Identification of novel universal housekeeping genes by statistical analysis of microarray data

Housekeeping genes are widely used as internal controls in a variety of study types, including real time RT-PCR, microarrays, Northern analysis and RNase protection assays. However, even commonly used housekeeping genes may vary in stability depending on the cell type or disease being studied. Thus, it is necessary to identify additional housekeeping-type genes that show sample-independent stability. Here, we used statistical analysis to examine a large human microarray database, seeking genes that were stably expressed in various tissues, disease states and cell lines. We further selected genes that were expressed at different levels, because reference and target genes should be present in similar copy numbers to achieve reliable quantitative results. Real time RT-PCR amplification of three newly identified reference genes, CGI-119, CTBP1 and GOLGAl, alongside three well-known housekeeping genes, B2M, GAPD, and TUBB, confirmed that the newly identified genes were more stably expressed in individual samples with similar ranges. These results collectively suggest that statistical analysis of microarray data can be used to identify new candidate housekeeping genes showing consistent expression across tissues and diseases. Our analysis identified three novel candidate housekeeping genes (CGI-119, GOLGA1, and CTBP1) that could prove useful for normalization across a variety of RNA-based techniques.     

On the nature of human housekeeping genes

Using a collection of expressed sequence tag (EST) data, we re-evaluated the correlation of tissue specificity with genomic structure, phyletic age, evolutionary rate and promoter architecture of human genes. We found that housekeeping genes are less compact and older than tissue-specific genes, and they evolve more slowly in terms of both coding and core promoter sequences. Housekeeping genes primarily use CpG-dependent core promoters, whereas the majority of tissue-specific genes possess neither CpG-islands nor TATA-boxes in their core promoters.     

Identification of genes associated with tumorigenesis of retinoblastoma by microarray analysis

There is no report on the gene expression profile of retinoblastoma (Rb). We analyzed the gene expression profile of Rb by the microarray technique. One thousand four genes were upregulated and 481 genes were downregulated. Microarray data were confirmed by semiquantitative RT-PCR for 5 genes in Rb samples: CDC25A, C17orf75, ERBB3, LATS2, and CHFR. Clusters of differentially expressed genes were identified on chromosomes 1, 16, and 17. Based on the expression profile, we hypothesized that the PI3K/AKT/mTOR (insulin signaling) pathway might be dysregulated in Rb. Our semiquantitative RT-PCR analysis of the PIK3CA, AKT1, FRAP1, and RPS6KB1 genes in Rb samples supported this hypothesis. We suggest that known inhibitors of this pathway could be evaluated for the treatment of Rb.     

Identification of appropriate housekeeping genes for gene expression studies in human renal cell carcinoma under hypoxic conditions

Molecular Biology Reports - Hypoxia pathways are deregulated in clear renal cell carcinoma (ccRCC) because of the loss of the von Hippel-Lindau tumor suppressor function. Quantitative PCR is a...     

Identification of genes associated with tumorigenesis of meibomian cell carcinoma by microarray analysis

Meibomian cell carcinoma (MCC) is a malignant tumor of the meibomian glands located in the eyelids. No information exists on the cytogenetic and genetic aspects of MCC. There is no report on the gene expression profile of MCC. Thus there is a need, for both scientific and clinical reasons, to identify genes and pathways that are involved in the development and progression of MCC. We analyzed the gene expression profile of MCC by the microarray technique. Forty-four genes were upregulated and 149 genes were downregulated in MCC. Differential expression data were confirmed for 5 genes by semiquantitative RT-PCR in MCC tumors: GTF2H4, RBM12, UBE2D3, DDX17, and LZTS1. We found dysregulation of two major pathways in MCC: MAPK and JAK/STAT. Clusters of genes on chromosomes 1, 12, and 19 were dysregulated in MCC. The data presented here will facilitate the identification of specific markers and therapeutic targets for the treatment of MCC patients.     

