DORSAL ROOT GANGLIA CHANGES IN A RAT MODEL OF PACLITAXEL PERIPHERAL NEUROTOXICITY
Cavaletti G., Schenone A., Nicolini G., Miloso M., Nobbio L., Oggioni N., Cavalletti E., Minoia C., Zucchetti M., D'Incalci M., Trecdici G. Fifth Dept. of Neurology and Inst. of Human Anatomy, Univ. of Milan; Dept. of Neurology and Rehabilitation, Univ. of Genoa; Boehringer Mannheim Italy Research Center, Monza "S. Maugeri" Foundation, Pavia; "M. Negri" Inst. for Pharmacological Research, Milan, Italy.
The features of paclitaxel (Taxol®)-induced peripheral neurotoxicity are not completely known. Some of its clinical aspects, however, suggest that not only peripheral nerves, but also dorsal root ganglia (DRG) neuron may be a target of paclitaxel action. In this study we examined in a well-established model of paclitaxel-induced peripheral neurotoxicity obtained by repeated iv administration of the drug 1) its tissue distribution in the brain, spinal cord, DRG and peripheral nerves and 2) the changes induced by the treatment in NGF, BDNF, NT-3 and cognate receptors mRNA expression. Although the morphological appearance of DRG neurons was completely normal, using HPLC/MS/MS we demonstrated for the first time that after treatment the amount of paclitaxel detectable in the DRG was 7-fold higher than in the peripheral nerves and 30-fold higher than in the spinal cord, while concentration of paclitaxel was extremely low in the brain. Also the semiquantitative evaluation of neurotrophin and related receptors mRNA expression obtained by RTPCR demonstrated that paclitaxel induced modifications in the DRG. In fact, NGF, BDNF, TrkA and TrkB mRNA relative values were significantly lower in treated rats than in controls, while NT-3 and TrkC were unchanged. Our data support the hypothesis that during paclitaxel treatment pathological changes can be induced also in the DRG and point to another mechanism (besides tubulin polymeration enhancement with axonal transport interference) to explain some of the features of its peripheral neurotoxicity.
INSULIN-LIKE GROWTH FACTOR-1 PREVENTS CASPASE MEDIATED PROGRAMMED CELL DEATH IN SCHWANN CELLS
Delaney C.L., Cheng H-L Daum J., Feldman E.L. University of Michigan, Department of Neurology, Ann Arbor, MI
Our laboratory has demonstrated that IGF-I protects SHSY5Y
neuroblastoma cells from glucose mediated programmed cell death.
IGF-I protects via a phosphatidylinositol 3-kinase (PI-3K), retards
proteolytic cleavage of caspases, and inhibits apoptotic changes
in bcl-2 and bcl-xL expression. Since neurons and glia interact
during both development and disease, we examined the ability of
IGF-I to protect Schwann cells (SC) from apoptosis following serum
withdrawal. SC are harvested from the sciatic nerve of postnatal
day 3 rats and grown in serum-containing media until confluency.
At the start of each experiment, cells are rinsed and then grown
in serum-free insulin-free defined media (DM). After 12-72 h,
cells are collected, stained with propidium iodide and processed
for FACS analysis. After 12 h approximately 14% of SC have undergone
PCD, increasing to 57% at 72 h. Since IGF-I and insulin are important
components of serum, we added back either IGF-I or insulin to
the DM. After 24 h in DM, +/-IGF-I or +/- insulin, IGF-I rescues
more than half of the moribund cells. To examine kinase activity
involved in IGF-I rescue, we added either the PI-3K inhibitor
LY294002 or the MEK inhibitor PD98059 prior to IGF-I. LY294002,
but not PD98059, blocks IGF-I protection. To examine caspase-3
activation in SC PCD we, included the caspase inhibitor bok-asp-fmk
(BAF) in the DM with or without LY294002. In both cases, SC did
not succumb to PCD. These data suggest that IGF-I prevents PCD
in SC predominantly via PI-3K which is upstream from caspase activation.
Western blot analysis identified expression of several apoptotic
proteins including bcl's, Apaf-1 and other caspases, Future studies
will further define the role of these proteins in SC apoptosis.
Supported by NIH T32 NS07222 (C.L.D.), ROI NS38849 and grants
from the ADA and the JDFI (E.L.F.)
ERYTHROPOIETIN-DERIVED PEPTIDES ATTENUATE NERVE DISORDERS IN DIABETIC RATS
Freshwater J.D. (1), Campana W.M.(2), Mizisin A.P.(1), O'Brien J.S.(2), Calcutt N.A.(1) Departments of Pathology(1) and Neuroscience(2), University of California San Diego, La Jolla, CA 92093, USA.
