Regulation of aberrant neurofilament phosphorylation in neuronal perikarya. IV. Evidence for the involvement of two signals

Brain Research, 626 (1993) 23-30

23

© 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00

BRES 19319

Regulation of aberrant neurofilament phosphorylation in neuronal perikarya. IV. Evidence for the involvement of two signals Bruce

G. Gold

a,b T o n i S t o r m - D i c k e r s o n

a and Daniel

R. Austin

a

a Center for Research on Occupational and Environmental Toxicology and b Department of Cell Biology and Anatomy, Oregon Health Sciences Unieersity, Portland, OR 97201-3098 (USA)

(Accepted 26 May 1993)

Key words: Axon reaction; Axotomy; Cell body; Dorsal root ganglion; Phosphorylated neurofilament; Ola mice; Trophic; Wallerian degeneration

Axonal regeneration over long distances is dependent upon events occurring both in the distal stump and in the neuronal cell body. Little is known concerning how events in the distal stump influence the cell body response to injury, or the axon reaction. In the present study, we examined this relationship for one component of the axon reaction (i.e. aberrant neurofilament (NF) phosphorylation) in the C57BL/OIa (Ola) mouse mutant, a model which exhibits delayed Wallerian degeneration (up to 3 weeks) and retarded regeneration of sensory neurons. Non-axotomized normal (C57/6J/BL) and Ola mice demonstrated modest immunostaining to phosphorylated NF (pNF) epitopes (using monoclonal antibody 06-17) in some (11%) L4 dorsal root ganglion (DRG) neuronal cell bodies. In normal mice, modest to intense immunoreactivity was present in 43% of DRG neurons at 1 week following a sciatic nerve crush (axotomy). The intensity and extent of staining declined with reinnervation, being reduced slightly at 2 weeks and more notably by 3 weeks following axotomy. In Ola mice, the intensity and extent (43%) of staining were not different from normal axotomized mice at 1 week following axotomy. However, the intensity was less and the extent of staining reduced by 28% at 2 weeks following axotomy. By 3 weeks, staining levels were again increased, being similar to that observed in Ola and normal mice at 1 week following axotomy. Taken together, the results suggest that aberrant expression of pNF epitopes in DRG neuronal cell bodies is regulated by at least two signals. The first signal is not dependent upon processes associated with Wallerian degeneration since pNF expression is fully developed in Ola mice when the vast majority of fibers remain intact. This suggests that induction arises from the loss of a target tissue-derived retrogradely transported trophic signal, The presence of a second signal is suggested by the failure of Ola mice to maintain (at 2 weeks) the level of pNF expression. This indicates that events in the distal stump may influence both the magnitude and duration of the axon reaction.

INTRODUCTION Nerve t r a n s e c t i o n (axotomy) elicits both W a l l e r i a n d e g e n e r a t i o n in the distal s t u m p a n d a stereotypic r e s p o n s e in the n e u r o n a l cell body (perikaryon) t e r m e d the axon r e a c t i o n (see refs. 21, 27). It is generally a s s u m e d that these events are causally related. F o r example, it has b e e n p r o p o s e d 16,z9'3° that W a l l e r i a n d e g e n e r a t i o n p r o d u c e s a t r o p h i c signal which, following r e t r o g r a d e axonal t r a n s p o r t to the cell body, induces the axon reaction. However, studies of axotomized nerves are complicated by the inability to differentiate b e t w e e n events associated with W a l l e r i a n d e g e n e r a t i o n per se a n d r e d u c e d trophic s u p p o r t from target tissue n o r m a l l y conveyed to the cell body by r e t r o g r a d e t r a n s p o r t (see ref. 13).

This latter possibility (i.e. i n d u c t i o n of the axon reaction by loss of a target tissue-derived retrogradely t r a n s p o r t e d trophic signal) is s u p p o r t e d by o u r previous studies 8'11 d e m o n s t r a t i n g that some c o m p o n e n t s of the axon reaction can be i n d u c e d in intact n e u r o n s (i.e. without axonal loss). However, this m o d e l does not p r e c l u d e the i n v o l v e m e n t of d e g e n e r a t i v e events in the r e g u l a t i o n of the axon reaction. I n this regard, r e c e n t studies from our laboratory 10 also s u p p o r t the role for a second signal which may serve to m a i n t a i n at least o n e c o m p o n e n t of the axon r e a c t i o n once initiated. T h e hypothesis that a second signal, p r e s e n t in i n j u r e d nerves, may i n f l u e n c e the axon reaction was b a s e d u p o n e x a m i n a t i o n of the r e g u l a t i o n of a b e r r a n t expression of p h o s p h o r y l a t e d n e u r o f i l a m e n t ( p N F ) epitopes in dorsal root g a n g l i o n ( D R G ) n e u r o n a l cell

Correspondence: B.G. Gold, Center for Research on Occupational and Environmental Toxicology, L606, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, Portland, OR 97201-3098, USA. Fax: (1) (503) 494-6831.

