Blockade of N-methyl-d -aspartate receptors by MK-801 (dizocilpine maleate) rescues motoneurones in developing rats

DEVELOPMENTAL BRAIN RESEARCH

ELSEVIER

Developmental Brain Research 81 (1994) 162-170

Research Report

Blockade of N-methyl-D-aspartate receptors by MK-801 (dizocilpine maleate) rescues motoneurones in developing rats L. Greensmith, GZ. Mentis, G. Vrbova

*

Department of Anatomy and Developmental Biology, Centre for Nellroscience, University College London, Gower Street, London, WCIE 6BT, UK (Accepted 22 March 1994)

Abstract

In rats following nerve injury at birth a large proportion of motoneurones to the soleus muscle dies. Blocking of N-methyl-D-aspartate (NMDA) receptors by MK-SOl (dizocilpine maleate) for 12 days after nerve injury at birth leads to rescue of a proportion of motoneurones destined to die. Retrograde labelling of soleus motoneurones shows that 6-S weeks after crushing the sciatic nerve in one hindlimb, only 10.9 ± 2.3% of the motoneurones have survived. In animals treated with an NMDA receptor blocker MK-SOI (2 mg/kg i.p., from birth to 12 days old) 50.6 ± 3.S% of soleus motoneurones survived. This neuroprotective effect of MK-SOI was dose dependant, since after treatment with lower doses (0.5 mg/kg; 1 mg/kg) fewer motoneurones survived (13.7% and 34.5%, respectively). To assess the effect of treatment with MK-S01 on survival of a-motoneurones only, the number of soleus motor units was established physiologically. After nerve injury alone only 4.2 ± 1.2 of the 29-30 soleus motor units were present, while in animals treated with MK-S01 (2 mg/kg) 14 ± 1.5 motor units were identified. The neuroprotective effect of MK-S01 was not confined to soleus motoneurones but was also apparent on motoneurones to the extensor digitorum longus (EDL). In untreated EDL muscles of the 40 motor units only 5.5 ± 1.7 motor units survived neonatal nerve injury and this number increased to IS ± 2.6 after treatment with MK-SOl. The neuroprotective effect of MK-SOI was apparent regardless of whether the nerve lesion was carried out close to or far from the soleus muscle. Key words: Excitotoxicity; Motoneuron; Axotomy; Target deprivation

1. Introduction Following injury to the sciatic nerve early after birth in rats, a large proportion of motoneurones dies [29,35,15,16,30,13]. The possibility that motoneurone death is caused by a lack of target derived trophic factors has been advocated by several groups of research workers (for reviews see [21,17]). Recent support for this idea was provided by experiments in which brain derived neurotrophic factor (BDNF) and ciliary neurotrophic factor (CNTF) were shown to reduce motoneurone death after axotomy in neonates [31,32,34] or during normally occurring cell death [25]. However, in the experiments carried out in neonates in vivo, the effects of these treatments were followed for only brief periods of time (6-7 days after axotomy) and the experiments terminated at a time when cell death

* Corresponding author. Fax: (44) (71) 380 7349. 0165.3806/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0165·3806(94)00061-4

is known to continue. After sciatic nerve section in rats, most of the motoneurones to lateral gastrocnemius muscle are still present 10-15 days after axotomy and die later during the second and third week [13}. Thus, the question of whether the rescue of motoneurones by BDNF or CNTF is permanent is still not resolved. Whether the supply and uptake of trophic factors from the target is indeed the way in which motoneurones are induced to survive is not certain. Recent results show that even when motoneurones are not physically separated from their target but their axons are prevented from interacting with the muscle fibres by blocking the acetylcholine receptor during early postnatal development, large numbers of motoneurones die [11]. Interestingly, similar results were obtained in the isthmo-optic nucleus, where preventing nerve-target interaction by tetrodotoxin caused large numbers of neurones in the isthmo-optic nucleus to die [26]. Therefore, it is not merely the presence of the

L. Greensmith et al. / Developmental Brain Research 81 (J994) 162-170

axons in their targets, but their active interaction with it that induces their ability to survive in the developing central nervous system. It has recently been proposed [17] that as a result of interaction with the target, the motoneurone acquires properties related to the distribution of its glutamate receptors, which allow it to become successfully integrated into the spinal cord circuitry of the developing eNS. During late embryonic and early postnatal development, the motoneurone is subjected to an ever increasing amount of excitatory influences [28,20,33]' Since glutamate is the main excitatory transmitter in the spinal cord, the motoneurone has to survive exposure to increased amounts of glutamate. Glutamate and its agonist N-methyl-D-aspartate (NMDA) are known to be toxic to neurones and cause them to die [3,4,18]. In support of this proposal is the finding that motoneurones disconnected from their target have been shown to die after application of the agonist NMDA while motoneurones that remained in contact with their muscles survived the same treatment [9,10]. It is therefore possible that as a result of the interaction of the motoneurone with the target, a redistribution of glutamate receptors and their restriction to synaptic sites occurs. In the absence of this developmental event the cell will remain excessively sensitive to glutamate which is released in increasing amounts from the maturing synaptic inputs. Cell death of target-deprived neurones could then be explained by the excitotoxic effect of glutamate and ought to be reduced in the presence of glutamate receptor blockers. In this study this possibility was examined. A preliminary report of some of this work has been published as a Letter to Neuroscience [19]

