- Email: [email protected]
Upregulation of the pro-opiomelanocortin gene in motoneurones after nerve section in mice
MOLECULAR BRAIN RESEARCH ELSEVIER
Molecular Brain Research 25 (1994) 41-49
Research Report
Upregulation of the pro-opiomelanocortin gene in motoneurones after nerve section in mice Sharon Hughes, Margaret E. Smith * Department of Physiology, Medical School, University of Birmingham, Birmingham B15 2T~, UK
Accepted 1 March 1994
Abstract In the normal adult mouse a few of the ventral horn motoneurones were slightly immunoreactive for the pro-opiomelanocortin (POMC)-derived peptides/3-endorphin (/3-EP) and ~-melanotropin (a-MSH). After unilateral section of the sciatic nerve the incidence of immunoreactive cells and the intensity of staining increased in both the ipsilateral and the contralateral ventral horns. The proportion of cells which expressed the peptides reached a maJximum at approximately 24 h after nerve section, and thereafter declined, but the proportion was still higher than normal at 7 days after the section. Using histochemical in situ hybridization with cDNA oligonucleotide probes for POMC transcript regions which encode for adrenocorticotropin (ACTH) 4-11 and/3-endorphin 1-8, POMC mRNA was detected in very few motoneurones in unoperated or sham operated mice. After unilateral sciatic nerve section increases in the proportion of cells w.hich expressed the POMC mRNA were seen in both the ipsilateral and contralateral ventral horns. The increase, like that seen for peptide immunoreactivity, was maximum at around 24 h and thereafter declined but at 7 days after the section the proportion of POMC mRNA-positive cells had returned to normal. Thus injury to the motoneurones is accompanied by synthesis of POMC-derived peptides via upregulation of the POMC gene in the motoneurone. Key words: Motoneuron; Pro-opiomelanocortin; In situ hybridization;/3-Endorphin; t~-Melanotropin; Neuropeptide
I. Introduction In previous work we used immunocytochemistry to demonstrate the presence of the pro-opiomelanocortin (POMC)-derived peptides/3-endorphin (/3-EP) and am e l a n o t r o p i n ( a - M S H ) or a d r e n o c o r t i c o t r o p i n , (ACTH), the parent of a-MSH, in motor nerve terminals and axons of rodents. In the normal adult very few motoneurones were immunoreactive for the peptides but after nerve transection [11] or in diseases where there is a primary motor neuropathy such as motor neurone disease [12] or a secondary peripheral neuropathy such as diabetes mellitus [14] or murine muscular dystrophy [10] there were significantly higher incidences of immunoreactive motor nerve axons. It was shown that in mice with muscular dystrophy ( C 5 7 B L / 6 J strain), in which there is a motor nerve defect, and after administration in rats of the neurotoxicant /3,if-
* Corresponding author. Fax: (44) 21-414-6924. 0169-328X/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0169-328X(94)00042-D
iminodipropionitrile (IDPN) which targets motoneurones there were increases in the incidence of peptideimmunoreactive ventral horn cells as well as immunoreactive nerve axons [7,15]. These peptides are trophic factors which promote neuromuscular transmission and nerve regeneration and they may be expressed as part of the regenerative response in damaged or diseased motoneurones whatever the cause of the lesion. The peptides in the motor nerves could have originated from the circulation via uptake by the nerve terminals followed by retrograde transport in the axons, or their presence in the nerves could be a consequence of the synthesis and processing of the P O M C prohormone in the neurones. If they are synthesised in the motoneurone the increased proportion of cells with the peptides, seen after nerve damage or in disease, could be due to (i) increased transcription of the P O M C m R N A , (ii) increased post-translational processing of POMC, (iii) decreased release resulting in accumulation of the peptides or (iv) defective axonal transport.
42
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
Here we use immunocytochemistry to demonstrate the time course of the increase in POMC-derived peptide immunoreactivity in the motoneurone perikarya after
motor nerve section and a histochemical in situ hybridization method to investigate whether the increases are due to increased transcription of POMC and syn-
7
•
~ i?~L:~j!!
L ~ ; ¸ ' ,~'~
Fig. 1. A: section from an unoperated mouse, after incubation with /3-EP antibody. Only a few faintly stained cells are visible (arrows). B-F: sections from spinal cords of mice which had undergone unilateral nerve section 24 h previously; (B) ipsilateral ventral horn and (C) contralateral ventral horn, after incubation with an antibody to /3-EP, (D) ipsilateral ventral horn after incubation with an antibody to a-MSH. Intensely stained immunoreactive cells are visible in B, C and D (arrows). E,F: sections incubated with the/3-EP and a-MSH antibodies which had been preadsorbed with /3-EP (E) and a-MSH (F), respectively. No stained cells are visible in E or F. Bars = 100/~m.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
thesis of the peptides in the ventral horn motoneurones. An abstract describing part of this work has already been published [13].
43
sections were examined, the cells were stained with toluidine blue [16] to determine the total number of cells present in the sections (although 'unstained' motoneurones were actually visible in the immunostained sections and in the hybridized sections). The cells were counted by eye using a hand counter. Only cells with a visible nucleus were included in the counts.
