Brain Research, 84 (1975) 279-291
279
© Elsevier ScientificPublishingCompany,Amsterdam- Printed in The Netherlands
BIOLOGICAL IMPORTANCE OF RETROGRADE AXONAL TRANSPORT OF NERVE GROWTH FACTOR IN ADRENERGIC NEURONS
U. PARAVICINI, K. STOECKEL AND H. THOENEN
Department of Pharmacology, Biocenter of the University, Basel (Switzerland) (Accepted September 25th, 1974)
SUMMARY
Previous studies have shown that nerve growth factor (NGF) produces a selective induction of tyrosine hydroxylase (TH) in peripheral adrenergic neurons and that NGF is transported retrogradely with a high selectivity from the adrenergic nerve terminals to the perikaryon. In order to investigate the biological importance of retrograde NGF transport, the following experiments have been performed: (a) effect of NGF on TH activity in superior cervical ganglia (SCG) after unilateral injection into the anterior eye chamber and the submaxillary gland; and (b) effect of systemic injection of NGF on TH activity in SCG after blockade of retrograde axonal transport by axotomy. After unilateral injection of NGF into the anterior eye chamber and submaxillary gland of both 8-10-day-old rats and adult mice, the increase in TH activity in the SCG was considerably larger on the injected than on the non-injected side although the adrenergic neurons supplying the two organs do not account for more than 25 Yo of the total number of adrenergic neurons in the SCG. A direct diffusion mechanism could be excluded by the fact that unilateral local injection of [125I]NGF produced no significant side difference in the accumulation of radioactivity in the SCG 2 h after injection whereas after 14 h there was a several-fold difference between the injected and non-injected side. Moreover, the nodose ganglia which are located very close to the SCG exhibited no statistically significant difference in the accumulation of radioactivity at any time. Forty-eight hours after subcutaneous injection of 10 mg/kg of NGF the increase in TH activity of the SCG amounted to 154 ~o on the intact side and to 92 Y/ooon the axotomized side. However, these experiments do not permit decisions about the extent that axotomy, as such, impaired the response to NGF. It is concluded that the biological effect of NGF results to a considerable extent, from the moiety which reaches the cell body by retrograde transport from the nerve terminals.
280 INTRODUCTION
Nerve growth factor (NGF) isolated from mouse submaxillary gland has profound growth-promoting properties on the peripheral sympathetic nervous system of neonatal animalslL Moreover, NGF also enhances the differentiation of the terminal adrenergic neurons. This is reflected by a selective induction of tyrosine hydroxylase (TH) and dopamine fi-hydroxylase2°, key-enzymes in the synthesis of norepinephrine, which are exclusively located in adrenergic neurons15, is. It seems then, that NGF, in addition to its key functions in the development of the peripheral sympathetic nervous system 12, is also of essential importance for the maintenance of a normal function of the peripheral sympathetic nervous system in adult animals t°,19. This can be deduced from the observation that surgical removal of the submaxillary glands in adult mice, which causes a drop in the plasma and tissue concentrations of NGF 8, is followed by a reduction in the activity of all the enzymes involved in norepinephrine synthesis in both superior cervical and stellate gangliaS0,19. After Hendry and Iversen suggested that the biological effect of NGF might depend on a retrograde axonal transport from the nerve terminals to the cell bodies s, direct evidence for the retrograde transport was provided 9. It has been shown that labeled NGF injected into the anterior eye chamber is retrogradely transported from the adrenergic nerve terminals in the iris to the corresponding cell bodies in the superior cervical ganglion. Moreover, it has been shown that this retrograde transport is relatively rapid (2.5 mm/h), that it is sensitive to colchicine 9, and that it is highly specific, i.e., other proteins of similar molecular weight are not transported to a measurable extent and minor chemical changes of the NGF molecule such as oxidation of the tryptophan moieties drastically reduce its retrograde transport 17. It was the aim of the present experiments to establish whether the growth promoting and differentiating effect of NGF depends on the moiety of this protein reaching the cell body by retrograde transport or directly by the blood stream. The aspects of retrograde NGF transport are of particular interest in view of the fact that NGF is synthesized not only in the submaxillary gland but also at other sites 10-12,19. It could be assumed that the retrograde transport of locally synthesized NGF could represent a means of transfer of 'trophic information' from the effector cells to the cell body of the innervating neuron. We wish to report that unilateral injection of NGF into the anterior eye chamber and the submaxillary gland produced a considerably larger increase in TH activity in the superior cervical ganglion of the injected than of the contralateral side. Moreover, the effect of subcutaneously injected NGF on the TH levels of the superior cervical ganglia was very markedly reduced either by transection of the postganglionic adrenergic fibers and by colchicine-mediated blockade of axonal transport. MATERIALS AND METHODS
Isolation and labeling of NGF NGF was isolated as the 2.5 S subunit from submaxillary glands of male Swiss
281 albino mice weighing 30-35 g, according to the procedure of Bocchini and Angeletti 1. The purity of the preparation was examined by means of gel electrophoresis. The biological activity was determined according to Fenton 5 in organ cultures of dorsal root ganglia of 7-9-day-old chicken embryos. The biological activity of our NGF preparations was 220-260 biological units (BU) per/,g of NGF. Labeling of N G F was performed with 1251according to Greenwood et al. 7 with modifications described in detail by Stoeckel et al. 17. A major difference between the modified and original method is the omission of metabisulfite (Na2SzOs) which has been shown to partially inactivate NGF 17. Na125I for N G F labeling was provided by EIR, Wiirenlingen, Switzerland at a specific activity of 8-15 Ci/mg iodine. The specific activity of [125I]NGF was 1.2/,Ci/#g 2.5 S NGF. Surgical procedures and injection schedules
For all the experiments we used either male albino mice of the Swiss strain weighing 30-35 g or 8-10-day-old Sprague-Dawley rats of either sex weighing 15-20 g. Several litters of rats were randomized to equal numbers of 6 and the baby rats were taken away from their mothers for operation and injection only. The adult mice and the baby rats with their mothers were kept in polystyrene cages at a constant temperature of 23 °C and were given usual lab chow diet (supplied by Nafag AG, Gossau, SG, Switzerland) and tap water ad libitum. The initial experiments were performed with adult male mice. As the experiments progressed we realized that the transection of the postganglionic fibers of the superior cervical ganglion and intraganglionic injection could be more easily accomplished in rats. Thus, the majority of the experiments were performed in 8-10-day-old rats. For all surgical procedures and for injections of NGF into the anterior eye chamber or the submaxillary gland, the animals were anesthetized with ether. [lZ5I]NGF was always injected at a concentration of 300 #g/ml of saline and at a specific activity of 1.2/~Ci/#g NGF. Native NGF was used at a concentration of 20 mg/ml saline for intraocular injection and for injection into the submaxillary gland. The same concentration (20 mg/ml saline) of cytochrome C was used for intraocular injection and for injection into the submaxillary gland. For subcutaneous injections the concentration of NGF used was 3 mg/ml. For intraocular injection the volume for both mice and rats was 2/,1; for injections into the submaxillary glands the volume was 5/~1. For subcutaneous injections the dose of NGF amounted to 10 mg/kg. For all the transscleral injections into the anterior eye chamber and the injections into the submaxillary glands a 10/~1 Hamilton glass syringe was used. Injection of colchicine and transection of the postganglionic adrenergic fibers of the superior cervical ganglion were performed under a binocular microscope (Nr. 92008, Wild AG, Heerbrugg, Switzerland). For colchicine injections we used a saturated solution (about 20 mg/ml) of colchicine (supplied by Sandoz AG, Basel, Switzerland) which was injected into the superior cervical ganglion at a volume of 0.4/,1. For this injection a 1/zl Hamilton glass syringe was connected by a polyethylene catheter
282 to an injection needle (Hamilton microliter syringe needle KF 731). Controls were injected in an identical way with the same volume of saline. For counting the radioactivity in superior cervical and nodose ganglia and for determining TH activity in superior cervical ganglia, the animals were killed by a blow on the head. The ganglia were dissected under a binocular microscope. After injecting [lZSI]NGF the radioactivity of the tissues to be studied was determined in a Packard ?-counter at a counting efficiency of 50~o. Assay of TH In both mice and rats the activity of this enzyme was determined separately in the ganglia of the left and right side. Each ganglion was homogenized in 2 0 0 / d of 0.005 M Tris buffer, pH 7.4, containing 0.1 ~ Triton X-100. The homogenates were centrifuged for 20 min at 10,000 × g in a refrigerated (4 °C) Sorvatl centrifuge. The activity of T H was determined in the supernatant portion according to the method of Levitt et al. is with modifications described in detail by Mueller et al. 16. The concentration of the substrate (L-tyrosine) was 15/zM and that of the cofactor (6,7-dimethyl5,6,7,8-tetrahydropteridine. HCl) was 720/~M. Protein concentrations were determined according to the method of Lowry et al. 14. RESULTS
Unilateral injection of NGF into the anterior eye chamber and the submaxillary gland," effect on T H activity in the superior cervical ganglion In a first attempt to get information on the relative biological importance of the N G F reaching the cell body of adrenergic neurons by either the blood stream or by
Controls tO
NGF [ ] non-injected side lira injected side
300-
cl~a
~u
t'OO-r
Rat
Mouse
Fig. 1. Effect of tmilateral injection of N G F into the anterior eye chamber and submaxillary gland on tyrosine hydroxylase (TH) activity in the superior cervical ganglion (SCG). Adult mice (20-30 g) and 8-10-day-old rats (15-20 g) were injected unilaterally into the anterior eye chamber (2/~1) and submaxillary gland (5/~1) with a solution of 20 mg/ml of N G F . Forty-eight hours later the animals were killed and the activity was determined separately in the ganglia of the left and right side. The TH activity in control rats amounted to 0.1835 nmole DOPA/h/ganglia and to 0.0656 nmole DOPA/h/ ganglia in mice. Results are expressed in percent of controls as means ± S.E.M. of groups of 5-8 animals. **, P < 0.01, as compared to both controls and non-injected side.
