The Role of Small Nucleolar RNAs As Putative Biomarkers of Chemoresistance in Pediatric Acute Lymphoblastic Leukemia

Background

Many previous studies have focused on the molecular landscape of pediatric acute lymphoblastic leukemia (ALL) with the aim to identify genetic changes related to disease development and outcome. However, the functional role of non-coding RNAs (ncRNAs) in leukemogenesis and treatment response is not fully understood. A group of ncRNAs known as small nucleolar RNAs (snoRNAs) regulates ribosomal biosynthesis, RNA splicing, and performs microRNA-like functions that affect post-translational gene expression. These functionalities can be dysregulated in cancer. The aim of this study was to explore the relationship between snoRNA expression and ex-vivo drug resistance in a cohort of pediatric ALL patients enrolled in the Nordic Society of Pediatric Hematology and Oncology (NOPHO) protocols.

Methods

We analyzed 119 B-ALL patients who were treated according to two consecutive NOPHO protocols (1992 and 2000). B-cells (CD19+) and T-cells (CD3+) from healthy blood donors were used as controls (n = 10). We stratified patients into low-, medium-, and high-resistance groups by ex-vivo drug screening using 10 commonly used chemotherapeutic drugs with the fluorometric microculture cytotoxicity assay (FMCA). RNA sequencing (RNA-seq) data were generated for all patients. Pairwise comparisons (Benjamini-Hochberg (BH) corrected Mann-Whitney U test) were conducted between the low and high resistance groups to detect differentially expressed genes. We selectively retained the differentially expressed snoRNAs for downstream analysis. BH corrected Kruskal-Wallis H and Mann-Whitney U tests were performed to compare the snoRNA expression levels for the three drug resistance groups, and between the ALL and control samples, respectively.

Results

We identified a set of 23 unique significantly upregulated snoRNAs (p-adjusted value < 0.05 and absolute log2FC>1) in the high-resistance group for one or multiple drugs: amsacrine (ams, n = 10), etoposide (eto, n = 10), tioguanine (thio, n = 8) and mitoxantrone (mito, n = 1). Among these genes, five snoRNAs ( SNORA30, SNORA33, SNORA36C, SNORA62 and SNORA66) were overexpressed in patients resistant to ams and eto, and one snoRNA ( SNORD27) in patients resistant to ams and mito. Gene annotation revealed 22 significantly differentially expressed snoRNAs residing within the intronic region of 20 host genes, either protein coding (including ribosomal genes) or long ncRNAs. Three snoRNAs ( SNORD22-eto, SNORD26-ams and SNORD27-ams and mito) arise from the same host gene ( SNHG1). Of the 23 differentially expressed snoRNAs, the majority (n = 18) were not differentially expressed between the ALL and control samples (BH corrected Mann-Whitney U test p-value ≥ 0.05).

The expression of the host genes did not differ among the resistance groups (BH corrected Kruskal-Wallis H test p-value ≥ 0.05). However, we found that 14 of the host genes were differentially expressed between the ALL samples and controls (BH corrected Mann-Whitney U test p-value < 0.05).

Conclusions

This study is one of the first to focus on the role of snoRNAs and ex-vivo drug resistance in pediatric ALL. Our findings highlight substantial variations in snoRNA expression across the three resistance groups, underscoring the potential role of snoRNAs in modulating drug response. Importantly, our analysis indicates that the differential expression of snoRNAs in the resistance groups cannot be solely attributed to host gene expression, implying that targeting pathways involving host genes might not be the most effective approach. Rather than concentrating on pathways involving host genes, our results suggest that understanding the mechanisms of action of snoRNAs could provide promising avenues for developing novel therapeutic targets to enhance drug response in pediatric ALL.