THE JOURNAL OF COMPARATIVE NEUROLOGY 370~97-104(1996)

Intrinsic Versus Extrinsic Factors in

Determining the Regeneration of the
Central Processes of Rat Dorsal Root
Ganglion Neurons: The Influence
of a Peripheral Nerve Graft
M.S. CHONG, C.J. WOOLF, M. TURMAINE, P.C. EMSON, AND P.N. ANDERSON
Department of Anatomy and Developmental Biology, University College London,
London WClE 6BT, United Kingdom

ABSTRACT
The relative contribution of intrinsic growth capacity versus extrinsic growth-promoting
factors in determining the capacity of transected dorsal root axons to regenerate long distances
was studied. L4 dorsal root axons regenerating into 4-cm peripheral nerve grafts on transected
dorsal roots were counted. Few dorsal root myelinated axons regenerated to the distal end of the
grafts by 10 weeks unless the sciatic nerve was also crushed. Regeneration of unmyelinated
axons was also increased by peripheral lesions.
Crush or transection of the dorsal roots without grafting did not alter GAP-43 mRNA
expression in L4 dorsal root ganglion (DRG)cells. Grafting a peripheral nerve onto the cut end
of an L4 dorsal root doubled the number of DRG cells expressing high levels of GAP-43 mRNA
after a delay of several weeks. Peripheral nerve crush at the time of nerve grafting resulted in a
very rapid rise in GAP-43 mRNA expression, which then declined to a steady level, twice that of
controls, by 7 weeks.
Thus, the rapid increase in the number of DRG neurons expressing high levels of GAP-43
mRNA after peripheral but not central axotomy correlates with the regeneration of central
axons through nerve grafts. Because GAP-43mRNA is slowly upregulated in a subpopulationof
sensory neurons in response to exposure of their central axons to a peripheral nerve
environment, environments favourable for axonal growth may act by increasing the intrinsic
growth response of neurons. Lack of intrinsic growth capacity may contribute to the failure of
dorsal root axons to regenerate into the spinal cord. o 1996 Wiley-Liss, Ine.

Indexing terms: nerve injury, primary sensory neurons, GAP-43, Schwann cells, spinal cord

The central and peripheral axons of primary afferent
neurons, although branches of a single process arising from
the cell body in the dorsal root ganglion (DRG), are distinct
in one important respect. The targets of the two axons are
quite different: the central nervous system for the central
axon and peripheral tissue for the peripheral axon. These
targets will present a very different array of soluble trophic
factors for retrograde transport to the cell body, the
presence or absence of which might influence the pheno-
type of the DRG cell in quite different ways, Nerve growth
factor, for example, is present in high concentration in the
periphery (Bandtlowet al., 1987)but not in the spinal cord
(Thoenen et al., 1987).Disrupting contact with the periph-
eral target following an injury to the peripheral axon
produces profound morphological and metabolic changes in
the cell body, the chromatolytic response, which are, how-
ever, absent following an equivalent injury to the central
axon (Lieberman, 1971; Aldskogius et al., 1992). This
differencemay reflect a differencein the signals to the DRG,
negative or positive, resulting from such lesions. A key
question is whether the absence of the cell body response
after dorsal root section is causally related to the reduced
regenerative capacity of the central axon (Oblinger and
Lasek, 1984)and whether this in turn is responsible for the
Accepted January 19,1996.
M.S. Chong’s present address: Department of Neurology, Selly Oak
Hospital, Birmingham, UK.
P.C. Emson’s present address: MRC Molecular Neuroscience Group, The
Babraham Institute, Cambridge, CB2 4AT, UK.
Address reprint requests to Dr. P.N. Anderson, Department of Anatomy
and Developmental Biology, University College London, Gower Street,
London, WClE 6BT, UK. E-mail: p.anderson@ucl.ac.uk

O 1996 WILEY-LISS, INC.