We have recently described a peptide fragment of the erythropoietin molecule that, when applied to neurons in vitro, initiated ERK phosphorylation, prevented cell death following serum starvation and induced neuritogenesis. Because loss of neurotrophic support may contribute to nerve dysfunction in diabetic rats, we treated control and streptozotocin-diabetic rats (8-10/group) with either a linear 16 amino acid derivative (Epopep 10: 1 mg/kg ip) or a retroinverso peptide (Dl: 0.2 mg/kg ip) of the neurotrophic region of erythropoietin thrice weekly for the duration of diabetes. After 8 weeks of diabetes, untreated rats showed reduced MNCV and SNCV, exhibited tactile allodynia and had a reduced distance of regeneration 7 days following sciatic nerve crush injury (all p<0.05 vs controls by ANOVA with Student Newman Keuls post-hoc test). Epopep 10 treatment of diabetic rats significantly attenuated MNCV and SNCV slowing and normalized regeneration distance (all p<0.05 vs untreated diabetics), but was without effect on tactile allodynia. D1 treatment of diabetic rats significantly attenuated SNCV slowing (p<0.05 v untreated diabetic rats) but was without effect on MNCV slowing, impaired regeneration distance or tactile allodynia. Neither compound altered NCV in control rats or altered nerve laser Doppler flux or polyol pathway activity in control or diabetic rats. However, D1 but not Epopep 10 induced tactile allodynia in control rats. Derivatives of the neurotrophic region of erythropoietin act on disorders of nerve conduction and regeneration in diabetic rats and may be of therapeutic value for diabetic neuropathy. Supported by Myelos Corporation and the UC BioSTAR program.
PROSAPOSIN IMMUNOSTAINING AND GENE EXPRESSION IN PERIPHERAL NERVE AND MUSCLE OF NORMAL AND DIABETIC RATS
Garrett D.N.(1), Garrett R.S.(1), Mizisin A.P.(1), Mohiuddin L.(1), O'Brien J.S.(2), Calcutt N.A.(1) Departments of Pathology(1) and Neuroscience(2) University of California San Diego, La Jolla, CA 92093, USA.
Prosaposin is the precursor of saposins A-D but is also secreted unprocessed into CSF and from transected peripheral nerve. Prosaposin has neurotrophic properties that are associated with the saposin C domain and suggest a potential role as an endogenous neuronal support molecule. In order to understand the potential role of prosa-posin in peripheral nerve we, undertook an immuno-cytochemical survey of the distribution of prosaposin in the peripheral nerve and muscle of normal rats followed by examination of prosa-posin gene expression in these tissues taken from normal rats and rats with streptozotocin-induced diabetes. Using an polyclonal antibody directed against the N terminal region of the saposin C domain of prosaposin, we found that immunostaining in the peripheral nerve was localized to endoneurial vessel walls consistent with a distribution in pericytes and smooth muscle. Immunostaining was also prominent in skeletal, smooth and cardiac muscle. Northern analysis was conducted on RNA extracted from sciatic nerve and both skeletal and cardiac muscle from normal and 8 week diabetic rats. Prosaposin mRNA was identified using a 650 bp fragment of the rat SGP-1 cDNA followed by densitometric quantification relative to the amount of 18S ribosomal RNA identified in the same sample. Diabetes induced a significant (p<0.01) increase in prosaposin mRNA levels present in both sciatic nerve and all muscle types examined compared to control samples. In contrast, mRNA levels of the myelin protein PO, were significantly (p<0.01) reduced in the sciatic nerve of diabetic rats compared to controls. Blood vessels may provide a source of secreted prosa-posin in peripheral nerve and prosaposin may be a muscle-derived neurotrophic factor. Supported by Myelos Corporation and the UC BioSTAR program.
MYELINOTROPHIC EFFECTS OF PROSAPOSIN IN DEVELOPING MICE
Hiraiwa M., Otero D.A.C., O'Brien J.S. Department of Neurosciences, University of California, San Diego, La Jolla, CA 92093-0634, USA.