.....~ ,' ~ ~~i !ii~i

i

¸~¸!~

|~.~

25 bodies; pNF epitopes are normally poorly expressed in neuronal cell bodies 2°'31. Aberrant expression of pNF epitopes is observed in a variety of human diseases (e.g. motor neuron disease 22'23) and in experimental animal models produced by neurotoxic chemicals (i.e. acrylamide tz,tS, aluminum7,22'34, fl,fl'-iminodipropionitrile [IDPN] 9 and organophosphate 32) most likely as a secondary response to injury 3'14'28. Thus, aberrant pNF expression in neuronal cell bodies appears to be a useful marker of the axon reaction 8'9. In the present study, we continue to explore the regulation of aberrant pNF expression as a means toward unraveling the complex mechanisms underlying production of the full spectrum of events associated with the axon reaction (see refs. 6, 13). The mouse mutant C57BL/OIa (Ola), which exhibits a marked slowing in the rate of Wallerian degeneration following axotomy 25'26, provides a novel approach to separate the roles of lost trophic support from Wallerian degeneration in the regulation of the aberrant expression of pNF epitopes. We have chosen to examine sensory neurons since this alteration is not induced in motor neurons by axotomy 8. Furthermore, in contrast to motor neurons, sensory axons demonstrate impaired regeneration in Ola mice 4. Our findings support the involvement of multiple signals in the regulation of the axon reaction 6'13 and indicate that separate signals underlie induction and maintenance of aberrant expression of pNF epitopes in rat DRG neurons. MATERIALS AND METHODS

Animals and surgicalprocedure Eleven 10-week-old C57BL/Ola mice (Harlan Olac Limited, Bicester, U.K.) and eleven 10-week-old C57/6J/BL (Harlan U.S.A.) were used in this study. Nine animals from both groups were anesthetized with 4% halothane, the sciatic nerves exposed, and crushed (at the level of the hip) twice (for a total of 30 s) using a no. 7 Dumont jeweler's forceps 17. A suture was tied in the adjacent muscle to mark the crush site. The two additional animals in each group served as unoperated, non-axotomized controls.

Tissue fixation and preparation At 1, 2 or 3 weeks following axotomy, the animals (3 per group) were anesthetized with 4% halothane, heparinized, and perfused through the ascending aorta with 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4); unoperated controls were perfused at the 2-week time point. The sciatic nerves were sampled at approximately 10 mm distal to the crush site, the tissues placed in

sodium phosphate-buffered 5% gluaraldehyde (pH 7.4) overnight, postfixed in 0.1 M phosphate-buffered 2% osmium tetroxide (pH 7.4), deh3,drated in ascending concentrations of ethanol, embedded in plastic (Spurr's), and sections (1 izm) stained with Toluidine blue. The dorsal root ganglia (DRG) at the 4th and 5th lumbar levels (L4 and L5) were dissected following overnight fixation in situ (4°C), dehydrated in graded series of ethanol, and embedded in paraffin for immunocytochemistry (see below).

lmmunocytochemistry The immunocytochemical procedures have been described previously 12. Briefly, 10 ~m sections were mounted on chrom-alum-subbed slides, deparaffinized, incubated for 1 h in 3% normal goat serum, and incubated overnight at 4°C with a 1 : 1000 dilution (in 1% normal goat serum) of primary antibody: 2-135 (which recognizes a non-phosphorylated epitope of the 200 kilodalton (kDa) NF polypeptide), or 06-17 (which recognizes a phosphorylated epitope shared by the 200 and 145 kDa polypeptides)14'24'31. Sections were incubated for 1 h in goat-antimouse secondary antibody (1:20), washed, incubated for 1 h in mouse peroxidase-antiperoxidase (1:200), and the immunoreactivity visualized with 0.05% diaminobenzidine tetrahydrochloride/0.01% hydrogen peroxidase (8 rain). On each slide, primary antibody was omitted from one tissue section to assertain background staining levels.

