Transcript-specific expression profiles derived from sequence-based analysis of standard microarrays

Abstract

Background: Alternative mRNA processing mechanisms lead to multiple transcripts (i.e. splice isoforms) of a given gene which may have distinct biological functions. Microarrays like Affymetrix GeneChips measure mRNA expression of genes using sets of nucleotide probes. Until recently probe sets were not designed for transcript specificity. Nevertheless, the re-analysis of established microarray data using newly defined transcript-specific probe sets may provide information about expression levels of specific transcripts.

Methodology/principal findings: In the present study alignment of probe sequences of the Affymetrix microarray HG-U133A with Ensembl transcript sequences was performed to define transcript-specific probe sets. Out of a total of 247,965 perfect match probes, 95,008 were designated "transcript-specific", i.e. showing complete sequence alignment, no cross-hybridization, and transcript-, not only gene-specificity. These probes were grouped into 7,941 transcript-specific probe sets and 15,619 gene-specific probe sets, respectively. The former were used to differentiate 445 alternative transcripts of 215 genes. For selected transcripts, predicted by this analysis to be differentially expressed in the human kidney, confirmatory real-time RT-PCR experiments were performed. First, the expression of two specific transcripts of the genes PPM1A (PP2CA_HUMAN and P35813) and PLG (PLMN_HUMAN and Q5TEH5) in human kidneys was determined by the transcript-specific array analysis and confirmed by real-time RT-PCR. Secondly, disease-specific differential expression of single transcripts of PLG and ABCA1 (ABCA1_HUMAN and Q5VYS0_HUMAN) was computed from the available array data sets and confirmed by transcript-specific real-time RT-PCR.

Conclusions: Transcript-specific analysis of microarray experiments can be employed to study gene-regulation on the transcript level using conventional microarray data. In this study, predictions based on sufficient probe set size and fold-change are confirmed by independent means.

Figure 1. Intensity histograms of several probe categories.

Gene-specific and transcript-specific probes show more high intensity signals as compared to non-perfect or no-match probes. a) Probe intensities of one microarray (LD2) were RMA-background-adjusted and logarithmized (base 2). Probes were assigned to categories shown in the legend as defined in the results section. The normalized intensities of each category are put into 100 evenly-spaced intervals and plotted as histogram. The probe intensity 95%-quantile of Affymetrix-defined negative control probe sets is plotted in yellow. b) As the number of probes varies between the different categories it is hard to compare the shapes of the distributions in a). Therefore, the histograms were normalized with their number of probes in this plot. Averaging over multiple arrays instead of analyzing one array gives similar results.

Figure 2. Comparison of alternative transcript abundance in microarray and real-time RT-PCR.

On the left microarray signal intensities are shown for the genes PPM1A and PLG. Confirmatory real-time RT-PCR data are shown on the right. Lowly and highly abundant transcripts of the genes PPM1A and PLG were measured using microarrays in LD tissue. Transcript-specific probe intensities were background-adjusted and quantile-normalized using RMA. These are shown as single probe values (triangles) and, furthermore, as summarized transcript-specific probe set values (dots). In addition, the transcripts were quantified in the unaffected part of TN tissue using real-time RT-PCR. Real-time RT-PCR data are normalized to the transcript with lower abundance.

Figure 3. Microarray and real-time RT-PCR transcript measurements of PLG and ABCA1 in two patient cohorts.

For the gene PLG, transcript PLMN_HUMAN was repressed in DD compared to LD controls, while Q5TEH_HUMAN showed no differential expression. Real-time RT-PCR measurements on the same tissues were in agreement with these findings. For gene ABCA1, transcript ABCA1_HUMAN was induced in DN compared to LD, and Q5VYS0_HUMAN was not regulated. Real-time RT-PCR measurement confirmed the induction of ABCA1_HUMAN. Q5VYS0_HUMAN was expressed at a too low level to be measured (value 0). Real-time data are normalized to the cohort with lower abundance.