283 TABLE[ EFFECT OF UNILATERAL INJECTION OF CYTOCHROME C INTO THE ANTERIOR EYE CHAMBER AND SALIVARY GLAND ON T H LEVELS IN IPSI- AND CONTRALATERAL SUPERIOR CERVICAL GANGLIA
The (total) TH activity in untreated control animals was 0.0733 nmole DOPA/h/ganglion. a: in the first experiment adult mice (30-35 g) were injected with a solution of cytochrome C (20 mg/ml) unilaterally into the anterior eye chamber (2/fl) and submaxillary gland (5/~1). b: in the second experiment cytochrome C was injected as before followed immediately by the subcutaneous injection of l0 mg/kg of NGF. In each case 48 h after the injection the TH activity was determined in the ganglia of the ipsi- and contralateral side. The values given represent the means 4- S.E.M. for groups of 6-10 animals. TH activity in percent of controls Injected s i d e
Non-injected side
Controls
100 4- 6%
100 4- 6%
(a) Cytochrome C (b) Cytochrome C + NGF
119 4- 8%
118 4- 9%
146 4- 2%
144 4- 2%
retrograde axonal transport, mice and rats were injected unilaterally with high doses of N G F into the anterior eye chamber and the submaxillary gland. Forty-eight hours later the animals were killed and the T H activity was determined separately in the superior cervical ganglia of the injected and non-injected side. In both mice and rats the T H activity was significantly increased (P < 0.001) on both the injected and non-injected side as compared to untreated controls (Fig. 1). However, in both species the increase on the injected side was significantly (P < 0.001) bigger than that on the non-injected side. In rats the increase amounted to 155 % above untreated controls on the injected and to 92 % above untreated controls on the noninjected side. The corresponding values for mice were 124% and 66 %. Unilateral injection of cytochrome C into the anterior eye chamber and submaxillary gland; effect on T H levels in ipsi- and contralateral superior cervical ganglia In order to exclude the possibility that the unilateral injection as such might be responsible for the side difference in T H induction we have performed the following experiments. (a) Unilateral injection of cytochrome C into the anterior eye chamber and submaxillary gland of adult mice. Determination of T H activity in the ipsi- and contralateral ganglia 48 h after the injection. (b) Subcutaneous injection of 10 mg/kg N G F after injection of cytochrome C as described above. Determination of the T H activity in the ipsi- and contralateral ganglia 48 h later. Cytochrome C was chosen on account of its general physico-chemical properties (molecular weight, isoelectric point) which are very similar to those of N G F 17. In both experiments the T H activity exhibited no statistically significant difference between the ipsi- and contralateral side (Table I). The increase in T H activity after the injection of cytochrome C alone amounted to 18 % on the non-injected side and 19 % on the injected side as compared to nontreated controls (100 ~o) (Table I). The corresponding values after the additional sys-
284 J
J Intact side
300-
300-
Axotomy
tO
tO
r
c-
o "5 200L.. tO 0 Q-U
>
100-
T
I I---
Axotomy alone
i
Axotomy + NGF
]
I Controls Cotchicine
200-
Q-O >,U
2. ,00--r
Colchicine alone
Colchicine + NGF
Fig. 2. Effects of subcutaneously injected NGF on TH activity in the SCG after blockade of retrograde axonal transport by axotomy. Eight- to 10-day-old rats (15-20 g) were injected immediately after unilateral transection of the postganglionic fibers of the SCG with 10 mg/kg of NGF. Forty-eight hours later the animals were killed and the TH activity was determined separately in the ganglia of the intact and axotomized side. Results are expressed in percent of controls (intact ganglion of untreated animals) as means ± S.E.M. of groups of 5-8 animals. *, P < 0.025; **, P < 0.001. Fig. 3. Effect of subcutaneously injected NGF on TH activity in the SCG after intraganglionic injection of colchicine. Eight- to 10-day-old rats (15-20 g) were injected subcutaneously with 10 mg/kg of NGF immediately after unilateral administration of 0.4/~1 of a saturated colchicine solution into the ganglion. Forty-eight hours later the animals were killed and the TH activity was determined separately in intact and in colchicine-injected ganglia. Results are expressed in percent of controls (intact ganglion in the untreated animals) as means i S.E.M. of groups of 6-8 animals. *, P > 0.05; **, P < 0.025. temic injection o f N G F (10 mg/kg) were 4 4 ~ and 4 6 ~ . Thus, it can be concluded that the side difference o f T H induction does not result f r o m the unilateral injection into the anterior eye c h a m b e r and submaxillary gland as such. However, the injection produced a small but statistically significant (P < 0.05) increase in T H activity which most probably results from trans-synaptic induction in consequence o f the operation and injection stress.
Blockade of retrograde axonal transport by axotomy or colehicine; response of TH activity in superior cervical ganglia to subcutaneously injected NGF In order to further elucidate the relative importance of N G F reaching the adrenergic cell bodies by blood stream as opposed to that arriving by retrograde axonal transport, we studied the effect o f a x o t o m y on the response of the superior cervical ganglion to subcutaneously injected N G F . Immediately after unilateral transection o f the postganglionic fibers o f the superior cervical ganglion, rats were injected subcutaneously with 10 mg/kg o f N G F . Forty-eight hours later the animals were killed and the T H activity was determined separately in intact and in axotomized ganglia. A x o t o m y alone produced a decrease in T H activity o f 1 6 ~ as c o m p a r e d to the intact side. The increase in T H activity effected by N G F on the intact side a m o u n t e d to 154 ~o above control and to 95 ~o above control on the axotomized side (Fig. 2).
285
g-6 200-
I
I Controls Axotomy
= ~200-
E i0 0 (DO
I I Control k~---~ Colchieine
o o (.9(..)
~ "6 IO0-
2
14 Time (h)
2
14
Time ( h )
Fig. 4. Effect of axotomy on accumulation of radioactivity in the superior cervical ganglia after subcutaneous injection of p2~I]-labeled NGF. Eight- to 10-day-old rats (15-20 g) were injected immediately after unilateral transection of the postganglionic fibers of the SCG with 10/~Ci of [x25I]labeled NGF in 20 #1 saline. Animals were killed 2 and 14 h after injection and the radioactivity of the removed ganglia was counted in a Packard 7-counter at a counting efficiency of 50 ~. Results are expressed in counts]min]ganglion as means 4- S.E.M. of groups of 6-7 animals. Intact and axotomized ganglia differ: *, P > 0.05; **, P < 0.001. Fig. 5. Effect of intraganglionic colchicine injection on accumulation of radioactivity in the superior cervical ganglia after systemic injection of [125I]-labeledNGF. Eight- to 10-day-old rats (15-20 g) were injected with 10 #Ci of [125I]-labeled NGF immediately after unilateral application of 0.4 #1 of a saturated colchicine solution into the ganglion. The animals were killed 2 and 14 h after injection and the radioactivity of the removed ganglia was counted in a Packard ),-counter at a counting efficiency of 50~. Results are expressed in counts/min/ganglion as means 4- S.E.M. of groups of 6--7 animals. Intact and colchicine-injected ganglia differ: *, P > 0.05; **, P < 0.001.
Since a x o t o m y alone caused a decrease in T H activity it could be assumed that the response to blood-borne N G F would also be impaired. Thus, we tried to block retrograde axonal transport by another means, i.e. by injecting colchicine unilaterally into the superior cervical ganglion. A l t h o u g h this procedure had the advantage that the response o f the adrenergic perikaryon to N G F was n o t impaired by axotomy, the results obtained were hampered by the fact that colchicine alone p r o d u c e d an increase in T H activity on the injected side. Also the blockade o f retrograde transport o f N G F was not complete, as will be shown below. The injection o f colchicine alone led to an increase in T H activity o f 24 ~o (Fig. 3). This has to be taken into account, together with the incomplete blockade o f retrograde axonal transport, if one considers the relatively small difference (190 versus 220~o of controls) in the response to N G F between the colchicine-injected and non-injected side.