98
inability of injured dorsal root axons to grow back into the
spinal cord.
Crush injuries of the dorsal root result in limited regen-
eration without axonal penetration into the dorsal horn
(Perkins et al., 1980; Carlstedt, 1985; Liuzzi and Lasek,
1987; Stensaas et al., 1987; Siegal et al., 1990), whereas
successful regeneration of the peripheral axon may be seen
after injury to a nerve trunk (Horch, 1979). The cessation
of axonal regeneration in the dorsal root entry zone as a
consequence of the formation of synapselike interactions
with astrocytes may be a factor preventing regeneration of
central axons into the spinal cord (Carlstedt, 1985; Liuzzi
and Lasek, 1987; Stensaas et al., 1987), but the inability to
induce a vigorous growth capacity in the dorsal root gan-
glion cells after central axon injury may also contribute.
The growth-associated protein GAP-43 is a useful marker
of the intrinsic growth state of neurons (Skene, 1989).
GAP-43 is expressed by neurons at high levels during
development (Jacobson et al., 1986; Kalil and Skene, 1986;
Karns et al., 1987; Moya et al., 1988; De la Monte et a].,
1989; Fitzgerald et al., 1991; Chong et al., 1992)and during
regeneration in the adult (Skene and Willard, 1981a-c;
Woolf et al., 1990; Campbell et al., 1991; Schreyer and
Skene, 1991; Vaudano et al., 1995). Injuries to the periph-
eral axon cause a rapid and prolonged upregulation of both
GAP-43 mRNA and protein (Bisby, 1988; Hoffman, 1989;
Van der Zee et al., 1989; Woolf et al., 1990; Chong et al.,
1992) but sectioning dorsal roots does not have this effect
(Schreyer and Skene, 1993; Chong et al., 1994a).
Peripheral nerve grafts constilute an extremely favour-
able environment for axonal regeneration (Anderson et al.,
1983;Aguayo, 1985; Vidal-Sanz et al., 1987; Villegas-Perez
et al., 1988; Nadim et al., 1990; Campbell et al., 1992). The
injured central processes of DRG neurons have been shown
to be capable of regenerating into short peripheral nerve
grafts (Zelena and Jirmanova, 1988).However, Richardson
and Issa (1984) reported that only a few of the injured
central axons of DRG cells regenerate into peripheral
nerves grafted into the dorsal columns of the spinal cord
unless the sciatic nerve is also transected as a conditioning
stimulus. Similarly, Oudega et al. (1994) showed that a
predegenerate graft promotes even more regeneration of
dorsal column fibres than a fresh nerve graft but only in
experiments when there was a sciatic nerve crush. In the
present study, we have set out to examine further the
relative contribution of the intrinsic growth response, as
measured by GAP-43expression, and extrinsic environmen-
tal factors in determining regeneration of dorsal root axons
by performing experiments where long segments of periph-
eral nerve were grafted directly onto transected dorsal roots
with and without peripheral conditioning lesions.
METHODS
Adult inbred Fischer rats were used. Donor animals were
exsanguinated under terminal pentobarbitone anaesthesia,
and a 4-cm length of tibial nerve was removed from the
thigh and leg. The grafts were cleaned of any muscle or fat
and kept in Hank’s balanced saline solution (Gibco).Recipi-
ent animals were anaesthetised with a mixture of enfluranel
nitrous oxide, and a laminectomy was performed on the left
side to expose the lumbar spinal nerve roots. Experimental
animals were divided into four groups. In group 1, the L4
dorsal root was transected 17-20 mm from the DRG, and
the peripheral stump was laid on the adventitial surface of a
M.S. CHONG ET AL.
patch of aorta from the donor animal. Using 1010 sutures, a
tibial nerve graft was attached to the peripheral stump, and
bovine fibrinogen was used to ensure additional stability of
the anastomosis. The wound was closed in layers, and the
unattached distal end of the graft was marked with a 1010
suture and left free under the deep fascia overlyingthe back
muscles. Animals in group 2 underwent the same procedure
with an additional left sciatic nerve crush performed with
watchmaker’s forceps at mid-thigh level. In group 3, roots
were sectioned but not joined to a graft, and, in group 4,
roots were crushed with watchmaker’s forceps.
After survival times of 3 (n = 1),7(n = 21, and 15 (n = 2)
weeks for group 1, 3 (n = 11, 7 (n = 21, 15 (n = 21, and 19
(n = 1)weeks for group 2 , 3 (n = l),9 (n = l), and 16 (n = 1)
weeks for group 3, and 7 (n = 2) weeks for group 4, the
animals were terminally anaesthetised with pentobarbi-
tone. The spinal roots and graft were carefully dissected
