Christine C. Stichel

The CNS lesion scar: new vistas on an old regeneration barrier.

Abstractu2002Scar formation represents a reaction of nervous tissue to any form of physical injury. Research over the past decade has demonstrated that the scar composed of glial cells and several extracellular matrix molecules constitutes an obstacle to axon regeneration in the CNS. This review briefly summarizes the current knowledge on (a) the structural and functional features of the lesion scar and (b) the development of therapeutic interventions to override this regeneration barrier.

The mouse MPTP model: gene expression changes in dopaminergic neurons

Parkinsons disease (PD) is a common neurodegenerative disorder, characterized by the progressive loss of dopaminergic neurons in the substantia nigra. Although valuable animal models have been developed, our knowledge of the aetiology and pathogenic factors implicated in PD is still insufficient to develop causal therapeutic strategies aimed at halting its progression. The neurotoxicity induced by 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP) is one of the most valuable models for analysing pathological aspects of PD. In this paper we studied the gene expression patterns underlying the pathogenesis of MPTP‐induced neurodegeneration. We treated young and old C57BL/6 mice with different schedules of MPTP to induce degenerative processes that vary in intensity and time‐course. During the first week after intoxication we used nonradioactive in situ‐hybridization to investigate the expression patterns of genes associated with (i) dopamine metabolism and signalling; (ii) familial forms of PD; (iii) protein folding and (iv) energy metabolism. MPTP injections induced different severities of neuronal injury depending on the age of the animals and the schedule of administration as well as a significant degeneration in the striatum. In situ hybridization showed that MPTP intoxication initiated a number of gene expression changes that (i) were restricted to the neurons of the substantia nigra pars compacta; (ii) were correlated in intensity and number of changes with the age of the animals and the severity of histopathological disturbances; (iii) displayed in each a significant down‐regulation by the end of one week after the last MPTP injection, but (iv) varied within one MPTP regimen in expression levels during the observation period. The subacute injection of MPTP into one‐year‐old mice induced the most severe changes in gene expression. All genes investigated were affected. However, α‐synuclein was the only gene that was exclusively up‐regulated in MPTP‐treated animals displaying cell death.

Inhibition of collagen IV deposition promotes regeneration of injured CNS axons

Scarring impedes axon regrowth across the lesion site and is one major extrinsic constraint to effective regeneration in the adult mammalian central nervous system. In the present study we determined whether specific biochemical or immunochemical modulation of one major component of the scar, the basal membrane (BM), would provide a means to stimulate axon regeneration in the mechanically transected postcommissural fornix of the adult rat. Basal membrane developed within the first 2u2003weeks after transection in spatiotemporal coincidence with the abrupt growth arrest of spontaneously regrowing axons. Local injection of anticollagen IV antibodies or α, α′‐dipyridyl, an inhibitor of collagen triple helix formation and synthesis, significantly reduced lesion‐induced BM deposition. This treatment allowed massive axon elongation across the lesion site. Anterograde tracing provided unequivocal evidence that regenerating axons follow their original pathway, reinnervate the appropriate target, the mammillary body, and become remyelinated with compact myelin. Presynaptic electrophysiological recordings of regenerated fibre tracts showed recovery to nearly normal conduction properties. Our results indicate that lesion‐induced BM is an impediment for successful axonal regeneration and its reduction is a prerequisite and sufficient condition for regrowing axons to cross the lesion site.

EXPERIMENTAL STRATEGIES TO PROMOTE AXONAL REGENERATION AFTER TRAUMATIC CENTRAL NERVOUS SYSTEM INJURY

A damage or pathological process that destroys the continuity of axons in the mature central nervous system (CNS) has devastating consequences and produces lasting functional deficits. One of the major challenges in this field is to stimulate the regrowth of severed axons and reconstruction of pathways. Recent progress in molecular and cell biology has resulted in an explosion of knowledge on factors in the adult CNS being nonsupportive or even actively inhibitory to axonal regrowth. The new findings have a strong impact on the development of new therapeutic concepts directed to stimulate axonal regeneration. They give rise to cautious optimism, showing that under some circumstances repair of a CNS lesion is possible. In this review the authors summarize the current knowledge on the factors and mechanisms involved in regeneration failure and provide an overview of the current therapeutic approaches that may enable effective CNS regeneration in the future.

Differential expression of the small chondroitin/dermatan sulfate proteoglycans decorin and biglycan after injury of the adult rat brain.