Effect of axotomy and intraganglionic injection of colehicine on accumulation oJ'radioactivity in superior cervical ganglia after systemic injection of [125I] NGF In order to obtain a bett~r understanding o f the results o f the previous section it seemed to be o f importance So establish whether a x o t o m y or intraganglionic colchicine injection influenced the accumulation o f blood-borne [125I]NGF in the superior cervical ganglion, and to what extent these two procedures blocked the retrograde axonal transport o f [I~SI]NGF.
286 I --~ 700o
I non-injected side
~--~ injected side ,,,.
(,..)
"6 500-
°c 300(,9
100Nodose
Ganglion
Sup. Cerv. Ganglion
Fig. 6. Comparison between the accumulation of radioactivity in the superior cervical and nodose
ganglia after unilateral injection of [125I]-labeledNGF into the anterior eye chamber and the submaxillary gland. Eight- to 10-day-oldrats (15-20 g) were injected unilaterally with 2/~1 of [I~5I]NGF (0.72/tCi) into the anterior eye chamber and 5/~1(1.5/tCi) into the submaxillary gland. Animals were killed 14 h after injection and the removed ganglia directly counted in a Packard 7-counter at a counting efficiency of 50~. Results are expressed in counts/min/ganglion as means ~ S.E.M, of groups of 8-10 animals. Ganglia oftheinjectedandnon-injected sidediffer: *,P > 0.05; **,P 0.05) from that on the control side (Figs. 4 and 5). (b) Effect of axotomy and intraganglionic colchicine injection on the retrograde accumulation of [125I]NGF in the superior cervical ganglion. Previous experiments had shown that after intraocular injection of [125I]NGF the largest difference between the radioactivity accumulated in the superior cervical ganglion of the injected and noninjected side was observed 12-18 h after the injection 9. Since this difference could be abolished both by axotomy and by prior intraocular injection of colchicine, it was concluded that this difference between injected and non-injected side represents the moiety of N G F reaching the superior cervical ganglion by retrograde axonal transport 9. Since it can be assumed that axotomy completely abolishes retrograde transport of N G F from the nerve terminals to the cell body, and since axotomy and intraganglionic injection of colchicine do not seem to interfere with the accumulation of blood-borne N G F (see previous section), the difference between the accumulation of radioactivity in axotomized and colchicine-injected ganglia 14 h after systemically
287 injected [125I]NGF can be taken as a measure of the completeness of the blockade of retrograde transport by colchicine. Fourteen hours after subcutaneous injection of 10 #Ci of [125I]NGF, the radioactivity found on the axotomized side was 40 ~ less than that on the intact side (Fig. 4). After intraganglionic colchicine injection the difference amounted to 15 ~o (Fig. 5) indicating that the blockade of retrograde transport was not complete. Higher single or repeated smaller doses of colchicine could not be given on account of the general toxic effect of this drug.
Compar&on between the accumulation of radioactivity in superior cervical and nodose ganglia after unilateral injection of [125I] NGF into the anterior eye chamber and the submaxillary gland In order to exclude the possibility that after unilateral injection of [125I]NGF into the submaxillary gland the preferential accumulation of radioactivity in the superior cervical ganglia of the homolateral side resulted from diffusion or transfer of [125I]NGF from the site of injection to the ganglion by means of transfer by lymphatic or blood vessels, we compared the accumulation of radioactivity in the superior cervical with that in the nodose ganglion. The nodose and the superior cervical ganglion are in close topographical proximity and it could be expected that the effect of direct diffusion or transfer of radioactivity by non-neuronal means would be very similar for both organs. Fig. 6 shows that in the superior cervical ganglion the radioactivity accumulated on the injected side was 250 ~o higher than on the non-injected side. In contrast, there was no difference between injected and non-injected side in the nodose ganglion, excluding the possibility that diffusion of [125I]NGF or transfer by blood and lymphatic vessels contributed materially to the difference. Moreover, it is noteworthy that the amount of radioactivity accumulated in the nodose ganglion of the non-injected side is 16~o of that accumulated in the corresponding superior cervical ganglion, although the size of the ganglia is about the same and both are composed of neuronal cell bodies, nerve fibers, and satellite cells. DISCUSSION
After it had been suggested that NGF may act as atrophic factor reaching the perikaryon of the adrenergic neuron by retrograde transport, recent experiments have provided direct evidence for the existence 9 and high selectivity of this process 17. It was the purpose of the present study to get information as to whether the biological effect of NGF on adrenergic neurons results preferentially from the moiety reaching the perikaryon by retrograde axonal transport or by the moiety reaching the cell bodies (directly) from the blood stream. The changes in the activity of TH, an enzyme selectively located in adrenergic neurons15, is and induced by NGF with high preference20, were taken as a measure of the trophic action of NGF on the adrenergic neurons. To establish the relative importance of retrograde transport and direct (blood-borne) effect of NGF two different procedures were chosen. (a) Unilateral administration of NGF to the adrenergic nerve terminals located in the anterior eye chamber and submaxillary gland, comparison between TH induction in the superior cervical ganglia of the injected and non-injected side.
288 (b) Comparison between the effect of systemically injected NGF on intact superior cervical ganglia and ganglia with retrograde axonal transport blocked or impaired by axotomy or intraganglionic injection of colchicine. Both procedures are certainly not ideal for getting absolute quantitative information on the two possible routes of NGF action. However, they allow us to decide whether the retrograde transport of NGF is of any biological importance at all, or whether the whole biological effect depends on the moiety reaching the adrenergic cell bodies by the blood stream. In this case the highly selective retrograde axonal transport of NGF would represent an interesting accidental observation without functional importance. The major disadvantage of the first procedure chosen-- unilateral injections of NGF into the anterior eye chamber and submaxillary gland - - is the fact that a considerable part of the locally injected NGF inevitably escapes into the general circulation and, consequently also reaches the adrenergic neurons of the superior cervical ganglia at the contralateral side. In spite of this disadvantage the present experiments have shown that 48 h after local (anterior eye chamber and submaxillary gland) administration of NGF the increase in TH activity on the injected side was 60-70 % higher than on the non-injected side. That this difference in TH induction is due to the moiety of NGF reaching the cell bodies of the adrenergic neurons in the superior cervical ganglion by retrograde axonal transport rather than by diffusion or nonneuronal local transfer of NGF can be deduced from the following observations. Two hours after injecting [125I]NGF into the anterior eye chamber and the submaxillary gland no difference between the radioactivity accumulated on the injected and noninjected side could be detected. However, there was a several-fold difference between the two sides 14 h after injection when the retrogradely transported moiety displayed its maximum effect. This difference could be abolished by axotomy or by preceding injection of colchicine to the subsequent site of [125I]NGF injection. That this difference between the accumulation of radioactivity on the injected and non-injected side after local injection of NGF is due to retrograde transport in the adrenergic neuron is further supported by the observation that in the nodose ganglion, which is located very close to the superior cervical ganglion, and for which very similar conditions for non-neural transfer of radioactivity would be expected, no statistically significant difference between injected and non-injected side was detectable at any time. In order to fully appreciate the importance of the 60-70~ difference in TH induction in the superior cervical ganglion between injected and non-injected side, it has to be borne in mind that the neurons of the superior cervical ganglia innervating the anterior eye chamber and the submaxillary gland do not account for more than 25 ~ of the total number of adrenergic neurons present in the superior cervical ganglion. This statement is based on the observation that after intraocular injection of [I~SI]NGF the number of heavily labeled cell bodies detected in autoradiographic pictures of the superior cervical ganglion amounts to about 7-9 ~, and those labeled after submaxillary injection to 16-18 ~ (Iversen et al., in preparation). Thus, if all the nerve terminals of the superior cervical ganglion were unilaterally exposed to similar concentrations of NGF as that achieved in the present experiments for the anterior
289 eye chamber and the submaxillary gland, a ratio of about 6:1 in place of 1.5:1 for TH induction between injected and non-injected side might be expected. Additionally it has also to be taken into account that the injection as such produces a general stress which causes a bilateral trans-synaptic induction of TH of about 15-20 ~o. The subtraction of this bilateral increase, not related to the effect of NGF, leads to a further increase in the relative difference between the injected and non-injected side to 8:1. Moreover, it has also to be considered that the effect of NGF on the TH level in the superior cervical ganglion of the non-injected side could result, at least partially, from NGF reaching the cell bodies by retrograde transport from the nerve terminals rather than from NGF directly delivered to the cell bodies by the blood stream. Unfortunately it is not possible to distinguish between the effect of NGF reaching the cell body by retrograde axonal transport and directly by the blood-borne route by comparing TH induction at different time intervals after NGF administration. This failure is due to the fact that the retrograde transport is relatively rapid (2.5 mm/ h) and TH induction is relatively slowTM. The maximal effect on TH after a single NGF dose of 10 mg/kg is not reached before 48 h after administration17a. Thus, the earliest effects on TH resulting from the NGF moiety directly delivered to the cell body are not detectable before the bulk of NGF transported retrogradely reaches the perikaryon. In a second series of experiments NGF (10 mg/kg) was administered systemically and the putative contribution of the retrogradely transported moiety to the total trophic effect was estimated by unilateral axotomy of the postganglionic fibers or by unilateral intraganglionic injection of colchicine. These procedures have the advantage that all the adrenergic nerve terminals originating from the superior cervical ganglion are reached to a more or less equal extent. However, the major disadvantage exists in the fact that axotomy as such could influence TH induction by NGF in the perikaryon2,3. The same is true for intraganglionic injection of colchicine. In the latter case the interpretation is even more complicated by the fact that the blockade of axonal transport is incomplete and that the intraganglionic injection of colchicine produces an increase in TH by impairing the orthograde transport of this enzyme synthesized in the perikaryon. It is well established that the chromatolytic changes and the corresponding biochemical reactions in the cell body occur with some delay after axotomyL In the present experiments axotomy was performed immediately before NGF administration. Thus, the functional changes of the cell body could be expected to be relatively small. This assumption is supported by the observation that the accumulation of radioactivity in the superior cervical ganglion 2 h after subcutaneous administration of [lZSI]NGF did not differ between intact and axotomized side. Thus, at least the accumulation of NGF in the ganglion independent of the retrograde transport was not impaired by axotomy. Furthermore, axotomy neither significantly reduced the level of TH in the superior cervical ganglion nor completely abolished the response to the systemic administration of 10 mg/kg of NGF. However, it cannot be decided whether this response really corresponds to the moiety of NGF reaching the cell body by the blood stream and the difference between intact and axotomized side represents the response to the moiety of NGF transferred by retrograde axonal transport.
