CALPAIN IS ACTIVATED EARLY IN THE COURSE OF WALLERIAN DEGENERATION
Glass J.D., Culver D.G., Wu Y., Gorke A. Emory University School of Medicine, Atlanta, Georgia, USA
Axonal degeneration is a common feature of a variety of degenerative disorders of the nervous system, including peripheral neuropathy. We have been using the model of Wallerian degeneration to study the mechanisms of axonal degeneration, and have focused on the role of calcium and the calcium activated protease calpain in this process. Previous work suggested that calpain activation is responsible for cytoskeletal degradation during Wallerian degeneration. Using a panel of highly sensitive and specific antibodies to the 80 kD subunit of m-calpain, we found a marked reduction in calpain protein within hours of nerve transection in rodent sciatic nerves. This reduction in calpain occurred days before the onset of axonal degeneration. In order to understand the mechanism of calpain protein loss, we injected into the nerve a specific calpain inhibitor, calpastatin, just prior to nerve transection. Ratios of calpain protein in transected versus untransected sciatic nerves (in the same animal) were compared in calpain injected, saline injected, and uninjected animals. Whereas calpain levels decreased significantly in saline or uninjected transected nerves, calpastatin injected nerves retained levels of calpain comparable to the untransected sciatic nerve. These findings suggest that early calpain loss after nerve injury results from calpain-mediated calpain proteolysis (autoproteolysis). We propose that calpain activation, as opposed to being the penultimate event leading to axonal degeneration, is an early and likely necessary event within a proteolytic cascade that is activated after nerve injury.
DIVERGENT MACROPHAGE RESPONSES IN WALLERIAN DEGENERATION OF THE PNS AND CNS: A NOVEL POPULATION OF CD8+ MACROPHAGES
Stoll G., Jander S. Heinr.-Heine-Univ., Düsseldorf, Germany
Transection or crush of a peripheral nerve leads to prompt infiltration of hematogenous macrophages with rapid removal of debris that is lacking after injury of CNS fibre tracts. Based on our findings in cerebral ischemia (Jander et al., Eur J Neurosci (1998)10:680-688) we asked whether microglia/ macrophage responses differ phenotypically in PNS and CNS Wallerian degeneration (WD). In confirmation of previous studies we found that resident macrophages constitutively expressed CD4-protein and -mRNA, which further increased on macrophages in the distal stump of the sciatic (SN) and on microglia in optic nerve (ON) during WD. However, in the distal nerve segment of SN an additional population of CD8+ cells appeared in the perineurium at day 2-4. CD8+ cells moreover infiltrated the distal stump parenchyma and were abundant between day 7 and 28. CD8+ cells, which were absent in the normal SN, could be identified as EDl/Ox42+macrophages by double immunofluorescence. The immunocyto-chemical data could be confirmed by PCR showing concomitant induction of CD8a- and CD8b-mRNA. In contrast, activated microglia in the distal stump of the transected ON did not express CD8. In conclusion, our study shows that the rapid hematogenous cell infiltration in WD of the PNS involves a novel population of CD8+ macrophages that is lacking after injury of CNS fibre tracts. The functional properties of this unique cell type are currently under investigation. Supported by the DFG (SFB 194 B6) and the Hermann- and Lilly-Schilling-Stiftung
DELAYED AXONAL DEGENERATION AND REGENERATION IN MICE LACKING IMMUNOLOGICAL (INDUCIBLE) NITRIC OXIDE SYNTHASE (iNOS)
Zochodne D.W., Levy D. University of Calgary Calgary, Canada
We have previously demonstrated that nitric oxide is released
within injured peripheral nerve trunks from early axonal endbulbs
containing eNOS (endothelial nitric oxide synthase) and later
from Schwann cells and macrophages that newly synthesize iNOS
(immunologic or inducible NOS). In this work we examined the potential
contribution of iNOS expression to axonal degeneration and regeneration
in mice with a targeted mutation of the iNOS gene. Three injury
models were studied: (i) Transection: After complete sciatic nerve
transection without resuturing, there was a delay in the recovery
of the M potential from sciatic-tibial reinnervated foot interosseous
muscles in mice lacking iNOS, still evident by 10 weeks. At this
time regenerated myelinated fibers were smaller in caliber, indicating
delayed maturation. (ii) Crush: At a fixed distance distal to
a sciatic nerve crush site, quantitative morphometry after 2 weeks
indicated that iNOS "knockout" mice had fewer regenerating
myelinated fibers and larger numbers of persistent degenerating
fiber profiles than wild-type controls. By 6 weeks, this deficit
had recovered. (iii) CCI: Following a chronic constriction injury
(CCI) with prolonged axonal degeneration from strangulation, "knockout"
mice had delays in myelinated fiber breakdown, in new axonal sprouting
and in the appearance of behavioural features of neuropathic pain.
iNOS appears to be important in the facilitation of axonal degeneration
and subsequent regeneration, resembling the deficit previously
described, but not understood, in the C57/Ola mouse. Supported
by MRC, MDAC and AHFMR
MECHANICAL ROLE FOR THE EXTRACELLULAR MATRIX IN MYELINATIION OF DORSAL ROOT GANGLION CULTURES
Podratz J.L., McDonald E.S., and Windebank A.J. Mayo Molecular Neuroscience Program, Rochester, MN 55905.
