a major effect on the putative target scope. Sequences of the identified canonical and non-canonical microRNA isoforms were used for biocomputational prediction of their targets. As expected, microRNA isoforms differing by a single extra nucleotide at the 5'-terminus had only slightly overlapping sets of targets. For example, the canonical hsa-miR-10a-5p and its 5'-isoform without one terminal nucleotide had only 11 common targets in a union of 267 (4,1%) (see Table). The maximal proportion of common targets was observed for hsa-miR-29a-3p and its longer isoform: 246 common targets in a union of 788 (31.2%).
Regulatory network of interactions among microRNA isoforms, their targets and transcription factors
At the next step of analysis, we constructed a regulatory network of interactions for CRC cells. The miRGTF-net algorithm enables the construction of such networks by integrating two types of data: (1) biologically substantiated interactions from biological knowledge databases and (2) mRNA and microRNA expression levels measured in tumor samples. The regulatory network contained four types of interactions:
- TF regulates protein-coding gene expression;
- TF regulates microRNA expression;
- microRNA isoform regulates protein-coding gene expression;
- protein-coding host gene is coexpressed with its intronic microRNA
The transcriptomic data for mRNA and microRNA isoforms retrieved from TCGA-COAD were used to look for interactions supported by significant correlations in the studied sample of tumor specimens.
The constructed network encompassed 333 molecules: 24 microRNA (5'-isoforms counted), 166 TF and 143 non-TF coding genes (Fig. 2). Of a total of 456 interactions, 42 represented inhibition of target genes by microRNA isoforms, 413 represented regulation of protein-coding and microRNA gene expression by TF and just a single interaction (between HOXB3 and hsa-mir-10a) represented coexpression of a host gene and its intronic microRNA.
The highest number of negatively correlated targets (seven) was identified for the canonical hsa-miR-148a-3p. The list included known oncogenes including proliferation regulators and markers (CSF1, ETS1, FLT1, MEIS1, MITF and RUNX2; G0:0008284 — positive regulation of cell proliferation) and integrin-encoding gene ITGA5 involved in regulation of cell proliferation, invasion and migration by transmitting signals into cells [17]. Importantly, hsa-miR-148a-3p was a member of the maximal strongly connected component of the regulatory network (i.e., a subgraph with a directed path between any two of the vertices) straightforwardly associated with EMT and estrogen signaling (Fig. 3). The plausible inhibition of protooncogenes by hsa-miR-148a-3p makes it a putative tumor suppressor in CRC.
A pair of canonical hsa-miR-203a-3p|0 and its 5'-isoform hsa-miR-203a-3p|+1 had the second highest number of putative targets. Amid the absence of negatively correlated common targets, both isoforms were likely to share an antitumor functionality by inhibiting expression of oncogenes. For example, expression of canonical hsa-miR-203a-3p|0
Fig. 2. Regulatory network of interactions among microRNA isoforms, their targets and transcription factors. Blue, green and red colors correspond to microRNA 5'-isoforms, transcription factors and target genes, respectively. The edges are colored in accordance with the type of regulator. The vertex sizes are linearly related to degrees
negatively correlated with expression of TF SNAI2 known to be a prominent genetic driver in EMT [18], whereas its non-canonical 5'-isoform hsa-miR-203a-3p|+1 was characterized as a putative regulator of TNC — an extracellular matrix protein-encoding gene which also plays a key role in EMT characteristic of CRC [19].
DISCUSSION
The study applied bioinformatics analysis to assess the functional activity of 5'-isomiR in malignant colorectal tumors. The results indicate that microRNA isoforms differing by a single extra nucleotide at the 5'-terminus have limited number of common targets with a maximum overlap of 31.2%. The most active regulators identified among microRNA included hsa-miR-148a-3p (canonical) and two 5'-isoforms of hsa-miR-203a-3p: hsa-miR-203a-3p|0 (canonical) and hsa-miR-203a-3p|+1. Noteworthy, all three microRNAs were identified as putative inhibitors of pro-tumor gene expression, though the intersection of anti-correlated targets for canonical miR-203a and its 5'-isoform was empty.