Prosaposin, recently characterized as a myelino-trophic factor,
is a 66 KDa glycoprotein capable of increasing sulfatide content
in both Schwann cells and oligodendrocytes in culture. We also
demonstrated that a 14-mer prosaptide, a peptide derived from
the trophic sequence of prosaposin, mimicked the trophic activity
of prosaposin and enhanced not only sulfatide content but also
the levels of PO mRNA and the UDP-galactose: ceramide galactosyltransferase
in rat primary Schwann cells at nano molar levels. In the present
study, we tested whether prosaptide enhances the levels of myelin
markers in developing rat brain and sciatic nerve using prosaptide
D5. D5 is an all D-retro-inverso 11-mer prosaptide that is stable
to proteases. Pharmacokinetic studies indicated that it crosses
the blood-brain barrier and is delivered to brain and sciatic
nerve in sufficient quantities for the trophic activity after
given systemically. The prosaptide was injected subcutaneously
to neonatal rats at 100 ng/gram every other day. There were no
differences in weight gain between controls and the treated group
during the period of treatment. After 15 days of treatment, brains
and sciatic nerves were removed to determine expressions of myelin
markers. TLC immunoassay of sulfatide using a monoclonal anti-sulfatide
antibody demonstrated increased sulfatide contents by D5 injection
in both brain and sciatic nerve obtained 15 days after first injection:
200% increase in brain and 150% increase in sciatic nerve over
controls. Recently, our studies demonstrated that D5 improved
clinical, histochemical and histological signs in experimental
autoimmune encephalomyelitis. Taken together, prosaptide D5 is
a potent myelinotrophic peptide in vivo, extending a possibility
of the prosaptide for developing novel therapeutic agent for treatment
of nerve demyelination. Supported in part by a grant from Myelos
Neurosciences to JSO
GDNF IS ABLE TO SUPPORT A NOVEL MOTOR MYELINATING NERVE CULTURE
Ho. T., Li Y., Rothstein J.(1), Griffin J.(1) Department of Neurology, and (1)Neuroscience, Johns Hopkins University, Baltimore, MD, U.S.A.
Myelinating nerve culture is typically made using sensory neuron and co-culture with Schwann cells. However, many neuropathies such as AMAN form of Guillain-Barré syndrome (GBS) and multifocal motor neuropathy (MMN) selectively affect motor axons. We have developed a novel myelinating motor nerve culture system that will allow us to study motor axon outgrowth, myelination and neuromuscular junction. Organotypic spinal cord cultures were prepared from eight day old rat pup lumbar spinal cords and sections into 350-mm slices and cultured on collagen coated inserts. Addition of GDNF, but not other TGF b trophic factors was able to induce motor axon outgrowth from the spinal cords. This induction of motor axons is mediated through MAP kinase pathway and inhibitor of the MAP kinase kinase (MEK), PD98059, inhibits the motor axon outgrowth. Addition of Schwann cells to the organotypic culture along with ascorbic acid will allow >95% of the induced motor axons to myelinate. Finally, when co-culture with muscle cells, active neuromuscular junctions are formed. This model will allow us to study the pathological changes in disorder such as GBS and MNN in vitro.
THE INFLUENCE OF SKIN INNERVATION ON KERATINOCYTES
Hsieh S.T. National Taiwan University College of Medicine, Taipei, 10018, TAIWAN
We evaluated the influence of skin innervation on the epidermis in mice. The rich innervation of skin was demonstrated by immunocytochemistry with protein gene product 9.5 (PGP), a ubiquitin carboxy hydrolase. PGP(+)-nerve fibers were in the epidermis, subepidermal plexus, dermal nerve trunks, and nerve terminals around sweat glands. Effects of denervation on the plantar surface of the hindfoot were assessed by comparing the thickness of the epidermis, which was innervated by the sciatic nerve. Within 48 h after sectioning of the sciatic nerve, PGP(+)-nerves in the territory of the sciatic nerve had completely degenerated. There was a significant thinning of the denervated epidermis 72 h post-transection (30.5 ± 1.1 vs. 41.4 ± 2.9 mm, p< 0.001). The reduction in epidermal thickness persisted when skin remained denervated (69%-75% of the control side). Incorporation of brodeoxyuridine (BrdU) was reduced after denervation. Reduction in incorporation of BrdU was most pronounced within 48 h after denervation (19.4% ± 6% of the control side, p < 0.0001). The decrease in BrdU-labeling followed a similar temporal course as the thinning of the epidermis. Both epidermal thinning and reduced BrdU-labeling were reversed by epidermal reinnervation. Patterns of keratinocyte differentiation and programmed cell death assessed by TUNEL were unaffected by skin denervation. These findings are consistent with the notion that skin innervation exerts an influence on the proliferation of keratinocytes and the thickness of the epidermis.
A ROLE FOR HEDGEHOG PROTEINS IN PERIPHERAL NERVE DEVELOPMENT AND FUNCTION
Hunter-Schaedle K.E.(1), Gan X.(1), Karavanov I.(1), Williams K.P.(2), Rich M.(3), Scherer S.S.(3), Mahanthappa N.K.(1), Rubin, L.(1) (1)Ontogeny, Inc. and (2)Biogen, Inc., Cambridge, MA, USA; (3)Dept. Neurology, U. Penn., Philadelphia, PA, USA.
