J. Embryol. exp. Morph. Vol. 61, pp. 289-301, 1981
Printed in Great Britain © Company of Biologists Limited 1981
289
Control of neurite extension by embryonic
heart explants
By TED EBENDAL 1
From the Department of Zoology, Uppsala University
SUMMARY
The dynamics of neurite outgrowth elicited by embryonic chick heart explants in
sympathetic, spinal, ciliary and Remak's ganglia were investigated in collagen gel cocultures.
Neurites emerged preferentially on the side facing the heart explants even after only 6 h and
continued to increase in density and length for the next 2 days. Removal of the heart explants
after only initial stimulation resulted in less-dense neurite outgrowth. Washing of such
cultures led to retraction or degeneration of neurites, effects which could be countered by
again adding heart explants. Addition of a second set of heart explants on the back of
ganglia initiated a second wave of neuritic outgrowth locally. Ganglia extracted from gels
separate from their fibre halos and transferred to a second gel did not regenerate neurites
unless again stimulated by heart explants. Neurites from additional, distally positioned
ganglia failed to advance into parts of the gel shadowed from the heart explants by proximal
ganglia. The asymmetry of neurite outgrowths may be explained by local chemokinetic
stimulation of extension, possibly in combination with chemotactic orientation of fibre tips
up concentration gradients. The results show that the extension of several categories of
ganglionic neurites was reversible, being controlled by the concentration of a soluble
neuronotrophic factor released from a developing end organ.
INTRODUCTION
The theory that developing tissues emit chemical signals which attract
outgrowing nerve axons (neurotropism) was presented by Ramon y Cajal
already in the last century (see Jacobson, 1978, for a review of nerve fibre
guidance). Such mechanisms received little experimental support until it was
demonstrated that implants of tumour tissues producing nerve growth factor
(NGF) cause hyperinnervation of viscera in the chick embryo (Levi-Montalcini
& Hamburger, 1953; for reviews of NGF see Levi-Montalcini & Angeletti,
1968; Bradshaw, 1978), and that NGF injected into the brain of newborn rats
attracts massive ingrowth of sympathetic fibres along central pathways
normally devoid of such axons (Menesini Chen, Chen & Levi-Montalcini, 1978).
Observations of favoured outgrowth of neurites towards certain tissues
(Chamley, Goller & Burnstock, 1973; Eranko & Lahtinen, 1978) and towards
localized sources of NGF in culture (Charlwood, Lamont & Banks, 1972;
1
Author's address: Department of Zoology, Uppsala University, Box 561, S-751 22
Uppsala, Sweden.
290
T. EBENDAL
Ebendal & Jacobson, 1977a; Campenot, 1977; Letourneau, 1978) were also
taken to support the notion that NGF may act as a neurotropic factor (Chamley
& Dowel, 1975; Levi-Montalcini, Menesini Chen & Chen, 1978). Little is
known, however, of the actual distribution of NGF in developing tissues
(see Harper & Thoenen, 1980).
Other growth factors besides NGF which affect survival and neurite
extension in neurons (neuronotrophic factors, Varon & Bunge, 1978), were
also implicated in embryonic tissues (Ebendal & Jacobson, 1977 a; Helfand,
Riopelle & Wessells, 1978; McLennan & Hendry, 1978; Collins, 1978; Adler,
Landa, Manthorpe & Varon, 1979). Several reports independently established,
through the use of ganglionic or single-cell bioassays of nerve growth activity
in embryonic tissues, a developmentally regulated appearance of such non-NGF
trophic factors in the heart, liver and intraocular tissues of the chick (Ebendal,
1979; Lindsay & Tarbit, 1979: Landa, Adler, Manthorpe & Varon, 1980).
Similar neurite-inducing activities were also present at various levels in other
embryonic chick organs (Ebendal & Jacobson, 19776). Partial purification of
the active material was carried out in the case of heart (Ebendal, Belew,
Jacobson & Porath, 1979) and intraocular tissues (Manthorpe et al. 1980).