2. Materials and methods The effect of treatment with MK-801 (dizocilpine maleate) on the survival of sciatic motoneurones was investigated. Two groups of animals were used, in which the severity of the nerve injury differed. These two groups of rats were then subdivided (see Table 1). In the first group of animals, the sciatic nerve was crushed far from the cell body (distal crush). The effect of increasing doses of MK-801 on the number of soleus motoneurones, soleus motor units and extensor

163

digitorum longus (EDL) motor units was then investigated. In the second group of animals, the nerve was crushed closer to the cell body (proximal crush). In this group of animals the effect of the maximum tolerated dose of MK-801 on the number of soleus and EDL motor units was studied. The survival of soleus motoneurones was assessed using the retrograde tracer HRP and the number of soleus and EDL motor units was established by recording muscle tension in response to stimulation of the motor nerve.

2.1. Surgery In the first group of animals, neonatal Wistar albino rats were operated on within 24 h of birth. The pups were anaesthetized with ether and under sterile conditions the sciatic nerve was exposed in one hindlimb. The nerve was crushed in the popliteal fossa, just above the division of the sciatic nerve into the tibial and common peroneal nerves, using a fine pair of Watchmakers' forceps (distal crush). Following the crush, the nerve was examined to ensure that the endoneurial sheath was intact but the nerve translucent and the wound was then closed. After recovery from anaesthesia, the pups were returned to their mother. In the second group of animals, the sciatic nerve in one hindlimb was crushed within 12 h of birth, approximately 4 mm higher than in the first group of animals, at the entry of the sciatic into the thigh, near the obturator tendon and therefore closer to the cell body (proximal crush). This injury would be expected to have more severe effects upon motoneurone survival than the distal nerve crush, described above for the first group of animals.

2.2. Administration of the N·methyl·o·aspartate receptor antagonist, MK·801 (dizocilpine maleate) Following nerve crush, the rat pups received daily intraperitoneal injections of the NMDA receptor antagonist, MK·801, until they were 12 days of age. Control animals were injected with sterile saline. In the first set of experiments, the effect of treatment with increasing doses of MK·801 on the survival of sciatic motoneurones which received a distal nerve injury was studied. Neonatal rats were injected with either 0.5 mg/kg, 1 mg/kg or 2 mg/kg of MK·801, daily for 12 days and the effect of these treatments on the survival of soleus motoneurones after nerve injury was assessed. Furthermore, the effect of these doses of MK-801 on the number of motor units surviving after nerve injury in both the soleus and extensor digitorum longus (EDL) muscles was also studied. The maximum tolerated and most effective dose of MK·801 was found to be 2 mg/kg. However, daily administration of this dose of MK-BOl beginning on the day of birth (day 0), caused the pups to be very drowsy, resulting in poor weight gain and a relatively high mortality rate (approximately 40%). At 12 days of age control rat

Table 1 In the first group of animals, which received a distal nerve crush, the effect of increasing doses of MK-801 on soleus motoneurones, soleus motor units and EDL motor units was studied. In the second group, in which the nerve was crushed more proximally, the effect of the maximum tolerated dose of MK-801 on soleus and EDL motor units was assessed. The number of animals used in each experimental group is shown in brackets Effects on doses of MK·801 on experimental groups used in this study

On Soleus motoneurones On Soleus motor units On EDL motor units

Group 1: distal crush (far from the cell body), increasing doses of MK-801

Group 2: proximal crush (close to the cell body), maximum dose of MK-801

yes (n = 20) yes (n = 16) yes (n = 16)

no yes (n '" 13) yes (n ,., 13)

L. Greensmith et af. / Developmental Brain Research 81 (J994) 162-170

164

pups weighed approximately 30 g, whereas the treated animals weighed approximately 12 g. Thus, a second regime of drug administration was employed in a set of eXperiments where the effect of MK-801 on the injured soleus and EDL muscles was investigated further. In this group of animals the dose of MK-801 was gradually increased over the first 4 days of life, so that on day 0 they received 0.75 mg/kg, 1 mg/kg on day 1, 1.5 mg/kg on day 2 and finally 2 mg/kg on day 3 and thereafter until they were 12 days old. This protocol caused no mortality, but still a considerable reduction in weight gain. Control animals in this group weighed 30 g at day 12 and treated animals approximately 15 g. When the treatment was discontinued, the treated animals eventually reached normal body weight. 2.3. Retrograde labelling of motonellrones