Three mammalian homologues of the D. melanogaster gene
hedgehog have been identified: Sonic hedgehog, Indian
hedgehog and Desert hedgehog. All three hedgehog proteins
have been implicated in induction and patterning in early embryogenensis.
Desert hedgehog (Dhh) is of special interest to the peripheral
nervous system: it is expressed in early Schwann cells(a), and
deletion of the Dhh gene(b) results specifically in a dramatic
structural perturbation of the perineurium(c), the mesenchymal
sheath of peripheral nerve. Here we i) examine the Dhh -/- sciatic
nerve phenotype in detail at light and EM level; ii) demonstrate
that there is a motor nerve conduction deficit in Dhh -/- sciatic
nerve; iii) examine the expression and distribution of Hh proteins,
their receptors and downstream signaling elements in peripheral
nerve cells; and iv) illustrate that perineurial cells are the
direct target of Dhh in peripheral nerve. Together, these findings
suggest that Dhh has an important role in regulating the normal
development and maturation, and subsequently function, of peripheral
nerve. In the light of this, Hh proteins might have potential
as therapeutics for treatment of peripheral nerve disorders. (Funded
by Ontogeny, Inc. and Biogen, Inc.).
(a)Bitgood and McMahon, Dev Biol 172:126-138 (1995)
(b)Bitgood et al, Curr.Biol., 6:298:304 (1996)
(c)Parmantier et al, Soc. Neurosci. Abs. 27.3 (1998)
LAR PROTEIN TYROSINE PROSPHATASE: EXPRESSION AND FUNCTION DURING SCIATIC NERVE REGENERATION
Xie Y.M., Yeo T.T., Zhang C., Yang T., Longo F.M. Department of Neurology, UCSF and VAMC, San Francisco, CA, USA
Protein tyrosine kinases (PTKs) mediate the effects of growth factors and extracellular matrix factors during nerve regeneration. Protein tyrosine phosphatases (PTPs) counterbalance, or in some cases augment, the actions of PTKs. Identification of the PTPs functioning during nerve regeneration would provide a novel class of signal transduction targets to promote regeneration. The presence of the leukocyte common antigen-related (LAR) PTP receptor in dorsal root ganglion (DRG) neurons and in growth cones of cultured neurons led to the hypothesis that LAR regulates sensory fiber regeneration. Sciatic nerve crush caused an increase in LAR mRNA and protein expression in DRG, lumbar spinal cord and nerve segments distal to the crush site. Nerve crush also induced marked changes in alternative splicing of functional LAR domains likely to regulate intracellular location and interactions with extracellular matrix molecules including laminin. In LAR-deficient transgenic mice, the distance of post-crush sensory fiber regeneration was markedly reduced in a gene dosage dependent manner. DRG neuron counts were not significantly changed post-crush. The finding that LAR is required for regeneration introduces PTPs as an important new gene family required for nerve regeneration. Supported by a Beeson Award from the American Federation for Aging Research and the VA.
EXPRESSION OF GDNF RECEPTOR (RET AND GFRa-1) mRNAs AND ITS PHOSPHORYLATED PROTEIN IN THE SPINAL CORD OF PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS
Mitsuma N., Yamamoto M., Li M., Ito Y., Sobue G. Department of Neurology, Nagoya University School of Medicine, Nagoya, Japan
The mRNA levels of RET and GFRa-1 were studied in the spinal cord of patients with amyotrophic lateral sclerosis (ALS) by reverse transcription followed by polymerase chain reaction (RT-PCR) and in situ hybridization (ISH). Semi quantitative RT-PCR analysis revealed that RET mRNA levels in the ALS spinal cord anterior horn were reduced to one fifth of controls in proportion to motor neuron loss, whereas GFRa-1 mRNA was unchanged. ISH analysis showed that RET mRNA was expressed in the anterior horn motor neurons of the spinal cord, but GFRa-1 mRNA was expressed widely in the spinal cord neurons and glial cells. The RET mRNA levels, measured using a CCD image analyzer, were substantially preserved in individual motor neurons of ALS, but varied among those neurons. Relatively high levels of RET mRNA were observed in a certain population of atrophic neurons. RET mRNA expression was not altered in neurons with an accumulation of phosphorylated high molecular weight neurofilament, Bunina body-bearing or central chromatolytic neurons in ALS. The GFRa-1 mRNA levels in the motor neurons were similar in ALS and controls. In addition, the RET protein and its phosphorylated form were expressed in individual motor neurons in ALS. These results indicate that GDNF receptor expression persists at mRNA, protein and its phosphorylation levels in the degenerating motor neurons in ALS, supporting the view that GDNF is a candidate for therapeutic approach to ALS.