290 The only safe conclusion to be drawn from the present experiments is the following. Blockade of retrograde axonal transport does not completely abolish the response to systemically administered NGF. Thus, although a considerable part of the trophic action of systemically administered N G F seems to result from the moiety reaching the perikaryon by retrograde transport, this is not the exclusive way of action. There also seems to exist a response to N G F which reaches the adrenergic cell body directly. Since the trophic effect of N G F appears to result from both the moiety of NGF reaching the perikaryon by retrograde transport and the moiety directly delivered by the blood stream to the cell bodies, the question arises as to whether the trophic action of N G F is effected by an interaction either with receptors at the outer surface of the neuronal membrane or with receptive structures inside the neuronal cell body. An interaction with receptive structures at the neuronal membrane could be assumed for the N G F moiety producing its effect by direct action on the cell body by a similar mechanism as described for the biological action of insulin 4. The assumption of such a mechanism of action of N G F is also supported by the recent observation of Frazier et al. that N G F linked to Sepharose beads retains its biological activity on chicken dorsal root ganglia, i.e., the NGF-induced axonal sprouting occurs also with N G F covalently linked to Sepharose 6. However, for the TH induction resulting from the N G F moiety reaching the perikaryon by retrograde transport such a mechanism of action does not seem to be very likely, since one has to assume that the N G F molecules reaching the cell body by retrograde transport had first to be transferred to the outer surface of the neuronal membrane before they can develop their biological action. Preliminary electron microscopic autoradiographic experiments did not provide any indication that N G F is transferred to the outer surface of the neuronal membrane. The biological significance of retrogradely transported N G F becomes of additional importance if one assumes that the biological function of N G F does not result from the part of N G F delivered from the site of production into the circulation but from that moiety which reaches the nerve terminals directly from the site of production in the effector organ. The production of N G F by the effector organ and the highly selective retrograde transport of N G F from the nerve terminals to the cell body would provide the possibility of transfer of (trophic) information from the effector organ to the cell bodies of the innervating neurons. Such a mechanism could be thought to determine the density of innervation of an organ or at least regulate the outgrowth of nerve fibers as a prerequisite for specific contacts between nerve terminals and effector cells. ACKNOWLEDGEMENT
This work was supported by the Swiss National Foundation for Scientific Research (Grant No. 365371).
REFERENCES 1 BOCCHINI,V., AND ANGELETTI, P. U., The nerve growth factor: purification as a 30,000molecular weight protein, Proc. nat. Acad. Sci. (Wash.), 64 (1969) 787-794.
291 2 BRIMIJOIN,S., AND MOLINOFF,P. B., Effects of 6-hydroxydopamine on the activity of tyrosine hydroxylase and dopamine fl-hydroxylase in sympathetic ganglia of the rat, J. Pharmacol. exp. TheE., 178 (1971) 417-424. 3 CRAGG,B. C., What is the signal for chromatolysis? Brain Research, 23 (1970) 1-21. 4 CUATRECASAS,P., Properties of the insulin receptor isolated from liver and fat cell membranes, J. biol. Chem., 247 (1972) 1980-1991. 5 FErCrON,E. L., Tissue culture assay of nerve growth factor and of the specific antiserum, Exp. Cell Res., 59 (1970) 383-392. 6 FRAZIER,W. A., BOYD,L. F., AND BRADSHAW,R. A., Interaction of nerve growth factor with surface membranes: biological competence of insolubilized nerve growth factor, Proc. nat. Acad. Sci. (Wash.), 70 (1973) 2931-2935. 7 GREENWOOD,F. C., HUNTER,W. M., AND GLOVER,J. S., The preparation of lalI-labelled human growth hormone of high specific radioactivity, Biochem. J., 89 (1963) 114-123. 8 HENDRY,I. A., AND IVERSEN,L. L., Changes in tissue and plasma concentrations of nerve growth factor following removal of the submaxillary glands in adult mice and their effects on the sympathetic nervous system, Nature (Lond.), 243 (1973) 500-504. 9 HENDRY,I. A., STOECKEL,K., THOENEN,H., AND IVERSEN,L. L., Retrograde axonal transport of the nerve growth factor, Brain Research, 68 (1974) 103-121. 10 HENDRY,I. A., AND THOENEN,H., Changes of enzyme pattern in the sympathetic nervous system of adult mice after submaxillary gland removal, response to exogenous nerve growth factor, J. Neurochem., 22 (1974) 999-1004. 11 JOHNSON,D. G., SILBERSTEIN,S. D., MANBAUER,I., AND KOPIN, I. J., The role of nerve growth factor in the ramification of sympathetic nerve fibers into rat iris in organ culture, J. Neurochem., 19 (1972) 2025-2029. 12 LEvI-MONTALCINI,R., AND ANGELETTI,P. U., Nerve growth factor, Physiol. Rev., 48 (1969) 534569. 13 LEVITT,M., GraB, J. W., DALY, J. W., LIPTON, M., AND UDENFRIEND,S., A new class of tyrosine hydroxylase inhibitors and a simple assay of inhibition in vivo, Biochem. Pharmacol., 16 (1967) 1313-1321. 14 LOWRY,O. H., ROSEBROUGH,N. J., FARR, A. L., AND RANDALL, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 15 MOLINOFE,P. B., ANDAXELROD,J., Biochemistry of catecholamines, Ann. Rev. Biochem., 40 (1971) 465-500. 16 MUELLER,R. A., THOEN~N,H., AND AXELROD,J., Increase in tyrosine hydroxylase activity after reserpine administration, J. Pharmacol. exp. Ther., 169 (1969) 74-79. 17 STOECKEL,K., PARAVICINI,U., AND THOENEN,H., Specificity of the retrograde axonal transport of nerve growth factor, Brain Research, 76 (1974) 413-421. 17a STOECgEL, K., SOLOMON,F., PARAVIONI, U., AND THOENEN, H., Dissociation between effects of nerve growth factor on tyrosine hydroxylase and tubulin synthesis in sympathetic ganglia, Nature (Lond.), 250 (1974) 150-151. 18 THOErqEN,H., Neuronally mediated enzyme induction in adrenergic neurons and adrenal chromaffin cells, Biochem. Soc. Symp., 36 (1972) 3-15. 19 THOENEN,H., HENDRY, I. A., STOECKEL,K., PARAVICINI,U., ANDOESCH,F., Regulation of enzyme synthesis by neuronal activity and nerve growth factor. In K. FUXE, L. OLSONAND Y. ZOTTERMAN (Eds.), Dynamics of Degeneration and Growth in Neurons, Pergamon Press, Oxford, 1974, in press. 20 THOENEN,H., ANGELETTI,P. U., LEvI-MONTALC1NI,R., AND KETTLER,R., Selective induction by nerve growth factor of tyrosine hydroxylase and dopamine fl-hydroxylase in the rat superior cervical ganglia, Proc. nat. A cad. Sci. (Wash.), 68 (1971) 1598-1602.