The functional activity of 5'-isomiR was previously studied in the context of breast cancer. Two 5'-isoforms of hsa-miR-183-5p were shown to exert different transcriptomic effects in MDA-MB-231 cells; moreover, certain target genes of top clinical significance (EGFR, NRAS) were indirectly regulated by these isoforms with the effects being opposite [20]. To our knowledge, the current study is the first to deal with target scopes of 5'-isomiR in CRC.
Of seven putative targets of hsa-miR-148a-3p selected by us in this study, four genes have been experimentally validated in vitro — CSF1, ITGA5 [21], MITF [22] and RUNX2 [23]. In our own experimental setting, the hypoxia-induced inhibition of hsa-miR-148a-3p expression in CRC cell lines HT-29 and Caco-2 resulted in increased expression of the target gene ITGA5 [24]. Other studies demonstrate the pro-apoptotic effect of miR-148a and the corresponding suppressive effects on proliferation, migration and invasiveness of CRC cells in vitro conferred through Bcl-2 [25], ErbB3 [26] and WNT10b [27] inhibition. Apart from CRC, the role of hsa-miR-148a-3p as a suppressor of tumor cell proliferation was confirmed in the contexts of breast, prostate and urothelial cancers [28]. Thus, our in silico findings on the anti-tumor role of miR-148a show good agreement with the published experimental evidence.
A similar picture is observed for the canonical form of miR-203a: the inferred interaction between miR-203a and SNAI2 have been already validated in vitro [29], whereas overexpression of miR-203a in CRC cell lines has been shown to suppress migration and invasiveness [30]. Thus, the newly identified putative target scope of the non-canonical hsa-miR-203a-3p|+1 is consistent with the established functional profile of the canonical prototype microRNA.
CONCLUSIONS
Construction and analysis of regulatory networks of interactions by tailored bioinformatics tools provide a useful key to the functional activity of 5'-isomiR in colorectal cancer cells. We
Fig. 3. Strongly connected subgraph of interactions between microRNA 5'-isoforms and transcription factors. Ellipses and rectangles denote microRNA isoforms and transcription factors, respectively. Arrows correspond to expression activation, blunt-end lines correspond to expression inhibition
demonstrate that 5'-isoforms of hsa-miR-203a-3p may exert of 5'-isoforms of hsa-miR-203a-3p and other microRNA in vitro similar anti-tumor effects by regulating completely different sets of and in vivo, will be required to understand molecular mechanisms target genes. Further experimental studies, e.g., overexpression of colorectal tumor development and progression.
Table. Numbers of putative targets for highly expressed canonical microRNAs and their 5'-lsoforms
Canonical microRNA 5'-Isoform Canonical microRNA, number of targets 5'-Isoform, number of targets Number of common targets
hsa-miR-10a-5p|0 hsa-miR-10a-5p|+1 175 103 11
hsa-miR-10b-5p|0 hsa-miR-10b-5p|+1 173 102 12
hsa-miR-22-3p|0 hsa-miR-22-3p|+1 235 235 42
hsa-miR-29a-3p|0 hsa-miR-29a-3p|-1 671 363 246
hsa-miR-101-3p|0 hsa-miR-101-3p|-1 632 694 267
hsa-miR-142-3p|0 hsa-miR-142-3p|+1 254 359 33
hsa-miR-143-3p|0 hsa-miR-143-3p|-1 351 205 118
hsa-miR-183-5p|0 hsa-miR-183-5p|+1 366 396 63
hsa-miR-192-5p|0 hsa-miR-192-5p|+1 68 76 16
hsa-miR-203a-3p|0 hsa-miR-203a-3p|+1 573 676 256
References
1. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005; 120 (1): 15-20. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15652477.
2. Agarwal V, Bell GW, Nam J-W, Bartel DP. Predicting effective microRNA target sites in mammalian mRNAs. Elife. 2015; 4. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26267216.
