Molecular analysis of long non-coding RNA GAS5 and microRNA-34a expression signature in common solid tumors: A pilot study

Accumulating evidence indicates that non-coding RNAs including microRNAs (miRs) and long non-coding RNAs (lncRNAs) are aberrantly expressed in cancer, providing promising biomarkers for diagnosis, prognosis and/or therapeutic targets. We aimed in the current work to quantify the expression profile of miR-34a and one of its bioinformatically selected partner lncRNA growth arrest-specific 5 (GAS5) in a sample of Egyptian cancer patients, including three prevalent types of cancer in our region; renal cell carcinoma (RCC), hepatocellular carcinoma (HCC) and glioblastoma (GB) as well as to correlate these expression profiles with the available clinicopathological data in an attempt to clarify their roles in cancer. Quantitative real-time polymerase chain reaction analysis was applied. Different bioinformatics databases were searched to confirm the potential miRNAs-lncRNA interactions of the selected ncRNAs in cancer pathogenesis. GAS5 was significantly under-expressed in the three types of cancer. However, levels of miR-34a greatly varied according to the tumor type; it displayed an increased expression in RCC [4.05 (1.003-22.69), p <0.001] and a decreased expression in GB [0.35 (0.04-0.95), p <0.001]. A weak negative correlation was observed between levels of GAS5 and miR-34a in GB [r = −0.39, p =0.006]. Univariate analyses revealed a correlation of GAS5 downregulation with poor disease-free survival (r = 0.31, p =0.018) and overall survival (r = 0.28, p =0.029) in RCC but not in GB, and a marginal significance correlation with a higher number of lesions in HCC. Hierarchical clustering analysis showed RCC patients among others, could be clustered by GAS5 and miR-34a co-expression profile. Our results confirm the tumor suppressor role of GAS5 in cancer and suggest its potential applicability to be a predictor of bad outcomes with other conventional markers for various types of cancer. Further functional validation studies are warranted to confirm miR-34a/GAS5 interplay in cancer.

Ensembl and GeneCards databases to identify its biological function in cancer.

127
Study population and sampling collection 128 In total, 230 samples were analyzed, including 60 formalin-fixed, paraffin-embedded (FFPE) RCC 129 samples and their paired adjacent non-cancer tissues, 50 FFPE GB specimens and 10 non-cancer 130 brain tissues as well as 30 HCC blood samples and 20 controls.
6 131 (a) RCC cohort 132 The archived FFPE renal samples were taken from sixty patients who underwent radical 133 nephrectomy for a primary RCC and dating back for three years. All retrieved cases were archived 134 in the Pathology laboratory of Mansoura Oncology Center, Mansoura, Egypt. No history of 135 receiving neoadjuvant chemotherapy or radiotherapy prior to sampling. Clinicopathological data, 136 including the survival were collected from patient medical records. Paired sixty cancer-free adjacent 137 tissues were examined and sectioned to serve as controls for molecular analysis.   − (C q ncRNA -C q internal control) control [51]. Univariate analysis for association between ncRNA expression profile and clinico-pathological features in cancer patients was run. multiple snoRNA genes within its introns. These genes encode for ten C/D box snoRNAs, which snoRNA, which is associated with pseudouridylation ( Fig. 1B).

236
GAS5 gene is shown to be highly polymorphic, enclosing around 300 thousand variants (75% 237 intronic, 21% exonic, and 4% splice region polymorphisms). Among all these polymorphisms, 14 238 SNPs, one deletion and one insertion are common variants with minor allele frequency (MAF) > 239 0.05 (Fig. 1C) and (Electronic Supplementary Table S2).  death, cell cycle, and response to stress thus highlighting its role in cancer cell growth. In addition, 277 miR-34a was significantly associated with membrane organization, cell junction organization, and 278 cell motility, hence may play a key role in cancer cell invasion and metastasis (Fig. 2).    between levels of GAS5 and miR-34a in GB [r = -0.39, p = 0.006] (Fig. 4).        Dendrograms for Two-way agglomerative hierarchical cluster analysis, were employed (Fig. 6). 399 The following cluster setup parameters were adjusted: Distance method: Sorensen (Bray-Curtis), In this study, we measured the expression of two ncRNAs; the lncRNA GAS5 and the miRNA 415 miR-34a in three of the most prevalent and high-incidence tumors in Egypt; hepatic, renal and brain 416 cancer. We chose more than one tumor type to assess if the same ncRNA could work differently 417 according to the tissue type. We also investigated the possible association between GAS5 and miR-418 34a in mediating carcinogenesis after detecting an interaction between the two through our 419 preliminary in silico analysis.

420
Our results show that GAS5 was under-expressed in the three types of cancer; RCC, HCC and GB.

421
On the other hand, levels of miR-34a greatly varied according to the tumor type. RCC patients had 422 a lower GAS5 level and a higher miR-34a level in tumor tissue compared to adjacent normal tissue. increased apoptosis, as well as inhibited tumor metastatic potential [24], which could explain the 427 correlation between reduced patient survival and GAS5 level in our study. As for miR-34a, several well as the predicted interaction between the two, we believe that a new pathway; the GAS5/miR-471 34a pathway might be involved in the previously indicated molecular network leading to RCC.

472
The interaction between lncRNAs and miRNAs can be multifaceted. Yoon