Assessment of immunostaining Since identical staining patterns where found in L4 and L5 DRG neuronal cell bodies (see Results), quantitation was performed in only the larger ganglion (i.e, L4 DRG). All cells containing immunoreactivity above background were counted. The proportion of cells demonstrating pNF epitopes (% phosphorylated) was determined by counting the total number of neuronal cell bodies stained with antibody 06-17 and dividing this value by the total number of neuronal cell bodies in the DRG (determined by counting the total number of neuronal cell bodies stained with antibody 2-135 and lightly counterstained with Cresyl violet)s'2~. To assess changes in the degree of immunoreactivity, neurons were divided into two groups based upon staining intensity: modest or intense (dark, virtually black). The percentage of modestly and intensely stained cells was determined out of the total number of neuronal cell bodies in each DRG.

Statistical analysis Two-way analyses of variance (ANOVA) were performed to test for the effects of animal type (i.e. normal and Ola mice) and time (1, 2 and 3 weeks) following axotomy. Following these analyses, Scheffe's post-hoc analysis was performed to test for differences between individual groups. All values are mean +_SEM. RESULTS

Light microscopy Sciatic nerves from Ola mice demonstrated very slow Wallerian degeneration following a crush lesion, as expected 25'26. For the present purposes, we present the pattern of axonal degeneration (Fig. 1A-H) observed in the present series of animals only as a reference for the immunocytochemical study (see below).

4---

Fig. 1. Light micrographs of fibers in the mid-sciatic nerve from normal (A-D) and Ola (E-H) mice in non-axotomized nerves (A & E), and approximately 10 mm distal to a nerve crush at 1 (B & F), 2 (C & G), and 3 (D & H) weeks following axotomy, respectively, In normal mice, all fibers are undergoing Wallerian degeneration at 1 week (B); note mitotic figure in Schwann cell (arrowhead). Numerous regenerating axons, many with thin myelin sheaths, are present at 2 (C) and 3 (D) weeks; macrophages are present throughout the nerves at 2 weeks (C) and are less numerous by 3 weeks (D). In Ola mice, few degenerating fibers are present at 1 (F) and 2 (G) weeks, and even by 3 weeks (H) many fibers still appear to be structurally intact. Magnification × 305.

26 Normal mice. In normal mice, all fibers in the sciatic nerve were undergoing Wallerian degeneration by 1 week following axotomy (Fig. 1B). Numerous regenerating axons, with relatively thin myelin sheaths, were

A

present at 2 weeks following axotomy (Fig. 1C). By 3 weeks, myelinated fibers were larger in size and, compared to 2 weeks, there were fewer myelin-filled macrophages (Fig. 1D).

B

Fig. 2. Peroxidase-antiperoxidase staining of L4 and L5 D R G to pNF epitopes using antibody 06-17. A: non-axotomized Ola mouse L5 DRG. Modest staining is present in some cell bodies. B: normal mouse L4 D R G , 1 week following axotomy. Modest to intense immunoreactivity is present in many (almost half) cell bodies. C: normal mouse L4 D R G , 3 weeks followin~ axotomy. T h e extent and intensity of staining are reduced relative to that observed at to 1 week (compare with B). D: Ola mouse L4 DRG, 1 week following axotomy. The extent and intensity of staining are similar to that observed in normal mice at 1 week (compare with B). E: Ola mouse L4 D R G , 2 weeks following axotomy. The extent and intensity of staining are reduced compared to normal (compare with B) and Ola (compare with D) mice at 1 week. F: Ola mouse L4 D R G , 3 weeks following axotomy. The extent and intensity of staining are again similar to that observed in normal (compare with B) and Ola (compare with D) mice at 1 week. (For quantitation of immunostained cell bodies, see Fig. 3). × 140.

27

Ola mice. In contrast, few (5-10%) degenerating profiles were present in the sciatic nerves from Ola mice at 1 week following axotomy (Fig. 1F). At 2 weeks, the vast majority of fibers (> 75%) in nerves from Ola mice still appeared normal (Fig. 1G). By 3 weeks, degeneration was widespread, although numerous normal appearing fibers were still present (Fig. 1H).