3. Garzon R, Calin GA, Croce CM. MicroRNAs in Cancer. Annu Rev Med. 2009; 60 (1): 167-79. Available from: http://www.ncbi.nlm. nih.gov/pubmed/19630570.
4. Nersisyan S, Shkurnikov M, Poloznikov A, Turchinovich A, Burwinkel B, Anisimov N, et al. Post-Processing Algorithm for miRNA Microarray Data. Int J Mol Sci. 2020; 21 (4). Available
from: http://www.ncbi.nlm.nih.gov/pubmed/32059403.
5. Turchinovich A, Tonevitsky AG, Cho WC, Burwinkel B. Check and mate to exosomal extracellular miRNA: new lesson from a new approach. Front Mol Biosci. 2015; 2 (APR): 11. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25988178.
6. Zhiyanov A, Nersisyan S, Tonevitsky A. Hairpin sequence and structure is associated with features of isomiR biogenesis. RNA Biol. 2021; 18 (sup1): 430-8. Available from: http://www.ncbi. nlm.nih.gov/pubmed/34286662.
7. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021; 71 (3): 209-49. Available
from: http://www.ncbi.nlm.nih.gov/pubmed/33538338.
8. Hill L, Browne G, Tulchinsky E. ZEB/miR-200 feedback loop: 20. at the crossroads of signal transduction in cancer. Int J cancer. 2013; 132 (4): 745-54. Available from: http://www.ncbi.nlm.nih. gov/pubmed/22753312.
9. Chen B, Xia Z, Deng Y-N, Yang Y, Zhang P, Zhu H, et al. Emerging microRNA biomarkers for colorectal cancer diagnosis and 21. prognosis. Open Biol. 2019; 9 (1): 180212. Available from: http:// www.ncbi.nlm.nih.gov/pubmed/30958116.
10. Zelli V Compagnoni C, Capelli R, Corrente A, Cornice J, Vecchiotti D, et al. Emerging Role of isomiRs in Cancer: State of the Art and Recent Advances. Genes (Basel) 2021; 12 (9). Available from: http://www.ncbi.nlm.nih.gov/pubmed/34573429. 22.
11. Galatenko VV, Galatenko AV, Samatov TR, Turchinovich AA, Shkurnikov MY, Makarova JA, et al. Comprehensive network of miRNA-induced intergenic interactions and a biological role of its
core in cancer. Sci Rep. 2018; 8 (1): 2418. Available from: http:// 23. www.ncbi.nlm.nih.gov/pubmed/29402894.
12. Nersisyan S, Galatenko A, Galatenko V, Shkurnikov M, Tonevitsky A. miRGTF-net: Integrative miRNA-gene-TF network analysis reveals
key drivers of breast cancer recurrence. PLoS One. 2021; 16 24. (4): e0249424. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/33852600.
13. Chen Y, Wang X. miRDB: an online database for prediction
of functional microRNA targets. Nucleic Acids Res. 2020; 48 25. (D1): D127-31. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/31504780.
14. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene 26. expression data. Bioinformatics. 2010; 26 (1): 139-40. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19910308.
15. Sherman BT, Hao M, Qiu J, Jiao X, Baseler MW, Lane HC,
et al. DAVID: a web server for functional enrichment analysis 27. and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/35325185.
16. Gene Ontology Consortium. The Gene Ontology resource: enriching a GOld mine. Nucleic Acids Res. 2021; 49 (D1): D325-34. 28. Available from: http://www.ncbi.nlm.nih.gov/pubmed/33290552.
17. Hamidi H, Ivaska J. Every step of the way: integrins in cancer progression and metastasis. Nat Rev Cancer. 2018; 18 (9): 533-48. 29. Available from: http://www.ncbi.nlm.nih.gov/pubmed/30002479.
