Functionally different GPI proteins are organized in different domains on the neuronal surface.

We have investigated the organization, on the plasma membrane and in detergent-insoluble membrane vesicles, of two neuronal glycosylphosphatidylinositol-anchored (GPI) proteins: Thy-1, a negative regulator of transmembrane signalling; and prion protein, whose rapid endocytosis and Cu(2+) binding suggest that it functions in metal ion uptake. Prion protein occurred on the neuronal surface at high density in domains, located primarily at the cell body, which were relatively soluble in detergent. Thy-1, although much more abundantly expressed on neurons, occurred at lower density over much of the surface of neurites (and in lower abundance at the cell body) in domains that were highly resistant to detergent solubilization. Detergent-insoluble membrane vesicles contained Thy-1 at a density similar to that on the neuronal surface. Vesicles containing each protein could be separated by immunoaffinity isolation; lectin binding showed that they were enriched in different glycoproteins. Our results demonstrate a structural diversity of the domains occupied by functionally different GPI proteins.

Regulation of hemopexin synthesis in degenerating and regenerating rat sciatic nerve.

In injured peripheral nerves, hemopexin mRNA is expressed by fibroblasts, Schwann cells, and invading blood macrophages, and the protein accumulates in the extracellular matrix. This and its absence of regulation in injured central optic nerve suggest that hemopexin could play a positive role in peripheral nerve repair. Here, we studied the regulation of hemopexin expression in degenerating and regenerating nerves. After a sciatic nerve injury, both the synthesis of hemopexin and the level of its mRNA increase sharply during the first 2 days, leading to an accumulation of hemopexin in the nerve. Afterward, hemopexin expression decreases progressively in regenerating nerves. In permanently degenerated nerves, it is again transiently increased and then strongly decreased, whereas hemopexin from blood origin is accumulating. As part of the elucidation of the complex regulation of hemopexin expression in injured nerves, we demonstrate that interleukin-6 increases hemopexin synthesis in intact nerves, whereas adult rat serum, but not purified hemopexin, inhibits it in degenerated nerves. Hemopexin, known as acute-phase protein, is therefore one of the molecules rapidly and specifically up-regulated in injured peripheral nerves. More generally, our findings suggest that the acute phase could be not only a systemic liver-specific response but also a reaction of injured tissues themselves.

Long-lasting accumulation of hemopexin in permanently transected peripheral nerves and its down-regulation during regeneration.

We have previously demonstrated that hemopexin is present in the intact sciatic nerve and is overproduced in the distal stump after nerve transection (Swerts et al.: J Biol Chem 267:10596-10600, 1992). To get further insight into the function of this hemoprotein in nervous tissue, we have documented long-term changes in hemopexin levels in permanently degenerated (transected) and regenerating (crush-lesioned) sciatic nerves of adult rats, using immunochemical techniques. As early as a couple of days after nerve transection, the amount of hemopexin was raised in the distal stump and at the end of the proximal stump. Similarly, after a crush lesion hemopexin was rapidly increased at the injury site and in the distal part of the nerve. Subsequently, in transected nerves the level of hemopexin rose steadily and remained elevated, representing, three months after injury, over 20 times the amount found in intact contralateral nerves. In contrast, in crush-lesioned nerves, hemopexin level declined progressively in a proximodistal direction and returned to basal values 2 months after injury, together with axonal regeneration. This long-term increase in hemopexin in permanently degenerated nerves and its progressive return to normal levels during nerve regeneration suggests that hemopexin content could be regulated negatively, directly or indirectly, by growing axons. In turn, these results support the idea that hemopexin could be involved in the process of Wallerian degeneration and/or in nerve repair.