Chondroitin sulfate proteoglycans are widespread extracellular matrix proteins and are specifically upregulated after CNS injury at the lesion site. Many proteoglycan core proteins have been described in the rat brain, but detailed analysis of individual proteoglycans expressed after injury are missing. The present study represents an initial attempt to assess the diversity and timing of lesion-induced expression of proteoglycans in order to elucidate their functional role in CNS injury and repair. Using immunocytochemical methods we analysed the expression of decorin and biglycan in the transected postcommissural fornix of the adult rat. Transection of the fornix induced the upregulation of both decorin and biglycan. However, their expression differed with respect to time course, regional extent and cellular localization. The rapid upregulation of decorin within a wide area around the lesion was followed by a massive appearance of biglycan that remained restricted to the transection site. Three months after lesion, differences of the area size of decorin- and biglycan-immunoreactivities were no longer detectable. Both proteoglycans were restricted to the lesion site and the fornix stumps. While decorin was primarily expressed by astrocytes, biglycan was deposited extracellularly in sheet-like structures. The upregulation of both proteoglycans persisted for at least up to 6 months after lesion. These strong but divergent lesion-induced expression patterns indicate important but different roles of decorin and biglycan in CNS injury.

Parkin expression in the adult mouse brain.

Mutations in a protein designated Parkin were shown to be involved in the pathogenesis of autosomal recessive juvenile parkinsonism. Nothing is known about its regional and subcellular distribution in the mouse. In order to elucidate the Parkin mRNA and protein distribution in the adult mouse, the mouse cDNA was cloned and polyclonal antisera were generated against the N‐terminal part of mouse Parkin. The antibodies were shown to be specific using Western blot analysis, immunostaining of cells transfected with mouse Parkin and pre‐absorption tests. The Parkin protein expression profile was studied using immunohistochemistry and Western blot analysis and was compared with that of the mRNA yielded by in situ hybridization and RT‐PCR analysis. Parkin protein was widely distributed in all subdivisions of the mouse brain. Low levels were found in the telencephalon and diencephalon, while the brainstem contained a large number of cells heavily expressing Parkin. Ultrastructural analysis and double immunohistochemistry revealed that the majority of Parkin‐expressing cells were neurons, while only single glial cells exhibited immunostaining. The protein was distributed nonhomogeneously throughout the entire cytoplasm. A subpopulation of Parkin‐immunopositive cells displayed speckled immunodeposits in the nucleus. Dopaminergic cells of the substantia nigra pars compacta exhibited high levels of Parkin mRNA but no Parkin protein, while the striatum contained immunopositive profiles but no mRNA signals. Our data indicate that Parkin is neither restricted to a single functional system nor associated with a particular transmitter system. The speckled nuclear distribution of Parkin immunoreactivity strongly suggests a role for Parkin in gene expression.

Non-motor behavioural impairments in parkin-deficient mice.

Mutations in the parkin gene are the major cause of early‐onset familial Parkinsons disease (PD). We previously reported the generation and analysis of a knockout mouse carrying a deletion of exon 3 in the parkin gene. F1 hybrid pa+/– mice were backcrossed to wild‐type C57Bl/6 for three more generations to establish a pa–/–(F4) mouse line. The appearance of tyrosine hydroxylase‐positive neurons was normal in young and aged pa–/– (F4) animals. Loss of parkin function in mice did not enhance vulnerability of dopaminergic neurons to 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP) toxicity. However, the pa–/– (F4) mice displayed impaired exploration and habituation to a new environment and exhibited thigmotaxis behaviour in the open field and Morris water maze. Abnormal anxiety‐related behaviour of pa–/– (F4) mice was also observed in the light/dark exploration test paradigm. Dopamine metabolism was enhanced in the striatum of pa–/– (F4) mice, as revealed by increased homovanillic acid (HVA) content and a reduced ratio of dihydroxyphenylacetic acid (DOPAC)/HVA. The alterations found in the dopaminergic system could be responsible for the behavioural impairments of pa–/– (F4) mice. Consistent with a recent observation of cognitive dysfunction in parkin‐linked patients with PD, our findings provide evidence of a physiological role of parkin in non‐motor behaviour, possibly representing a disease stage that precedes dopaminergic neuron loss.