Immunocytochemistry The pattern of immunoreactivity observed in nonaxotomized animals and the alterations induced by axotomy in both normal and Ola mice were identical in neuronal cell bodies from L4 and L5 DRG. Non-axotomized animals. Unoperated, non-axotomized normal (not shown) and Ola (Fig. 2A) mice demonstrated modest immunoreactivity to antibody 06-17 directed against pNF epitopes in some IA and L5 DRG neuronal cell bodies; no cells exhibited intense immunostaining. This finding is in marked contrast to our experience with DRG tissue from normal, unoperated rats where significant (i.e. above background) staining is not found using this antibodyS-m; this difference is likely due to non-specific binding to mouse DRG tissue by these mouse origin antibodies 14'24'31. Nevertheless, this level of staining was readily discernible from the marked increased expression of pNF epitopes produced by axotomy (see below). Axotomized animals. In normal mice, modest to intense immunoreactivity to antibody 06-17 was observed in almost half of L4 and L5 DRG neuronal cell bodies at 1 week (Fig. 2B). The extent and intensity of immunoreactivity appeared to be reduced at 2 weeks following axotomy (not shown). The reduction in intensity of staining was more apparent by 3 weeks following axotomy (Fig. 2C). In Ola mice, immunostaining to antibody 06-17 at 1 week following axotomy appeared identical to that observed in normal mice (Fig. 2D). In contrast, both the intensity and extent of staining appeared to be markedly reduced in L4 and L5 DRG from Ola mice at 2 weeks following axotomy (Fig. 2E). By 3 weeks (Fig. 2F), staining intensity was similar to that present in both normal (compare with Fig. 2A) and Ola (compare with Fig. 2B) mice at 1 week following axotomy.

% Intense

50 0 I¢I

$

40

z

30

% Modest

Im NRL

[]

•

•

o~

NRL OLA .

m

,.x.\,

, x x .

o .J

>r~ 0 -r

20

13.

10

co 0

i iI

, x x ,

....

ii

O.

•

0

, .....

NRL

OLA

NRL

1

I

•

OLA

NRL

OLA 3

2 WEEKS

Fig. 3. Bar graphs showing numbers of immunostained neuronal cell bodies in L4 D R G to p N F epitopes (% phosphorylated) from normal (NRL) and Ola (OLA) mice at 1, 2, and 3 weeks following axotomy. The total n u m b e r s of stained cell bodies (i.e. cells demonstrating modest to intense immunoreactivity) are not different in normal and Ola mice at 1 week. In normal mice, there is a reduction in the total numbers of stained cells at 2 and 3 weeks with fewer intensely stained cells present by 3 weeks. In Ola mice, the total number of stained cells is markedly decreased at 2 weeks. By 3 weeks, the total numbers of stained cells is again increased (to 1 week levels) due to a greater number of intensely stained cells. Standard error bars and P values refer to total numbers of stained cells only. * p < 0.05; compared to normal mice at 1 week. ** p < 0.001; compared to Ola mice at one and 3 weeks. (For full statistical analyses, see Table I).

At 1 week following axotomy, the extent of aberrant pNF expression was similar in normal and Ola mice (Fig. 3; Table I). The percentage of L4 DRG neurons demonstrating staining to pNF epitopes was significantly reduced in Ola mice at 2 weeks compared to normal mice at this time, and Ola mice at either one or 3 weeks following axotomy (Fig. 3; Table I); the extent TABLE I

Incidence of DRG neuronal cell bodies demonstrating phosphorylated NF epitopes in OLA and normal mice following axotomy Percentage (mean -+ S.E.M.) of neuronal cell bodies modestly stained, intensely stained, and the total (modestly and darkly) stained with antibody 06-17 directed against phosphorylated NF epitopes (see Materials and Methods). n = number of animals.

Time (weeks)

% Phosphorylated Modestly stained

Intensely stained

Total stained

32_+4.6 30-+1.6 34-+1.6

11_+4.9 8-+1.2 2_+0.9

43+0.7 38+0.6 * 35_+0.6 *

31_+5,7 24_+1,3 24+0,6

12+5.1 7_+0.7 20-+1.3"*

43-+0.9 31_+0.6 * * * , t 45+1.9

Normal 1 (n=3) 2 (n=3) 3 (n=3)

Quantitative analysis

Ola

Quantitation was performed to ascertain the percentage of neuronal cell bodies showing modest or intense immunoreactivity to pNF epitopes (see Materials and Methods). In unoperated animals, staining was present in 9-13% (mean 11%) of L4 DRG neurons from both normal and Ola mice.

