Human cancer long non-coding RNA transcriptomes

Abstract

Once thought to be a part of the 'dark matter' of the genome, long non-coding RNAs (lncRNAs) are emerging as an integral functional component of the mammalian transcriptome. LncRNAs are a novel class of mRNA-like transcripts which, despite no known protein-coding potential, demonstrate a wide range of structural and functional roles in cellular biology. However, the magnitude of the contribution of lncRNA expression to normal human tissues and cancers has not been investigated in a comprehensive manner. In this study, we compiled 272 human serial analysis of gene expression (SAGE) libraries to delineate lncRNA transcription patterns across a broad spectrum of normal human tissues and cancers. Using a novel lncRNA discovery pipeline we parsed over 24 million SAGE tags and report lncRNA expression profiles across a panel of 26 different normal human tissues and 19 human cancers. Our findings show extensive, tissue-specific lncRNA expression in normal tissues and highly aberrant lncRNA expression in human cancers. Here, we present a first generation atlas for lncRNA profiling in cancer.

Figure 1. Tissue-type distribution of the 272 SAGE libraries with a minimum raw tag count of 50,000.

(CL) indicates one SAGE library that was generated from a mixture of human cell lines.

Figure 2. LncRNA discovery pipeline using SAGE analysis.

Numbers indicate programs or filtering steps as follows: (1) filtering to retain only those libraries with a minimum of 50,000 raw tag counts, (2) identifying unique SAGE tags and constructing SAGE tag expression matrix, (3) mapping SAGE tags to Unigene IDs using SAGE Genie mapping files, (4) filtering lists to retain only tags with ≥2 raw counts in a ≥3 of 272 libraries, (5) determining gene identity using SAGE Genie, (6) separating Unigene tags mapping to lncRNAs and ambiguous transcripts, (7) pooling ambiguous tags and unmapped tags, (8) mapping sequence tags to the reference list of 9,891 lncRNAs using SeqMap, a tag-to-gene mapping program, (remaining tags may map to unannotated lncRNAs or antisense transcripts not included in our reference list) (9) filtering tag matches for strand sense, (10) pooling forward mapping tags and tags determined from Unigene, and (11) confirming tag-to-lncRNA matches and summing tag counts for lncRNAs with multiple tag matches. A complete list of lncRNAs is provided as Table S5 and tag-to-lncRNA matches are provided as Table S6.

Figure 3. Distribution and levels of lncRNA expression in normal human tissues.

(A) Number of distinct lncRNAs expressed in normal human tissues, white blood cells and embryonic stem cells with a minimum average TPM of 20. The values in brackets indicate the number of SAGE libraries for each tissue. (B) Examples of lncRNAs detected exclusively in a single normal human tissue or in embryonic stem cells (ESC) with a minimum expression level of 10 TPM. For tissues with two or more libraries, the TPM values were averaged. LncRNAs without names are labeled with an Ensembl ID.

Figure 4. Expression patterns of lncRNAs in normal human tissues.

(A) LncRNAs with the highest overall expression (B) LncRNAs with the highest variance by a coefficient of variation (CV) test. Heatmaps indicate the relative intensity (normalized TPM) of each lncRNA across seventeen human tissues, white blood cells and human embryonic stem cells. Where more than one SAGE library was available, the TPM values were averaged. For the heatmap, the maximum threshold was set at 300 TPM. LncRNAs without names are labeled with an Ensembl ID.

Figure 5. Expression patterns of lncRNAs in human cancers.

(A) LncRNAs with the highest overall expression (B) LncRNAs with the highest variance by a coefficient of variation (CV) test. Heatmaps indicate the relative intensity (normalized TPM) of each lncRNA across seventeen human cancers and human embryonic stem cells. Where more than one SAGE library was available, the TPM values were averaged. For the heatmap, the maximum threshold was set at 300 TPM. LncRNAs without names are labeled with an Ensembl ID.

Figure 6. Aberrantly expressed lncRNAs in human cancers.

(A) Number of lncRNAs showing significant expression changes. The number of lncRNAs determined to have significant (BH p-value <0.05) differential expression of 2-fold or greater reported. Solid bars indicate upregulated genes, while bars with hatch marks indicate downregulated genes (B) Venn diagram of differentially expressed lncRNAs in human carcinomas.

Figure 7. Chromosomal distribution of protein-coding genes, microRNAs and long non-coding RNAs in the human genome.

Protein-coding gene (n = 20,655), microRNA (n = 1,746) and long non-coding RNA (n = 9,891) coordinates were downloaded from Ensembl v62 using BioMart.