Assembly of the extracellular matrix (ECM) in vitro requires serum and ascorbic acid. This has been studied extensively in the myelinating dorsal root ganglion (DRG) culture model (Wood Brain Res 115:361-375, 1976). Previously it had appeared that ascorbic acid was required for ECM assembly and that this must occur for myelination to proceed (Fernandez-Valle et al: J Neurobiol 25:1207-1226, 1994). We have shown that myelin formation occurs in the absence of both serum and ascorbic acid when cultured in B27, a fully defined serum free medium (Podratz; et al: Glia 23:383-388, 1998). Myelin in cultures without ascorbic acid appeared normal with both light and electron microscopy. Without ascorbic acid, neither basement membrane nor extracellular collagen assembly occurred. Cultures treated with ascorbic acid consistently had a greater abundance of myelinated fibers than those cultured in the absence of ascorbic acid. To further understand the role of ECM we cultured DRG neurons in B27 with (+) or without (-) ascorbic acid while agitating the cultures on a slow rotating rocker. DRG neurons cultured in B27(+) had excellent myelin assembly in both normal and agitated conditions. Neurons cultured in B27(-) had normal myelin formation in still cultures but significantly reduced myelination in the agitated cultures. These results indicate that the role for ECM may provide stabilization during the formation of myelin. This would be consistent with the important mechanical role of the ECM in the adult nerve.
THE NEUROTROPHIC ACTION OF FK506 IS MEDIATED BY THE IMMUNOPHILIN FKBP-52, A COMPONENT OF MATURE STEROID RECEPTOR COMPLEXES.
Gold B.G. (1), Bogardus B. Center Res. Occup. & Environ. Toxicology, (1) Dept. Cell & Developmental Biology, Oregon Health Sciences University, Portland, OR, USA.
The mechanism by which the immunosuppressant FK506 (tacrolimus) and related nonimmuno-suppressant neuroimmunophilin ligands increase nerve regeneration is unknown (see Gold Mol Neurobiol, 15:285, 1997; Snyder et al. Neuron, 21:283, 1998), but is believed to depend on the 12-kD FK506-binding-protein (FKBP-12) the immunophilin mediating immunosuppression. We now show that FK506 maintains its neurotrophic activity in primary hippocampal cell cultures from FKBP-12 knockout mice. Using human neuroblastoma SH-SY5Y cells, we show that the neurotrophic action of FK506 (10 pM - 10 nM) is completely prevented by addition of a monoclonal antibody (50-100 nM) to the immunophilin FKBP-52 (also known as FKBP-59 or hsp-56), a component of mature steroid receptor complexes. Surprisingly, the FKBP-52 antibody is also neurotrophic. Furthermore, the neurotrophic activity of dexamethasone (50 nM) is potentiated by FK506, while that of b-estradiol (50 nM) is not altered, suggesting a common mechanism of action. Geldanamycin (which disrupts mature steroid receptor complexes, releasing p23 from hsp-90) is also neurotrophic (0.1 - 10 nM), whereas it reduces the neurotrophic activity of FK506 and steroid hormones (dexamethasone and b-estradiol). Conversely, 20 mM molybdate (which prevents disruption of mature steroid receptor complexes) decreases the neurotrophic activity of FK506, FKBP-52 antibody, dexamethasone and b-estradiol. Finally, we show that the mitogen-associated protein (MAP) kinase inhibitor PD 098059 (0.1 - 10 mm) completely blocks the neurotrophic action of FK506. We propose that dissociation of a component(s) of steroid receptor complexes (perhaps p23) elicits a gain-of-function, augmenting known signal transduction pathways for neurotrophins (e.g., NGF). Components of steroid receptor complexes are novel targets for the design of neurotrophic drugs.
MOUSE MODELS OF MOTONEURON DISEASE: A COMPARISON OF WR, PMN, SOD-1(G93A), AND MND-2 WITH RESPECT TO ASSESSMENT OF THERAPEUTIC BENEFIT
Schmalbruch H.(1), Haase G.(3), Krarup C.(2), Kahn A.(3), Meisler M.H.(4), Jockusch H.(5), Castelnau-Ptakhine L.(3) Depts. (1)Med. Physiol. and (2)Clin. Neurophysiol., Copenhagen, Denmark; (3)Inserm U129 Paris France; (4)Dept. Hum. Gen. Ann Arbor, USA; (5)Dev. Biol. Unit, Bielefeld, Germany
Mice with motoneuron disease are used to test treatment regimes for ALS or SMA. Verification of a beneficial effect demands reliable and objective parameters. Lifespan is variable and often long in wr, brief and constant in pmn (41 days) and mnd-2 (33 days), and long and constant in SOD-1 (162 days). Degeneration of somata of spinal motoneurons is pronounced in wr and SOD-1, negligible in pmn (but abundant in the facial nucleus), and absent in mnd-2. In pmn and SOD-1, the number of myelinated fibres in the phrenic nerve eventually decreases from 250 to 100 which may be the cause of death; phrenic nerves are normal in wr and mnd-2. Peripheral nerves are affected, and muscle histology and EMG indicates denervation in wr, SOD-1 and pmn but not in mnd-2. Silver staining shows collateral reinnervation by sprouting in wr, almost no reinnervation in pmn and SOD-1, and no denervation in mnd-2. Sensory neurons are affected in wr and SOD-1 but not in pmn and mnd-2. The pyramidal tract is normal in all 4 mutants. Lifespan and phrenic nerve fibres are objective parameters for a therapeutic effect in pmn and SOD-1. Effects on wr are difficult to assess, because disease progression varies, and mnd-2 needs further investigation.