2. Materials and methods 2.2. lmmunocytochemistry 2.1. Animals Mice of the C57BL/6J strain, 10-12 weeks of age, were used. The sciatic nerve was sectioned unilaterally in the midthigh region, under halothane anaesthesia. In control 'sham-operated' mice the same unilateral operative procedure was followed except that the sciatic nerve was not transected. The mice were killed by cervical dislocation at 3, 6, 24 or 48 h, or 7 days after the operation and part of the spinal cord, including segments T13 to L2, was removed. The tissues were washed thoroughly in phosphate-buffered saline (0.1 M) containing phenylmethylsulphonyl fluoride (0.1 /xM) and cyclohexamide (0.1 ttM), pH 7.4, and quickly frozen in isopentane cooled in liquid nitrogen. Every third section was examined for the presence of motoneurones which stained for peptide antibody or POMC transcript. The results were expressed as the number of cells which stained for peptide immunoreactivity or POMC mRNA as a proportion of the total number of cells from the left or the right ventral horn in spinal cord sections. In some experiments, in which serial
Twenty/xm cryostat sections were prepared from the spinal cord segments T13 to L2 and immunoreactivity for/3-EP and a-MT was detected in ventral horn cells using the indirect peroxidase-antiperoxidase technique. The immunocytochemistry was performed using antibodies to /3-EP and to a-MSH as described previously [11]. Control sections were incubated with either pre-immune rabbit serum, or with the specific antiserum which had been preadsorbed for 18 h with 20/.tg/ml of the appropriate antigen, in order to detect non-specific staining. No immunostaining was observed with either the/3-EP or the a-MSH antiserum after these procedures indicating that non-specific staining of the nervous tissue was not being detected.
2.3. In situ hybridization Synthetic cDNA oligonucleotide probes complementary to the ACTH 4-11 and the fl-EP 1-8 encoding regions of rat POMC
Fig. 1 (continued).
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
44
m R N A (Affiniti Research products LTD, Nottingham, UK) were used to detect P O M C transcript via in situ hybridization. The 24 base c D N A oligonucleotides used were: 5' CTT G C C C C A G C G G A A G T G CTC C A T 3' (antisense strand) which probes for the A C T H 4 - l l encoding region of P O M C m R N A and the corresponding sense strand 5'ATG G A G C A C T T C C G C T G G G G C A A G 3', and 5' CTC G G A G G T C A T G A A G C C G C C G T A 3' (antisense strand) which probes for the /3-EP 1-8 encoding region and the corresponding sense strand 5' T A C G G C G G C TTC A T G A C C TCC G A G 3' [1]. The oligonucleotides were covalently conjugated to calf intestinal alkaline phosphatase and the presence of P O M C transcript in ventral horn cells was detected using the colorimetric histochemical method of McGadey [17] to visualise the enzyme reaction product. The alkaline phosphatase-conjugated sense strands were used as specific controls for the corresponding probes. Twenty ~.m transverse cryostat sections were prepared from spinal cord segments T13 to L2. The sections were prehybridized in a solution of 0.3 M NaCI, 0.03 M trisodium citrate solution (SSC) containing 50% formamide, for 1 h at 37°C. They were then incubated with the hybridization mixture containing the probe conjugate (antisense strand, 20 pM). The hybridization mixture contained TrisHCI (10 raM, pH 7.5), Denhardt's solution (12.5 x concentrate), 0.3 M NaCI, 0.03 M trisodium citrate, sodium pyrophosphate (5 m g / m l ) , 0.5% SDS, 50% deionised formamide, 10% dextran sulphate and sheared salmon sperm (250 p~g/ml). The sections were covered with dimethyldichlorosilane-coated coverslips and the sections were incubated in a sealed humid container overnight at 42°C. The coverslips were then removed in a solution of NaC1 (0.6 M) and sodium citrate (0.06 M). The slides were then washed in two changes of SSC (10 rain at room temperature) and then in 5% SSC ( 2 × 2 0 min at 42°C), and then in SSC for 10 rain at room temperature, and then incubated in a solution consisting of 7.5 ml of substrate buffer (50 m M Tris, 100 mM NaCI, 50 m M MgC12, pH 9.5 containing 80 m g / m l levamisole) + 33 #1 Nitro-blue tetrazolium solution (75 m g / m l in 70% dimethylf o r m a m i d e ) + 2 5 p.l 5-bromo-4-chloro-3-indolyl phosphate solution (50 m g / m l in dimethylformamide) for 4 h. The progress of the reaction was checked under the microscope. The control sections were treated similarly but the conjugated sense strand was used instead of the conjugated antisense strand. The sections were mounted with aquamount.
50
40 kJ s0 ka
v E
Io
o 0
Ik2
2'4 3; 4'8 Time After Nerve Section (h)
6'0
I;8
Fig. 2. Time course of the expression of (i) a - M S H and ( 0 ) / 3 - E P in the ipsilateral ventral horn after motor nerve section. The results are expressed as the percentage of the total number of motoneurones which were immunoreactive for the peptides. The values are given as the mean ( ± S.E.M.)for the results from at least 3 mice in each case.
50-
40
-~- 30-
2O
~
10
0 1'2
2~4
3'6
48
60
1;8
T i m e After Nerve Section (h) Fig. 3. Time course of the expression of (i) a - M S H immunoreactivity and ( o ) /3-EP immunoreactivity in the contralateral ventral horn after motor nerve section. The results are expressed as the percentage of the total number of motoneurones which were immunoreactive for the peptides. The values are given as the mean ( + S.E.M.) for the results from at least 3 mice in each case.