279
© Elsevier ScientificPublishingCompany,Amsterdam- Printed in The Netherlands
BIOLOGICAL IMPORTANCE OF RETROGRADE AXONAL TRANSPORT OF NERVE GROWTH FACTOR IN ADRENERGIC NEURONS
U. PARAVICINI, K. STOECKEL AND H. THOENEN
Department of Pharmacology, Biocenter of the University, Basel (Switzerland) (Accepted September 25th, 1974)
SUMMARY
Previous studies have shown that nerve growth factor (NGF) produces a selective induction of tyrosine hydroxylase (TH) in peripheral adrenergic neurons and that NGF is transported retrogradely with a high selectivity from the adrenergic nerve terminals to the perikaryon. In order to investigate the biological importance of retrograde NGF transport, the following experiments have been performed: (a) effect of NGF on TH activity in superior cervical ganglia (SCG) after unilateral injection into the anterior eye chamber and the submaxillary gland; and (b) effect of systemic injection of NGF on TH activity in SCG after blockade of retrograde axonal transport by axotomy. After unilateral injection of NGF into the anterior eye chamber and submaxillary gland of both 8-10-day-old rats and adult mice, the increase in TH activity in the SCG was considerably larger on the injected than on the non-injected side although the adrenergic neurons supplying the two organs do not account for more than 25 Yo of the total number of adrenergic neurons in the SCG. A direct diffusion mechanism could be excluded by the fact that unilateral local injection of [125I]NGF produced no significant side difference in the accumulation of radioactivity in the SCG 2 h after injection whereas after 14 h there was a several-fold difference between the injected and non-injected side. Moreover, the nodose ganglia which are located very close to the SCG exhibited no statistically significant difference in the accumulation of radioactivity at any time. Forty-eight hours after subcutaneous injection of 10 mg/kg of NGF the increase in TH activity of the SCG amounted to 154 ~o on the intact side and to 92 Y/ooon the axotomized side. However, these experiments do not permit decisions about the extent that axotomy, as such, impaired the response to NGF. It is concluded that the biological effect of NGF results to a considerable extent, from the moiety which reaches the cell body by retrograde transport from the nerve terminals.
280 INTRODUCTION
Nerve growth factor (NGF) isolated from mouse submaxillary gland has profound growth-promoting properties on the peripheral sympathetic nervous system of neonatal animalslL Moreover, NGF also enhances the differentiation of the terminal adrenergic neurons. This is reflected by a selective induction of tyrosine hydroxylase (TH) and dopamine fi-hydroxylase2°, key-enzymes in the synthesis of norepinephrine, which are exclusively located in adrenergic neurons15, is. It seems then, that NGF, in addition to its key functions in the development of the peripheral sympathetic nervous system 12, is also of essential importance for the maintenance of a normal function of the peripheral sympathetic nervous system in adult animals t°,19. This can be deduced from the observation that surgical removal of the submaxillary glands in adult mice, which causes a drop in the plasma and tissue concentrations of NGF 8, is followed by a reduction in the activity of all the enzymes involved in norepinephrine synthesis in both superior cervical and stellate gangliaS0,19. After Hendry and Iversen suggested that the biological effect of NGF might depend on a retrograde axonal transport from the nerve terminals to the cell bodies s, direct evidence for the retrograde transport was provided 9. It has been shown that labeled NGF injected into the anterior eye chamber is retrogradely transported from the adrenergic nerve terminals in the iris to the corresponding cell bodies in the superior cervical ganglion. Moreover, it has been shown that this retrograde transport is relatively rapid (2.5 mm/h), that it is sensitive to colchicine 9, and that it is highly specific, i.e., other proteins of similar molecular weight are not transported to a measurable extent and minor chemical changes of the NGF molecule such as oxidation of the tryptophan moieties drastically reduce its retrograde transport 17. It was the aim of the present experiments to establish whether the growth promoting and differentiating effect of NGF depends on the moiety of this protein reaching the cell body by retrograde transport or directly by the blood stream. The aspects of retrograde NGF transport are of particular interest in view of the fact that NGF is synthesized not only in the submaxillary gland but also at other sites 10-12,19. It could be assumed that the retrograde transport of locally synthesized NGF could represent a means of transfer of 'trophic information' from the effector cells to the cell body of the innervating neuron. We wish to report that unilateral injection of NGF into the anterior eye chamber and the submaxillary gland produced a considerably larger increase in TH activity in the superior cervical ganglion of the injected than of the contralateral side. Moreover, the effect of subcutaneously injected NGF on the TH levels of the superior cervical ganglia was very markedly reduced either by transection of the postganglionic adrenergic fibers and by colchicine-mediated blockade of axonal transport. MATERIALS AND METHODS
Isolation and labeling of NGF NGF was isolated as the 2.5 S subunit from submaxillary glands of male Swiss
281 albino mice weighing 30-35 g, according to the procedure of Bocchini and Angeletti 1. The purity of the preparation was examined by means of gel electrophoresis. The biological activity was determined according to Fenton 5 in organ cultures of dorsal root ganglia of 7-9-day-old chicken embryos. The biological activity of our NGF preparations was 220-260 biological units (BU) per/,g of NGF. Labeling of N G F was performed with 1251according to Greenwood et al. 7 with modifications described in detail by Stoeckel et al. 17. A major difference between the modified and original method is the omission of metabisulfite (Na2SzOs) which has been shown to partially inactivate NGF 17. Na125I for N G F labeling was provided by EIR, Wiirenlingen, Switzerland at a specific activity of 8-15 Ci/mg iodine. The specific activity of [125I]NGF was 1.2/,Ci/#g 2.5 S NGF. Surgical procedures and injection schedules
For all the experiments we used either male albino mice of the Swiss strain weighing 30-35 g or 8-10-day-old Sprague-Dawley rats of either sex weighing 15-20 g. Several litters of rats were randomized to equal numbers of 6 and the baby rats were taken away from their mothers for operation and injection only. The adult mice and the baby rats with their mothers were kept in polystyrene cages at a constant temperature of 23 °C and were given usual lab chow diet (supplied by Nafag AG, Gossau, SG, Switzerland) and tap water ad libitum. The initial experiments were performed with adult male mice. As the experiments progressed we realized that the transection of the postganglionic fibers of the superior cervical ganglion and intraganglionic injection could be more easily accomplished in rats. Thus, the majority of the experiments were performed in 8-10-day-old rats. For all surgical procedures and for injections of NGF into the anterior eye chamber or the submaxillary gland, the animals were anesthetized with ether. [lZ5I]NGF was always injected at a concentration of 300 #g/ml of saline and at a specific activity of 1.2/~Ci/#g NGF. Native NGF was used at a concentration of 20 mg/ml saline for intraocular injection and for injection into the submaxillary gland. The same concentration (20 mg/ml saline) of cytochrome C was used for intraocular injection and for injection into the submaxillary gland. For subcutaneous injections the concentration of NGF used was 3 mg/ml. For intraocular injection the volume for both mice and rats was 2/,1; for injections into the submaxillary glands the volume was 5/~1. For subcutaneous injections the dose of NGF amounted to 10 mg/kg. For all the transscleral injections into the anterior eye chamber and the injections into the submaxillary glands a 10/~1 Hamilton glass syringe was used. Injection of colchicine and transection of the postganglionic adrenergic fibers of the superior cervical ganglion were performed under a binocular microscope (Nr. 92008, Wild AG, Heerbrugg, Switzerland). For colchicine injections we used a saturated solution (about 20 mg/ml) of colchicine (supplied by Sandoz AG, Basel, Switzerland) which was injected into the superior cervical ganglion at a volume of 0.4/,1. For this injection a 1/zl Hamilton glass syringe was connected by a polyethylene catheter
282 to an injection needle (Hamilton microliter syringe needle KF 731). Controls were injected in an identical way with the same volume of saline. For counting the radioactivity in superior cervical and nodose ganglia and for determining TH activity in superior cervical ganglia, the animals were killed by a blow on the head. The ganglia were dissected under a binocular microscope. After injecting [lZSI]NGF the radioactivity of the tissues to be studied was determined in a Packard ?-counter at a counting efficiency of 50~o. Assay of TH In both mice and rats the activity of this enzyme was determined separately in the ganglia of the left and right side. Each ganglion was homogenized in 2 0 0 / d of 0.005 M Tris buffer, pH 7.4, containing 0.1 ~ Triton X-100. The homogenates were centrifuged for 20 min at 10,000 × g in a refrigerated (4 °C) Sorvatl centrifuge. The activity of T H was determined in the supernatant portion according to the method of Levitt et al. is with modifications described in detail by Mueller et al. 16. The concentration of the substrate (L-tyrosine) was 15/zM and that of the cofactor (6,7-dimethyl5,6,7,8-tetrahydropteridine. HCl) was 720/~M. Protein concentrations were determined according to the method of Lowry et al. 14. RESULTS
Unilateral injection of NGF into the anterior eye chamber and the submaxillary gland," effect on T H activity in the superior cervical ganglion In a first attempt to get information on the relative biological importance of the N G F reaching the cell body of adrenergic neurons by either the blood stream or by
Controls tO
NGF [ ] non-injected side lira injected side
300-
cl~a
~u
t'OO-r
Rat
Mouse
Fig. 1. Effect of tmilateral injection of N G F into the anterior eye chamber and submaxillary gland on tyrosine hydroxylase (TH) activity in the superior cervical ganglion (SCG). Adult mice (20-30 g) and 8-10-day-old rats (15-20 g) were injected unilaterally into the anterior eye chamber (2/~1) and submaxillary gland (5/~1) with a solution of 20 mg/ml of N G F . Forty-eight hours later the animals were killed and the activity was determined separately in the ganglia of the left and right side. The TH activity in control rats amounted to 0.1835 nmole DOPA/h/ganglia and to 0.0656 nmole DOPA/h/ ganglia in mice. Results are expressed in percent of controls as means ± S.E.M. of groups of 5-8 animals. **, P < 0.01, as compared to both controls and non-injected side.
