DRPLA aggregation and transglutaminase, revisited

Cells that express proteins with long amino-terminal stretches of glutamines, such as those employed by Igarashi et al.¹, should be useful for investigating the question of whether crosslinking by transglutaminase², might cause precipitation of pathological proteins of this type, and whether the enzyme is responsible for the induction of apoptosis. However, in the absence of proper controls³, the study by Igarashi et al. still leaves the issue unresolved.

The authors' claims are based on the effects of cystamine⁴ and monodansylcadaverine⁵ (MDC), two compounds that can be used to probe the reactions of transglutaminases. Cystamine is known to inactivate the enzyme³, probably by forming a mixed disulfide. This raises the question of whether similar interference with other thiol enzymes in COS cells (such as the caspases) might also occur. As primary amines, both cystamine and MDC are good substrates for transglutaminase⁶, and are useful for blocking the glutamine residues in proteins which would otherwise participate in forming N^ε(γ-glutamyl)lysine bridges. However, important hurdles must be overcome before making a reasonable case for suggesting that these substrates might specifically inhibit crosslinking by transglutaminase in a cellular setting. Precautions should be taken to inhibit oxidases^3,7 so as to prevent possible conversions of the substrates to aldehydes, which would certainly modify a variety of cell constituents. To prove that the transglutaminase-reactive, primary amino group in MDC was needed for inhibition, a minimal requirement would be to compare the effects of MDC with those of its dimethylated tertiary amine analogue^3,7. Of course, direct labelling of the DRPLA protein by MDC should also be provided^7,8,9. In addition, it is important to bear in mind that the chemical structure of MDC is quite similar to compounds (W5, W7) that are widely marketed as calmodulin inhibitors, raising the possibility that MDC might interfere with the functions of calmodulin in COS cells. It was rather puzzling to find in the article of Igarashi et al. that MDC prevented nuclear fragmentation, that is, apoptotic cell death (Fig. 8e), but was ineffective in blocking the formation of aggregate bodies supposedly produced by the transglutaminase-catalysed crosslinking of the DRPLA protein (Fig. 8b).

Howard Green's original suggestion²— not just as an alternative, but perhaps in concert with the polar zipper^10,11 mechanism of precipitation—still remains an attractive and viable possibility. It may very well turn out that the reaction of transglutaminases with the N-terminal polyglutamine extensions in the DRPLA protein, and also in the other abnormal gene products of related disorders¹², has a role in the development of the neurodegenerative process. Further experiments with COS cells expressing the truncated DRPLA protein could help to examine the validity of this notion.