* **

1 (n=3) 2 (n=3) 3 (n=3) P P 3 *** P t P

< 0.05 (compared to normal mice at 1 week). < 0.05 (compared to Ola mice at 2 weeks and normal mice at weeks). < 0.001 (compared to Ola mice at 1 and 3 weeks). < 0.01 (compared to normal mice at 1 and 2 weeks).

28 of staining in Ola mice at 2 weeks was reduced by 28(} compared to Ola mice at I week following axotomy. Between 2 and 3 weeks, the alterations in staining levels demonstrated very different patterns in normal and Ola mice (Fig. 3). In normal mice, the extent of immunostaining at 2 and 3 weeks following axotomy were significantly decreased (Fig. 3; Table I); the percentage of stained cells at 3 weeks was reduced by 19% compared to corresponding values obtained at 1 week. The visual impression that the reduction in staining levels was greater by 3 weeks (Fig. 2C) was reflected by a decrease, albeit not significant, in the numbers of intensely stained cells at this time (Table 1). In contrast, the percentage of stained neuronal cell bodies at 3 weeks in Ola mice was not significantly different from that found in these animals at 1 week (Fig. 3; Table I); the increase in total numbers of stained cells between 2 and 3 weeks being due to a significant increase in the numbers of intensely stained cells (Table l). DISCUSSION The Ola mouse phenotype manifests as a dramatic retardation in the rate of Wallerian degeneration 25'2" and, for sensory neurons, axonal regeneration 4. Surprisingly, the Ola mice used in our study exhibited an even greater degree of impairment in axonal degeneration than expected based upon the initial reports by Perry and co-workers 25'26. This very slow rate of Wallerian degeneration may be due to the influence of unknown environmental factors 26. However, our animals were not housed individually, one factor which has been shown to retard the rate of degeneration in Ola mice 2". Alternatively, this exacerbation in the survival rate of severed axons may be a consequence of differences in the severity of the nerve injury employed. In our study the sciatic nerves were crushed, which leaves the Schwann cell basal lamina intact, in contrast to previous studies where the nerve was cut 252~'. One possibility is that the apparent difference in the rates of degeneration between these two situations is due to the rate of recruitment of blood-borne macrophages 2 in crushed versus cut nerves. The Ola mutant mouse model provides a unique opportunity to examine the influence of the axon and its environment on the neuronal cell body. In the present study, we have examined the consequence of delayed axonal d e g e n e r a t i o n / r e g e n e r a t i o n on one component of the axon reaction; i.e., the aberrant expression of p N F epitopes in sensory neuronal cell bodies. Our findings demonstrate that in Ola mice the pattern of aberrant expression of pNF epitopes in

D R G neurons exhibits an atypical time course follow~ ing axotomy: aberrant expression is fully developed in Ola mice at 1 week, as in normal mice, but is markedly reduced at 2 weeks, being present again at 3 weeks following axotomy; the reduction in staining intensity observed in normal mice beginning at 2 weeks following axotomy is most likely a consequence of reinnervation of target tissue by these regenerating nerves (see Fig. 1). The simplest interpretation of the data is that Wallerian degeneration is not necessary to induce aberrant pNF expression in D R G neurons but that its continued maintenance is dependent upon degenerative a n d / o r regenerative processes occurring in the distal stump. This proposal is supported by our previous findings in two separate studies~l( First, we have shown that blockade of retrograde axonal transport by colchicine application to the sciatic nerve is sufficient to induce aberrant p N F staining in intact nerves s. This indicates that aberrant p N F expression in D R G neurons is not induced by a signal generated in the distal stump during Wallerian degeneration (see Introduction), but by a loss of a retrogradely transported trophic signal. Additional components of the axon reaction appear to be similarly regulated in D R G neurons ~~. For example, we ~1 and others 35 have shown that in some D R G neurons nerve growth factor (NGF) appears to be the target tissue-derived trophic factor whose loss initiates at least some components of the axon reaction (e.g. reduction in neurofilament synthesis and content in the axon). Furthermore, we have recently found that N G F infusion prevents the development of pNF expression in D R G neurons following either axotomy or 1DPN administration ~3~'. Taken together, these findings strongly support a model whereby the axon reaction (including the production of abnormal pNF expression in D R G neurons) is initiated by loss of target-derived trophic factor(s) (e.g. NGF). In regard to the Ola mouse model, it is interesting to note that a preliminary report describes the production of the axotomyinduced decrease in neurofilament synthesis in Ola mice at 1 week following axotomy 5. The observation that neurofilament synthesis is reduced in the Ola mutant prior to the onset of significant Wallerian degeneration lends further support to our model for induction of the axon reaction; i.e., loss of retrogradely transported NGF. Thus, it will be important to determine whether N G F administration can prevent this axotomy-induced reduction in neurofilament synthesis in Ola mice. Second, the role of a maintenance signal in the regulation of aberrant p N F expression was suggested by the reduction in staining intensity in axotomized