283 TABLE[ EFFECT OF UNILATERAL INJECTION OF CYTOCHROME C INTO THE ANTERIOR EYE CHAMBER AND SALIVARY GLAND ON T H LEVELS IN IPSI- AND CONTRALATERAL SUPERIOR CERVICAL GANGLIA
The (total) TH activity in untreated control animals was 0.0733 nmole DOPA/h/ganglion. a: in the first experiment adult mice (30-35 g) were injected with a solution of cytochrome C (20 mg/ml) unilaterally into the anterior eye chamber (2/fl) and submaxillary gland (5/~1). b: in the second experiment cytochrome C was injected as before followed immediately by the subcutaneous injection of l0 mg/kg of NGF. In each case 48 h after the injection the TH activity was determined in the ganglia of the ipsi- and contralateral side. The values given represent the means 4- S.E.M. for groups of 6-10 animals. TH activity in percent of controls Injected s i d e
Non-injected side
Controls
100 4- 6%
100 4- 6%
(a) Cytochrome C (b) Cytochrome C + NGF
119 4- 8%
118 4- 9%
146 4- 2%
144 4- 2%
retrograde axonal transport, mice and rats were injected unilaterally with high doses of N G F into the anterior eye chamber and the submaxillary gland. Forty-eight hours later the animals were killed and the T H activity was determined separately in the superior cervical ganglia of the injected and non-injected side. In both mice and rats the T H activity was significantly increased (P < 0.001) on both the injected and non-injected side as compared to untreated controls (Fig. 1). However, in both species the increase on the injected side was significantly (P < 0.001) bigger than that on the non-injected side. In rats the increase amounted to 155 % above untreated controls on the injected and to 92 % above untreated controls on the noninjected side. The corresponding values for mice were 124% and 66 %. Unilateral injection of cytochrome C into the anterior eye chamber and submaxillary gland; effect on T H levels in ipsi- and contralateral superior cervical ganglia In order to exclude the possibility that the unilateral injection as such might be responsible for the side difference in T H induction we have performed the following experiments. (a) Unilateral injection of cytochrome C into the anterior eye chamber and submaxillary gland of adult mice. Determination of T H activity in the ipsi- and contralateral ganglia 48 h after the injection. (b) Subcutaneous injection of 10 mg/kg N G F after injection of cytochrome C as described above. Determination of the T H activity in the ipsi- and contralateral ganglia 48 h later. Cytochrome C was chosen on account of its general physico-chemical properties (molecular weight, isoelectric point) which are very similar to those of N G F 17. In both experiments the T H activity exhibited no statistically significant difference between the ipsi- and contralateral side (Table I). The increase in T H activity after the injection of cytochrome C alone amounted to 18 % on the non-injected side and 19 % on the injected side as compared to nontreated controls (100 ~o) (Table I). The corresponding values after the additional sys-
284 J
J Intact side
300-
300-
Axotomy
tO
tO
r
c-
o "5 200L.. tO 0 Q-U
>
100-
T
I I---
Axotomy alone
i
Axotomy + NGF
]
I Controls Cotchicine
200-
Q-O >,U
2. ,00--r
Colchicine alone
Colchicine + NGF
Fig. 2. Effects of subcutaneously injected NGF on TH activity in the SCG after blockade of retrograde axonal transport by axotomy. Eight- to 10-day-old rats (15-20 g) were injected immediately after unilateral transection of the postganglionic fibers of the SCG with 10 mg/kg of NGF. Forty-eight hours later the animals were killed and the TH activity was determined separately in the ganglia of the intact and axotomized side. Results are expressed in percent of controls (intact ganglion of untreated animals) as means ± S.E.M. of groups of 5-8 animals. *, P < 0.025; **, P < 0.001. Fig. 3. Effect of subcutaneously injected NGF on TH activity in the SCG after intraganglionic injection of colchicine. Eight- to 10-day-old rats (15-20 g) were injected subcutaneously with 10 mg/kg of NGF immediately after unilateral administration of 0.4/~1 of a saturated colchicine solution into the ganglion. Forty-eight hours later the animals were killed and the TH activity was determined separately in intact and in colchicine-injected ganglia. Results are expressed in percent of controls (intact ganglion in the untreated animals) as means i S.E.M. of groups of 6-8 animals. *, P > 0.05; **, P < 0.025. temic injection o f N G F (10 mg/kg) were 4 4 ~ and 4 6 ~ . Thus, it can be concluded that the side difference o f T H induction does not result f r o m the unilateral injection into the anterior eye c h a m b e r and submaxillary gland as such. However, the injection produced a small but statistically significant (P < 0.05) increase in T H activity which most probably results from trans-synaptic induction in consequence o f the operation and injection stress.
Blockade of retrograde axonal transport by axotomy or colehicine; response of TH activity in superior cervical ganglia to subcutaneously injected NGF In order to further elucidate the relative importance of N G F reaching the adrenergic cell bodies by blood stream as opposed to that arriving by retrograde axonal transport, we studied the effect o f a x o t o m y on the response of the superior cervical ganglion to subcutaneously injected N G F . Immediately after unilateral transection o f the postganglionic fibers o f the superior cervical ganglion, rats were injected subcutaneously with 10 mg/kg o f N G F . Forty-eight hours later the animals were killed and the T H activity was determined separately in intact and in axotomized ganglia. A x o t o m y alone produced a decrease in T H activity o f 1 6 ~ as c o m p a r e d to the intact side. The increase in T H activity effected by N G F on the intact side a m o u n t e d to 154 ~o above control and to 95 ~o above control on the axotomized side (Fig. 2).
285
g-6 200-
I
I Controls Axotomy
= ~200-
E i0 0 (DO
I I Control k~---~ Colchieine
o o (.9(..)
~ "6 IO0-
2
14 Time (h)
2
14
Time ( h )
Fig. 4. Effect of axotomy on accumulation of radioactivity in the superior cervical ganglia after subcutaneous injection of p2~I]-labeled NGF. Eight- to 10-day-old rats (15-20 g) were injected immediately after unilateral transection of the postganglionic fibers of the SCG with 10/~Ci of [x25I]labeled NGF in 20 #1 saline. Animals were killed 2 and 14 h after injection and the radioactivity of the removed ganglia was counted in a Packard 7-counter at a counting efficiency of 50 ~. Results are expressed in counts]min]ganglion as means 4- S.E.M. of groups of 6-7 animals. Intact and axotomized ganglia differ: *, P > 0.05; **, P < 0.001. Fig. 5. Effect of intraganglionic colchicine injection on accumulation of radioactivity in the superior cervical ganglia after systemic injection of [125I]-labeledNGF. Eight- to 10-day-old rats (15-20 g) were injected with 10 #Ci of [125I]-labeled NGF immediately after unilateral application of 0.4 #1 of a saturated colchicine solution into the ganglion. The animals were killed 2 and 14 h after injection and the radioactivity of the removed ganglia was counted in a Packard ),-counter at a counting efficiency of 50~. Results are expressed in counts/min/ganglion as means 4- S.E.M. of groups of 6--7 animals. Intact and colchicine-injected ganglia differ: *, P > 0.05; **, P < 0.001.
Since a x o t o m y alone caused a decrease in T H activity it could be assumed that the response to blood-borne N G F would also be impaired. Thus, we tried to block retrograde axonal transport by another means, i.e. by injecting colchicine unilaterally into the superior cervical ganglion. A l t h o u g h this procedure had the advantage that the response o f the adrenergic perikaryon to N G F was n o t impaired by axotomy, the results obtained were hampered by the fact that colchicine alone p r o d u c e d an increase in T H activity on the injected side. Also the blockade o f retrograde transport o f N G F was not complete, as will be shown below. The injection o f colchicine alone led to an increase in T H activity o f 24 ~o (Fig. 3). This has to be taken into account, together with the incomplete blockade o f retrograde axonal transport, if one considers the relatively small difference (190 versus 220~o of controls) in the response to N G F between the colchicine-injected and non-injected side.