In order to confirm the developmental role, if any, of these factors, it is
vital to discover not only their intraembryonic distribution and chemical
properties but also the extent to which they can control density and direction
of neurite outgrowth. In an earlier paper (Ebendal, 1979), I reported that chick
heart explants co-cultured with different chick ganglia in collagen gels elicited
outgrowth of neurites preferentially towards, rather than away from, the
stimulatory explants. The present experiments were designed to study the
interaction between the heart and ganglia by removing, inserting or repositioning
explants at various intervals, in order to gain a better understanding of the
possibilities and limitations for a peripheral tissue to control its innervation
by way of neuronal growth factors.
MATERIALS AND METHODS
Sympathetic ganglia, ciliary ganglia, spinal ganglia and Remak's ganglia
were collected from 8- to 10-day-old chick embryos (White Leghorn), Ganglia
were exposed from one side to three explants of 16- to 18-day-old embryonic
chick heart ventricle in collagen gels (Elsdale & Bard, 1972) as described earlier
(Ebendal, 1979; see e.g. Figs 1 and 2). The chosen age combination results in
maximum or near maximum outgrowth of neurites (Ebendal, 1979). The
distance between ganglionic and heart explants was about 1 mm. Eagle's basal
medium plus 1 % fetal calf serum was used for the gels. The 35 mm culture
dishes containing the gels were incubated for 2 days at 37 °C in a humidified
atmosphere with 5 % CO 2 . Cultures were examined at intervals through an
inverted microscope.
Control o neurite extensifon
291
Fig. 1. Sympathetic chick ganglion cultured for 2 days in a collagen gel. Three heart
explants were present (left) for the first 24 h and then removed (arrows). An
asymmetrical outgrowth of neurites is evoked by the heart tissue.
A technique of extricating explants from, or inserting new explants into, an
already-precipitated gel with the aid of fine watchmaker's forceps, was used
in this study. After pieces were removed from the gel (thickness about 0-2-0-5
mm) a hole was left (Fig. 1). The gel readily accommodated new explants if
these were inserted with care. The method also allowed transfer of explants
from one gel to another (see Fig. 7). For some experiments the gels were gently
washed by several changes of extra culture medium, superimposed on the
collagen layer.
Neurite outgrowth was scored after 1 or 2 days by counting the number of
intersections seen between nerve fibre bundles and a straight line of an eyepiece
graticule positioned through the outgrowth at half the distance between explant
boundary and leading fibre tips as shown, e.g. in Fig. 2. Using dark-field
illumination, outgrowth was determined, independently for the side of the
ganglion facing the original heart explants (side a in Figs 2, 6 and 7) and the
opposite side (side b in corresponding Figs). It was previously shown that this
method is a reliable measure of neurite densities evoked in ganglia by a series
of NGF concentrations (Ebendal, 1979). Measurements were carried out in
eight dishes for each combination and are given as means ± S.E.M. in the graphs.
For statistical analysis the individual scores were compared using the MannWhitney U test (Siegel, 1956).
EXPERIMENTS AND RESULTS
Neuritic outgrowth and the presence of heart explants
In order to examine how neurite outgrowth depends upon the initial
exposure to heart tissue, the heart explants were removed from the gels after
various periods (Fig. 2). The outgrowth observed after 2 days on the exposed
side (a side) was increased when the heart explants were present for a longer
period. This held good for all four types of ganglia tested (Fig. 2). A similar
292
T. EBENDAL
•—•
o—o
• — •
D
D
Remak's G.
Ciliary G.
Spinal G.
Sympathetic G.
200
O 150
100
50
0 3 6 1 2
24
48
0 3 6 1 2
24
48
Presence of heart on side a (h)
Fig. 2. Effects on density of neuritic outgrowth of gradually increasing the period
when heart tissue was present in the gel during 2 days of culture. The heart explants
(HT) were present initially, then removed at the intervals indicated. Neuritic outgrowth from the ganglion (G) was measured as number of intersections between
fibres and an eye-piece grid line on the side facing (line a) or opposite (line b) the
heart explants. Each point represents the mean ± S.E.M. of eight ganglia of each type.
