Methods of transcriptome analysis

The primary method of transcriptome analysis is next-generation sequencing (NGS). Billions of DNA (or RNA) strands can be decoded in parallel. See table 2 for a summary of the techniques, advantages and disadvantages of NGS, microarray and real-time quantitative PCR.

Understanding of function of healthy cells and tissues and insights into pathogenesis of disease

Biologically active proteins are constantly interacting and changing within their tissue or cell,4 making their measurement challenging. RNA sequencing of a specific tissue provides a means of determining how much and what types of proteins are in production. Transcriptome databases catalogue quantitative data on spatiotemporal patterns of gene expression in human tissues and organs at different stages of normal prenatal and postnatal life. Examples include the Human Gene Spatial Expression Database,5 Brainspan6 and the Genotype-Tissue Expression project.7

Clues to the mechanisms of disease can be found by comparing gene expression in normal tissue controls with pathological specimens. Skeletal muscle transcripts from 50 patients with rare muscle diseases who were undiagnosed after whole exome sequencing were analysed and compared with 180 control samples. Disease causing variants were identified in 35% of cases.8

Environmental factors such as low socioeconomic status in childhood have been shown to affect cardiovascular disease and cancer risk in later life. Reproducible findings of upregulation of proinflammatory genes and reduced expression of genes involved in antiviral response and antibody production hint at possible mechanisms9 10 and could guide future strategies for early diagnosis and disease prevention.

Interpretation of genetic results

NGS has allowed us to decode the entire genome, which detects differences in DNA code, but it does not tell us about the activity or function of genes. Analysis of transcripts allows further information to be gleaned about suspected pathological variants. Patients in the 100 000 Genomes Project are having samples taken to augment their genomic data with information about their transcriptome.1

In practice, identical mutations have different effects in different people due to variation in interaction between genes, proteins, metabolites and environment.11 The non-coding DNA plays a part in regulating dynamic processes of DNA transcription and protein production. The system is incredibly complex, with one gene encoding many different isoforms of messenger RNA (mRNA). This is achieved primarily through alternative splicing. During mRNA production, introns between the exons are chopped out, and the exons are spliced together. Different protein isoforms vary in function and are often specific to different cell or tissue types and to different stages of development.

When genetic variants affect only specific transcripts of a gene, having information about where and when those transcripts are produced can help determine whether the variant is likely to cause disease or not.

Variants in splice sites and the spliceosomal machinery interfere with normal transcript formation and account for a significant proportion of disease-causing variants in many single gene disorders. For example, analysis of RNA transcripts in patients with retinitis pigmentosa can be used to confirm pathogenicity.

Analysis of RNA transcripts can be used to prove if variants in locations related to intron–exon boundaries are likely to affect splicing. This evidence can be invaluable for confirming a diagnosis and allowing a genetic variant to be used for prenatal diagnosis and screening in other family members where indicated.

New diagnostic techniques

Biomarkers are tools for diagnosing, staging and classifying disease. They can also be used to monitor clinical response to intervention.12 Commercial assays have been developed to measure RNA transcripts of selected genes in tumour tissue to assess prognosis and guide stratification of treatment.13

Fusion transcripts are important drivers of paediatric tumours. They occur when two genes become fused together due to chromosomal rearrangements and can result in inactivation of tumour suppressor genes and activation of proto-oncogenes. MYC is a proto-oncogene that exerts its effect by controlling multiple splicing factors.14 Cancers that overexpress MYC can be targeted by specific splice-modifying agents.15 RNA sequencing is more sensitive for detecting fusion products and could potentially supersede conventional karyotyping (looking at banded chromosomes down the microscope) and be used to guide treatment and prediction of outcomes.

A potential source of new biomarkers are microRNAs; they are short non-coding RNAs that modify mRNA post transcription. One of the earliest miRNAs discovered was miR-21, which is overexpressed in bowel cancer and Crohn’s disease,16 and hence, it can be used as a biomarker for inflammatory bowel disease. The possibility of correcting functional abnormality with therapeutic interventions that antagonise or mimic miR-21 is being explored.16

New treatments

Transcriptomics has the potential to identify RNA molecules amenable to therapeutic manipulation. For example, microRNA-122 (miR-122) is a liver-specific microRNA that has been shown to enhance the survival and replication of the hepatitis C virus (HCV) within the liver.17 Clinical trials are ongoing, looking at whether inhibition of miR-122 by a drug known as Miravirsen could reduce the viral RNA replication in liver cells infected by HCV.17

In disease states where a deficiency of microRNA is seen, direct replacement can be used as treatment. Significantly reduced expression of microRNA 132 was identified in diabetic ulcers. It is speculated that replacement accelerates wound healing by regulation of inflammatory signalling pathways for example tumour necrosis factor signalling.18