3. Results In unoperated mice, a few of the ventral horn cells were slightly immunoreactive for/7-EP or a-MSH. Fig. 1A shows a section from a control unoperated mouse which had been incubated with an antibody to /3-EP. The incidence of immunoreactive cells increased after nerve section. Fig. 1B, C and D show typical sections of spinal cord from mice which had undergone sciatic nerve section 24 h previously. The sections had been incubated with an antibody to /3-EP (Fig. 1B,C) or a-MSH (Fig. 1D). Intense staining for the peptide antibodies is present in some cells in the ipsilateral ventral horn (Fig. 1B,D). Both large diameter and small diameter motoneurones exhibited the immunoreactivity. Fig. 2 shows the time course for the increase in expression of the peptides after nerve section. The incidence of immunoreactive cells increased up to 24 h after nerve section and thereafter declined. At 7 days after the section, the incidence of immunoreactive cells was still higher than in the unoperated mice. The findings with the two different peptide antibodies were similar. There was also an increase in the proportion of immunoreactive cells in the contralateral ventral horn after nerve section (see Fig. 1C) and the time course paralleled that for the ipsilateral horn (Fig. 3). There were no stained cells in sections which had been incubated with preimmune serum or antibody preadsorbed with peptide (Fig. 1E,F) indicating that it was not nonspecific staining that was being detected. Very few motoneurones expressed the POMC m R N A in spinal cord sections from unoperated or sham operated mice and only faint staining was visible in any of those cells (see Fig. 4A). After nerve section,
s. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49 h o w e v e r , t h e r e was a d r a m a t i c i n c r e a s e in t he incid e n c e o f cells wh i c h e x p r e s s e d t h e transcript. Fig. 4B and C show sections o f spinal cords f r o m a m o u s e
45
w h i ch h ad u n d e r g o n e u n i l a t e r a l sciatic n e r v e section 24 h p r i o r to t h e e x p e r i m e n t . T h e sections ha d b e e n h y b r i d i z e d with t h e p r o b e to t h e A C T H 4 - 1 1 r e g i o n o f
~p~r~'¸~~
Fig. 4. Sections of spinal cord from mice 24 h after (A) unilateral sham operation, incubated with the probe for the ACTH 4-11 region of POMC. A few cells which stained faintly for the hybrid are visible (arrows). B,C: unilateral nerve transection, incubated with the probe for ACTH 4-11 region of POMC (B) or the probe for the/3-EP 1-8 region of POMC, ipsilateral side (C). LHS transected side, RHS contralateral unoperated side. Cells which stained intensely for the hybrid are visible (arrows). D,E: unilateral nerve section, incubated with the sense strands for the ACTH 4-11 region of POMC (D) or the/3-EP 1-8 region of POMC, ipsilateral side (E). Bars = 100/zm.
46
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
Fig. 4 (continued).
P O M C (Fig. 4B) or the probe to the /3-EP 1-8 region of P O M C (Fig. 4C). Intense staining can be seen in some ventral horn motoneurones indicating the presence of P O M C transcript. Both small and large diameter motoneurones stained for the m R N A . The incidence of cells which expressed the P O M C transcript in the ipsilateral ventral horn increased up to 24 h after the section and then declined. Seven days after the section the incidence of motoneurones which stained for the hybrid had declined to the level seen in the unoperated mice. Some of the cells in the contralateral ventral horn also stained for P O M C transcript. The incidence of stained cells in the contralateral ventral horn was similar to that in the ipsilateral ventral horn. Interestingly the stained cells did not appear to be in the same locations in the two ventral horns (Fig. 4B,C). Figs. 5 and 6 show the time course for the expression of the m R N A , using the probe to A C T H 4-11 and /3-EP 1-8, in the ipsilateral and contralateral ventral horns, respectively, after nerve section. There was a significant increase in the inci-
50
40 (O 30 0-
<7 2o rr E 10-
o
1'2
~4
~e
~8
~o
tl
,
168
Time After Nerve Section (h)
Fig. 5. Time course of the expression of POMC m R N A in ipsilateral ventral spinal cord after nerve section, using the probe for the ACTH 4-11 region (o) or the /3-EP 1-8 region (©). The results are expressed as the percentage of the total number of motoneurones which expressed POMC mRNA. The values are given as the mean (+_ S.E.M.) for the results from at least 3 mice in each case.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49 40-
_~ 30.
no2o. Z hE
~10.
¢2
~4
~6
~8
do II
,
168
Time After Nerve Section (h)
Fig. 6. Time course of the expression of P O M C m R N A in contralateral ventral spinal cord after nerve section using the probe for the A C T H 4 - 1 0 region (e) or the fl-EP 1 - 8 region (©). T h e results are expressed as the percentage of the total n u m b e r of motoneurones which expressed P O M C m R N A . T h e values are given as the mean (5: S.E.M.) for the results from at least 3 mice in each case.
dence of cells with the m R N A compared to the unoperated mice at 3 h after the nerve transection. The maximum increase in both the ipsilateral and contralateral horns was seen, with both probes, at 24 h after nerve section. The incidence thereafter declined and 7 days after the section the incidence on both sides was similar to that seen in the unoperated mice. The resuits obtained with the two different probes were similar. In sections from sham-operated mice examined at 6 and 24 h after the operation the proportions of cells which stained for POMC m R N A (using the probe to the A C T H 4-11 region of POMC) in the ipsilateral and contralateral ventral horns were not significantly different to the proportion in unoperated mice. No staining was seen in control sections, from the operated, sham-operated, or unoperated mice, which had been incubated with either of the conjugated sense strands (Fig. 4D,E).