29 nerves in which axonal regeneration was prevented either mechanically 28 or by local injection of acrylamide 12. These studies showed that maintenance of aberrant pNF expression in D R G neurons is dependent upon continued axonal elongation in the distal stump. In sum, our previous findings suggested that aberrant expression of pNF epitopes in rat D R G neurons is regulated by at least two signals during regeneration. The present results support and extend our previous studies by providing evidence for the involvement of both induction and maintenance signals in a single model (i.e. Ola mice). Specifically, aberrant expression of pNF epitopes in D R G neurons from Ola mice appears to be regulated, as in rat DRG, by at least two signals: 1) a retrogradely transported target tissue-derived trophic factor lost to the cell body following axotomy (e.g. NGF), and 2) a maintenance signal which is dependent upon events occurring in the distal stump. The nature of this latter signal is unclear. Since sensory neurons in Ola mice demonstrate a defect in axonal regeneration as early as five days following axotomy 4, the reduction in staining intensity may arise directly from the failure of the axonal sprouts to elongate ~2'28. This could be mediated by a mechanism intrinsic to the neuron enabling the cell body to somehow sense that its axon is elongating (see ref. 19). Alternatively, impaired axonal regeneration 12'28 could lead to a failure of non-elongating sprouts to properly access a new signal generated from the products of Wallerian degeneration in the distal stump (see ref. 13). In Ola mice, the retardation in the onset of Wallerian degeneration could also impede the production of such a putative signal. This raises the possibility that the processes transpiring in sensory nerves from Ola mice (i.e. lack of an appropriate maintenance signal) are analogous to those occurring in the mammalian central nervous system (CNS) where axonal regeneration is limited. The seminal studies by Aguayo and co-workers 1 showing that a peripheral nerve graft enables CNS axons to regenerate over long distances following axotomy clearly establish that the lack of functional recovery in the CNS is not due to an intrinsic property of the neuron (see ref. 13). Furthermore, recent studies 33 indicate that in some CNS neurons (e.g. rubrospinal) this defect in regeneration is due, not to an inability to respond to axotomy, but to a failure to maintain the axon reaction. Thus, the Ola mouse model may provide a means to identify a putative signal, perhaps lacking in peripheral nerves from Ola mice at early times (up to 2 weeks) following axotomy, with therapeutic potential for the promotion of functional recovery in the CNS.

Regardless of the underlying mechanism resulting in the failure of D R G neurons from Ola mice to maintain aberrant pNF expression, our studies clearly indicate that in an individual sensory neuron at least two signals regulate this component of the axon reaction. Moreover, the presence of a second signal originating from the distal stump offers a possible explanation for the well established (see refs. 21, 27) inverse relationship between the magnitude of the axon reaction and distance of the lesion from the neuronal cell body; more proximal lesions would be expected to enhance production of the putative signal due to the presence of a larger distal stump. Taken together, our findings support a model whereby events in the distal stump regulate the magnitude and duration of the axon reaction initiated by lost target tissue-derived trophic support. Acknowledgement. Supported by funds from the U.S. Public Health Service Grant NIH NS19611.