18. Dudas J, Ladanyi A, Ingruber J, Steinbichler TB, Riechelmann H. Epithelial to Mesenchymal Transition: A Mechanism that Fuels Cancer Radio/Chemoresistance. Cells. 2020; 9 (2). Available from: http://www.ncbi.nlm.nih.gov/pubmed/32059478.
19. Takahashi Y, Sawada G, Kurashige J, Matsumura T, Uchi R, 30. Ueo H, et al. Tumor-derived tenascin-C promotes the epithelial-mesenchymal transition in colorectal cancer cells. Anticancer Res. 2013; 33 (5): 1927-34. Available from: http://www.ncbi.nlm.nih.
gov/pubmed/23645740.
Telonis AG, Loher P, Jing Y, Londin E, Rigoutsos I. Beyond the one-locus-one-miRNA paradigm: microRNA isoforms enable deeper insights into breast cancer heterogeneity. Nucleic Acids Res. 2015; 43 (19): 9158-75. Available from: http://www.ncbi. nlm.nih.gov/pubmed/26400174.
Cimino D, De Pitta C, Orso F, Zampini M, Casara S, Penna E, et al. miR148b is a major coordinator of breast cancer progression in a relapse-associated microRNA signature by targeting ITGA5, R0CK1, PIK3CA, NRAS, and CSF1. FASEB J. 2013; 27 (3): 1223-35. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/23233531.
Haflidadottir BS, Bergsteinsdottir K, Praetorius C, Steingrimsson E. miR-148 regulates Mitf in melanoma cells. PLoS One. 2010; 5 (7): e11574. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/20644734.
Liu H, Su H, Wang X, Hao W. MiR-148a regulates bone marrow mesenchymal stem cells-mediated fracture healing by targeting insulin-like growth factor 1. J Cell Biochem. 2018. Available from: http://www.ncbi.nlm.nih.gov/pubmed/30335895. Nersisyan S, Galatenko A, Chekova M, Tonevitsky A. Hypoxia-Induced miR-148a Downregulation Contributes to Poor Survival in Colorectal Cancer. Front Genet. 2021; 12: 662468. Available from: http://www.ncbi.nlm.nih.gov/pubmed/34135940. Zhang H, Li Y, Huang Q, Ren X, Hu H, Sheng H, et al. MiR-148a promotes apoptosis by targeting Bcl-2 in colorectal cancer. Cell Death Differ. 2011; 18 (11): 1702-10. Available from: http://www. ncbi.nlm.nih.gov/pubmed/21455217.
Zhao W, Zheng J, Wei G, Yang K, Wang G, Sun X. miR-148a inhibits cell proliferation and migration through targeting ErbB3 in colorectal cancer. Oncol Lett. 2019; 18 (3): 2530-6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/31402949. Shi L, Xi J, Xu X, Peng B, Zhang B. MiR-148a suppressed cell invasion and migration via targeting WNT10b and modulating p-catenin signaling in cisplatin-resistant colorectal cancer cells. Biomed Pharmacother. 2019; 109: 902-9. Available from: http:// www.ncbi.nlm.nih.gov/pubmed/30551544. Li Y, Deng X, Zeng X, Peng X. The Role of Mir-148a in Cancer. J Cancer. 2016; 7 (10): 1233-41. Available from: http://www.ncbi. nlm.nih.gov/pubmed/27390598.
Ma X, Li L, Jia T, Chen M, Liu G, Li C, et al. miR-203a controls keratinocyte proliferation and differentiation via targeting the stemness-associated factor ANp63 and establishing a regulatory circuit with SNAI2. Biochem Biophys Res Commun. 2017; 491 (2): 241-9. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/28754589.
Qian Z, Gong L, Mou Y, Han Y, Zheng S. MicroRNA-203a-3p is a candidate tumor suppressor that targets thrombospondin 2 in colorectal carcinoma. Oncol Rep. 2019; 42 (5): 1825-32. Available from: http://www.ncbi.nlm.nih.gov/pubmed/31545460.
Литература
1. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005; 120 (1): 15-20. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15652477. 7.