4. Discussion
It has previously been shown in this laboratory that immunoreactivity for the POMC-derived peptides/3-EP and a - M S H can be detected in axons of mechanically damaged or diseased motor nerves as well as in chemically intoxicated motor nerves. An increased incidence of peptide-immunoreactive ventral horn cells were also seen in murine muscular dystrophy in mice [7] and in rats treated with IDPN [15], a neurotoxicant which targets motoneurones Furthermore we have shown increases in/3-EP and a - M S H immunoreactivity in intra-
47
muscular nerves in many conditions where motor neuropathy may be present. Edwards et al. [5] also demonstrated the presence of an a-MSH-like peptide in crushed motoneurones but they speculated that it was due to a degeneration product of neurofilament protein which shares a common amino acid sequence with a-MSH, enabling it to cross-react with antibodies to the peptide [3]. Oren et al. [18] demonstrated a-MSHlike immunoreactivity associated with neurofilaments in frog motor nerve axons. However, immunoreactivity for /3-EP which is formed from a different region of POMC is also expressed in the ventral horn cells [7]. It seemed more likely therefore that processing of POMC occurs in these neurones and that A C T H as well as /~-EP peptides are produced via processing of the prohormone in the nerves, rather than from neurofilament protein. In this study we used in situ hybridization to demonstrate POMC m R N A and to demonstrate its localisation in the ventral horn motoneurones thereby indicating that POMC-derived peptides can be synthesised in the motoneurones. We used oligoprobes complementary to two different regions of POMC m R N A to demonstrate the presence of POMC transcript. It seems likely therefore that it was indeed POMC m R N A which was detected and not the m R N A for some other protein which has a common amino acid sequence. Plantinga et al. [19], using polymerase chain reaction and northern blot analysis demonstrated the presence of low levels of POMC transcript in the T13 to L2 region of the rat spinal cord. They could not demonstrate increased synthesis of POMC m R N A in the ventral spinal cord after sciatic nerve crush. It is possible that the expression of the peptides in the rat ventral horn is less marked than that in the mouse. Indeed in earlier work where we investigated peptide immunoreactivity in intramuscular nerves, we observed a markedly lower incidence of immunoreactive nerves after nerve section in the rat [11] than in the mouse [10]. There are however, other reasons why the findings of their study and ours might be different. For example nerve crush is a relatively mild insult to the neurone compared to nerve transection and the pathological changes are less marked and more short-lived. In addition several biochemical procedures were necessary in their study and incomplete recovery of the nucleotide hybrids may therefore have resulted. It is incidently important to note that the sensitive histochemical method used in the present study enabled the m R N A to be precisely localised to the perikarya so that the motoneurone could be positively identified as the site of synthesis of POMC-derived peptides. The incidence of both peptide-immunoreactive cells and mRNA-positive cells showed parallel increases up to 24 h after nerve section and the incidence of both declined thereafter. The time course of the m R N A expression and of the peptide expression was initially
48
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
similar, the time interval between maximum transcription and maximum peptide synthesis indicating a difference of a few hours at most between transcription and synthesis of the peptides. At 7 days after nerve section the proportion of cells expressing the m R N A had returned to the levels in the unoperated controls. Interestingly, however, at this time after nerve section, the proportion of immunoreactive cells was still appreciably higher than in the control mice. The apparent delay in the time course of the expression of the immunoreactivity compared to that of the synthesis of the transcript at later times may be due to the slow rate of transport of the peptides from the perikarya to the axons after their synthesis in the motoneurones. It is not clear why increased peptide expression occurs in motoneurones of the contralateral ventral horn after nerve section. However previous work has shown that immunoreactivity for the peptides could be detected in contralateral intramuscular nerves after nerve section. Evidence from experiments where the nerve section was located at different distances from the spinal cord indicated that the contralateral effect was due to a transneuronal signal originating at the site of injury and travelling to the contralateral motoneurones via the spinal cord [11]. The results of the present study showing staining apparently in motoneurones in different locations in the two horns indicate that if this is the case then the 'crosstalk' between neurones in the two ventral horns of the spinal cord, at least at the same segmental level, may involve neurones of different muscle pools in the two sides. It is not known whether there are direct connections between motoneurones in the two ventral horns at the same segmental level and the nature of the message transferred is also unclear. The POMC-derived peptides probably act in concert to exert a t r o p h i c control of a number of different processes at the developing or regenerating neuromuscular junction. These processes include nerve growth [23,4], maturation of the endplate [6], production of the different acetylcholinesterase molecular forms [9], development of muscle strength [22] and neuromuscular transmission [2]. We have shown that at least one POMC-derived peptide,/3-EP, is released upon electrical stimulation of the motor nerve in the developing rat [8]. Furthermore receptors for these peptides were demonstrated in skeletal muscles [21]. It seems possible therefore that the trophic effects are mediated by the neuronal peptides, probably in conjunction with the circulating peptides. The studies of Strand and Kung [23] and Edwards et al. [4] have shown that exogenous ACTH-derived peptides can increase the rate of nerve regeneration after sciatic nerve transection. The time course of the upregulation of the P O M C gene shown in the present study is very much shorter than that for re-innervation after
motor nerve section, having returned to normal levels by 7 days after the section. Thus the suppression of P O M C transcription in the motoneurones is not controlled solely by the target muscle. The brevity of the time course is interesting in the light of the findings of others [20,24] that the susceptibility of nerve regeneration to the synthetic A C T H derivative, O R G 2766, is restricted to the first few days after nerve injury. However, the actions of peptides on nerve growth at late or early times after nerve injury may depend partly on their uptake by the nerve terminals from the blood or the medium surrounding the nerve and administered exogenous peptides could gain access to the nerves from the blood. In previous work increases in P O M C peptide immunoreactivity were seen in the distal stump or 'ghost' of the transected sciatic nerve after the axonal material (investigated using a silver staining technique) had disappeared, in both mice and rats [10,11]. Furthermore, in the rat the time course of the expression of the peptides in the intramuscular nerve terminals of both the transected and contralateral sides was more prolonged than that of the time course of the expression of P O M C m R N A in the mouse perikarya in the present study. Thus there was an initial increase up to approximately 24 h followed by a slight decrease to a plateau which remained high for at least 12 days [11]. In this case the immunoreactivity at later times could reflect a specis difference or it could have been due partly to uptake of the peptides from the blood. On the other hand the ventral horn cell may not be the only relevant site where increased synthesis of P O M C occurs after nerve damage. It is also interesting in this context that in the normal unoperated mouse immunoreactivity was present in a higher proportion of the nerve terminals than the proportion seen in the ventral horn cells of unoperated mice in the present study. It is possible therefore that increased expression of P O M C also occurs in extraneuronal elements such as Schwann cells or macrophages. POMC-derived peptides synthesised in elements at the neuromuscular junction (or circulating in the blood) may take over the functions of the neuronally synthesised peptides after they are no longer synthesised in the nerves. Our results indicate that damage to the motor nerve can result in increased synthesis of the P O M C precursor and the POMC-derived peptides. It seems likely that the peptides are synthesised at these locations whenever their actions are required to facilitate the repair processes or to augment neuromuscular transmission.