EPIDERMAL DENERVATION IN FABRY DISEASE CORRELATES WITH INCREASED EPIDERMAL LEVELS OF NGF AND GDNF: REGULATION OF NEUROTROPHIN EXPRESSION BY SENSORY INNERVATION
Scott L.J.C.(1), and Schiffmann R.(2) (1)University of Vermont, Burlington, VT, USA, (2)The National Institutes of Health, Bethesda, MD, USA
How denervated target tissues influence nerve regeneration in peripheral neuropathies is difficult to evaluate, since systemic diseases (e.g., diabetes) which cause neuropathy may also alter target tissue biology. In Fabry disease, however, a selective loss of intraepidermal innervation occurs early in the disease in the absence of systemic disease. To study the proximate effects of denervation on epidermal biology, we analyzed skin punch biopsies and epidermal sheets removed by suction blister from 18 Fabry patients with normal kidney function and 6 controls. Innervation density was quantitated by skin punch biopsy. In Fabry patients with epidermal denervation, immunohistochemistry of skin biopsies showed increased epidermal staining for NGF and GDNF compared with controls. Extracts of isolated epidermal sheets were analyzed by ELISA for NGF and GDNF. NGF levels were 1.6- to 4.8-fold higher on a per mg protein basis in Fabry patients as compared with controls. Low levels of GDNF were detectable in control epidermis, but the levels were 3.2- to 3.5-fold higher in the Fabry patients. The results show that in Fabry patients with epidermal denervation the density of innervation of the epidermis regulates the levels of epidermal neurotrophic factors. It also documents the presence of low levels of GDNF in normal, innervated adult epidermis, with increased GDNF levels in denervated epidermis. This parallels what is seen with NGF, which is known to have an important role in epidermal innervation, suggesting a novel role for GDNF in the epidermis.
TYPE I INSULIN-LIKE GROWTH FACTOR RECEPTOR ACTIVATION AND BCL-2 OVEREXPRESSION PREVENT MANNITOL MEDIATED APOPTOSIS IN HUMAN NEUROBLASTOMA CELLS
van Golen C.M., Castle V.P.(1), Feldman E.L. University of Michigan, Neuroscience Program, Departments of Neurology and (1)Pediatrics, Ann Arbor MI
Diabetic neuropathy has been correlated to the severity and
duration of hyperglycemia. Our laboratory is investigating the
effects of hyperglycemic, hyperosmotic exposure on neuronal cells.
In the current study we examine the effects of high mannitol exposure
and IGF-I on apoptosis and mitochondrial membrane depolarization
(M-MD) in SHEP neuroblastoma cells. We also investigate the effect
of type I IGF receptor (IGF-IR) and Bcl-2 overexpression on MMD
± IGF-I. In these studies SHEP human neuroblastoma cells
are used as they produce little IGF-IR and no Bcl-2. Cells are
serum deprived overnight, then exposed to 300 mM mannitol ±
IGF-I, and then analyzed by flow cytometry for MMD using Rho-damine
123 fluorescence or apoptosis using propidium iodide. Normal SHEP
cells show apop-tosis in response to serum deprivation, and only
high doses of mannitol increase the number of apoptotic cells
over that induced by serum withdrawal. The number of cells with
MMD increases with increasing concentrations of mannitol. IGF-I
addition prevents both MMD and apoptosis in these cells. When
SHEP cells are transfected with the IGF-IR, the serum withdrawal
effect is abolished, and IGF-I protects a greater number of cells
from MMD and apoptosis. Less than 5% of SHEP cells undergo mannitol
induced apoptosis when transfected with Bcl-2. SHEP/Bcl-2 cells
also show very little MMD. Bcl-2 transfection prevents apoptosis
induced by serum withdrawal as well, even after four days of serum
deprivation. Collectively these results demonstrate that IGF-I
can protect SHEP neuroblastoma cells from both serum withdrawal
and mannitol induced apoptosis and MMD. Transfection with the
IGF-IR enhances the effect of IGF-I and abolishes the serum withdrawal
effect. Bcl-2 transfection completely protects SHEP cells from
both apoptosis and MMD, even at the highest dose of mannitol tested.
In summary, both IGF-I:IGF-IR and Bcl-2 can regulate events at
the mitochondria in SHEP cells which protects these cells from
apoptosis. Supported by NIH RO1 S32843, RO1 NS38849 and grants
from the ADA and JDFI.