out, and the ipsilateral and contralateral L4 DRGs were
removed and frozen in liquid nitrogen. Serial sections
20-pm thick were cut on a cryostat and mounted on
gelatinised slides. The method for nonisotopic in situ
hybridisation has been described elsewhere (Weise and
Emson, 1991; Chong et al., 1992). Briefly, the sections were
fixed in 4% paraformaldehyde, acetylated, and delipidated
before hybridisation with a 39-mer alkaline-phosphatase-
linked cDNA antisense probe corresponding to positions
119-157 of the rat GAP-43 mRNA. The bound probe was
visualised by the colour reaction produced by alkaline
phosphatase on the substrates nitroblue tetrazolium and
5-bromo-4-chloro-3-indoyl-phosphate. In control experi-
ments, it was shown that a 100-fold excess of unlabelled probe
and pretreatment with RNAase eliminated all staining.
Analysis of the results was performed with the aid of the
TPL-V4 (Seescan) image analysis system. Only highly
stained profiles, whose relative intensity levels did not
overlap with that of the background levels in specimens
processed without probe, were counted as positive. These
positively labelled profiles were then counted by using the
physical dissector method (Chong et al., 1992). This in-
volves the reconstruction of the DRG and the calculation of
volume by using the Cavalieri method, followed by serial
sampling to obtain the number of stained profiles per unit
volume (Sterio, 1984; Gundersen et al., 1988; Coggeshall,
1992). Total number of positively stained cells is then
obtained by multiplying the average number of stained
profiles per unit volume with volume of the tissue.
Some animals in group 1,at survival times of 3.5 (n = l),
6.5-7 (n = 31, 10 (n = 2) and 20 (n = 2) weeks and in group
2, at survival times 3.5 (n = l),6.5-7 (n = 3),10 (n = 2), 11
(n = l), and 19 (n = l), weeks were terminally anaesthe-
tised and perfused through the left ventricle of the heart
with 2% paraformaldehyde/4% glutaraldehyde in phos-
phate buffer, pH 7.4. The grafts were removed, cut into
sections, and processed into Araldite. Axons were counted
in sections 1-2 mm from the site of anastomosis (proximal
graft) and 1-3 mm from the distal ends of the grafts.
Counts of myelinated axons were made by using montages
of micrographs taken from 1-pm transverse sections of the
grafts stained with toluidine blue. Counts of unmyelinated
axons were carried out by systematic scanning of ultrathin
sections mounted on films covering slotted grids by using a
Philips EM 300. Unmyelinated axons could only be counted
in optimally processed segments of nerve because of the
difficulty in distinguishing between regenerating axons and
Schwann cell processes.
GAP-43 AND DORSAL ROOT REGENERATION 99
TABLE 1. Numbers of L4 Dorsal Root Ganglion Cells Expressing High L4 DRG was similar at 3 weeks to that after a dorsal root
Levels of GAP-43 mRNA After Dorsal Root Rhizotomy and Rhizotomy transection without grafting, which in turn was identical to
Combined With GraftingWith and Without
a Concurrent Sciatic Nerve Crush that in the contralateral, unoperated, ganglion (Figs. 1-2,
Table 1). However, 7 weeks after the operation, 6,500-
Ipsilateral Contralateral
Time (weeks) L4 count L4 count 6,800 cells expressing high levels of GAP-43 mRNA were
seen in the ipsilateral DRG, approximately double the
Dorsal rhizotomy number present in unoperated ganglia and higher than that
3 3,684 3,540
9 4,110 found after dorsal root transection alone. This increase in
16 4,330 the number of sensory neurons expressing high levels of
Dorsal root crush
7 2,905 GAP-43 mRNA in the group with grafts persisted until at
7 3,167 least 15 weeks (Figs. 1-2, Table 1).
Tibial nerve graft onto dorsal roots
3 3,822 3,366 The production of a peripheral nerve crush lesion at the
7 6,840 3,555 same time as the dorsal root graft resulted in large numbers
7 6,557 of L4 DRG cells expressing high levels of GAP-43 mRNA 3
15 6,160
15 6,891 weeks after the operation (Figs. 1-2). This result is similar
Tibial nerve graft onto dorsal root and sciatic crush to that found after a sciatic nerve crush performed alone
3 14,066 3,858
7 7,410 (Chong et al., 1992). Seven weeks postoperation, however,
7 7,419 3,379 there were still more than 7,000 L4 DRG cells expressing
14 7,026 3,033
19 6,594 4.235 high levels of GAP-43 mRNA, which is higher than the
numbers present after a sciatic nerve crush alone, when the
numbers begin to return to control levels at 5 weeks
rhizotomy
postcrush (Chong et al., 1992). This increased level of
GAP-43 mRNA expression in the grafted group persisted