Acknowledgements We are grateful to the Wellcome Trust and the Motor Neurone Disease Association for supporting this work.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
References [1] Chang, A.C.Y., Cochet, M. and Cohen, S.N., Structural organisation of human genomic DNA encoding the pro-opiomelanocortin peptide, Proc. Natl. Acad. Sci. USA, 77 (1980) 4890-4894. [2] Davies, D.A. and Smith, M.E., ACTH (4-10) increases increases quantal content at the mouse neuromuscular junction, Brain Res., 637 (1994) 328-330. [3] Drager, U.C., Edwards, L.D. and Kleinschmidt, J., Neurofilaments contain a-melanocyte-stimulating hormone (a-MSH)-like immunoreactivity, Proc. Natl. Acad. Sci. USA, 80 (1983) 64086412. [4] Edwards, P.M., Kuiters, R.R.F., Boer, G.J. and Gispen, W.H., Recovery from peripheral nerve transection is accelerated by local application of a-MSH by means of microporous Accurel polypropylene tubes, J. Neurol. Sci., 74 (1986) 171-176. [5] Edwards, P.M., Van der Zee, C.E.E.M., Verhaagen, J., Schotman, P., Jennekens, F.G.I. and Gispen, W.H., Evidence that the neurotrophic actions of ot-MSH may derive from its ability to mimic the actions of a peptide formed in degenerating nerve stumps, J. Neurol. Sci., 64 (1984) 333-340. [6] Frischer, R.E., E1-Kawa, N.M. and Strand, F.L., ACTH peptides as organisers of neuronal patterns in development: maturation of the rat neuromuscular junction as seen by scanning electron microscopy, Peptides, 6, Suppl. 2 (1985) 13-18. [7] Haynes, L.W. and Smith, M.E., Presence of immunoreactive a-melanotropin and/3-endorphin in spinal motoneurones of the dystrophic mouse, Neurosci. Lett., 58 (1985) 13-18. [8] Haynes, L.W., Smith, M.E. and Li, C.H., The regulation by /3-endorphin and related peptides of collagen-tailed acetylcholinesterase forms in the skeletal muscles of vertebrates. In G.B. Stefano (Ed.), CRC Handbook of Comparative Aspects of Opioid and Related Neuropeptides Mechanisms, Vol. II, Ch. 31, CRC Press, San Francisco, 1986, pp. 65-79. [9] Haynes, L.W., Smith, M.E. and Smyth, D.G., Evidence for the neurotrophic regulation of collagen-tailed acetylcholinesterase in immature skeletal muscle by/3-endorphin, J. Neurochem., 42 (1984) 1542-1551. [10] Hughes, S. and Smith, M.E., Effect of nerve section on /3-endorphin and a-melanotropin immunoreactivity in motoneurones of normal and dystrophic mice, Neurosci. Lett., 92 (1988) 1-7. [11] Hughes, S. and Smith, M.E., Pro-opiomelanocortin-derived pep-
49
tides in transected and contralateral motor nerves of the rat, J. Chem. Neuroanat., 2 (1989) 227-237. [12] Hughes, S. and Smith, M.E., Proopiomelanocortin-derived peptides in mice with motoneurone disease, Neurosci. Lett., 103 (1989) 169-173. [13] Hughes, S and Smith, M.E., Expression of the POMC gene in motoneurones of the mouse, J. Physiol., 473 (1993) 40P [14] Hughes, S., Smith, M.E. and Bailey, C.J., POMC-derived peptides in the neuromuscular system of streptozotocin-diabetic mice, Peptides, 13 (1992) 873-877. [15] Hughes, S., Smith, M.E., Simpson, M.G. and Allen, S.L., Effect of IDPN on the expression of POMC-derived peptides in rat motoneurones, Peptides, 13 (1992) 1021-1023. [16] Kramer, H and Windrum, G.M., The metachromatic staining reaction, J. Histochem. Cytochem., 3 (1955) 227-237. [17] McGadey, J., A tetrazolium method for non-specific alkaline phosphatase, Histochemie, 23 (1970) 180-184. [18] Oren, N., Micevych, P.E. and Letinsky, M.S., Presence of amelanocyte-stimulating hormone-like immunoreactivity in the innervation of amphibian skeletal muscle, Z Neurosci. Res., 23 (1989) 225-233. [19] Plantinga, L.C., Verhaagen, J., Edwards, P.M., Schrama, L.H., Burbach, J.P.H. and Gispen, W.H., Expression of the proopiomelanocortin gene in dorsal root ganglia, spinal cord and sciatic nerve after sciatic nerve crush in the rat, Mol. Brain Res., 16 (1992) 135-142. [20] Saint-Come, C. and Strand, F.L., ACTH 4-9 analogue (Org 2766) improves qualitative and quantitative aspects of motor nerve regeneration, Peptides, 8, Suppl. 1 (1988) 215-221. [21] Smith, M.E. and Hughes, S., Pro-opiomelanocortin neuropeptide receptors on developing and dystrophic muscle fibres, MoL Chem. Neuropathol., 19 (1993) 137-145. [22] Smith, C.M. and Strand, F.L., Neuromuscular response of the immature rat to A C T H / M S H 4-10, Peptides, 2 (1981) 197-206. [23] Strand, F.L. and Kung, T.T., ACTH accelerates recovery of neuromuscular function following crushing of peripheral nerve, Peptides, 1 (1980) 135-138. [24] Verhaagen, J., Edwards, P.M.,Jennekens, F.G.I., Schotman, P. and Gispen, W.H., Early effect of an ACTH 4-9 analog (Org 2766) on regenerative sprouting demonstrated by the use of neurofilament binding antibodies isolated from a serum raised by a-MSH immunization, Brain Res., 404 (1987) 147-150.
