Impact of RNA degradation on gene expression profiling

Background

Cancer is a heterogeneous disease caused by genomic aberrations and characterized by significant variability in clinical outcomes and response to therapies. Several subtypes of common cancers have been identified based on alterations of individual cancer genes, such as HER2, EGFR, and others. However, cancer is a complex disease driven by the interaction of multiple genes, so the copy number status of individual genes is not sufficient to define cancer subtypes and predict responses to treatments. A classification based on genome-wide copy number patterns would be better suited for this purpose.

Method

To develop a more comprehensive cancer taxonomy based on genome-wide patterns of copy number abnormalities, we designed an unsupervised classification algorithm that identifies genomic subgroups of tumors. This algorithm is based on a modified genomic Non-negative Matrix Factorization (gNMF) algorithm and includes several additional components, namely a pilot hierarchical clustering procedure to determine the number of clusters, a multiple random initiation scheme, a new stop criterion for the core gNMF, as well as a 10-fold cross-validation stability test for quality assessment.

Result

We applied our algorithm to identify genomic subgroups of three major cancer types: non-small cell lung carcinoma (NSCLC), colorectal cancer (CRC), and malignant melanoma. High-density SNP array datasets for patient tumors and established cell lines were used to define genomic subclasses of the diseases and identify cell lines representative of each genomic subtype. The algorithm was compared with several traditional clustering methods and showed improved performance. To validate our genomic taxonomy of NSCLC, we correlated the genomic classification with disease outcomes. Overall survival time and time to recurrence were shown to differ significantly between the genomic subtypes.

Conclusions

We developed an algorithm for cancer classification based on genome-wide patterns of copy number aberrations and demonstrated its superiority to existing clustering methods. The algorithm was applied to define genomic subgroups of three cancer types and identify cell lines representative of these subgroups. Our data enabled the assembly of representative cell line panels for testing drug candidates.     

RNA quality control: degradation of defective transfer RNA

The distinction between stable (tRNA and rRNA) and unstable (mRNA) RNA has been considered an important feature of bacterial RNA metabolism. One factor thought to contribute to the difference between these RNA populations is polyadenylation, which promotes degradation of unstable RNA. However, the recent discovery that polyadenylation also occurs on stable RNA led us to examine whether poly(A) might serve as a signal for eliminating defective stable RNAs, and thus play a role in RNA quality control. Here we show that a readily denaturable, mutant tRNA(Trp) does not accumulate to normal levels in Escherichia coli because its precursor is rapidly degraded. Degradation is largely dependent on polyadenylation of the precursor by poly(A) polymerase and on its removal by polynucleotide phosphorylase. Thus, in the absence of these two enzymes large amounts of tRNA(Trp) precursor accumulate. We propose that defective stable RNA precursors that are poorly converted to their mature forms may be polyadenylated and subsequently degraded. These data indicate that quality control of stable RNA metabolism in many ways resembles normal turnover of unstable RNA.     

The environmental contribution to gene expression profiles

Microarray analysis provides a bridge between the molecular genetic analysis of model organisms in laboratory settings and studies of physiology, development, and adaptation in the wild. By sampling species across a range of environments, it is possible to gain a broad picture of the genomic response to environmental perturbation. Incorporating estimates of genetic relationships into study designs will facilitate genomic analysis of environmental plasticity by aiding the identification of major regulatory loci in natural populations.     

Quantitative assessment of filter-based cDNA microarrays: gene expression profiles of human T-lymphoma cell lines

MOTIVATION: While the use of cDNA microarrays for functional genomic analysis has become commonplace, relatively little attention has been placed on false positives, i.e. the likelihood that a change in measured radioactive or fluorescence intensity may reflect a change in gene expression when, in fact, there is none. Since cDNA arrays are being increasingly used to rapidly distinguish biomarkers for disease detection and subsequent assay development (Wellman et al., Blood, 96, 398-404, 2000), the impact of false positives can be significant. For the use of this technology, it is necessary to develop quantitative criteria for reduction of false positives with radioactively-labeled cDNA arrays. RESULTS: We used a single source of RNA (HuT78 T lymphoma cells) to eliminate sample variation and quantitatively examined intensity ratios using radioactively labeled cDNA microarrays. Variation in intensity ratios was reduced by processing microarrays in side-by-side (parallel mode) rather than by using the same microarray for two hybridizations (sequential mode). Based on statistical independence, calculation of the expected number of false positives as a function of threshold showed that a detection limit of [log(2)R] >0.65 with agreement from three replicates could be used to identify up- or down-modulated genes. Using this quantitative criteria, gene expression differences between two related T lymphoma cell lines, HuT78 and H9, were identified. The relevance of these findings to the known functional differences between these cell types is discussed.     