REFERENCES 1 Aguayo, A.J., David, S. and Gray, G.M., Influences of the glial environment on the elongation of axons after injury transplantation studies in adult rodents, J. Exp. Biol., 95 (1981) 231-240. 2 Beuche, W. and Friede, R.L., The role of non-resident cells in Wallerian degeneration, J. Neurocytol., 13 (1984) 767-796. 3 Bignami, A. and Gambeni, P., Neurofilament phosphorylation in peripheral nerve regeneration, Brain Res., 375 (1986) 73-82. 4 Bisby, M.A. and Chen, S., Delayed Wallerian degeneration in sciatic nerves of C57BL/OIa mice is associated with impaired regeneration of sensory axons, Brain Res., 530 (1990) 117-120. 5 Bisby, M.A. and Quarrington, J.M., Regulation of the cell body response to axotomy in facial motoneurons of the C57BL/Ola mouse, Proc. Can. Fed. Biol. Socs., 35 (1992) 88. 6 Bisby, M.A. and Tetzlaff, W., Changes in cytoskeletal protein synthesis following axon injury and during axon regeneration, Mol. Neurobiol., 6 (1992) 107-123. 7 Bizzi, A. and Gambetti, P., Phosphorylation of neurofilaments is altered in aluminum intoxication, Acta Neuropathol., 71 (1986) 154-158. 8 Gold, B.G. and Austin, D.R., Regulation of aberrant neurofilament phosphorylation in neuronal perikarya I. Production following colchicine application to the sciatic nerve, J. Neuropathol. Exp. Neurol., 50 (1991) 615-626. 9 Gold, B.G. and Austin, D.R., Regulation of aberrant neurofllament phosphorylation in neuronal perikarya III. Alterations following single and continuous fl,fl'-iminodipropionitrile administrations, Brain Res., 563 (1991) 151-162. 10 Gold, B.G., Austin, D.R. and Griffin, J.W., Regulation of aberrant neurofilament phosphorylation in neuronal perikarya II. Correlation with continued axonal elongation following axotomy, J. Neuropathol. Exp. Neurol., 50 (1991)627-648. 11 Gold, B.G., Mobley, W.C. and Matheson, S.F., Regulation of axonal caliber, neurofilament content and nuclear localization in mature sensory neurons by nerve growth factor, J. Neurosci., 11 (1991) 943-955. 12 Gold, B.G., Price, D.L., Griffin, J.W., Rosenfeld, J., Hoffman, P.N., Sternberger, N.H. and Sternberger, L.A., Neurofilament antigens in acrylamide neuropathy, J. Neuropathol. Exp. Neurol., 47 (1988) 145-157. 13 Gold, B.G. and Spencer, P.S., Neurotrophic function in normal nerve and in peripheral neuropathies. In A. Gorio (Ed.), Neuroregeneration, Raven Press, New York, 1993, pp. 101-122.

3O 13aGold, B.G., Austin, D., Mobley, W.C. and Storm-Dickerson, T.. Reversal of axonal injury-induced aberrant neurofilament phosphorylation in sensory neuronal cell bodies by NGF, Soc. Neurosci. Abstr., 18 (1992) 14. 14 Goldstein, M.E., Copper, H.S., Bruce, J., Carden, M.J., Lee, V.M.-Y. and Schlaepfer, W.W., Phosphorylation of neurofilament proteins and chromatolysis following transection of rat sciatic nerve, .L Neurosci., 7 (1987) 1586-1594. 15 Goldstein, M.E., Sternberger, L.A. and Sternberger, N.tt., Microheterogeneity ('neurotypy') of neurofilament proteins, Proc. NatL Acad. Sci. U.S.A.. 228 (1983) 459-473. 16 Greeson, D.M., Moix, l_,., Meier, M., Armstrong, D.M. and Wiley, R.G., A continuing signal maintains NGF receptor expression in hypoglossal motor neurons after crush injury, Brain Res., 594 (1992) 351 355. 17 Griffin, J.W., Drachman, D.B. and Price, D.L., Fast axonal transport in motor nerve regeneration, J. Neurobiol., 7 (1976) 355-370. 18 Howland, R.D. and Alli, P., Altered phosphorylation of rat neuronal cytoskeletal proteins in acrylamide induced neuropathy, Brain Res., 363 (1986) 333-339. 19 Lasek, R.J., Studying the intrinsic determinants of neuronal form and function. In R.J. Lasek and M.M. Black (Eds.), Intrinsic Determinants of Neuronal Form and Function, Alan R. Liss, New York, 1988, pp. 3-58. 20 Lee, V.M.-Y., Carden, M.J., Schlaepfer, W.W. and Trojanowski J.Q., Monoclonal antibodies distinguish several differentially phosphorylated states of the two largest rat neurofilament subunits (NF-H and NF-M) and demonstrate their existence in the normal nervous system of adult rats, J. Neurosci., 7 (1987) 34743488. 21 Lieberman, A.R., The axon reaction: a review of the principal features of perikaryal responses to axon injury. In C.C. Pfeiffer and J.R. Smythies (Eds.), International ReL,iew of Neurobiology. Vol. 14, Academic Press, New York, 1971, 49-124. 22 Munoz, D.G., Greene, C., Perl, D.P. and Selkoe, D.J., Accumulation of phosphorylated neurofilaments in anterior horn motoneutons of amyotrophic lateral sclerosis patients, J. Neuropathol. L~p. Neurol., 47 (1988) 9-18. 23 Munoz-Garcia. D., Pendlebury, W.W., Kessler, J.B. and Perl, D.P., An immunocytochemical comparison of cytoskeletal proteins in aluminum-induced and Alzheimer-type neurofibrillary tangles, Acta Neuropathol., 70 (1986) 243-248. 24 Ostermann, E., Sternberger, N.H. and Sternberger, L.A., Immunocytochemistry of brain-reactive monoclonal antibodies in peripheral tissues, Cell Tissue Res., 228 (1983) 459-472.