Molecular Brain Research 25 (1994) 41-49
Research Report
Upregulation of the pro-opiomelanocortin gene in motoneurones after nerve section in mice Sharon Hughes, Margaret E. Smith * Department of Physiology, Medical School, University of Birmingham, Birmingham B15 2T~, UK
Accepted 1 March 1994
Abstract In the normal adult mouse a few of the ventral horn motoneurones were slightly immunoreactive for the pro-opiomelanocortin (POMC)-derived peptides/3-endorphin (/3-EP) and ~-melanotropin (a-MSH). After unilateral section of the sciatic nerve the incidence of immunoreactive cells and the intensity of staining increased in both the ipsilateral and the contralateral ventral horns. The proportion of cells which expressed the peptides reached a maJximum at approximately 24 h after nerve section, and thereafter declined, but the proportion was still higher than normal at 7 days after the section. Using histochemical in situ hybridization with cDNA oligonucleotide probes for POMC transcript regions which encode for adrenocorticotropin (ACTH) 4-11 and/3-endorphin 1-8, POMC mRNA was detected in very few motoneurones in unoperated or sham operated mice. After unilateral sciatic nerve section increases in the proportion of cells w.hich expressed the POMC mRNA were seen in both the ipsilateral and contralateral ventral horns. The increase, like that seen for peptide immunoreactivity, was maximum at around 24 h and thereafter declined but at 7 days after the section the proportion of POMC mRNA-positive cells had returned to normal. Thus injury to the motoneurones is accompanied by synthesis of POMC-derived peptides via upregulation of the POMC gene in the motoneurone. Key words: Motoneuron; Pro-opiomelanocortin; In situ hybridization;/3-Endorphin; t~-Melanotropin; Neuropeptide
I. Introduction In previous work we used immunocytochemistry to demonstrate the presence of the pro-opiomelanocortin (POMC)-derived peptides/3-endorphin (/3-EP) and am e l a n o t r o p i n ( a - M S H ) or a d r e n o c o r t i c o t r o p i n , (ACTH), the parent of a-MSH, in motor nerve terminals and axons of rodents. In the normal adult very few motoneurones were immunoreactive for the peptides but after nerve transection [11] or in diseases where there is a primary motor neuropathy such as motor neurone disease [12] or a secondary peripheral neuropathy such as diabetes mellitus [14] or murine muscular dystrophy [10] there were significantly higher incidences of immunoreactive motor nerve axons. It was shown that in mice with muscular dystrophy ( C 5 7 B L / 6 J strain), in which there is a motor nerve defect, and after administration in rats of the neurotoxicant /3,if-
* Corresponding author. Fax: (44) 21-414-6924. 0169-328X/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0169-328X(94)00042-D
iminodipropionitrile (IDPN) which targets motoneurones there were increases in the incidence of peptideimmunoreactive ventral horn cells as well as immunoreactive nerve axons [7,15]. These peptides are trophic factors which promote neuromuscular transmission and nerve regeneration and they may be expressed as part of the regenerative response in damaged or diseased motoneurones whatever the cause of the lesion. The peptides in the motor nerves could have originated from the circulation via uptake by the nerve terminals followed by retrograde transport in the axons, or their presence in the nerves could be a consequence of the synthesis and processing of the P O M C prohormone in the neurones. If they are synthesised in the motoneurone the increased proportion of cells with the peptides, seen after nerve damage or in disease, could be due to (i) increased transcription of the P O M C m R N A , (ii) increased post-translational processing of POMC, (iii) decreased release resulting in accumulation of the peptides or (iv) defective axonal transport.
42
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
Here we use immunocytochemistry to demonstrate the time course of the increase in POMC-derived peptide immunoreactivity in the motoneurone perikarya after
motor nerve section and a histochemical in situ hybridization method to investigate whether the increases are due to increased transcription of POMC and syn-
7
•
~ i?~L:~j!!
L ~ ; ¸ ' ,~'~
Fig. 1. A: section from an unoperated mouse, after incubation with /3-EP antibody. Only a few faintly stained cells are visible (arrows). B-F: sections from spinal cords of mice which had undergone unilateral nerve section 24 h previously; (B) ipsilateral ventral horn and (C) contralateral ventral horn, after incubation with an antibody to /3-EP, (D) ipsilateral ventral horn after incubation with an antibody to a-MSH. Intensely stained immunoreactive cells are visible in B, C and D (arrows). E,F: sections incubated with the/3-EP and a-MSH antibodies which had been preadsorbed with /3-EP (E) and a-MSH (F), respectively. No stained cells are visible in E or F. Bars = 100/~m.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
thesis of the peptides in the ventral horn motoneurones. An abstract describing part of this work has already been published [13].
43
sections were examined, the cells were stained with toluidine blue [16] to determine the total number of cells present in the sections (although 'unstained' motoneurones were actually visible in the immunostained sections and in the hybridized sections). The cells were counted by eye using a hand counter. Only cells with a visible nucleus were included in the counts.