I
16000 t. nocmsh
for at least 19 weeks (Table 1).From 7 weeks postoperation
&4 sciatic crush onwards, the numbers of GAP-43-positivecells were similar
after a dorsal root graft whether or not a peripheral nerve
14000 t lesion had been carried out.
\
\
12000 \ Axonal counts
c
u)

2
2 10000 i
3
\
\
\
\
Three and a half weeks after a 4-cm peripheral nerve had
been grafted onto a transected dorsal root, many myelinated
axons were present in the proximal part of the grafts (Fig.
3), but no myelinated axons had reached the distal end.
After 6.5-7 weeks, only a very few myelinated axons reached
-- A the distal end of the graft (65 2 20 S.E.M., n = 3). By 10
weeks, the numbers had increased (Fig. 3) to a mean of 530
(n = 2) with a further substantial increase at 20 weeks to
1,124 (n = 2; Table 2, Fig. 3). The numbers ofunmyelinated
axons more than doubled from a mean of 852 (n = 2) at

A
2ooo0 0 2 4 6 8 10 12
Weeks after grafting
Fig. 1. Graph drawn from data in Table 1 showing the changes in
the numbers of dorsal root ganglion (DRG) neurons expressing high
levels of GAP-43 mRNA following dorsal rhizotomy alone or the
attachment of a tibial nerve graft to the injured dorsal root with and
without a concurrent sciatic nerve crush.
RESULTS
GAP-43 mRNA
An L4 dorsal root section or crush lesion by itself failed to
produce any detectable upregulation of GAP-43 mRNA in
L4 DRG cells compared with either naive L4 DRGs or with
DRGs contralateral to the central axotomy when examined
3,9, and 16 weeks (transection) or 7 weeks (crush) after the
surgery (Table 1,Fig. 11, a result similar to that reported at
shorter survival times (Chonget al., 1994a).
When a peripheral nerve graft was attached to the cut
end of the L4 dorsal root, the number of DRG cells
expressing high levels of GAP-43 mRNA in the ipsilateral
6.5-7 weeks to 2,310 (n = 2) at 20 weeks (Table 2).
When a sciatic nerve crush lesion was performed concur-
rently with the tibial graft to the dorsal root, there were, at
14 16 18 20 all time points examined, more myelinated axons at the
distal ends of the grafts than when the graft was performed
without the peripheral conditioning lesion. (Table 2, Figs.
3,4). For example, at 6.5-7 weeks, there were more than 20
times as many myelinated axons at the distal ends of grafts
in animals with a sciatic nerve crush (1,383 & 295, n = 3) as
in those without a sciatic nerve injury (65 t 20, n = 3). The
data for unmyelinated axons (Table 2) show a similar
phenomenon; at each time point, they were more numerous
at the distal end of the grafts in animals that had a sciatic
nerve crush performed.
The numbers of myelinated axons in the proximal part of
the graft showed smaller changes with time and were less
influenced by a sciatic nerve conditioning lesion (Table 2).
DISCUSSION
Injury to the central processes of primary sensory neu-
rons in the form of a dorsal root section or crush did not
influence GAP-43 mRNA expression in DRG cells for up to
16 weeks after the axotomy, confirming the results of
previous experiments with shorter duration (Schreyer and
Skene, 1993; Chong et al., 1994a). Grafting of a peripheral
100 M.S. CHONG ET AL.

Graft & Sciatic Crush Dorsal Root Graft

Fig. 2. Photomicrographs showing in situ hybridization for GAP-43 graphs a and d show ganglia 3 weeks after operation, b and d show
mRNA in sections of L4 DRG at various times after attaching a nerve ganglia 7 weeks after operation, c shows a ganglion 14 weeks after
graft to the severed root with concurrent ipsilateral sciatic nerve injury operation, and f shows ganglia 15 weeks after operation. Scale bars =
(a+) and in the absence of sciatic nerve injury (d-0. Photomicro- 150 Fm.
 Fig. 3. Photomicrographs of 1-pm sections through tibial nerve
grafts attached to severed L4 dorsal roots. A,B: Proximal graft 3.5
weeks after operation. C,D: Distal graft 10 weeks after operation. A and
C are from an experiment in which the sciatic nerve was undamaged,
and B and D are from an experiment in which the ipsilateral sciatic
nerve was crushed when the graft was attached. Sciatic nerve crush has
little effect on the growth of myelinated axons (arrows) into the
proximal graft but increases the numbers found in the distal graft.
Plastic embedded. Scale bar = 10 pm.
102 M.S. CHONG ET AL.
TABLE 2. Counts ofAxons Present in the Proximal and Distal (4cm) End nerve crush was probably entirely the result of the sciatic
of Tibial Nerves Grafted Onto L4 Dorsal Roots nerve injury; the values found were almost identical to
at Various Time Intervals After Surgery
those found following sciatic nerve crush alone (Chong et
Myelinated Distal al., 1992,199413).Beyond this time, however, the determin-
Time (weeks) Proximal Distal Unmyelinated Total ing influence on the number of GAP-43-positive neurons
appeared to be only the presence or absence of the periph-
Tibial nerve graft without sciatic
nerve crush eral nerve graft because sciatic nerve crush alone results in
3.5 - 0 - - a decay in GAP-43 upregulation soon after peripheral
6.5 1,877 68 1,490 1,558
6.5 2,386 98 - - reinnervation (Chong et al., 1992). It appears, therefore,
7 2,457 28 213 241 that a subpopulation of primary sensory neurons in the L4
10 1,642 394 611 1,005 DRG can respond, after a considerable delay, to the expo-
10 3,421 667 - -
20 1,457 779 2,278 3,057 sure of their central axons to some aspect of the peripheral
20 3,761 1,469 2,342 3,811 nerve environment by upregulating GAP-43 expression.
Tibial nervegraftwithsciaticcrush
3.5 - 3 - - The upregulation of GAP-43 in DRG cell bodies as their
6.5
6.5
2,739
2,227
1,932
1,293
3,281
-
5,231
-
axons regenerate through the peripheral nerve would sug-
7 3,446 925 1,499 2,424 gest a retrograde signal triggered by the interaction be-
10 5,221 1,697 3,869 5,566 tween axon and graft, which is not present after a dorsal
10 3,379 1,350 2,825 4,175 root section or crush. The nature of this signal is unknown
11 3,416 1,506 2,984 4,490
19 - 4,280 6,330 10,610 but could include events induced by contact between the
growing axons and Schwann cells in the graft, trophic
factors secreted by cells in the graft, or even some substrate
nocrush molecule present in the graft. To look for this signal, unique
to, or concentrated in, peripheral nerves, it would be
B sciatic crush reasonable to examine the difference between denervated
dorsal roots and peripheral nerves. Although Schwann cells
in the dorsal roots are thought to be similar to those in
peripheral sensory nerves, there are differences such as the