Analysis of midgut gene expression profiles from different silkworm varieties after exposure to high temperature

The silkworm is a poikilothermic animal, whose growth and development is significantly influenced by environmental temperature. To identify genes and metabolic pathways involved in the heat-stress response, digital gene expression analysis was performed on the midgut of the thermotolerant silkworm variety ‘932’ and thermosensitive variety ‘HY’ after exposure to high temperature (932T and HYT). Deep sequencing yielded 6,211,484, 5,898,028, 5,870,395 and 6,088,303 reads for the 932, 932T, HY and HYT samples, respectively. The annotated genes associated with these tags numbered 4357, 4378, 4296 and 4658 for the 932, 932T, HY and HYT samples, respecti'vely. In the HY-vs-932, 932-vs-932T, and HY-vs-HYT comparisons, 561, 316 and 281 differentially expressed genes were identified, which could be assigned to 179, 140 and 123 biological pathways, respectively. It was found that some of the biological pathways, which included oxidative phosphorylation, related to glucose and lipid metabolism, are greatly affected by high temperature and may lead to a decrease in the ingestion of fresh mulberry. When subjected to an early period of continuous heat stress, HSP genes, such as HSP19.9, HSP23.7, HSP40-3, HSP70, HSP90 and HSP70 binding protein, are up-regulated but then reduced after 24 h and the thermotolerant ‘932’ strain has higher levels of mRNA of some HSPs, except HSP70, than the thermosensitive variety during continuous high temperature treatment. It is suggested that HSPs and the levels of their expression may play important roles in the resistance to high temperature stress among silkworm varieties. This study has generated important reference tools that can be used to further analyze the mechanisms that underlie thermotolerance differences among silkworm varieties.     

Impact of chronic intermittent hypoxia on the long non-coding RNA and mRNA expression profiles in myocardial infarction

Chronic intermittent hypoxia (CIH) is the primary feature of obstructive sleep apnoea (OSA), a crucial risk factor for cardiovascular diseases. Long non-coding RNAs (lncRNAs) in myocardial infarction (MI) pathogenesis have drawn considerable attention. However, whether CIH participates in the modulation of lncRNA profiles during MI is yet unclear. To investigate the influence of CIH on MI, cardiac damage was assessed by histology and echocardiography, and lncRNA and mRNA integrated microarrays were screened. MI mouse model showed myocardial hypertrophy, aggravated inflammation and fibrosis, and compromised left ventricle function under CIH. Compared with normoxia, 644 lncRNAs and 1084 differentially expressed mRNAs were identified following CIH for 4 weeks, whereas 1482 lncRNAs and 990 mRNAs were altered at 8 weeks. Strikingly, reoxygenation after CIH markedly affected 1759 lncRNAs and 778 mRNAs. Of these, 11 lncRNAs modulated by CIH were restored after reoxygenation and were validated by qPCR. The GO terms and KEGG pathways of genes varied significantly by CIH. lncRNA-mRNA correlation further showed that lncRNAs, NONMMUT032513 and NONMMUT074571 were positively correlated with ZEB1 and negatively correlated with Cmbl. The current results demonstrated a causal correlation between CIH and lncRNA alternations during MI, suggesting that lncRNAs might be responsible for MI aggravation under CIH.     

Mining gene expression profiles: expression signatures as cancer phenotypes

Many examples highlight the power of gene expression profiles, or signatures, to inform an understanding of biological phenotypes. This is perhaps best seen in the context of cancer, where expression signatures have tremendous power to identify new subtypes and to predict clinical outcomes. Although the ability to interpret the meaning of the individual genes in these signatures remains a challenge, this does not diminish the power of the signature to characterize biological states. The use of these signatures as surrogate phenotypes has been particularly important, linking diverse experimental systems that dissect the complexity of biological systems with the in vivo setting in a way that was not previously feasible.     

A semi-parametric statistical model for integrating gene expression profiles across different platforms

Determining differentially expressed genes (DEGs) between biological samples is the key to understand how genotype gives rise to phenotype. RNA-seq and microarray are two main technologies for profiling gene expression levels. However, considerable discrepancy has been found between DEGs detected using the two technologies. Integration data across these two platforms has the potential to improve the power and reliability of DEG detection. We propose a rank-based semi-parametric model to determine DEGs using information across different sources and apply it to the integration of RNA-seq and microarray data. By incorporating both the significance of differential expression and the consistency across platforms, our method effectively detects DEGs with moderate but consistent signals. We demonstrate the effectiveness of our method using simulation studies, MAQC/SEQC data and a synthetic microRNA dataset. Our integration method is not only robust to noise and heterogeneity in the data, but also adaptive to the structure of data. In our simulations and real data studies, our approach shows a higher discriminate power and identifies more biologically relevant DEGs than eBayes, DEseq and some commonly used meta-analysis methods.     

A comparison of different methods to determine cell proliferation by flow cytometry

Abstract. The proliferation of human melanoma cells (MeWo) in vitro was studied with a number of different techniques. In particular, we compared the expression of PCNA and the Ki-67 antigen on the one hand with BrdU pulse and continuous labelling on the other. Two-dimensional flow cytometry (with DNA content as a second parameter) was employed to discriminate between cycling and non-cycling cells as well as cells in the G₁, S and G₂ phases of the cycle. Cell cultures in different stages of growth were analyzed. We found that the percentage of anti-PCNA and Ki-67 positive cells agreed very well with the BrdU pulse and continuous labelling index, respectively. Our data further support the assumption that under certain conditions PCNA is a marker of S-phase cells, whereas Ki-67 can be used to quantify the growth fraction. Possible pitfalls of the techniques are discussed.     

Large-scale analysis of gene expression profiles

The accumulation of DNA microarray data has now made it possible to use gene expression profiles to analyse expression data. A gene expression profile contains the expression data for a given gene over various samples, and can be contrasted with an expression signature, which contains the expression data for a single sample. Gene expression profiles are most revealing when samples are grouped appropriately, either by standard clinical or pathological categories or by categories discovered through cluster analysis techniques. Expression profiles can exist at various levels of abstraction, yielding information across various tissues or across diseases within a particular tissue. Hypothesis tests may be applied to expression profiles on a large scale to identify candidate genes of interest.