2. Materials and methods 2.2. lmmunocytochemistry 2.1. Animals Mice of the C57BL/6J strain, 10-12 weeks of age, were used. The sciatic nerve was sectioned unilaterally in the midthigh region, under halothane anaesthesia. In control 'sham-operated' mice the same unilateral operative procedure was followed except that the sciatic nerve was not transected. The mice were killed by cervical dislocation at 3, 6, 24 or 48 h, or 7 days after the operation and part of the spinal cord, including segments T13 to L2, was removed. The tissues were washed thoroughly in phosphate-buffered saline (0.1 M) containing phenylmethylsulphonyl fluoride (0.1 /xM) and cyclohexamide (0.1 ttM), pH 7.4, and quickly frozen in isopentane cooled in liquid nitrogen. Every third section was examined for the presence of motoneurones which stained for peptide antibody or POMC transcript. The results were expressed as the number of cells which stained for peptide immunoreactivity or POMC mRNA as a proportion of the total number of cells from the left or the right ventral horn in spinal cord sections. In some experiments, in which serial
Twenty/xm cryostat sections were prepared from the spinal cord segments T13 to L2 and immunoreactivity for/3-EP and a-MT was detected in ventral horn cells using the indirect peroxidase-antiperoxidase technique. The immunocytochemistry was performed using antibodies to /3-EP and to a-MSH as described previously [11]. Control sections were incubated with either pre-immune rabbit serum, or with the specific antiserum which had been preadsorbed for 18 h with 20/.tg/ml of the appropriate antigen, in order to detect non-specific staining. No immunostaining was observed with either the/3-EP or the a-MSH antiserum after these procedures indicating that non-specific staining of the nervous tissue was not being detected.
2.3. In situ hybridization Synthetic cDNA oligonucleotide probes complementary to the ACTH 4-11 and the fl-EP 1-8 encoding regions of rat POMC
Fig. 1 (continued).
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
44
m R N A (Affiniti Research products LTD, Nottingham, UK) were used to detect P O M C transcript via in situ hybridization. The 24 base c D N A oligonucleotides used were: 5' CTT G C C C C A G C G G A A G T G CTC C A T 3' (antisense strand) which probes for the A C T H 4 - l l encoding region of P O M C m R N A and the corresponding sense strand 5'ATG G A G C A C T T C C G C T G G G G C A A G 3', and 5' CTC G G A G G T C A T G A A G C C G C C G T A 3' (antisense strand) which probes for the /3-EP 1-8 encoding region and the corresponding sense strand 5' T A C G G C G G C TTC A T G A C C TCC G A G 3' [1]. The oligonucleotides were covalently conjugated to calf intestinal alkaline phosphatase and the presence of P O M C transcript in ventral horn cells was detected using the colorimetric histochemical method of McGadey [17] to visualise the enzyme reaction product. The alkaline phosphatase-conjugated sense strands were used as specific controls for the corresponding probes. Twenty ~.m transverse cryostat sections were prepared from spinal cord segments T13 to L2. The sections were prehybridized in a solution of 0.3 M NaCI, 0.03 M trisodium citrate solution (SSC) containing 50% formamide, for 1 h at 37°C. They were then incubated with the hybridization mixture containing the probe conjugate (antisense strand, 20 pM). The hybridization mixture contained TrisHCI (10 raM, pH 7.5), Denhardt's solution (12.5 x concentrate), 0.3 M NaCI, 0.03 M trisodium citrate, sodium pyrophosphate (5 m g / m l ) , 0.5% SDS, 50% deionised formamide, 10% dextran sulphate and sheared salmon sperm (250 p~g/ml). The sections were covered with dimethyldichlorosilane-coated coverslips and the sections were incubated in a sealed humid container overnight at 42°C. The coverslips were then removed in a solution of NaC1 (0.6 M) and sodium citrate (0.06 M). The slides were then washed in two changes of SSC (10 rain at room temperature) and then in 5% SSC ( 2 × 2 0 min at 42°C), and then in SSC for 10 rain at room temperature, and then incubated in a solution consisting of 7.5 ml of substrate buffer (50 m M Tris, 100 mM NaCI, 50 m M MgC12, pH 9.5 containing 80 m g / m l levamisole) + 33 #1 Nitro-blue tetrazolium solution (75 m g / m l in 70% dimethylf o r m a m i d e ) + 2 5 p.l 5-bromo-4-chloro-3-indolyl phosphate solution (50 m g / m l in dimethylformamide) for 4 h. The progress of the reaction was checked under the microscope. The control sections were treated similarly but the conjugated sense strand was used instead of the conjugated antisense strand. The sections were mounted with aquamount.
50
40 kJ s0 ka
v E
Io
o 0
Ik2
2'4 3; 4'8 Time After Nerve Section (h)
6'0
I;8
Fig. 2. Time course of the expression of (i) a - M S H and ( 0 ) / 3 - E P in the ipsilateral ventral horn after motor nerve section. The results are expressed as the percentage of the total number of motoneurones which were immunoreactive for the peptides. The values are given as the mean ( ± S.E.M.)for the results from at least 3 mice in each case.
50-
40
-~- 30-
2O
~
10
0 1'2
2~4
3'6
48
60
1;8
T i m e After Nerve Section (h) Fig. 3. Time course of the expression of (i) a - M S H immunoreactivity and ( o ) /3-EP immunoreactivity in the contralateral ventral horn after motor nerve section. The results are expressed as the percentage of the total number of motoneurones which were immunoreactive for the peptides. The values are given as the mean ( + S.E.M.) for the results from at least 3 mice in each case.