a=
4500

4000

3500
1
J
-
?
expression of the LBIHNK-1 epitope, which in adult ani-
mals is predominantly expressed in motor nerves (Martini
et al., 1992).An interaction between regenerating axon and
Schwann cells expressing LBIHNK-1, or some other periph-
eral marker, may bring about the reexpression of GAP-43
EUJ in DRG cells. It is likely that many regenerating central
-
J 3000 - processes will come into contact with Schwann cells express-
._
U
lo ing the LBIHNK-1epitope in the graft but not after a dorsal
.G 2500 - root crush, and it would be interesting to find out if the
fastest regenerating axons are both GAP-43 positive and
:
c
u)

U
8 2000 found in the bands of Biingner previously occupied by
motor fibres. Although much is known about the expression
3
.? 1500
- 4 of surface markers and growth factors such as nerve growth
% factor, brain-derived neurotrophic factor, and ciliary neuro-
= 1000 1 trophic factor in denervated Schwann cells in injured nerve

500

0-c
4 trunks (Bixbyet al., 1988; Daniloff et al., 1989; Meyer et al.,
0 2 4 6 8 10 12 14 16
Weeks after grafting
Fig. 4. Graph drawn from data in Table 2 showing the number of
myelinated axons present in sections through the distal parts of grafts
attached to severed L4 dorsal roots. Counts from experiments of 6.5-7
weeks’ duration have been averaged, as have counts from experiments
of 10-11 weeks’ duration.
nerve onto a severed L4 dorsal root had no effect on the
number of DRG neurons expressing high levels of GAP-43
mRNA at 3 weeks. However, after longer postoperative
periods, the grafts attached to the L4 dorsal roots did
appear to cause an upregulation of GAP-43 mRNA in some
DRG neurons, with approximately a doubling of the num-
ber of GAP-43 mRNA-positiveneurons in the L4 DRG at 7
weeks. The upregulation of GAP-43 found 3 weeks after
attaching a graft to the dorsal root in animals with a sciatic
1992; Martini, 19941, there have been few comparable
studies on injured dorsal roots.
Crushing the sciatic nerve at the time when a graft was
attached to the L4 dorsal root produced a massive, but
18 20 temporary, increase in GAP-43 mRNA expression and
increased the rate of regeneration of myelinated and of
unmyelinated axons through the grafts. Our results con-
firm the findings of Richardson and Verge (1987) and Lu
and Richardson (1991) who showed that the rate of regen-
eration of crushed dorsal roots was increased by a sciatic
nerve injury and are readily compatible with the results of
Richardson and Issa (1984) and Oudega et al. (1994),where
a peripheral conditioning stimulus was shown to increase
the regeneration of central axons of primary afferents into
nerve grafts placed in the dorsal columns.However, Oblinger
and Lasek (1984) reported that a peripheral conditioning
lesion did not increase the rate of dorsal root regeneration
but doubled the speed of regeneration in the sciatic nerve.
The explanation of this apparent discrepancy is not clear,
but different methods were used to assess regeneration in
each case, and the experiments of Oblinger and Lasek
(1984) were confined to the first 7 days after rhizotomy. It
may be germane that in the present study a conditioning
GAP-43 AND DORSAL ROOT REGENERATION
lesion also had comparatively little effect on the number of
myelinated axons, which regenerated into the proximal
graft. This may indicate that many injured dorsal root
axons have the capacity to regenerate short distances but
that a particular growth state is required to maintain long
neurite outgrowth. Nonetheless, a sciatic nerve injury, with
its associated increase in GAP-43 expression in the sensory
neurons, enabled a much more vigorous regenerative re-
sponse by injured dorsal root axons in the environment
provided by a peripheral nerve graft, and this advantage
persisted for 3 months after the lesion, with no detectable
catching up by the group with a graft but no conditioning
lesion.
Although the availabilityof a favourableexternal environ-
ment (Aguayo, 1985; Chong et al., 1994b) is a factor in
promoting successful regeneration, it is becoming clear that
neurons differ dramatically in their ability to regenerate
axons (Morrow et al., 1993) and that there are important
intrinsic determinants of axonal growth (Fawcett, 1992).
We have demonstrated that, although the provision of a
peripheral nerve environment for regenerating central
processes of primary sensory neurons can eventually trigger
an enhanced cell body response in some injured neurons, a