25 Perry, V.It., Brown, M.C., Lunn, E.R., 'Free, t'. and Gordon, S., Evidence that very slow Wallerian degeneration in C57BL/OIa mice is an intrinsic property of the peripheral nerve, E u r . . I Neurosci., 2 (1990) 802-8{)8. 26 Perry, V.H., Lunn, E.R., Brown, M.C., Cahusac, S. and Gordon, S.. Evidence that the rate of Wallerian degeneration is controlled by a single autosomal dominant gene, Etu..l. ,%'eurosci, 2 (1990) 408-413. 27 Price, D.L., Griffin, J.W., Hoffman, P.N., Cork, L.('. and Spencer, P.S., The response of motor neurons to injury and disease. In P.J. Dyck, P.K. Thomas, E.H. Lambert and R. Bunge (Eds.), Peripheral Neuropathy, Vol. l, W.B. Saunders, Philadelphia, 1984. pp. 732-759. 28 Rosenfeld, J., Dorman, M.E., Griffin, J.W., Gold, B.G., Sternberger, L.A., Sternberger, N.H. and Price, D.I.,., Distribution of ncurofilament antigens after axonal injury, J. NeuropathoL Exp. ,N~,utv/., 40 (1987) 269-282. 29 Singer, P.A., Mehler, S. and Fernandez, H.L., Blockade of retro--rade axonal transport delays the onset of metabolic and morphoi~,ical changes induced by axotomy, J. Neurosci., 2 (1982) 1299130~. 30 Singer, P.A., Mehler, S. and Fernandez, H.I_,., Effect of extracts of injured nerve on initiating the regenerative response in the hypoglossal nucleus in the rat, Neurosci. Lett., 84 (1988) 155-160. 31 Sternbergcr, L.A. and Sternberger, N.H., Monoclonal antibodies distinguish phosphorylated and nonphosphorylated forms of neurofilaments in situ, Proc. Natl. Acad. Sci. USA, 80 (1983) 61266130. 32 Suwita, E., Lapadula, D.M. and Abou-Donia, M.B., Calcium and calmodulin-enhanced in vitro phosphorylation of hen brain coldstable microtubules and spinal cord neurofilament triplet proteins after a single oral dose of tri-o-cresyl phosphate, Proc. Natl. Acad. Sci. USA, 83 (1986) 6174-6178. 33 Tetzlaff, W., Alexander, S., Miller, F.D. and Bisby, M.A., Response of facial and rubrospinal neurons to axotomy: Changes in mRNA expression for cytoskeletal proteins and GAP-43, .L Neurosci., 11 (1991) 2528-2544. 34 Troncoso, J.C., Sternberger, N.H., Sternberger, L.A., Hoffman, P.N. and Price, D.L., Immunocytochemical studies of neurofilament antigens in the neurofibrillary pathology induced by aluminum, Brain Res., 364 (1986) 295-300. 35 Verge, V.M.K., Tetzlaff, W., Bisby, M.A. and Richardson, P.M., Influence of nerve growth factor on neurofilament gene expression in mature primary sensory neurons, J. Neurosci., 10 (1990) 2018-2025.

N o t e a d d e d in p r o o f A recent paper (Chert, S. and Bisby, M.A., J. Comp. Neurol., 333 (1993) 449-454) demonstrated that motor axons in Ola mice also exhibit impaired regeneration.