In one group of animals, which received a distal nerve crush at birth, the effect of treatment with MK-801 on the survival of soleus motoneurones was assessed. The rats were examined when they were 6-10 weeks old. Using halothane anaesthesia and sterile conditions, the retrograde tracer Horseradish peroxidase (HRP; Sigma Type vI) was injected into the soleus muscles of both hindlimbs using a Hamilton microsyringe [15,16)' Twenty-four h later, the animals were re-anaesthetized, the soleus muscles removed and weighed and the animals perfused transcardially with a fixative containing glutaraldehyde (2.5% in Millonigs phosphate buffer, pH 7.3). The spinal cords were removed, the lumbar region post-fixed for 2 h in the same fixative, then cryoprotected in sucrose. Frozen, free floating sections were cut on a cryostat at 50 p.m, processed for HRP histochemistry

o

using the modified Hanker-Yates method, [12] and lightly counterstained with a Nissl stain (gallocyanin). The number of HRP-Iabelled motoneurones in each ventral horn was counted under a light microscope. In order to avoid counting the same cell twice in consecutive sections, only those cells in which the nucleolus was clearly visible at high magnification were counted. As an index of motoneurone survival, the number of labelled cells on the operated side was expressed as a percentage of that on the contralateral control side. 2.4. Tension recordings

In a second series of experiments the effect of MK-801 treatment on the number of motor units surviving in the soleus and EDL muscles after nerve injury at birth was investigated. When the animals were 6-10 weeks old, they were anaesthetized with chloralhydrate (4.5%, 1 mljlOO g body weight i.p.) and prepared for tension recording of their muscles. The distal tendon of the fast extensor digitorum longus (EDL) muscle of both legs was dissected free. The soleus muscle was separated from the surrounding musculature and its nerve exposed. The deep branch of the peroneal nerve was dissected and prepared for stimulation. All nerves to other muscles of the leg were cut. Both legs of the animal were fixed to a table by stainless steel pins. The muscle tendons were then attached to strain gauges (Dynamometer UFI, Devices) via a silk thread. Isometric contractions were elicited by stimulating the cut end of the motor nerve via bipolar silver chloride electrodes, using a pulse width of 0.02 ms and the contractions were displayed on an oscilloscope screen (Tetronix R5113). The muscle length was adjusted to obtain maximum twitch tension at supramaximal stimulus intensity.

c

Fig. 1. A cross-section of the lumbar region of an 8 week old rat spinal cord after injecting both soleus muscles with HRP is shown. The spinal cord was taken from an animal that received a nerve crush at birth and was subsequently treated with MK-801 (2 mg/kg) for 12 days. The sciatic motoneurone pool is encircled by the broken white line. The arrows indicate the location of HRP labelled cells on both the control (C) and operated (0) sides of the spinal cord. The scale bar is 500 p.m.

165

L. Greensmith et al. / Developmental Brain Research 81 (1994) 162-170

2.5. Motor unit numbers To estimate the number of motor units in each muscle, the nerves to the operated and contralateral soleus and EDL muscles were stimulated by single pulses every 4 s. The stimulus strength was gradually increased to obtain stepwise increments of twitch tension, as individual motor axons are recruited. The number of stepwise increments was counted to give an estimate of the number of motor axons present in the nerve. Using this method, the assessment of motor unit numbers in muscles that contain more than 20 motor units is difficult. We were therefore able to count the numbers of motor units in only a few control muscles. In the five control muscles tested the number of motor units in the soleus muscle varied between 26-32 and in EDL varied between 36-45. These results compare well with findings of other authors [5]. The mean number of motor units in soleus was found to be 30 and for EDL, 40 motor units. We have therefore used these numbers for all control muscles. After all tension experiments had been completed, the control and operated soleus and EDL muscles were dissected out from the rats and weighed.

3. Results

3.1. Behavioural observations

Following injection of MK-801 (2 mg/kg), the neonatal rats became very drowsy. During the 12-day period of treatment, the toe-spreading reflex was absent, not only on the side of the nerve crush, but also on the contralateral control side. When placed on their backs, the righting-response was also delayed. The MK-801-treated rats appeared to be generally clumsy when compared to their control litter mates. However, these behavioural effects were only temporary as 2 days after treatment was discontinued, both the toespreading and righting-reflex was similar to that observed in normal rats.