but less marked tendency was seen for the side facing away from the heart
(b side). The 12 h scores for the b side are thus significantly higher than for
background outgrowth (P < 0-01 for all ganglia) while significantly lower than
those for the a side (P < 0-01 for all ganglia). Nearly maximum outgrowth was
reached with the explants present during the first day of the two-day culture
period (Figs 1 and 2). These results suggest that the density of neurite outgrowth
is determined by the level of a soluble stimulant, most probably that partly
defined biochemically by Ebendal et al. (1979), and that this stimulant probably
accumulates in the gel with time in consequence of release from the heart
explants. Observations of rapid and vigorous neurite outgrowth in ganglia
inserted into gels preincubated with heart explants for one or two days support
this notion. In the case of preincubated gels the stimulus can indeed be maintained within a local conditioned volume for at least 24 h between removal of
heart and insertion of ganglia.
Intermittent observations of cultures showed that normally the first fibres
in coculture appeared after 6 h and even then projected towards the heart
Control ofneurite extension
293
Fig. 3. Outgrowth of neurites from a sympathetic ganglion after 12 h of co-culture
with heart explants (left). A number of short neurites extends into the gel from the
side of the ganglion (side a) facing the heart tissue whereas no neurites leave the
opposite side of the ganglion (side b).
Fig. 4. The same ganglion cultured for a further 12 h with the heart explants removed
{arrows) and the gel carefully washed and left with culture medium on top. The
neurites, initially extending toward the heart explants, are retracting.
Fig. 5. The ganglion after a further 24 h with new heart explants inserted (arrows)
in the original position and the fluid culture medium poured off the gel. Neurite
outgrowth is again supported towards the heart tissue (some single migrating cells,
but not neurites, can be seen on the opposite side of the ganglion). Phase contrast
of living cultures. All at same magnification.
explants. After 12 h a fairly dense outgrowth of short neurites had normally
emerged on the a side (Fig. 3). In order to test whether the outgrowth of
neurites progresses autonomously after initial excitation such gels were carefully
294
T. EBENDAL
•—•
o—o
A
D
A
D
Remak's G.
Ciliary G.
Spinal G.
Sympathetic G.
200
150
100
50
24
36 424548
0
24
36
424548
Presence of heart on side b (h)
Fig. 6. Effects on neurite outgrowth of introduction of a second set of heart explants.
Determination of fibre density and symbols as in Fig. 2.
washed with several changes of culture medium after removal of heart explants.
After the initial 12 h of stimulation a subsequent 12 h of washing led to
a marked cessation of extension and even to a retraction of the neurites (Fig. 4).
After a further 24 h of washing of the gel, outgrowth was totally abolished.
Degenerative changes were, furthermore, manifest in the neurite outgrowth
after 24 h of stimulation followed by 24 h of washing. However, the negative
effects of washing could invariably be overcome by introducing new heart
explants to the gel (Fig. 5). In this situation neurite extension was again
stimulated and outgrowths formed readily within 24 h toward the new heart
explants whether inserted in the original position or on the opposite side of the
ganglion. These facts suggest that the stimulative factor can be eliminated from
the gel by washing and is thus likely to exert its effects without forming insoluble
precipitates on the substratum. Furthermore, neurite extension appears to be
a reversible process which requires the continuous presence of the trophic
factor to persist.
On the other hand, it was also evident that without initial stimulation by
heart tissue the ability of ganglia to extend neurites rapidly diminished: the
response to heart explants inserted after a 6 h delay was thus lowered in all
ganglia and when 12 h had elapsed the ability to respond was seriously impaired
(data not shown).
Control ofneurite extension
295
b
a
0 L I I I I
50 A •
•
•
D
o
D O
A •
D O
Spinal G.
Remak's G.
Sympathetic G.
Ciliary G.
\ \ \
0 L
A
+i
nn
•
n O
A •
[] ()
Fig. 7. Effects on neurite outgrowth of transfer of initially stimulated ganglia to
control gels or gels preincubated with heart for 24 h. Outgrowths were measured
after the initial 24 h (left). Ganglia were then transferred and neurite extension
scored again after a further 24 h of control culture (centre) or of co-culture with
heart tissue (right). Ganglia taken to the gels preincubated with heart tissue were
given a reverse orientation with reference to the heart explants. Symbols as in Fig. 2.