Effect of axotomy and intraganglionic injection of colehicine on accumulation oJ'radioactivity in superior cervical ganglia after systemic injection of [125I] NGF In order to obtain a bett~r understanding o f the results o f the previous section it seemed to be o f importance So establish whether a x o t o m y or intraganglionic colchicine injection influenced the accumulation o f blood-borne [125I]NGF in the superior cervical ganglion, and to what extent these two procedures blocked the retrograde axonal transport o f [I~SI]NGF.
286 I --~ 700o
I non-injected side
~--~ injected side ,,,.
(,..)
"6 500-
°c 300(,9
100Nodose
Ganglion
Sup. Cerv. Ganglion
Fig. 6. Comparison between the accumulation of radioactivity in the superior cervical and nodose
ganglia after unilateral injection of [125I]-labeledNGF into the anterior eye chamber and the submaxillary gland. Eight- to 10-day-oldrats (15-20 g) were injected unilaterally with 2/~1 of [I~5I]NGF (0.72/tCi) into the anterior eye chamber and 5/~1(1.5/tCi) into the submaxillary gland. Animals were killed 14 h after injection and the removed ganglia directly counted in a Packard 7-counter at a counting efficiency of 50~. Results are expressed in counts/min/ganglion as means ~ S.E.M, of groups of 8-10 animals. Ganglia oftheinjectedandnon-injected sidediffer: *,P > 0.05; **,P 0.05) from that on the control side (Figs. 4 and 5). (b) Effect of axotomy and intraganglionic colchicine injection on the retrograde accumulation of [125I]NGF in the superior cervical ganglion. Previous experiments had shown that after intraocular injection of [125I]NGF the largest difference between the radioactivity accumulated in the superior cervical ganglion of the injected and noninjected side was observed 12-18 h after the injection 9. Since this difference could be abolished both by axotomy and by prior intraocular injection of colchicine, it was concluded that this difference between injected and non-injected side represents the moiety of N G F reaching the superior cervical ganglion by retrograde axonal transport 9. Since it can be assumed that axotomy completely abolishes retrograde transport of N G F from the nerve terminals to the cell body, and since axotomy and intraganglionic injection of colchicine do not seem to interfere with the accumulation of blood-borne N G F (see previous section), the difference between the accumulation of radioactivity in axotomized and colchicine-injected ganglia 14 h after systemically
287 injected [125I]NGF can be taken as a measure of the completeness of the blockade of retrograde transport by colchicine. Fourteen hours after subcutaneous injection of 10 #Ci of [125I]NGF, the radioactivity found on the axotomized side was 40 ~ less than that on the intact side (Fig. 4). After intraganglionic colchicine injection the difference amounted to 15 ~o (Fig. 5) indicating that the blockade of retrograde transport was not complete. Higher single or repeated smaller doses of colchicine could not be given on account of the general toxic effect of this drug.
Compar&on between the accumulation of radioactivity in superior cervical and nodose ganglia after unilateral injection of [125I] NGF into the anterior eye chamber and the submaxillary gland In order to exclude the possibility that after unilateral injection of [125I]NGF into the submaxillary gland the preferential accumulation of radioactivity in the superior cervical ganglia of the homolateral side resulted from diffusion or transfer of [125I]NGF from the site of injection to the ganglion by means of transfer by lymphatic or blood vessels, we compared the accumulation of radioactivity in the superior cervical with that in the nodose ganglion. The nodose and the superior cervical ganglion are in close topographical proximity and it could be expected that the effect of direct diffusion or transfer of radioactivity by non-neuronal means would be very similar for both organs. Fig. 6 shows that in the superior cervical ganglion the radioactivity accumulated on the injected side was 250 ~o higher than on the non-injected side. In contrast, there was no difference between injected and non-injected side in the nodose ganglion, excluding the possibility that diffusion of [125I]NGF or transfer by blood and lymphatic vessels contributed materially to the difference. Moreover, it is noteworthy that the amount of radioactivity accumulated in the nodose ganglion of the non-injected side is 16~o of that accumulated in the corresponding superior cervical ganglion, although the size of the ganglia is about the same and both are composed of neuronal cell bodies, nerve fibers, and satellite cells. DISCUSSION
After it had been suggested that NGF may act as atrophic factor reaching the perikaryon of the adrenergic neuron by retrograde transport, recent experiments have provided direct evidence for the existence 9 and high selectivity of this process 17. It was the purpose of the present study to get information as to whether the biological effect of NGF on adrenergic neurons results preferentially from the moiety reaching the perikaryon by retrograde axonal transport or by the moiety reaching the cell bodies (directly) from the blood stream. The changes in the activity of TH, an enzyme selectively located in adrenergic neurons15, is and induced by NGF with high preference20, were taken as a measure of the trophic action of NGF on the adrenergic neurons. To establish the relative importance of retrograde transport and direct (blood-borne) effect of NGF two different procedures were chosen. (a) Unilateral administration of NGF to the adrenergic nerve terminals located in the anterior eye chamber and submaxillary gland, comparison between TH induction in the superior cervical ganglia of the injected and non-injected side.
288 (b) Comparison between the effect of systemically injected NGF on intact superior cervical ganglia and ganglia with retrograde axonal transport blocked or impaired by axotomy or intraganglionic injection of colchicine. Both procedures are certainly not ideal for getting absolute quantitative information on the two possible routes of NGF action. However, they allow us to decide whether the retrograde transport of NGF is of any biological importance at all, or whether the whole biological effect depends on the moiety reaching the adrenergic cell bodies by the blood stream. In this case the highly selective retrograde axonal transport of NGF would represent an interesting accidental observation without functional importance. The major disadvantage of the first procedure chosen-- unilateral injections of NGF into the anterior eye chamber and submaxillary gland - - is the fact that a considerable part of the locally injected NGF inevitably escapes into the general circulation and, consequently also reaches the adrenergic neurons of the superior cervical ganglia at the contralateral side. In spite of this disadvantage the present experiments have shown that 48 h after local (anterior eye chamber and submaxillary gland) administration of NGF the increase in TH activity on the injected side was 60-70 % higher than on the non-injected side. That this difference in TH induction is due to the moiety of NGF reaching the cell bodies of the adrenergic neurons in the superior cervical ganglion by retrograde axonal transport rather than by diffusion or nonneuronal local transfer of NGF can be deduced from the following observations. Two hours after injecting [125I]NGF into the anterior eye chamber and the submaxillary gland no difference between the radioactivity accumulated on the injected and noninjected side could be detected. However, there was a several-fold difference between the two sides 14 h after injection when the retrogradely transported moiety displayed its maximum effect. This difference could be abolished by axotomy or by preceding injection of colchicine to the subsequent site of [125I]NGF injection. That this difference between the accumulation of radioactivity on the injected and non-injected side after local injection of NGF is due to retrograde transport in the adrenergic neuron is further supported by the observation that in the nodose ganglion, which is located very close to the superior cervical ganglion, and for which very similar conditions for non-neural transfer of radioactivity would be expected, no statistically significant difference between injected and non-injected side was detectable at any time. In order to fully appreciate the importance of the 60-70~ difference in TH induction in the superior cervical ganglion between injected and non-injected side, it has to be borne in mind that the neurons of the superior cervical ganglia innervating the anterior eye chamber and the submaxillary gland do not account for more than 25 ~ of the total number of adrenergic neurons present in the superior cervical ganglion. This statement is based on the observation that after intraocular injection of [I~SI]NGF the number of heavily labeled cell bodies detected in autoradiographic pictures of the superior cervical ganglion amounts to about 7-9 ~, and those labeled after submaxillary injection to 16-18 ~ (Iversen et al., in preparation). Thus, if all the nerve terminals of the superior cervical ganglion were unilaterally exposed to similar concentrations of NGF as that achieved in the present experiments for the anterior
289 eye chamber and the submaxillary gland, a ratio of about 6:1 in place of 1.5:1 for TH induction between injected and non-injected side might be expected. Additionally it has also to be taken into account that the injection as such produces a general stress which causes a bilateral trans-synaptic induction of TH of about 15-20 ~o. The subtraction of this bilateral increase, not related to the effect of NGF, leads to a further increase in the relative difference between the injected and non-injected side to 8:1. Moreover, it has also to be considered that the effect of NGF on the TH level in the superior cervical ganglion of the non-injected side could result, at least partially, from NGF reaching the cell bodies by retrograde transport from the nerve terminals rather than from NGF directly delivered to the cell bodies by the blood stream. Unfortunately it is not possible to distinguish between the effect of NGF reaching the cell body by retrograde axonal transport and directly by the blood-borne route by comparing TH induction at different time intervals after NGF administration. This failure is due to the fact that the retrograde transport is relatively rapid (2.5 mm/ h) and TH induction is relatively slowTM. The maximal effect on TH after a single NGF dose of 10 mg/kg is not reached before 48 h after administration17a. Thus, the earliest effects on TH resulting from the NGF moiety directly delivered to the cell body are not detectable before the bulk of NGF transported retrogradely reaches the perikaryon. In a second series of experiments NGF (10 mg/kg) was administered systemically and the putative contribution of the retrogradely transported moiety to the total trophic effect was estimated by unilateral axotomy of the postganglionic fibers or by unilateral intraganglionic injection of colchicine. These procedures have the advantage that all the adrenergic nerve terminals originating from the superior cervical ganglion are reached to a more or less equal extent. However, the major disadvantage exists in the fact that axotomy as such could influence TH induction by NGF in the perikaryon2,3. The same is true for intraganglionic injection of colchicine. In the latter case the interpretation is even more complicated by the fact that the blockade of axonal transport is incomplete and that the intraganglionic injection of colchicine produces an increase in TH by impairing the orthograde transport of this enzyme synthesized in the perikaryon. It is well established that the chromatolytic changes and the corresponding biochemical reactions in the cell body occur with some delay after axotomyL In the present experiments axotomy was performed immediately before NGF administration. Thus, the functional changes of the cell body could be expected to be relatively small. This assumption is supported by the observation that the accumulation of radioactivity in the superior cervical ganglion 2 h after subcutaneous administration of [lZSI]NGF did not differ between intact and axotomized side. Thus, at least the accumulation of NGF in the ganglion independent of the retrograde transport was not impaired by axotomy. Furthermore, axotomy neither significantly reduced the level of TH in the superior cervical ganglion nor completely abolished the response to the systemic administration of 10 mg/kg of NGF. However, it cannot be decided whether this response really corresponds to the moiety of NGF reaching the cell body by the blood stream and the difference between intact and axotomized side represents the response to the moiety of NGF transferred by retrograde axonal transport.