3.2. The effect of different doses of MK-801 (dizocilpine maleate) on the survival ofaxotomized soleus motoneurones In the first set of experiments, the effect of treatment with increasing doses of MK-801 on the survival of soleus motoneurones after nerve injury far from the cell body was studied. Following sciatic nerve crush within 24 h of birth, the rat pups received daily intraperitoneal injections of either sterile saline or solutions containing different amounts of MK-801 until they were 12 days of age. Six to 10 weeks later the animals were anaesthetised and HRP injected into both soleus muscles to be retrogradely transported to their motoneurones. Fig. 1 shows an example from a section of a spinal cord from an animal after distal nerve crush at birth treated with 2 mg/kg of MK-801 that had its soleus muscles injected with HRP 8 weeks later. Counts of retrogradely labelled motoneurones from groups of animals treated with saline or different amounts of MK-801 are summarised in Table 2 and Fig. 2. It is clear from the summary of these results that the number of surviving soleus motoneurones was greater in animals that were treated with either 1 or 2 mg/kg of MK-801 during the 12 days following nerve crush at birth. Moreover, Table 2 and Fig. 2 indicate that the effect of the NMDA receptor inhibitor was dose dependent. To confirm this result and establish whether the rescued motoneurones were able to make and maintain functional contacts with the muscle fibres, a series of experiments was carried out where the number of motor units in the soleus muscle was established using physiological criteria. In a separate group of rats treated shortly after birth in a similar way as described above,

Table 2 The effect of MK-801 treatment on the survival of soleus motoneurones and motor units after nerve crush at birth Expt. NC

+ saline

NC

+ MK·801 0.5 mg

NC NC

+ MK·801 1 mg + MK-801 2 mg

MK-801 2 mg

No. of MNs Can

No. ofMNs Op

%Op/Con MNs

No. of MUs Can

S1.4 ± 2.1 (n = 5) 46 ± 0.6 (n = 3) 54.6 ± 2.6 (n = 3) 56.1 ± 2.1 (n = 6) 48 ± 1.5 (n = 3)

5.8 ± 1.5 (n =5) 6.3 ± 0.6

10.9 ± 2.3

30

13.7 ± 2.5

30

(II = 3)

18.6 ± 4.6 (II = 3) 29.3 ± 2.6 (n = 6) 47.6 ± 1.2 (n = 3)

No. of MUs Op 4.2 ± 1.2 = S) 4.S

en (n

*

34.5 ± 8.8

*

30

%Op/CoMUs 13.9 ± 4.1

IS.1

= 2)

9.5

31.61

(n = 2)

**

50.6 ± 3.8 99.5 ± 3.7

**

30

14 ± 1.5 (n = 4)

***

46.71

± 5.2 ***

29 ± 0.5 (n = 3)

Table 2 summarizes results from experiments of retrogradely labelled soleus motoneurones (No. of MNs) where nerve crush at birth (NC) was followed either by daily intraperitoneal injection of saline or solutions containing different amounts of MK·801. The Table also shows estimates of motor unit numbers (MUs) in soleus muscle after nerve crush at birth and treatment with either saline or MK-801.Statistical comparisons were made between saline and MK-80J·treated motoneurones or motor units using the Mann-Whitney U-test. 'I< p < 0.025, ** P < 0.004, *** P < 0.001. The numbers in brackets show the number of animals used in each experimental group.

L. Greensmitlz et af. / Developmental Brain Research 81 (1994) 162-170

166

the number of motor units was estimated 6-10 weeks after the initial operation. This was carried out by following the stepwise increments of twitch tension in response to stimulation of the motor nerve by stimuli of increasing intensity. Table 2 summarises the results and shows that as with retrograde labelling, the numbers of motor units were higher in rat pups treated with MK-801 after nerve crush at birth (see also Fig. 3A,B). Thus, using either method, the results show that after treatment with MK-801, more motoneurones survive nerve injury at birth. In this series of experiments the injury to the sciatic nerve was inflicted far from the cell body so that the motoneurones were disconnected from the muscle for only 5-7 days. Injuring the sciatic nerve further away from the muscle ought to cause a greater loss of motoneurones. The next set of experiments was carried out to see whether after an injury that causes almost complete loss of motoneurones, treatment with MK-801 will still be effective. In this set of experiments only the most effective dose of MK-801 (2 mg/kg) was used. Nerve crush was carried out within the first 12 h of birth and closer to the cell body. The results are illustrated in Fig. 3B and show that this injury leads to a greater reduction of motor units than that seen in the previous experiments (see Fig. 3A). In this more severely affected group only 1.0 (± 0.4 S.E.M., n = 5) motor unit survived the injury in the saline-treated animals, as compared to 4.2 (± 1.2 S.E.M., n = 5) in the previous group. Littermates in both groups were given high doses (2 mg/kg) of MK-801. Fig. 3A,B show

(4)

(4)

Fig. 3. The block diagrams illustrate the number of motor units in the soleus muscle after nelVe crush (A) far from (distal) and (B) close to (proximal) the cell body. The empty bars represent the results from the group which received nelVe crush at birth followed by treatment with saline; the hatched bars are results from the group treated with MK-801 (2 mg/kg) after nelVe crush at birth. The filled bars show results from normal, untreated animals. The error bars are ± S.E.M .. The number of animals in each group is shown in brackets. In (A) and (B) the MK-801-treated groups are significantly different from the saline-treated groups at the level of * P < 0.01 and ** P < 0.009, respectively. It can be seen that in both groups the number of motor units was significantly higher after treatment with MK-801.