concomitant peripheral nerve lesion has a much earlier and
more dramatic effect on a marker of growth capacity, i.e.,
GAP-43 expression. GAP-43 expression in DRG neurons
correlates very well with the growth of injured axons, and
the most important finding of the present series of experi-
ments is that a peripheral nerve conditioning lesion leads to
a more rapid and sustained growth of central axons for long
distances. The inability of DRG neurons to upregulate
GAP-43 in response to dorsal rhizotomy may be a contribu-
tory reason for the failure of sensory axons to grow back
into the spinal cord after such lesions. Fetal DRG neurons
grafted into the position of excised lumbar DRG in adult
rats can grow axons back into the spinal cord (Kozlova et
al., 1995); such cells would be expected to express high
levels of growth-associated molecules including GAP-43.
However, the existence of a strain of transgenic mice
lacking the GAP-43 gene (Strittmatter et al., 1995) must
call into question the necessity of GAP-43 for axonal
growth. Although very few of these animals survive the
neonatal period, their nervous system is not grossly abnor-
mal, and the abnormalities that exist suggest problems in
axonal pathfinding in the visual system. This is in keeping
with other evidence that GAP-43is involved in the response
of growth cones to environmental signals (Aigner and
Caroni, 1995). Nonetheless, no evidence was found in the
present study to suggest that regenerating dorsal root
axons had difficulty in finding their way across the junction
zone between the severed roots and the nerve grafts, even
in experiments when there was no sciatic nerve crush to
increase the numbers of DRG neurons expressing GAP-43
during the first few weeks after operation. Within the
grafts, axonal growth took place inside bands of Bungner,
where path finding would not be expected to pose great
difficulties, yet axonal regeneration was slower in the
animals in which there was no sciatic nerve crush. Further-
more, it is not yet known whether axonal regeneration is
normal in the GAP-43-deficient mice; the development of
axonal pathways, which is largely normal, occurs over
much smaller distances than those over which axons can
regenerate.
Only certain populations of neurons respond to axotomy
or the presence of a peripheral nerve graft by upregulating
GAP-43 (Chong et al., 1994a; Vaudano et al., 1995). The
103
data indicate furthermore that the capacity to upregulate
GAP-43 correlates with the ability to grow long distances.
This clearly has important implications for attempts to
encourage central nervous system axonal regeneration:
intrinsic growth response needs to be promoted in addition
to the provision of a favourable environment if successful
axonal elongation is to be achieved. Favourable environ-
ments by themselves may not be sufficient.
ACKNOWLEDGEMENTS
This work was supported by the Royal Society, MRC, and
the ISRT. C.J.W. is a recipient of a Bristol-Myers Squibb
Unrestricted Pain Research grant.
LITERATURE CITED
Aguayo, A.J. (1985) Axonal regeneration from injured neurons in the adult
mammalian central nervous system. In C.W. Cotman (ed): Synaptic
Plasticity. New York Guilford Press, pp. 457484.
Aigner, L., and P. Caroni (1995)Absence of persistent spreading, branching,
and adhesion in GAP-43-depleted growth cones. J. Cell Biol. 128:647-
660.
Aldskogius, H., J. Arvidsson, and G. Grant (1992) Axotomy-induced changes
in primary sensory neurons. In S.A. Scott (ed): Sensory Neurons.
Diversity, Development and Plasticity. Oxford: Oxford University Press,
pp. 363-383.
Anderson, P.N., J. Mitchell, D. Mayor, andV.V. Stauber (1983)An ultrastruc-
t u r d study of the early stages of axonal regeneration through rat nerve
grafts. Neuropathol. Appl. Neurobiol. 9:455466.
Bandtlow, C, R. Heumann, M.E. Schwab, and H. Thoenen (1987) Cellular
localization of nerve growth factor synthesis by in situ hybridization.
EMBO J. 6:891-899.
Bisby, M.A. (1988) Dependence of GAP-43 (B50,Fl)transport on axonal re-
generation in rat dorsal root ganglion neurons. Brain Res. 458:157-161.
Bixby, J.L., J. Lilien, and L.F. Reichardt (1988) Identification of the major
proteins that promote neuronal process outgrowth on Schwann cells in
vitro. J. Cell Biol. 107:353-361.
Campbell, G., P.N. Anderson, M. Turmaine, and A.R. Lieberman (1991)
GAP-43 in the axons of mammalian CNS neurons regenerating into
peripheral nerve grafts. Exp. Brain Res. 87:67-74.
Campbell, G., A.R. Lieberman, P.N. Anderson, and M. Turmaine (1992)