Effects of additional heart explants
To examine the response to a second competing source of stimulation,
additional sets of heart explants were inserted on the other side of the ganglia
at various intervals (Fig. 6). This evoked a marked additional outgrowth
towards the new explants but did not interrupt outgrowth towards the original
heart explants (the values for outgrowth on the a side at 0 and 48 h do not
differ significantly for any of the ganglia, P > 0-1). The sooner the new heart
explants were introduced (i.e. the longer they were present during the 2-day
culture period) the longer and denser became the extra outgrowth of neurites
(Fig. 6). The same effects were found even if the original set of heart explants
was removed from the gel after the first 24 h.
These results demonstrate that an increased level of the trophic factor on
the back of ganglia rapidly evokes a massive extra outgrowth of neurites which
would not normally have emerged. Nevertheless it is not clear whether these
neurites belong to a separate population of neurons or are collaterals to neurites
which were already present in the original outgrowth.
296
T. EBENDAL
Fig. 8. Sympathetic ganglia co-cultured for 24 h in tandem with heart explants (left).
The proximal ganglion shows the normal asymmetrical pattern of outgrowth. This
pattern is not repeated in the distal ganglion which fails to extend fibres towards
the adjacent rear but shows some scattered neurites at both ends. The left and right
framed areas indicate positions similar to those shown in section in Figs 11 and 12,
respectively.
Fig. 9. A ganglion at a distance (about 1-8 mm) from the heart explants comparable
to that for the distal ganglion of Fig. 8 but with no shielding, proximal ganglion.
Although at this distance fibre outgrowth is less dense, it is not impaired as seen in
tandem combinations. Darkfield.
Fig. 10. Model experiment showing the absorption of methylene blue by a ganglion
in the gel. Methylene blue was allowed to spread from a square offilterpaper {arrow)
for about 4 h before photography. Staining of cells is heavy at the front and both
ends of the proximal ganglion (dark areas). It is also evident in the microscope
(but not seen well in the photograph) that the gel behind the ganglion is only lightly
stained. Brightfield. Figs 8 to 10 at same magnification.
Fig. 11. Horizontal section of a sympathetic ganglion co-cultured with heart explants
for 2 days. The side facing the heart tissue is shown (a side; see Fig. 8, left frame,
for a corresponding position). Numerous surviving neurons are visible together
with fibre bundles projecting into the gel.
Fig. 12. A corresponding section through the part of ganglion facing away from the
heart explants (b side; cf. Fig. 8, right frame). Neuronal survival is as good as in
Fig. 11 but no fibres extend into the gel. Phase contrast of toluidine-blue-stained
2/im plastic sections. Same magnification in Figs 11 and 12.
Control of neurite extension
297
Transfer of stimulated ganglia
In one series of cultures the ability of ganglia to reform fibre halos lost after
initial stimulation was studied. Ganglia responding to heart explants were thus
transferred to new gels after 24 h of co-culture (Fig. 7). Upon transfer the fibre
halos remained attached to the original gel. Prior to transfer, the outgrowth
of neurites was significantly higher towards the heart tissue (P < 0-01 in all
ganglia). During an additional 24 h period in the new gel without heart explants,
the ganglia failed to generate neurites (Fig. 7) despite prior stimulation
(a similar lack of fibre outgrowth was evident also in sympathetic and spinal
ganglia which had been incubated floating in NGF-solutions before insertion
into control gels). On the other hand, ganglia transferred to newly prepared
gels for a second exposure to heart explants regenerated fibre outgrowths of
normal appearance, preferentially directed towards the heart tissue irrespective
of whether this was positioned on the original a or b side (data not shown).
Ganglia transferred to gels preincubated with heart explants for one day, also
developed within 24 h dense outgrowths of neurites directed toward the source
of stimulation, irrespective of its position relative to the original a and b sides
(Fig. 7). However, the outgrowth away from the preincubated heart explants
was always higher than from newly inserted heart tissue. Thus, in the situation
illustrated in Fig. 7 the shift in fibre outgrowth mainly from the originally
exposed a side to the b side exposed to the preincubated heart explants after
transfer is significant for the sympathetic and Remak's ganglia (P < 0-01 and
0-05, respectively) but not wholly so for the spinal and ciliary ganglia (P ~ 0-1
and 02, respectively). These results suggest firstly that the ganglia retained no
memory of the signal which initiated and directed, neurite outgrowth in the
first place and, secondly, that the outgrowth of neurites from the back of
ganglia is higher if the stimulant has been spread in advance without being
hindered by the bulk of ganglionic tissue.