60

50

Co

u 0.40

.£ (f)

Q)

5 GO :; Q)

c 0

'0 ::2

20

~ 0

10

!

0 0

0.5

1.5

2

mg/kg MK Fig. 2. The graph shows the effect of different doses of MK-801 (abscissa) on the sUlVival of soleus motoneurones after nelVe injury at birth. For each animal the number of motoneurones on the operated side is expressed as a percentage of controL In each group the means from 3-6 animals were taken and the vertical bars show ±S.E.M.. The full circle represents results from saline-treated and the empty squares from MK-801-treated animals.

that in the treated animals the number of motor units was significantly higher both in animals that had their nerve injured close or far from the cell body. Thus MK-801 can rescue a significant proportion of motoneurones even after an extremely severe nerve injury, where otherwise almost all motoneurones would have died. Since the effects of the MK-S01 treatment was effective in both groups of experiments, the results from all the experimental animals were pooled together. These pooled results show that after nerve injury alone 2.6 (± 0.8 S.E.M., n = 10) motor units survived. Treatment with MK-S01 increased this survival to 10.8 (± L5 S.E.M., n = S). Soleus is a slow twitch muscle and it could be that motoneurones to fast muscles may react differently to treatment with NMDA inhibitors. Therefore, the number of motor units of extensor digitorum longus (EDL), a fast twitch muscle was studied following nerve injury at birth.

L. Greensmith et af. / Developmental Brain Research 81 (]994) 162-170 40

a

167

A

b 30

20

Fig. 4. Isometric single contractions elicited by stimulation of the peroneal nerve from EDL muscles of 2 different animals aged 2 months are shown. The stepwise increments of force were elicited by gradually increasing the stimulus intensity to the nerve. Both animals had their sciatic nerve crushed at birth, but one animal (a) was given daily injections of saline and the other (b) MK-801 (2 mg/kg) for 12 days after birth. The number of increments in (a) is 11-12 and in (b) 17-18. The vertical bar represents 4 g and the horizontal 40 ms.

3.3. Changes in the number of motor units of extensor digitorum longus muscles after nerve injury at birth and treatment with MK-80 1 (dizocilpine maleate)

10

.~ c:

~

...

0

:2:

40

.8 a

'0

0 (5)

40

(4) SALINE

(2)

o 5111g

(2) 1111g

(4) 2mg

(4)

MK

Fig. 5. The block diagram illustrates the number of motor units in the EDL muscle following nerve crush at birth and treatment with either saline (empty block) or MK-801. The dose of MK-801 shown below the bars is mg/kg body weight. The last group received no nerve injury and the animals were treated with MK-801 (2 mg/kg). Error bars are S.E.M .. The number of animals in each group is shown in brackets. It can be seen that the reduction of motor unit number induced by nerve injury at birth is less in animals treated WIth MK-801 and that this effect is dose dependant. The group * whieh received MK-801 at a dose of 2 mg/kg is significantly different from the saline-treated group at the level of P < 0.025.

(4)

B

a

Z

30

20

10

In this set of experiments, the effect of increasing doses of MK-801 on EDL motor units after a distal nerve injury was studied. An example from an experiment where the number of motor units was assessed is shown in Fig. 4, which compares the result obtained after nerve crush at birth and saline treatments (Fig. 4a) with that obtained from a different animal after treatment with MK-801 (Fig. 4b). Fig. 5 summarizes the results from experiments like these and shows that

(4)

0

~ (5)

0 (4)

(4)

Fig. 6. The block diagrams illustrate the number of motor units in the EDL muscle after nerve crush (A) far from (distal) and (B) close to (proximal) the cell body. The empty bars represent the results from the group which received nerve crush at birth followed by treatment with saline; the hatched bars are results from the group treated with MK-801 (2 mg/kg) after nerve crush at birth. The filled bars show results from normal, untreated animals. The error bars are S.E.M .. The number of animals in each group is shown in brackets. Both in (A) and (B) the MK-801-treated groups are significantly different from the saline-treated groups at the level of * P < 0.025 and ** P < 0.009, respectively.

similarly to soleus motor units, the reduction of motor unit numbers that follow nerve injury at birth is less in animals treated with MK-801 and that the drug is effective only if given at doses higher than 1 mg/kg. Additional experiments were carried out on animals where the nerve injury was modified so as to cause a more severe loss of motoneurones. In these experiments the sciatic nerve was crushed within 12 h of birth closer to the cell body and the animals were treated with either 2 rug/kg of MK-801 or saline. In the saline treated group, animals with this type of nerve lesion had a significantly greater loss of EDL motoneurones than those of the previous group (compare Fig. 6A,B). The number of motor units was reduced to 1.6 (± 1.0 S.E.M., n = 5), as compared to 10.5 (± 0.6 S.E.M., n = 4) in the previous group (see Fig. 6A). This difference is statistically significant. Nevertheless, even after this severe injury, treatment with high doses of MK-801 significantly increased the number of motor units in EDL muscles (see Fig. 6A,B). Since the effect of MK·801 was similar in both groups, the results from the two groups of experiments