Regeneration of adult rat CNS axons into peripheral nerve autografts:
Ultrastructural studies of the early stages of axonal sprouting and
regenerative axonal growth. J. Neurocytol. 21:755-787.
Carlstedt, T. (1985) Regenerating axons form nerve terminals at astrocytes.
Brain Res. 347:18%191.
Chong, M.S., M. Fitzgerald, J. Winter, M. Hu-Tsai, P.C. Emson, U. Wiese,
and C.J. Woolf (1992) Gap-43 mRNA in rat spinal cord dorsal root
ganglion neurons: Developmental changes and re-expression following
peripheral nerve injury. Eur. J. Neurosci. 49383-895.
Chong, M.S., M.L. Reynolds, N. Irwin, R.E. Coggeshall, P.C.Emson, L.I.
Benowitz, and C.J. Woolf (1994a) GAP-43 expression in primary sensory
neurons followingcentral axotomy. J. Neurosci. 14.43754384.
Chong, M.S., C.J. Woolf, P. Andrews, M. Turmaine, D.J. Schreyer, and P.N.
Anderson (1994b) The downregulation of GAP-43 is not responsible for
the failure of regeneration in freeze-killed nerve grafts in the rat. Exp.
Neurol. 129:311-320.
Coggeshall, R.E (1992) A consideration of neural counting methods. TINS
15t9-13.
Daniloff, J.K., K.L. Crossin, M. Pincon-Raymond, M. Murawsky, F. Rieger,
and G.M. Edelman (1989) Expression of cytotactin in the normal and
regenerating neuromuscular system. J. Cell Biol. 108:625435.
De la Monte, S.M., H.J. Federoff, S.-C. Ng, E. Grabczyk, and M.C. Fishman
(1989) GAP-43 gene expression during development Persistence in a
distinctive set of neurons in the mature central nervous system. Dev.
Brain Res. 46:161-168.
Fawcett, J.W. (1992)Intrinsic neuronal determinants of regeneration. TINS
15:5-8.
Fitzgerald, M., M.L. Reynolds, and L.I. Benowitz (19911GAP-43 expression
in the developing rat lumbar spinal cord. Neuroscience 41:187-199.
Gundersen, H.J.G., P. Bagger, T.F. Bendtsen, S.M. Evans, L. Korbo, N.
Marcussen, A. Mgller, K. Nielsen, J.R. Nyengaard, B Pakkenberg, F.B.
104
Sgrensen, A. Vesterby, and M.J. West (1988)The new stereologicaltools:
Disector, fractionator, nucleator and point sampling intercepts and their
use in pathological research and diagnosis. Acta Pathol. Microbiol.
Immunol. Scand. 96.857-879.
Horch, K. (1979) Guidance of r e g r o ~ n gsensory axons after cutaneous nerve
lesion in the cat. J. Neurophysiol. 421437-1447.
Jacobson, R.D., I. Virag, and J.H.P. Skene (1986) A protein associated with
axonal growth, GAP-43, is widely distributed and developmentally
regulated in rat CNS. J. Neurosci. 6:1843-1855.
Kalil, K., and J.H.P. Skene (1986) Elevated synthesis of an axonally
transported protein correlates with axonal outgrowth in normal and
injured pyramidal tracts. J. Neurosci. 6.2563-2570.
Karns, L.R., S.-C. Ng, J.A. Freeman, and M.C. Fishman (1987) Cloning of
complementary DNA for GAP-43, a neuronal growth-related protein.
Science 236r597-560.
Kozlova, E.N., I. Stromherg, M. Bygdeman, and H. Aldskogius (1995)
Peripherally grafted human foetal dorsal root ganglion cells extend
axons into the spinal cord of adult host rats by circumventing dorsal root
entry zone astrocytes. NeuroReport 5.269-272.
Lieberman, A.R. (1971) The axon reaction: A review of the principal features
of perikaryon responses to axon injury. Int. Rev. Neurobiol. 14t49-124.
Liuzzi, F.J., and R.J. Lasek (1987) Astrocytes block axonal regeneration in
mammals by activating the physiological stop pathway. Science 237:642-
645.
Lu, X., and P.M. Richardson (1991) Inflammation near the nerve cell body
enhances axonal regeneration. J. Neurosci. 11.972-978.
Martini, R. (1994) Expression and functional roles of neural cell surface
molecules and extracellular matrix components during development and
regeneration of peripheral nerves. J. Neurocytol. 23.1-28.
Martini, R., Y. Xin, B. Schmitz, and M. Schachner (1992) The LPIHNK-1
carbohydrate epitope is involved in the preferential outgrowth of motor
neurons on ventral roots and motor nerves. Eur. J. Neurosci. 4:62&639.
Meyer, M., I. Matsuoka, C. Wetmore, L. Olsen, and H. Theonen (1992)
Enhanced synthesis of BDNF in the lesioned peripheral nerve: Different
cellular mechanisms are respoonsible for the regulation of BDNF and
NGF mRNA. J. Cell Biol. 119r45-54.
Morrow, D.R., G. Campbell, A.R. Lieberman, and P.N. Anderson (1993)
DilTerential regenerative growth of CNS axons into tibia1 and peroneal
nerve grafts in the thalamus of adult rats. Exp. Neurol. 120:60-69.
Moya, K.L., L.I. Benowitz, S. Jhaveri, and G. Schneider (1988) Changes in
rapidly transported proteins in developing hamster retinofugal axons. J.
Neurosci. 8r44454454.
Nadim, W., P.N. Anderson, and M. Turmaine (1990) The role of Schwann