2. Agarwal V, Bell GW, Nam J-W, Bartel DP. Predicting effective microRNA target sites in mammalian mRNAs. Elife. 2015; 4. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26267216.
3. Garzon R, Calin GA, Croce CM. MicroRNAs in Cancer. Annu Rev Med. 2009; 60 (1): 167-79. Available from: http://www.ncbi.nlm. 8. nih.gov/pubmed/19630570.
4. Nersisyan S, Shkurnikov M, Poloznikov A, Turchinovich A, Burwinkel B, Anisimov N, et al. Post-Processing Algorithm for miRNA Microarray Data. Int J Mol Sci. 2020; 21 (4). Available 9. from: http://www.ncbi.nlm.nih.gov/pubmed/32059403.
5. Turchinovich A, Tonevitsky AG, Cho WC, Burwinkel B. Check and mate to exosomal extracellular miRNA: new lesson from a new approach. Front Mol Biosci. 2015; 2 (APR): 11. Available from: 10. http://www.ncbi.nlm.nih.gov/pubmed/25988178.
6. Zhiyanov A, Nersisyan S, Tonevitsky A. Hairpin sequence and
structure is associated with features of isomiR biogenesis. RNA Biol. 2021; 18 (sup1): 430-8. Available from: http://www.ncbi. nlm.nih.gov/pubmed/34286662.
Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021; 71 (3): 209-49. Available from: http://www.ncbi.nlm.nih.gov/pubmed/33538338. Hill L, Browne G, Tulchinsky E. ZEB/miR-200 feedback loop: at the crossroads of signal transduction in cancer. Int J cancer. 2013; 132 (4): 745-54. Available from: http://www.ncbi.nlm.nih. gov/pubmed/22753312.
Chen B, Xia Z, Deng Y-N, Yang Y, Zhang P, Zhu H, et al. Emerging
microRNA biomarkers for colorectal cancer diagnosis and
prognosis. Open Biol. 2019; 9 (1): 180212. Available from: http://
www.ncbi.nlm.nih.gov/pubmed/30958116.
Zelli V, Compagnoni C, Capelli R, Corrente A, Cornice J, Vecchiotti D,
et al. Emerging Role of isomiRs in Cancer: State of the Art and
Recent Advances. Genes (Basel) 2021; 12 (9). Available from:
http://www.ncbi.nlm.nih.gov/pubmed/34573429.
11. Galatenko VV, Galatenko AV, Samatov TR, Turchinovich AA, Shkurnikov MY, Makarova JA, et al. Comprehensive network of miRNA-induced intergenic interactions and a biological role of its core in cancer. Sci Rep. 2018; 8 (1): 2418. Available from: http:// www.ncbi.nlm.nih.gov/pubmed/29402894.
12. Nersisyan S, Galatenko A, Galatenko V, Shkurnikov M, Tonevitsky A. miRGTF-net: Integrative miRNA-gene-TF network analysis reveals key drivers of breast cancer recurrence. PLoS One. 2021; 16 (4): e0249424. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/33852600.
13. Chen Y, Wang X. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res. 2020; 48 (D1): D127-31. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/31504780.
14. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010; 26 (1): 139-40. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19910308.
15. Sherman BT, Hao M, Qiu J, Jiao X, Baseler MW, Lane HC, et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/35325185.
16. Gene Ontology Consortium. The Gene Ontology resource: enriching a GOld mine. Nucleic Acids Res. 2021; 49 (D1): D325-34. Available from: http://www.ncbi.nlm.nih.gov/pubmed/33290552.
17. Hamidi H, Ivaska J. Every step of the way: integrins in cancer progression and metastasis. Nat Rev Cancer. 2018; 18 (9): 533-48. Available from: http://www.ncbi.nlm.nih.gov/pubmed/30002479.
18. Dudas J, Ladanyi A, Ingruber J, Steinbichler TB, Riechelmann H. Epithelial to Mesenchymal Transition: A Mechanism that Fuels Cancer Radio/Chemoresistance. Cells. 2020; 9 (2). Available from: http://www.ncbi.nlm.nih.gov/pubmed/32059478.