168

L Greensmith et ai. / Deueiopmentai Brain Research 81 (1994) 162-170

were pooled together. Of the 40-42 EDL motor units only 5.5 (± 1.7 S.E.M., n = 9) survived nerve injury at birth. In animals treated with MK-801 18.2 (± 2.6 S.E.M., n = 7) motor units were present. Using the Mann-Whitney U~test this difference was shown to be highly significant (P < 0.001).

4. Discussion

The present results confirm earlier findings that injury to the motor nerve shortly after birth leads to permanent loss of soleus motoneurones [35,16]. The number of retrogradely labelled motoneurones from control soleus muscles obtained in this study was similar to that reported by others in this laboratory [16,2,11] and is in agreement with estimates from other groups [21,27]. However, the reduction of motoneurones after neonatal nerve injury seen in the present experiments was greater than reported by Lowrie et a1. (1987) [16]. The reason for this may be either the nature of the inflicted injury or the exact position of the lesion in relation to the soleus muscle. Results presented here show that when the nerve is damaged close to the cell body, thus prolonging the period of denervation, more motoneurones are lost. This is in agreement with previous results [14,13] which show that the length of time that the motoneurone is separated from its target is an important factor for its survival. Thus, it can be expected that the reduction of motoneurones after axotomy will be variable, dependent upon the exact location of the lesion. However, regardless whether the injury was close to or far from the cell body, systemic administration of a non-competitive NMDA receptor blocker, MK-801, during a 12-day period after nerve injury at birth, rescued a high proportion of motoneurones destined to die. Treatment with MK-801 reduced the rate of growth of these animals and it could be argued that this effect may be responsible for the reduction in motoneurone death after nerve injury. However, recent results from this laboratory have found that treatment of rats for a similar period of time with cytotoxic drugs after nerve injury at birth led to a similar reduction in the rate of growth, but, in contrast to the effect of MK-801, this treatment resulted in an increased level of motoneurone' death [6]. Therefore, it is unlikely that the improvement in motoneurone survival is due to the reduction in the rate of growth of these treated rats and is probably a consequence of the known actions of MK-801 in blocking NMDA receptors. The effect of MK-801 appeared to be dose dependent. When the highest dose of the NMDA receptor blocker was used, 4-5 times as many motoneurones survived axotomy as in control, saline-treated rats. Retrograde labelling probably reveals both 11- and y-

motoneurones, [16] and these may be differently affected by nerve injury or influenced by NMDA receptor blockers. To test this the number of 11motoneurones that survive the various procedures was established by recording increments of tension in response to application of stimuli of increasing intensity to the motor nerve. This method applied to normal soleus muscles in this study gave results showing that control soleus muscles contained about 30 motor units, which is in good agreement with other estimates of numbers of 11 motor units in rat soleus muscles using this or other methods. [35,5,1,7] Following nerve injury at birth the number of soleus motor units was reduced to 14% of that in control muscles, a value not significantly different from that obtained with retrograde labelling. Thus it appears that 11- and 'Y-motoneurones are equally affected by the insult. Similarly, treatment of the animals with NMDA receptor blockers rescued a similar proportion of motor units estimated physiologically to that revealed by counts of retrogradely labelled motoneurones, indicating that treatment with an NMDA receptor blocker rescued both 11- and 'Ymotoneurones. Soleus is a slow twitch muscle and it could be that its motoneurones are affected by nerve injury in a different way from those to fast twitch muscles. Previous [16] and present results show that this is not the case and that the reduction of motor unit numbers after sciatic crush at birth is equally great in EDL, a fast twitch muscle as in the soleus muscle. In addition, like soleus Il-motoneurones EDL Il-motoneurones are also rescued by treatment with MK-801. Thus there does not seem to be a great deal of difference between the responses of these two populations of motoneurones either to neonatal nerve injury or to the protective effects of NMDA receptor blockers. It is interesting to speculate why not all the motoneurones destined to die are rescued by the NMDA receptor blocker. There are several explanations to this question. Our results indicate that the effect of treatment with MK-801 is dose dependant and we have restricted the dose given to the animals to 2 mg/kg, since a higher dose rendered them too drowsy to suckle and they may not have survived the treatment. Therefore, a topical application to the spinal cord may be more effective than systemic injection which restricts the dosage by unwanted side effects. It is also possible that treatment for 12 days is not sufficiently long to rescue all the injured motoneurones, since it could be that some motoneurones still die after this period. It would therefore be important to prolong the period of treatment. Whether treatment with MK-801 protects the motoneurones destined to die by blocking their NMDA receptors. pr by reducing excitatory inputs from other sources, either supraspinal or interneuronal is uncer~