cells and basal lamina tubes in the regeneration of axons through long
lengths of freeze-killed nerve grafts. Neuropathol. Appl. Neurobiol.
16:411421.
Oblinger, M.M., and R.J. Lasek (1984) A conditioning lesion of the periph-
eral axons of dorsal root ganglion cells accelerates regeneration of only
their peripheral axons. J. Neurosci. 4:1736-1744.
Oudega, M., S. Varon and T. Hagg (1994) Regeneration of adult rat sensory
axons into intraspinal nerve grafts: Promoting effects of conditioning
lesion and graft predegeneration. Exp. Neurol. 129.194-206.
Perkins, C.S., T. Carlstedt, K. Mizuno, and A.J. Aguayo (1980) Failure of
regenerating dorsal root axons to regrow into the spinal cord. Can. J.
Neurol. Sci. 7:323.
Richardon, P.M., and V.M.K. Issa (1984) Peripheral nerve injury enhances
central regeneration of primary sensory neurones. Nature 309391-793.
Richardson, P.M., and V.M.K. Verge (1987) Axonal regeneration in dorsal
spinal roots is accelerated by peripheral axonal transection. Brain Res.
411:406408.
M.S. CHONG ET AL.
Schreyer, D.J., and J.H.P. Skene (1991) Fate of GAP-43 in ascending spinal
axons of DRG neurons after peripheral nerve injury: Delayed accumula-
tion and correlation with regenerative potential. J. Neurosci. 1113738-
3751.
Schreyer, D.J., and J.H.P. Skene (1993) Injury induced GAP-43 expression
displays axon branch specificity in adult dorsal root ganglion neurons. J.
Neurobiol. 24t959-970.
Siegal, J.D., M. Kliot, G.M. Smith, and J. Silver (1990) A comparison of the
regenerating potential of dorsal root fibers into grey or white matter of
the adult rat spinal cord. Exp. Neurol. 109:90-97.
Skene, J.H.P. (1989) Axonal growth-associated proteins. Annu. Rev. Neuro-
sci. 12127-156.
Skene, J.H.P., and M. Willard (1981a) Changes in axonally transported
protein during axon regeneration in toad retinal ganglion cells. J. Cell
Biol. 89.86-95.
Skene, J.H.P., and M. Willard (1981b) Axonally transported proteins
associated with axon growth in rabbit cental and peripheral nervous
system. J. Cell Biol. 89r96-103.
Skene, J.H.P., and M. Willard (1981~)Characteristics of growth associated
polypeptides in regenerating toad retinal ganglion cell axons. J. Neuro-
sci. k419-426.
Stensaas, L.J., L.M. Partlow, P.R. Burgess, and K.W. Horch (1987) Inhihi-
tion of regeneration: The ultrastructure of reactive astrocytes and
abortive axon terminals in the transitional zone of the dorsal root. Prog.
Brain Res. 71t457467.
Sterio, D.C. (1984)The unbiased estimation of number and sizes of arbitrary
particles using the disector. J. Microsc. 134t127-136.
Strittmatter, S.M., C. Fankhauser, P.L. Huang, H. Mashimo, and M.C.
Fishman (1995) Neuronal pathfinding is abnormal in mice lacking the
neuronal growth cone protein GAP-43. Cell 80:445452.
Thoenen, H., C. Bandtlow, and R. Heumann (1987) The physiological
function of nerve growth factor in the central nervous system: Compari-
son with the periphery. Rev. Physiol. Biochem. Pharmacol. 109.146-178.
Van der Zee, C.E.E.M., H.B. Nielander, J.P. Vos, S.L. da Silva, J. Verhaagen,
A.B. Oestreicher, L.H. Schrama, P. Schotman, and W.H. Gispen (1989)
Expression of growth-associated protein B-50 (GAP43) in dorsal root
ganglia and sciatic nerve during regenerative sprouting. J. Neurosci.
9:3505-35 12.
Vaudano, E, G. Campbell, P.N. Anderson, A.P. Davies, C. Woolhead, D.J.
Schreyer, and A.R. Lieberman (1995) The effects of a lesion or a
peripjheral nerve graft on GAP-43 upregulation in adult rat brain: An in
situ hybridization and immunocytochemicalstudy. J. Neurosci. 15.3594-
3611.
Vidal-Sanz, M., G.M. Bray, M.P. Villegas-Perez, S . Thanos, and A.J. Aguayo
(1987) Axonal regeneration and synapse formation in the superior
colliculus by retinal ganglion cells in the adult rat. J. Neurosci. 7.2894-
2907.
Villegas-Perez, M.P., M. Vidal-Sanz, G.M. Bray, and A.J. Aguayo (1988)
Influences of peripheral nerve grafts on the survival and regrowth of
axotomized retinal ganglion cells in adult rats. J. Neurosci. 8.265-280.
Wiese, U.H., and P.C. Emson (1991) Intraspinal cellular responses after
spinal cord transection in rats: Neuronal expression of growth-associated
protein (GAP-93). Res Neurol. Neurosci. 2.199-204.
Woolf, C.J., M. Reynolds, C. Molander, C. O’Brien, R.M. Lindsay, and L.I.
Benowitz (1990) GAP-43 appears in the rat dorsal horn following
peripheral nerve injury. Neuroscience 34.465478.
Zelena, J., and I. Jirmanova (1988) Grafts of pacinian corpuscles reinervated
hy dorsal root axons. Brain Res. 438; 1655174.