Research focus: gene expression profiling

Global gene expression profiling has taken our understanding of specific disease and biological processes from that of one gene at a time to a view encompassing thousands of genes. Indeed, major pharmaceutical companies and academic centers have established internal microarray programs and routinely utilize gene expression data generated by microarrays by applying these technologies to clinical trials, basic research and biomarker discovery programs. The latter endeavor, biomarker discovery, has particular promise, as the successful implementation of a biomarker for either safety or efficacy in drug development can allow better informed project decisions and improve the selection and prioritization of ongoing therapeutic programs. In some cases, a biomarker can serve as a prognostic or diagnostic marker for clinical use.

Company Gene Logic is uniquely positioned to bring the power of gene expression profiling to biomarker discovery and development programs. Gene Logic’s warehouse of gene expression data from normal and diseased human tissue, as well as tissues from pre-clinical species, can provide in silico verification of the usefulness across multiple species and tissues. This forms the foundation of the company’s support for biomarker discovery, which also can leverage a mechanistic understanding of the target, disease and toxicity. Testing potential candidate biomarkers for tissues and disease specificity early in discovery and/or development provides a clear sense if these markers are applicable and, if so, in which species. Project teams can better assess non-clinical and clinical environments when specificity of the candidate biomarkers is established and characterized.

Here we describe a workflow schema that can quickly demonstrate tissue, disease and species specificity of selected candidate biomarkers, in this case Arginase 1. This analysis found that Arginase 1 has expression induction specific to normal liver tissues in rat, canine and human species. However, it was found to lack specificity for human liver diseases but may have expanded utility for other human disorders including Psoriasis and Wegener’s Granulomatosis. A full case study describing the discovery of specific markers is available from Gene Logic upon request.

To demonstrate how specific potential candidate biomarkers could be tested for tissue, species and disease specificity, Arginase 1, a known biomarker used as a diagnostic in the clinic since 1959 for human liver disease, was chosen. Normal tissue panels for rat, canine and human were examined for Arginase 1 expression and to test for tissue specificity. A set of human diseased tissues were then examined for their expression of Arginase 1 to test for disease specificity.

Arginase 1 is a cytoplasmic urea cycle enzyme found mainly in the liver. It is present in both human and rat sera and is Indicative of chemically-induced liver damage. Arginase 1 sera levels can precede ALT and AST changes and have been used to monitor liver transplantations and (Clin Chim Acta (1998) 271:11-13) for acute and chronic hepatic disease, as well as partial hepatectomy (Clin Biochem (2001) 34:455-61)

Normal samples of rat, canine and human tissues and human diseased tissues from ToxExpress and BioExpress Systems were examined.

When Arginase 1 expression level in normal rat tissues was evaluated, the highest expression was found in normal rat liver tissue as compared to other normal rat tissues. This is illustrated in Figure 1 where the blue circle indicates high level of expression for Arginase 1 in normal rat liver tissue.

As in the rat, expression is the highest for Arginase 1 in normal canine liver tissue as compared to other normal canine tissues (as illustrated with the blue circle).

Expression of Arginase 1 across normal human tissues suggests that this marker is liver-specific, although the response shows a wide degree of variation across this sample set. This is indicated by the length of the circle illustrating the response of Arginase

The panel of normal tissues suggests that Arginase 1 would be liver specific across the three species.

Would this tissue specificity differ when human diseased tissues are examined?
When Arginase 1 expression level was surveyed against human normal and diseased tissues, an interesting finding was discovered. The human diseased tissues showing pathological evidence of steatosis and cirrhosis expressed Arginase 1 at the same level as human normal liver tissues. By contrast, Arginase 1 expression in other human tissues suggests that this marker appears to have better specificity for psoriasis during active inflammation and for Wegener’s Granulomatosis (WG).

These findings suggest that Arginase 1 is not a specific marker of liver toxicity as suggested in the literature. While Arginase 1 expression in normal tissues appears to be liver specific, it is significantly modulated in skin from patients with Psoriasis and in white blood cells from patients with Wegener’s Granulomatosis. The modulation in these disease states suggest that this marker may offer extended utility as a diagnostic for these disorders.

This type of database survey for tissue, species and disease-state specificity can be extended to novel candidate markers, providing evidence to support further validation.