Identification and characterization of novel and unknown mouse epididymis-specific genes by complementary DNA microarray technology

To examine epididymal function, we attempted to identify highly expressed genes in mouse epididymis using a cDNA microarray containing PCR products amplified from a mouse epididymal cDNA library. We isolated one novel and four known genes-lymphocyte cytosolic protein 1 (Lcp1), complement subcomponents C1r/C1s, Uegf protein, and bone morphogenetic protein and zona pellucida-like domains 1 (Cuzd1), transmembrane epididymal protein 1 (Teddm1), and whey acidic protein 4-disulfide core domain 16 (Wfdc16)-with unknown functions in the epididymis. The novel gene, designated Serpina1f (serine peptidase inhibitor [SERPIN], clade A, member 1f), harbors an open reading frame of 1 233 bp encoding a putative protein of 411 amino acids, including a SERPIN domain. These five genes were predominantly expressed in the epididymis as compared to other organs. In situ hybridization analysis revealed their epididymal region-specific expression patterns. Real-time RT-PCR analysis revealed a significant increase in mRNA expression of these genes around puberty. Castration decreased their expression, except forLcp1. Testosterone (T) restored these reduced expressions, except forTeddm1; however, this restoration was not observed with 17 beta-estradiol (E2). Administration of T and E2 combination recovered the Serpina1f mRNA concentration; this recovery was also observed with T alone. However, the recovery of Cuzd1and Wfdc16mRNA concentrations was inadequate. Neonatal diethylstilbestrol treatment suppressed the Cuzd1, Wfdc16, and Serpina1f mRNA expression in the epididymis of 8-week-old mice; this was not observed with E2. These results suggest that our microarray system can provide a novel insight into the epididymal function on a molecular basis, and the five genes might play important roles in the epididymis.     

Evidence based selection of housekeeping genes

For accurate and reliable gene expression analysis, normalization of gene expression data against housekeeping genes (reference or internal control genes) is required. It is known that commonly used housekeeping genes (e.g. ACTB, GAPDH, HPRT1, and B2M) vary considerably under different experimental conditions and therefore their use for normalization is limited. We performed a meta-analysis of 13,629 human gene array samples in order to identify the most stable expressed genes. Here we show novel candidate housekeeping genes (e.g. RPS13, RPL27, RPS20 and OAZ1) with enhanced stability among a multitude of different cell types and varying experimental conditions. None of the commonly used housekeeping genes were present in the top 50 of the most stable expressed genes. In addition, using 2,543 diverse mouse gene array samples we were able to confirm the enhanced stability of the candidate novel housekeeping genes in another mammalian species. Therefore, the identified novel candidate housekeeping genes seem to be the most appropriate choice for normalizing gene expression data.     

Identification and optimization of classifier genes from multi-class earthworm microarray dataset

Monitoring, assessment and prediction of environmental risks that chemicals pose demand rapid and accurate diagnostic assays. A variety of toxicological effects have been associated with explosive compounds TNT and RDX. One important goal of microarray experiments is to discover novel biomarkers for toxicity evaluation. We have developed an earthworm microarray containing 15,208 unique oligo probes and have used it to profile gene expression in 248 earthworms exposed to TNT, RDX or neither. We assembled a new machine learning pipeline consisting of several well-established feature filtering/selection and classification techniques to analyze the 248-array dataset in order to construct classifier models that can separate earthworm samples into three groups: control, TNT-treated, and RDX-treated. First, a total of 869 genes differentially expressed in response to TNT or RDX exposure were identified using a univariate statistical algorithm of class comparison. Then, decision tree-based algorithms were applied to select a subset of 354 classifier genes, which were ranked by their overall weight of significance. A multiclass support vector machine (MC-SVM) method and an unsupervised K-mean clustering method were applied to independently refine the classifier, producing a smaller subset of 39 and 30 classifier genes, separately, with 11 common genes being potential biomarkers. The combined 58 genes were considered the refined subset and used to build MC-SVM and clustering models with classification accuracy of 83.5% and 56.9%, respectively. This study demonstrates that the machine learning approach can be used to identify and optimize a small subset of classifier/biomarker genes from high dimensional datasets and generate classification models of acceptable precision for multiple classes.     

Variable expression of some "housekeeping" genes during human keratinocyte differentiation

We investigated the expression levels of four cellular "housekeeping" genes during epithelial differentiation. Differentiation is a dynamic process and various cellular RNAs have been targeted for use as internal controls during differentiation of human keratinocytes, but the consistent expression of such standards has not been previously validated. We used the organotypic (raft) culture system to grow stratified and differentiated epithelium in vitro. We compared cellular RNAs from epithelial tissues of both normal human keratinocytes and keratinocytes whose differentiation scheme is altered by the replication of human papillomavirus. Using ribonuclease protection assays to quantify RNA expression levels, we found that beta-actin and glyceraldehyde-3-phosphate dehydrogenase levels fluctuated during epithelial differentiation, whereas cyclophilin RNA and 28S-ribosomal RNA were the most consistently expressed during epithelial differentiation. These stably expressed cellular RNAs can be targeted as controls to permit quantitative expression analyses of cellular and pathogen RNAs during epithelial differentiation under various experimental conditions.