290 The only safe conclusion to be drawn from the present experiments is the following. Blockade of retrograde axonal transport does not completely abolish the response to systemically administered NGF. Thus, although a considerable part of the trophic action of systemically administered N G F seems to result from the moiety reaching the perikaryon by retrograde transport, this is not the exclusive way of action. There also seems to exist a response to N G F which reaches the adrenergic cell body directly. Since the trophic effect of N G F appears to result from both the moiety of NGF reaching the perikaryon by retrograde transport and the moiety directly delivered by the blood stream to the cell bodies, the question arises as to whether the trophic action of N G F is effected by an interaction either with receptors at the outer surface of the neuronal membrane or with receptive structures inside the neuronal cell body. An interaction with receptive structures at the neuronal membrane could be assumed for the N G F moiety producing its effect by direct action on the cell body by a similar mechanism as described for the biological action of insulin 4. The assumption of such a mechanism of action of N G F is also supported by the recent observation of Frazier et al. that N G F linked to Sepharose beads retains its biological activity on chicken dorsal root ganglia, i.e., the NGF-induced axonal sprouting occurs also with N G F covalently linked to Sepharose 6. However, for the TH induction resulting from the N G F moiety reaching the perikaryon by retrograde transport such a mechanism of action does not seem to be very likely, since one has to assume that the N G F molecules reaching the cell body by retrograde transport had first to be transferred to the outer surface of the neuronal membrane before they can develop their biological action. Preliminary electron microscopic autoradiographic experiments did not provide any indication that N G F is transferred to the outer surface of the neuronal membrane. The biological significance of retrogradely transported N G F becomes of additional importance if one assumes that the biological function of N G F does not result from the part of N G F delivered from the site of production into the circulation but from that moiety which reaches the nerve terminals directly from the site of production in the effector organ. The production of N G F by the effector organ and the highly selective retrograde transport of N G F from the nerve terminals to the cell body would provide the possibility of transfer of (trophic) information from the effector organ to the cell bodies of the innervating neurons. Such a mechanism could be thought to determine the density of innervation of an organ or at least regulate the outgrowth of nerve fibers as a prerequisite for specific contacts between nerve terminals and effector cells. ACKNOWLEDGEMENT
This work was supported by the Swiss National Foundation for Scientific Research (Grant No. 365371).
REFERENCES 1 BOCCHINI,V., AND ANGELETTI, P. U., The nerve growth factor: purification as a 30,000molecular weight protein, Proc. nat. Acad. Sci. (Wash.), 64 (1969) 787-794.
291 2 BRIMIJOIN,S., AND MOLINOFF,P. B., Effects of 6-hydroxydopamine on the activity of tyrosine hydroxylase and dopamine fl-hydroxylase in sympathetic ganglia of the rat, J. Pharmacol. exp. TheE., 178 (1971) 417-424. 3 CRAGG,B. C., What is the signal for chromatolysis? Brain Research, 23 (1970) 1-21. 4 CUATRECASAS,P., Properties of the insulin receptor isolated from liver and fat cell membranes, J. biol. Chem., 247 (1972) 1980-1991. 5 FErCrON,E. L., Tissue culture assay of nerve growth factor and of the specific antiserum, Exp. Cell Res., 59 (1970) 383-392. 6 FRAZIER,W. A., BOYD,L. F., AND BRADSHAW,R. A., Interaction of nerve growth factor with surface membranes: biological competence of insolubilized nerve growth factor, Proc. nat. Acad. Sci. (Wash.), 70 (1973) 2931-2935. 7 GREENWOOD,F. C., HUNTER,W. M., AND GLOVER,J. S., The preparation of lalI-labelled human growth hormone of high specific radioactivity, Biochem. J., 89 (1963) 114-123. 8 HENDRY,I. A., AND IVERSEN,L. L., Changes in tissue and plasma concentrations of nerve growth factor following removal of the submaxillary glands in adult mice and their effects on the sympathetic nervous system, Nature (Lond.), 243 (1973) 500-504. 9 HENDRY,I. A., STOECKEL,K., THOENEN,H., AND IVERSEN,L. L., Retrograde axonal transport of the nerve growth factor, Brain Research, 68 (1974) 103-121. 10 HENDRY,I. A., AND THOENEN,H., Changes of enzyme pattern in the sympathetic nervous system of adult mice after submaxillary gland removal, response to exogenous nerve growth factor, J. Neurochem., 22 (1974) 999-1004. 11 JOHNSON,D. G., SILBERSTEIN,S. D., MANBAUER,I., AND KOPIN, I. J., The role of nerve growth factor in the ramification of sympathetic nerve fibers into rat iris in organ culture, J. Neurochem., 19 (1972) 2025-2029. 12 LEvI-MONTALCINI,R., AND ANGELETTI,P. U., Nerve growth factor, Physiol. Rev., 48 (1969) 534569. 13 LEVITT,M., GraB, J. W., DALY, J. W., LIPTON, M., AND UDENFRIEND,S., A new class of tyrosine hydroxylase inhibitors and a simple assay of inhibition in vivo, Biochem. Pharmacol., 16 (1967) 1313-1321. 14 LOWRY,O. H., ROSEBROUGH,N. J., FARR, A. L., AND RANDALL, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 15 MOLINOFE,P. B., ANDAXELROD,J., Biochemistry of catecholamines, Ann. Rev. Biochem., 40 (1971) 465-500. 16 MUELLER,R. A., THOEN~N,H., AND AXELROD,J., Increase in tyrosine hydroxylase activity after reserpine administration, J. Pharmacol. exp. Ther., 169 (1969) 74-79. 17 STOECKEL,K., PARAVICINI,U., AND THOENEN,H., Specificity of the retrograde axonal transport of nerve growth factor, Brain Research, 76 (1974) 413-421. 17a STOECgEL, K., SOLOMON,F., PARAVIONI, U., AND THOENEN, H., Dissociation between effects of nerve growth factor on tyrosine hydroxylase and tubulin synthesis in sympathetic ganglia, Nature (Lond.), 250 (1974) 150-151. 18 THOErqEN,H., Neuronally mediated enzyme induction in adrenergic neurons and adrenal chromaffin cells, Biochem. Soc. Symp., 36 (1972) 3-15. 19 THOENEN,H., HENDRY, I. A., STOECKEL,K., PARAVICINI,U., ANDOESCH,F., Regulation of enzyme synthesis by neuronal activity and nerve growth factor. In K. FUXE, L. OLSONAND Y. ZOTTERMAN (Eds.), Dynamics of Degeneration and Growth in Neurons, Pergamon Press, Oxford, 1974, in press. 20 THOENEN,H., ANGELETTI,P. U., LEvI-MONTALC1NI,R., AND KETTLER,R., Selective induction by nerve growth factor of tyrosine hydroxylase and dopamine fl-hydroxylase in the rat superior cervical ganglia, Proc. nat. A cad. Sci. (Wash.), 68 (1971) 1598-1602.