Stimulation of ganglia arranged in tandem
To test further the notion that ganglionic explants may hinder spreading
of the trophic factor, doublets of sympathetic ganglia were arranged in tandem
facing a triplet of heart explants (Fig. 8). The proximal ganglia presented
completely normal neurite outgrowths with fibres extending from the side
facing the heart and from both ends of the ganglion. In contrast, the distal
ganglia emitted only a few fibres, mainly from their protruding ends, but consistently failed to send neurites into the gel behind the proximal ganglia (Fig. 8).
The distal ganglia were positioned less than 2 mm away from the heart explants.
This is within a distance normally bridged by the spreading factor although
the density of fibres of the asymmetric outgrowth is reduced with an increase
in spacing between the co-cultured explants (Fig. 9; Ebendal, 1979).
These observations support the idea that ganglia themselves act as barriers
298
T. EBENDAL
locally shadowing the gel from the trophic factor spreading from the heart
explants. A model experiment in which methylene blue was allowed to spread
through the gel from a piece of filter paper confirmed that only cells at the
front and the ends of ganglia were heavily stained (Fig. 10) and that, furthermore, the ganglia effectively prevented dense staining of the gel immediately
behind them (not seen well in photographs as in Fig. 10). One possibility
suggested by such patterns of labelling is that selective neuronal survival in
favoured parts of the ganglia exposed to the heart tissue may account for the
asymmetry of neurite outgrowth. To test this, horizontal sections were cut
of four ganglia, drawn from a pool of eight ganglia showing preferential
outgrowth towards heart explants, and read on a blind basis. The distribution
of neurons did not match the asymmetry in outgrowth of fibres but, rather,
neuronal survival was excellent throughout the stimulated ganglia (examples
of areas are given in Figs 11 and 12). Fibre bundles run parallel to the surface
at the back of ganglia whereas in parts more exposed to the heart explants they
followed courses perpendicular to the ganglionic surface and extended out into
the gel (Figs 11 and 12). In control ganglia cultured without the heart tissue
almost no neurons survive. Hence it seems likely, as was concluded also in an
earlier study on silver-impregnated whole mounts of ganglia showing preferential outgrowths (Ebendal & Jacobson, 19776), that a direct influence on
fibre guidance causes the asymmetrical outgrowth of fibres.
DISCUSSION
The present results demonstrate the strict dependence of ganglionic neurite
extension on the presence of a factor spreading from the embryonic chick heart
explants. A heat-labile molecular species with an apparent weight of about
40000 dalton extracted from the chick heart presumably mediates this stimulation (Ebendal et al. 1979). Accumulation of the stimulus in the gel is evidently
time-dependent (Fig. 2), and once the gel is conditioned the heart explants are
no longer needed to evoke the graded neurite outgrowth responses.
Moreover, neurite outgrowth in response to the heart factor would seem
reversible since washing of the gels, which is likely to remove the factor,
resulted in a neurite retraction (Figs 3 and 4) and this retraction could be
reversed by again introducing heart explants (Fig. 5). The progressive development of neurites thus depended upon a continuous trophic support and did
not persist after initial triggering. This is analogous to the need to renew the
supply of NGF repeatedly in order to maintain ectopic sympathetic fibres
which invade the brain of newborn rats upon intracerebral injection of NGF
(Levi-Montalcini et al. 1978; Menesini Chen et al. 1978). The observation that
extra heart explants will provoke additional outgrowth of neurites on the rear
of the ganglia (Fig. 6) indicates that the stimulation of neurites is exerted
locally. This accords with the concept of local control of neurite development
Control ofneurite extension
299
by NGF presented by Campenot (1977). The fact that either the heart factor
or NGF must be present in the gels to propagate neurites in transferred ganglia
despite prior stimulation (Fig. 7) lends further support to their functional
equivalence in local control of neurite extension.
