Microarray analysis of gene expression profiles of cardiac myocytes and fibroblasts after mechanical stress, ionising or ultraviolet radiation

Abstract

Background: During excessive pressure or volume overload, cardiac cells are subjected to increased mechanical stress (MS). We set out to investigate how the stress response of cardiac cells to MS can be compared to genotoxic stresses induced by DNA damaging agents. We chose for this purpose to use ionising radiation (IR), which during mediastinal radiotherapy can result in cardiac tissue remodelling and diminished heart function, and ultraviolet radiation (UV) that in contrast to IR induces high concentrations of DNA replication- and transcription-blocking lesions.

Results: Cultures enriched for neonatal rat cardiac myocytes (CM) or fibroblasts were subjected to any one of the three stressors. Affymetrix microarrays, analysed with Linear Modelling on Probe Level, were used to determine gene expression patterns at 24 hours after (the start of) treatment. The numbers of differentially expressed genes after UV were considerably higher than after IR or MS. Remarkably, after all three stressors the predominant gene expression response in CM-enriched fractions was up-regulation, while in fibroblasts genes were more frequently down-regulated. To investigate the activation or repression of specific cellular pathways, genes present on the array were assigned to 25 groups, based on their biological function. As an example, in the group of cholesterol biosynthesis a significant proportion of genes was up-regulated in CM-enriched fractions after MS, but down-regulated after IR or UV.

Conclusion: Gene expression responses after the types of cellular stress investigated (MS, IR or UV) have a high stressor and cell type specificity.

Figure 1

Accuracy of the linear model. Dot plot of all PM probe signals as calculated by the linear model, against actual PM probe signals determined from fibroblasts after UV (A). The data presented in Figure 1A were used to calculate correlation coefficients (R²) between the signals calculated by the linear model and the actual PM signals of fibroblasts after UV (B).

Figure 2

Numbers of differentially expressed genes. Numbers of differentially expressed genes (q < 0.005) with a unique LocusLink ID in cultures of fibroblasts (dotted) and CM-enriched cultures after MS (A), in cultures of fibroblasts (dotted) and CM-enriched fractions after IR (B), or in cultures of fibroblasts (dotted) and CM-enriched fractions after UV (C). Overlapping parts of the circles represent genes that show differential expression both in CM-enriched cultures/fractions and cultures of fibroblasts.

Figure 3

Percentages of differentially expressed genes. Percentage of total number of genes within functional groups that are differentially expressed in CM-enriched cultures/fractions and cultures of fibroblasts after MS, IR or UV. Numbers between brackets represent total numbers of genes within a functional group. For example, of the 22 MAPkinases and phosphatases found to be represented by the array, 27% were differentially expressed in CM-enriched fractions after UV. AA: amino acid. *Hypergeometric probability P < 0.005

Figure 4

Percentages of up-regulated genes. Percentage of changed genes that were up-regulated per functional group after IR or UV. Numbers between brackets represent numbers of differentially expressed genes in CM-enriched fractions and cultures of fibroblasts, respectively. For example, of the 8 MAPkinases and phosphatases that were differentially expressed in CM-enriched fractions after UV, 63% were up-regulated. *Hypergeometric probability P < 0.005

Figure 5

Representative result of the Tenascin C PCR Total RNA was isolated from CM-enriched cultures at 24 hours after MS or control treatment. After cDNA synthesis, semi-quantitative PCR was used to determine Tenascin C gene expression. Lane 1: smart ladder; lane 2: negative control; lane 3: positive control; lanes 4 and 5: CM-enriched after control treatment; lanes 6 and 7: CM-enriched after MS.