Unravelling the myriad genetic and environmental factors that contribute to the pathogenesis of complex diseases such as cancer and heart disease remains an outstanding biomedical challenge. Now, Solveig Gretarsdottir et al., reporting in Nature Genetics, have provided a new piece of the puzzle of stroke — the third leading cause of death in western countries — through the identification of the gene encoding phosphodiesterase 4D (PDE4D) as a risk factor for ischaemic stroke.

Although there are a few known mutations that cause rare Mendelian forms of stroke, the genes involved are probably not contributing factors to the common forms of stroke, which are usually related to atherosclerosis. In a previous search for the genetic basis of common strokes, the same research team identified the chromosomal region 5q12 as a stroke-risk locus. In the present study, this locus has been finely mapped, leading to the characterization of PDE4D as a gene strongly associated with two atherogenic forms of stroke, carotid and cardiogenic stroke.

After establishing PDE4D as a risk factor for stroke, Gretarsdottir et al. looked for functional variants of PDE4D that might play a causal role in stroke formation. No significant genetic variants in the coding exons of PDE4D were found, and so attention was turned to the expression of PDE4D isoforms. An examination of messenger RNA levels showed that the overall level of PDE4D mRNA was decreased in randomly selected affected individuals, a difference that was mainly attributable to reduced levels of the PDE4D1, PDE4D2 and PDE4D5 isoforms. Association studies using single-nucleotide polymorphisms around the PDE4D gene also revealed three haplotypes in the study: wild type, at-risk and protective. Interestingly, affected individuals who possess the at-risk haplotype showed decreased levels of just the PDE4D7 and PDE4D9 isoforms.

PDE4D is expressed in the principal cell types involved with atherosclerosis, and is known to selectively degrade the second messenger cyclic AMP, which is ubiquitous in signal transduction and reduced levels of which are implicated in atherosclerosis. Reciprocally, the PDE4D isoforms identified as under-expressed in affected individuals — PDE4D1, PDE4D2 and PDE4D5 — are upregulated by cAMP. Taken together, these results indicate that dysregulation at the level of cAMP could be a key causal factor. The authors suggest that decreased cAMP concentrations, caused by greater activity of one or a few different isoforms of PDE4D, could lead to the lower levels of expression of PDE4D isoforms observed in stroke patients. As such, a small molecule that reduced the overall activity of PDE4D, or which specifically targeted the overactive isoforms, could result in elevated levels of cAMP, and a decreased risk of atherosclerosis and stroke.