The neurite outgrowth is easily observed to be stimulated in the main
direction towards the heart explants (Figs 1 and 8). A series of definitions
relating to migratory behaviour in response to chemical stimuli were proposed
by Keller et al. (1977). These referred chiefly to polymorphonuclear leucocytes
but they should apply to the locomotory behaviour of extending neurites not
withstanding that complications arise in that the stimulants may directly
affect survival of neurons. Positive chemokinesis thus denotes an increased
rate of locomotion, in random directions, in the presence of a chemical stimulus
while positive chemotaxis signifies directional movement up a concentration
gradient of a chemical attractant. A combination of these two responses is also
feasible and the best way to clarify a specific situation was considered to be
direct microscopic observation of the behaviour of individual cells in relation
to gradients of chemical stimulants (Zigmond, 1977, 1978). A series of circumstances favours biased chemokinesis as a major mechanism causing the
asymmetrical outgrowth of neurites seen in the present co-cultures. Firstly, as
noted above (see e.g. Fig. 2) the rate of extension is obviously related to the
level of the heart factor present in the gel. Secondly, all fibres extend more or
less radially, possibly due to contact inhibition (Dunn, 1971), which means
that some fibres seek the source of stimulation whereas others, at the ends of
the ganglion, extend perpendicular to this direction (Figs 1 and 8). This pattern
hardly supports the theory that neuronal growth cones actively orient up
a gradient which should have its highest concentration close to the heart
explants. Thirdly, a number of observations (see e.g. Figs 8 and 9) show that
the concentration of the heart factor is low behind the ganglia, whether acting
as barriers or sinks, which would itself suffice to explain why neurites do not
occupy this region. In line with this view, outgrowths in ganglia only partially
inserted into the gel, and thus probably less effective as barriers, were markedly
less asymmetrical (not shown).
Chemotactic orientation cannot, however, be excluded. It may be that the
barrier effect (Fig. 10) by the ganglionic tissue itself locally sets up a gradient
of the trophic factor that is steep enough for the neurites to detect. According
to observations in leucocytes, chemotaxis is a rapid response, occurring within
15 to 30 min after stimulation and may, for example be at work during initial
stages of extension before neurites are visible outside the ganglia. Especially
after the recent demonstration of chemotactic orientation in sensory growth
cones within 9 to 21 min after application of gradients of NGF (Gundersen &
Barrett, 1979) chemotaxis combined with chemokinesis may perhaps be
postulated in directed fibre outgrowth.
The present results suggest that developing peripheral tissues have the ability
300
T. EBENDAL
locally to control neurite extension by releasing neuronal trophic factors
(Varon & Bunge, 1978). The mechanism in the observed behaviour of neurites
may be envisaged as involving binding of the heart factor, at the tips of growing
neurites, to cell surface receptors in some way exerting transmembrane control
of the mesh work of actin microfilaments (Yamada, Spooner & Wessells, 1971;
Kuczmarski & Rosenbaum, 1979) and microtubular system both of which
may be expected to be directly involved in extension/retraction of neurites
(Roisen & Murphy, 1973). Moreover, the time and distance scales (h and mm,
respectively) found effective in the present experiments are those considered
realistic by Crick (1970) for setting up gradients of morphogens in an embryo.
Finally, the control of fibre extension seen here is in general agreement with
the direct observations by Speidel (1933) of neuronal growth cones seemingly
attracted towards specific regions in the tails of living frog tadpoles.
This work was supported by the Swedish Natural Science Research Council. The cultures
were prepared by Ms Annika Jordell-Kylberg and Ms Stine Soderstrom. Thanks are due to
Ms Kerstin Ahlfors for mounting the photographs and to Ms Vibeke Nilsson for secretarial
help.
REFERENCES
R., LANDA, K. B., MANTHORPE, M. & VARON, S. (1979). Cholinergic neuronotrophic
factors: Intraocular distribution of trophic activity for ciliary neurons. Science 204,
1434-1436.
BRADSHAW, R. A. (1978). Nerve growth factor. Ann. Rev. Biochem. 47, 191-216.