Basal membrane-depleted scar in lesioned CNS: characteristics and relationships with regenerating axons

The lesion scar formed after CNS injury is an impediment to axonal regeneration and leads to growth arrest or misrouting of sprouting axons. Our previous study showed that pharmacological reduction of basal membrane formation within the scar can overcome this scar impermeability [Stichel C. C. et al. (1999) Eur. J. Neurosci. 11, 632-646]. The aim of the present study was to characterize the basal membrane-depleted scar and to analyse its relationships with penetrating axons. The experiments comprised two groups of animals in which the left postcommissural fornix was transected; in addition, one group received a local immediate injection of the collagen IV-reducing agent dipyridyl, while the other group received an injection of phosphate-buffered saline. Immunohistochemical methods were used to characterize scar formation and scar-axon relationships. Animals receiving dipyridyl showed reduction of collagen IV-immunopositive basal membrane in the lesion center, which was accompanied by: (i) a decrease in laminin, as well as chondroitin and heparan sulfate proteoglycan, deposition in the lesion center; (ii) an increase in chondroitin and keratan sulfate proteoglycan expression in the perilesional area; (iii) a typical activation pattern of astrocytes and microglia/macrophages; (iv) axons regenerating through this modified scar were closely associated with various glial cell types and crossed a prominent chondroitin/keratan sulfate proteoglycan matrix. Our results suggest that neither the formation of a reactive astroglial network nor the accumulation of microglia/macrophages or the deposition of chondroitin and keratan sulfate proteoglycans in the perilesional area represent a barrier to regrowing axons. The present approach demonstrates that the lesion-induced basal membrane itself is the primary determinant of scar impermeability.

Parvalbumin in Cat Brain: Isolation, Characterization, and Localization

Because of the increasing evidence that Ca2+‐binding proteins have important regulating functions in nerve cells and because of the indications that there are species differences in the structures of these proteins, parvalbumin was purified from cat brain and muscle. Brain and muscle parvalbumins were found to be indistinguishable from each other in their biochemical and im‐munological properties. However, cat parvalbumin differs from all other mammalian parvalbumins by its apparently lower Mr on sodium dodecyl sulfate‐polyacryl‐amide gel electrophoresis of 10–11K (compared to rat parvalbumin, 12K), and a lower pI of 4.6 (rat parvalbumin, 4.9), in the tryptic peptide maps, and in the immu‐nological properties, indicating a distinct primary structure. With the purified parvalbumin as antigen, polyclonal antibodies were raised in rabbits and these were subsequently used for immunohistochemical localizations of parvalbumin in the cat brain. In the visual cortices of adult cats immunoreactive neurons were present throughout layers II and IV. In cerebellar cortex. Pur‐kinje, basket, and stellate cells were immunoreactive. Comparison with staining patterns obtained with anti‐serum against rat parvalbumin revealed some cross‐reactivity but confirmed the existence of species differences in the antigenic structure of rat and cat parvalbumin.

Restricted appearance of tenascin and chondroitin sulphate proteoglycans after transection and sprouting of adult rat postcommissural fornix

SummaryTransected fibres of the adult rat postcommissural fornix sprout over short distances but fail to traverse the lesion site and terminate in close vicinity to the wound. As a step in defining the molecular environment responsible for regeneration failure at the lesion site, we have used immunocytochemistry to analyse the spatio-temporal expression pattern of two putative growth-inhibitory extracellular matrix components, tenascin and chondroitin sulphate proteoglycans and their topographical relationship to the sprouting axons. Both tenascin and chondroitin sulphate proteoglycan labelling appeared after fornix transection and were confined to the immediate vicinity of the lesion site. While tenascin-labelling was associated with astrocytes and microglia/macrophages, which accumulate preferentially at the tract borders, chondroitin sulphate proteoglycan labelling appeared as a homogeneous meshwork around the wound. Tenascin-like immunoreactivity disappeared between 17 days and 4 weeks, but chondroitin sulphate proteoglycan staining persisted at least up to 14 months after transection. Regrowing fornix fibres invaded and elongated within the chondroitin sulphate proteoglycanimmunopositive region up to the lesion site, where they terminated. This zone of axonal growth inhibition was neither characterized by an increase of chondroitin sulphate proteoglycan immunoreactivity nor by the presence of tenascinimmunopositive structures. The spatio-temporal distribution patterns of tenascin and chondroitin sulphate proteoglycan and the permeability of the chondroitin sulphate proteoglycan-immunopositive region for sprouting axons do not support the hypothesis that chondroitin sulphate proteoglycan alone and/or tenascin inhibit the advance of sprouting fornix fibres.