Two distinct types of reactive astrocytes were studied in rat CNS. Reactive astrocytes secondary to penetrating trauma (anisomorphic gliosis) were induced by stab wounds to the brain. Reactive astrocytes secondary to Wallerian degeneration (isomorphic gliosis) were induced in spinal cord dorsal columns by dorsal rhizotomy proximal to dorsal root ganglia. Anisomorphic glial scars did not stain with antibodies to the glial hyaluronate‐binding protein (GHAP), a structural glycoprotein of white matter extracellular matrix. Conversely, isomorphic glial scars were still GHAP‐positive 3 months after dorsal root transection. Only after 5 months did GHAP immunoreactivity start to disappear from the isomorphic glial scar. Extensive dorsal rhizotomy was performed at the lumbar level to produce Wallerian degeneration of spinal cord dorsal columns. One month later, the rats were reoperated and two thoracic dorsal roots were implanted in the degenerated dorsal columns. The rats were examined 1 month after grafting. As expected, there was a dense anisomorphic glial scar at the site of surgery, while the dorsal columns above the graft showed isomorphic gliosis. Extensive axonal growth was observed in the dense glial scar surrounding the graft. Conversely, no axonal growth was observed in the degenerated dorsal columns undergoing isomorphic gliosis above the implant. The findings suggested that GHAP‐negative astrocytes responding to traumatic injury are permissive for axonal growth and that GHAP‐positive astrocytes responding to Wallerian degeneration are not permissive.

Neurofilament phosphorylation in regenerating rat sciatic nerve was studied by indirect immunofluorscence with monoclonal antibodies reacting with phosphorylated epitopes of the 2 large polypeptides of the neurofilament protein triplet (NF 150K, NF 200K). One group of antibodies decorated axons early in the process. In fact, no differences were seen in double labeled sections between these antibodies and polyclonal neurofilament antibodies as to their reactivity with the distal stump of transected sciatic nerves. Another group stained axons after they had completed their elongation i.e., after they had reached the distal part of the denervated sciatic nerve. In general, the epitopes recognized by antibodies in this group appeared more sensitive to phosphatase digestion as compared to the first group. Furthermore, there was a good correlation between the thickness of the regenerated axons and staining with these monoclonal antibodies. Thick axons (like those observed in normal nerves) were stained, while bundles of thin axons remained unstained. Monoclonal II32 stained regenerated axons in a remarkable segmental pattern. With this antibody, continuous decoration of the axons was still not observed 7 weeks after transection, the longest follow-up period in this study. We suggest that some neurofilament phosphorylation events may contribute to the stabilization of the axonal cytoskeleton and that abnormalities persist in regenerated axons as to the extent of neurofilament phosphorylation.

Rat spinal cord was stained by indirect immunofluorescence with 11 neurofilament monoclonal antibodies that recognize phosphorylated epitopes. All monoclonals were axon‐specific in this location. The large motoneurons containing bundles of neurofilaments did not stain and the pattern remained unchanged after transection of the sciatic nerve in the thigh. With nine monoclonals, stained motoneurons were observed in the ventral horns 3 days, 5 days, 1 week, and 2 weeks after transection of the ventral roots close to the spinal cord. The abnormal motoneurons were typically scattered among normal (i.e., nonstained) cells. Even in animals showing the most severe reaction, the whole motoneuron population at the site of rhizotomy was not affected, stained and nonstained perikarya often coexisting side by side. Stained motoneurons were no longer observed 3 weeks after ventral root transection. Changes in neuronal immunoreactivity were also observed after dorsal root transection. However, a different population was affected, i.e., middle‐sized neurons in dorsal horns and at the base of ventral horns. With two monoclonals (A9 and D21), cell bodies remained negative following all operations. It is concluded that axotomy in proximity of the cell body may induce certain neurofilament phosphorylation events in motor neuron perikarya, whereas other phosphorylation events remain confined to the axons under these experimental conditions. The absence of changes after transection of the sciatic nerve in the thigh suggests that neurofilament phosphorylation is a reaction to cell injury rather than a cellular event related to nerve regeneration.

Rat dorsal root ganglia and spinal cord were stained with 12 monoclonal antibodies reacting with phosphorylated epitopes of two neurofilament proteins (NF 150K and NF 200K). Three monoclonal antibodies were axon‐specific in both locations; neuronal perikarya were not stained. Nine monoclonal antibodies stained a subpopulation of neurofilament‐positive sensory neurons, as indicated by double labeling experiments with polyclonal antibodies reacting with phosphorylated and dephosphorylated forms of the neurofilament protein triplet. Of these nine antibodies, two stained motor neuron perikarya in the spinal cord, while the remaining seven antibodies were axon‐specific in this location. Subpopulations of stained and unstained motor neurons were not observed. With all 12 antibodies, the staining pattern in the lumbar dorsal root ganglia and spinal cord remained unchanged following sciatic nerve crush and ligature. The findings suggest that, in the neurofilament, some phosphorylated epitopes are axon specific, while other phosphorylated epitopes are present in both axons and perikarya. Furthermore, they suggest that differences exist between neuronal populations as to the presence of phosphorylated epitopes in perikaryal neurofilaments. It remains to be seen whether phosphorylation events in perikarya and axons have similar or different effects on neurofilament structure and function.