CAMPENOT, R. B. (1977). Local control of neurite development by nerve growth factor.
Proc. natn. Acad. Sci. U.S.A. 74, 4516-4519.
CHAMLEY, J. H. & DOWEL, J. J. (1975). Specificity of nerve fibre 'attraction' to autonomic
effector organs in tissue culture. Expl Cell Res. 90, 1-7.
CHAMLEY, J. H., GOLLER, I. & BURNSTOCK, G. (1973). Selective growth of sympathetic
nerve fibers to explants of normally densely innervated autonomic effector organs in
tissue culture. Devi Biol. 31, 362-379.
CHARLWOOD, K. A., LAMONT, D. M. & BANKS, B. E. C. (1972). Apparent orientating effects
produced by nerve growth factor. In Nerve Growth Factor and its Antiserum (ed. E. Zaimis
& J. Knight), pp. 102-107. London: Athlone Press.
COLLINS, F. (1978). Induction of neurite outgrowth by a conditioned-medium factor bound
to the culture substratum. Proc. natn. Acad. Sci. U.S.A. 75, 5210-5213.
CRICK, F. H. C. (1970). Diffusion in embryogenesis. Nature, Lond. 225, 420-422.
DUNN, G. A. (1971). Mutual contact inhibition of extension of chick sensory nerve fibres
in vitro. J. comp. Neurol. 143, 491-508.
EBENDAL, T. (1979). Stage-dependent stimulation of neurite outgrowth exerted by nerve
growth factor and chick heart in cultured embryonic ganglia. Devi Biol. 72, 276-290.
EBENDAL, T. & JACOBSON, C.-O. (1977a). Tests of possible role of NGF in neurite outgrowth
stimulation exerted by glial cells and heart explants in culture. Brain Res. 131, 373-378.
EBENDAL, T. & JACOBSON, C.-O. (19776). Tissue explants affecting extension and orientation
of axons in cultured chick embryo ganglia. Expl Cell Res. 105, 379-387. .
EBENDAL, T., BELEW, M., JACOBSON, C.-O. & PORATH, J. (1979). Neurite outgrowth elicitep
by embryonic chick heart: Partial purification of the active factor. Neurosci. Lett. 14,
91-95.
ELSDALE, T. & BARD, J. (1972). Collagen substrata for studies on cell behavior. /. Cell Biol.
54, 626-637.
ERANKO, O. & LAHTINEN, T. (1978). Attraction of nerve fiber outgrowth from sympathetic
ganglia to heart auricles in tissue culture. Acta physiol. Scand. 103, 394-403.
ADLER,
Control ofneurite extension
301
R. W. & BARRETT, J. N. (1979). Neuronal chemotaxis: Chick dorsal-root
axons turn toward high concentrations of nerve growth factor. Science 206, 1079-1080.
HARPER, G. P. & THOENEN, H. (1980). Nerve growth factor: Biological significance,
measurement, and distribution. / . Neurochem. 34, 5-16.
HELFAND, S. L., RIOPELLE, R. J. & WESSELLS, N. K. (1978). Non-equivalence of conditioned
medium and nerve growth factor for sympathetic, parasympathetic, and sensory neurons.
Expl Cell Res. 113, 39-45.
JACOBSON, M. (1978). Developmental Neurobiology. 2nd ed., pp. 562. New York, London:
Plenum Press.
GUNDERSEN,
KELLER, H. U., WILKINSON, P. C, ABERCROMBIE, M., BECKER, E. L.,
MILLER, M. E., SCOTT RAMSEY, W. & ZIGMOND, S. H. (1977). A proposal
HIRSCH, J. G.,
for the definition
of terms related to locomotion of leucocytes and other cells. Cell. Biol. Int. Rep. 1, 391-397.
KUCZMARSKI, E. R. & ROSENBAUM, J. L. (1979). Studies on the organization and localization
of actin and myosin in neurons. /. Cell Biol. 80, 356-371.
LANDA, K. B., ADLER, R., MANTHORPE, H. & VARON, S. (1980). Cholinergic neuronotrophic
factors. III. Developmental increase of trophic activity for chick embryo ciliary ganglion
neurons in their intraocular target tissues. Devi Biol. 74, 401-408.