L. Greensmith et al. / Developmental Brain Research 81 (1994) 162-170

tain. Although there is no direct evidence that MK-801 blocks NMDA receptors in this situation, the behavioural observations following treatment with MK801 and the finding that the effects of MK-801 are dose dependant, suggests that NMDA-receptor mediated events may be involved. Motoneurones do have NMDA receptors [22,23] but it is possible that the systemic application of the drug affects not only NMDA receptors on the motoneurone cell body, but also NMDA receptors of other structures that mediate excitatory influences to the motoneurones in question. It is possible that those motoneurones that have not been protected by the treatment receive inputs from sources that are not influenced by MK-801 and continue to transmit their excitatory impulses to the axotomized motoneurones. Alternatively, the population of motoneurones that is protected by the treatment is perhaps more mature. In a recent study [8] the effect of neurotoxins released by microglial cells on mixed spinal cord cultures was demonstrated and this effect could be blocked by MK-801. It could be that loss of contact with the target prevents maturation of motoneurones and the treatment with MK-801 may have to be prolonged to allow the maturation of all the cells. The protective effect provided by MK-801 to a large proportion, if not all the motoneurones destined to die, is nevertheless clear from our results. The most likely explanation is that this protective effect is due to a reduction of excitatory influences to the motoneurones in question. Finally the question of whether various neurotrophic factors that seem to reduce motoneurone loss after nerve injury at birth, at least for 6-7 days, may act by influencing the synthesis or distribution of glutamate receptors has to be considered. It is conceivable that the mode of action of these growth factors is through regulating the synthesis and distribution of glutamate receptors. However, the evidence for a permanent beneficial effect of neurotrophic factors is at present not available, for in all the experiments the animals were followed for only short periods of time. Such short-term experiments may be misleading. Recent results on the isthmo-optic nucleus, show that treatment with tetrodotoxin first increases the numbers of surviving neurones, but a few days later leads to massive neuronal death [26]. In our previous experiments too, treatment of soleus muscles with abungarotoxin for a brief period during postnatal life did not initially lead to any loss of motoneurones, but at later stages, i.e., 10 weeks after birth, a substantial proportion of motoneurones died [11]. Only when long-term effects of treatment with neurotrophic agents is available, will it be possible to assess their effectiveness and discuss their mode of action. Thus the present results provide evidence that NMDA receptor mediated events may be involved in

169

the death of target-deprived motoneurones. The finding that treatment with NMDA receptor inhibitors leads to a permanent rescue of motoneurones destined to die indicates that this approach could be considered for treatment of diseases where muscle weakness is caused by loss of motoneurones.

Acknowledgements We would like to thank Merck, Sharp and Dohme for the gift of MK-801 and Jim Dick and Francois Tyc for their invaluable help. This work was supported by Action Research and The Wellcome Trust.

References [1] Brown, M.C., Jansen, 1.K.S. and Van Essen, D., Polyneuronal innervation of skeletal muscle in newborn rats and its elimination during maturation, J. Physiol., 261 (1976) 387-422. [2] Burls, A., Subramaniam Krishnan, Lowrie, M.B. and Vrbova, G., Absence of nerve-muscle interaction influences the survival of developing motoneurones, Eur. J. NeuroseL, 3 (1991) 216-221. [3] Choi, D.W., Glutamate neurotoxicity in cortical cell culture is calcium dependant, Neurosci. Lett., 58 (1985) 293-297. [4] Choi, D.W., Glutamate neurotoxicity and diseases of the nervous system, Neuron, 1 (1988) 623-634. [5] Close, R.I., Properties of motor units in fast and slow skeletal muscles of the rat, J. Physioi., 193 (1967) 45-55. [6] Clowry, G., Sen, P. and Vrbova, G., Effect of an RNA synthesis inhibitor on the survival of injured neonatal motoneurones, Brain Res. Assoc. Abstr., 10 (1993) 4.4. [7] Fisher, TJ., Vrbova, G. and Wijetunge, A., Partial de nervation of the rat soleus muscle at two different developmental stages, Neuroscience, 28 (1989) 755-763. [8] Giulian, D., Vaco, K. and Corpuz, M., Brain glia release factors with opposing actions upon neuronal survival, J. Neurosci., 13 (1993) 29-37. [9] Greensmith, L., Hasan, H.1. and Vrbova, G .. Nerve injury in anaesthetized neonatal rats increases the subsequent susceptibility of motoneurones to NMDA-induced neurotoxicity, 1. Physiol., 434 (1991) 20. [10] Greensmith, L., Hasan, H.1. and Vrbova, G., Nerve injury increases the susceptibility of motoneurones to NMDA-induced neurotoxicity in the developing rat, Neuroscience, 58 (1994) 727-733. (11] Greensmith, L. and Vrbova, G., Alterations of nerve-muscle interaction during postnatal development influence motoneurone survival in rats, Dev. Br. Res., 69 (1992) 125-131. [12] Hanker, 1.S., Yates, P.E., Metz, C.B. and Rustioni, A., A new specific, sensitive and non-carcinogenic reagent for the demonstration of horseradish peroxidase, Histochem. J., 9 (1977) 789-