19. Takahashi Y, Sawada G, Kurashige J, Matsumura T, Uchi R, Ueo H, et al. Tumor-derived tenascin-C promotes the epithelial-mesenchymal transition in colorectal cancer cells. Anticancer Res. 2013; 33 (5): 1927-34. Available from: http://www.ncbi.nlm.nih. gov/pubmed/23645740.
20. Telonis AG, Loher P, Jing Y, Londin E, Rigoutsos I. Beyond the one-locus-one-miRNA paradigm: microRNA isoforms enable deeper insights into breast cancer heterogeneity. Nucleic Acids Res. 2015; 43 (19): 9158-75. Available from: http://www.ncbi. nlm.nih.gov/pubmed/26400174.
21. Cimino D, De Pitta C, Orso F, Zampini M, Casara S, Penna E, et al. miR148b is a major coordinator of breast cancer progression in a relapse-associated microRNA signature by targeting ITGA5, ROCK1, PIK3CA, NRAS, and CSF1. FASEB J. 2013; 27 (3): 1223-35. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/23233531.
22. Haflidadottir BS, Bergsteinsdottir K, Praetorius C, Steingrimsson E. miR-148 regulates Mitf in melanoma cells. PLoS One. 2010; 5 (7): e11574. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/20644734.
23. Liu H, Su H, Wang X, Hao W. MiR-148a regulates bone marrow mesenchymal stem cells-mediated fracture healing by targeting insulin-like growth factor 1. J Cell Biochem. 2018. Available from: http://www.ncbi.nlm.nih.gov/pubmed/30335895.
24. Nersisyan S, Galatenko A, Chekova M, Tonevitsky A. Hypoxia-Induced miR-1 48a Downregulation Contributes to Poor Survival in Colorectal Cancer. Front Genet. 2021; 12: 662468. Available from: http://www.ncbi.nlm.nih.gov/pubmed/34135940.
25. Zhang H, Li Y, Huang Q, Ren X, Hu H, Sheng H, et al. MiR-148a promotes apoptosis by targeting Bcl-2 in colorectal cancer. Cell Death Differ. 2011; 18 (11): 1702-10. Available from: http://www. ncbi.nlm.nih.gov/pubmed/21455217.
26. Zhao W, Zheng J, Wei G, Yang K, Wang G, Sun X. miR-148a inhibits cell proliferation and migration through targeting ErbB3 in colorectal cancer. Oncol Lett. 2019; 18 (3): 2530-6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/31402949.
27. Shi L, Xi J, Xu X, Peng B, Zhang B. MiR-148a suppressed cell invasion and migration via targeting WNT10b and modulating p-catenin signaling in cisplatin-resistant colorectal cancer cells. Biomed Pharmacother. 2019; 109: 902-9. Available from: http:// www.ncbi.nlm.nih.gov/pubmed/30551544.
28. Li Y, Deng X, Zeng X, Peng X. The Role of Mir-148a in Cancer. J Cancer. 2016; 7 (10): 1233-41. Available from: http://www.ncbi. nlm.nih.gov/pubmed/27390598.
29. Ma X, Li L, Jia T, Chen M, Liu G, Li C, et al. miR-203a controls keratinocyte proliferation and differentiation via targeting the stemness-associated factor ANp63 and establishing a regulatory circuit with SNAI2. Biochem Biophys Res Commun. 2017; 491 (2): 241-9. Available from: http://www.ncbi.nlm.nih.gov/ pubmed/28754589.
30. Qian Z, Gong L, Mou Y, Han Y, Zheng S. MicroRNA-203a-3p is a candidate tumor suppressor that targets thrombospondin 2 in colorectal carcinoma. Oncol Rep. 2019; 42 (5): 1825-32. Available from: http://www.ncbi.nlm.nih.gov/pubmed/31545460.