LETOURNEAU, P. C. (1978). Chemotactic response of nerve fiber elongation to nerve growth
factor. Devi Biol. 66, 183-196.
LEVI-MONTALCINI, R. & ANGELETTI, P. U. (1968). Nerve growth factor. Physiol. Rev. 48,
534-569.
LEVI-MONTALCINI, R. & HAMBURGER, V. (1953). A diffusible agent of mouse sarcoma,
producing hyperplasia of sympathetic ganglia and hyperneurotization of viscera in the
chick embryo. /. exp. Zool. 123, 233-288.
LEVI-MONTALCINI, R., MENESINI CHEN, M. G. & CHEN, J. S. (1978). Neurotropic effects of
the nerve growth factor in chick embryos and in neonatal rodents. In Formshaping
Movements in Neurogenesis (ed. C.-O. Jacobson & T. EbendaI),Zoon 6,201-212. Stockholm :
Almqvist & Wiksell International.
LINDSAY, R. M. & TARBIT, J. (1979). Developmentally regulated induction of neurite
from immature chick sensory neurons (DRG) by homogenates of avian or mammalian
heart, liver and brain. Neurosci. Lett. 12, 195-200.
MANTHORPE, M., SKAPER, S., ADLER, R., LANDA, K. & VARON, S. (1980). Cholinergic
neuronotrophic factors: Fractionation properties of an extract from selected chick
embryonic eye tissues. /. Neurochem. 34, 69-75.
MCLENNAN, I. S. & HENDRY, I. A. (1978). Parasympathetic neuronal survival induced by
factors from muscle. Neurosci. Lett. 10, 269-273.
MENESINI CHEN, M. G., CHEN, J. S. & LEVI-MONTALCINI, R. (1978). Sympathetic nerve
fibers ingrowth in the central nervous system of neonatal rodent upon intracerebral NGF
injections. Archo Ital. Biol. 116, 53-84.
ROISEN, F. J. & MURPHY, R. A. (1973). Neurite development in vitro. II. The role of microfilaments and microtubules in dibutyryl adenosine 3',5'-cyclic monophosphate and nerve
growth factor stimulated maturation. /. Neurobiol. 4, 397-412.
SIEGEL, S. (1956). Nonparametric Statistics for the Behavioral Sciences. New York: McGrawHill.
SPEIDEL, C. C. (1933). Studies of living nerves. II. Activities of ameboid growth cones,
sheath cells, and myelin segments, as revealed by prolonged observation of individual
nerve fibres in frog tadpoles. Am. J. Anat. 52, 1-79.
VARON, S. S. & BUNGE, R. P. (1978). Trophic mechanisms in the peripheral nervous system.
Ann. Rev. Neurosci. 1, 327-361.
YAMADA, K. M., SPOONER, B. S. & WESSELS, N. K. (1971). Ultrastructure and function of
growth cones and axons of cultured nerve cells. /. Cell Biol. 49, 614-635.
ZIGMOND, S. H. (1977). Ability of polymorphonuclear leukocytes to orient in gradients of
chemotactic factors. J. Cell Biol. 75, 606-616.
ZIGMOND, S. (1978). Chemotaxis by polymorphonuclear leukocytes. /. Cell Biol. 11, 269-287.
(Received 10 June 1980, revised 2 September 1980)
VIII b
Fragment
D
Non-R
6
Illb
Fragment
R
9
IXa
D
Non-R
4
11
R
+
Non-R
17
D
IVb
R
11
R
9
+
D
Non-R
10
IXb
R
0
Table 3 {continued) p. 311
D
Non-R
19
IVa
Table 2 {continued) p. 309
Xa
D
D
Non-R
5
+
Non-R
7
K
- -,
Va
R
6
R
10
D
Non-R
15
Xb
Vb
R
0
R
4
Volume 61, pp. 303-316.
KARLSSON, JANE: The distribution of regenerative potential in the wing disc of Drosophila
The headings to the second parts of Tables 2 and 3 (pages 309 and 311 respectively) were inadvertently omitted. These
headings are supplied below so they can be appended to the appropriate tables.
Erratum