792. [13] Kashihara, Y., Kuno, M. and Miyata, Y., Cell death of rumtomized motoneurones in neonatal rats and its prevention by peripheral reinnervation, J. Physioi., 386 (1987) 135-148. [14] LaVelle, A., Critical period of neuronal maturation, Prog. Br. Res., 9 (1964) 93-96. (15] Lowrie, M.B., Subramaniam Krishnan and Vrbovii, G., Recovery of slow and fast muscles following nerve injury during early postnatal development in the rat, J. Physioi., 331 (1982) 51-66. [16] Lowrie, M.B., Subramaniam Krishnan and Vrbova, G., Permanent changes in muscle and motoneurones induced by nerve

1. Greensmith et al. / Developmental Brain Research 81 (1994) 162-170

170

injury during a critical period of development of the rat, Dev.

Br. Res., 310 (1987) 91-101. [17J Lowrie, M.B. and Vrbova, G., Dependance of postnatal motoneurones on their targets: review and hypothesis, Trends Neural. Sci., 15 (1992) 80-84. [18] Meldrum, B. and Garthwaite, 1., Excitatory amino acid neurotoxicity and neurodegenerative disease, Trends Pharm. Sci., 11 (1990) 379-387. [19] Mentis, G.Z., Greensmith, L, Vrbova, G., Motoneurones destined to die are rescued by blocking NMDA receptors by MK-801, Neuroscience, 54 (1993) 283-285. [20] Navarrete, Rand Vrbova, G., Changes of activity patterns in slow and fast muscles during postnatal development, Dev. Br. Res., 8 (1983) 11-19. [21] Nicolopou]os-Stournaras, S. and Iles, J.F., Motor neurone columns in the lumbar spinal cord of the rat, J. Compo Neuro!., 217 (1983) 75-85. [22] O'Brien, R.J. and Fischbach, G.D., Characterisation of excitatory amino acid receptors expressed by embryonic chick motoneurones in vitro, J. Neurosci., 6 (1986) 3275-3283. (23) O'Donovan, R.I. and Landmesser, L, The development of hindlimb motor activity studied in the isolated spinal cord of the chick embryo, J. Neurosci., 7 (1987) 3256-3283. [24) Oppenheim, R.W., Cell death during development of the nervous system, Ann. Rev. Neurosci., 14 (1991) 453-501. [25] Oppenheim, R.W., Qin-Wei, Y., Prevette, D. and Yan Q., Brain derived neurotrophic factor rescues developing avian motoneurones from cell death, Nature, 360 (1992) 755-756. [26] Pequignot, Y. and Clarke, P.G.H., Changes in lamination and

[27J

[28]

[29]

[30] [31]

[32]

[33] [34]

[35]

neuronal survival in the Isthmo-Optic Nucleus following the intraocular injection of tetrodotoxin in chick embryos, J. Camp. Neural., 321 (1992) 336-350. Peyronnard, J.M., Charron, LF., LaVoire, 1. and Messier, J.P., Motor, sympathetic and sensory innervation of rat skeletal muscles, Brain Res., 373 (1986) 288-302. Provine, R., Embryologic spinal cord: synchrony and spatial distribution of polyneuronal burst discharge, Brain Res., 29 (1971) 155-158. Romanes, G.l., Motor localisation and the effects of nerve injury on the ventral horn cells of the spinal cord, J. Anal., 80 (1946) 11-l31. Schmalbruch, H., Motoneurone death after sciatic nerve section in newborn rats, J. Camp. Neuro!., 224 (1984) 252-258. Sendtner, M., Kreutzberg, G.W. and Thoenen, H., Ciliary neurotrophic factor prevents the degeneration of motoneurones after axotomy, Nature, 345 (1990) 440-441. Sendtner, M., Holtmann, B., Kolbeck, R, Thoenen, H., and Barde, Y.A, Brain derived neurotrophic factor prevents the death of motoneurones in newborn rats after nerve section, Nature, 360 (1992) 757-759. Westerga, J. and Gramsbergen, A, The development of locomotion in the rat, Dev. Br. Res., 57 (1990) 163-174. Yan, Q., Elliott, J. and Snider, W.O., Brain derived neurotrophic factor rescues neurones from axotomy-induced cell death, Nature, 360 (1992) 753-755. Zelena, 1. and Hnlk, P., Motor and receptor units in the soleus muscle after nerve regeneration in very young rats, Physiol. Bohemoslov., 12 (1963) 227-289.