3. Results In unoperated mice, a few of the ventral horn cells were slightly immunoreactive for/7-EP or a-MSH. Fig. 1A shows a section from a control unoperated mouse which had been incubated with an antibody to /3-EP. The incidence of immunoreactive cells increased after nerve section. Fig. 1B, C and D show typical sections of spinal cord from mice which had undergone sciatic nerve section 24 h previously. The sections had been incubated with an antibody to /3-EP (Fig. 1B,C) or a-MSH (Fig. 1D). Intense staining for the peptide antibodies is present in some cells in the ipsilateral ventral horn (Fig. 1B,D). Both large diameter and small diameter motoneurones exhibited the immunoreactivity. Fig. 2 shows the time course for the increase in expression of the peptides after nerve section. The incidence of immunoreactive cells increased up to 24 h after nerve section and thereafter declined. At 7 days after the section, the incidence of immunoreactive cells was still higher than in the unoperated mice. The findings with the two different peptide antibodies were similar. There was also an increase in the proportion of immunoreactive cells in the contralateral ventral horn after nerve section (see Fig. 1C) and the time course paralleled that for the ipsilateral horn (Fig. 3). There were no stained cells in sections which had been incubated with preimmune serum or antibody preadsorbed with peptide (Fig. 1E,F) indicating that it was not nonspecific staining that was being detected. Very few motoneurones expressed the POMC m R N A in spinal cord sections from unoperated or sham operated mice and only faint staining was visible in any of those cells (see Fig. 4A). After nerve section,
s. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49 h o w e v e r , t h e r e was a d r a m a t i c i n c r e a s e in t he incid e n c e o f cells wh i c h e x p r e s s e d t h e transcript. Fig. 4B and C show sections o f spinal cords f r o m a m o u s e
45
w h i ch h ad u n d e r g o n e u n i l a t e r a l sciatic n e r v e section 24 h p r i o r to t h e e x p e r i m e n t . T h e sections ha d b e e n h y b r i d i z e d with t h e p r o b e to t h e A C T H 4 - 1 1 r e g i o n o f
~p~r~'¸~~
Fig. 4. Sections of spinal cord from mice 24 h after (A) unilateral sham operation, incubated with the probe for the ACTH 4-11 region of POMC. A few cells which stained faintly for the hybrid are visible (arrows). B,C: unilateral nerve transection, incubated with the probe for ACTH 4-11 region of POMC (B) or the probe for the/3-EP 1-8 region of POMC, ipsilateral side (C). LHS transected side, RHS contralateral unoperated side. Cells which stained intensely for the hybrid are visible (arrows). D,E: unilateral nerve section, incubated with the sense strands for the ACTH 4-11 region of POMC (D) or the/3-EP 1-8 region of POMC, ipsilateral side (E). Bars = 100/zm.
46
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
Fig. 4 (continued).
P O M C (Fig. 4B) or the probe to the /3-EP 1-8 region of P O M C (Fig. 4C). Intense staining can be seen in some ventral horn motoneurones indicating the presence of P O M C transcript. Both small and large diameter motoneurones stained for the m R N A . The incidence of cells which expressed the P O M C transcript in the ipsilateral ventral horn increased up to 24 h after the section and then declined. Seven days after the section the incidence of motoneurones which stained for the hybrid had declined to the level seen in the unoperated mice. Some of the cells in the contralateral ventral horn also stained for P O M C transcript. The incidence of stained cells in the contralateral ventral horn was similar to that in the ipsilateral ventral horn. Interestingly the stained cells did not appear to be in the same locations in the two ventral horns (Fig. 4B,C). Figs. 5 and 6 show the time course for the expression of the m R N A , using the probe to A C T H 4-11 and /3-EP 1-8, in the ipsilateral and contralateral ventral horns, respectively, after nerve section. There was a significant increase in the inci-
50
40 (O 30 0-
<7 2o rr E 10-
o
1'2
~4
~e
~8
~o
tl
,
168
Time After Nerve Section (h)
Fig. 5. Time course of the expression of POMC m R N A in ipsilateral ventral spinal cord after nerve section, using the probe for the ACTH 4-11 region (o) or the /3-EP 1-8 region (©). The results are expressed as the percentage of the total number of motoneurones which expressed POMC mRNA. The values are given as the mean (+_ S.E.M.) for the results from at least 3 mice in each case.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49 40-
_~ 30.
no2o. Z hE
~10.
¢2
~4
~6
~8
do II
,
168
Time After Nerve Section (h)
Fig. 6. Time course of the expression of P O M C m R N A in contralateral ventral spinal cord after nerve section using the probe for the A C T H 4 - 1 0 region (e) or the fl-EP 1 - 8 region (©). T h e results are expressed as the percentage of the total n u m b e r of motoneurones which expressed P O M C m R N A . T h e values are given as the mean (5: S.E.M.) for the results from at least 3 mice in each case.
dence of cells with the m R N A compared to the unoperated mice at 3 h after the nerve transection. The maximum increase in both the ipsilateral and contralateral horns was seen, with both probes, at 24 h after nerve section. The incidence thereafter declined and 7 days after the section the incidence on both sides was similar to that seen in the unoperated mice. The resuits obtained with the two different probes were similar. In sections from sham-operated mice examined at 6 and 24 h after the operation the proportions of cells which stained for POMC m R N A (using the probe to the A C T H 4-11 region of POMC) in the ipsilateral and contralateral ventral horns were not significantly different to the proportion in unoperated mice. No staining was seen in control sections, from the operated, sham-operated, or unoperated mice, which had been incubated with either of the conjugated sense strands (Fig. 4D,E).
4. Discussion
It has previously been shown in this laboratory that immunoreactivity for the POMC-derived peptides/3-EP and a - M S H can be detected in axons of mechanically damaged or diseased motor nerves as well as in chemically intoxicated motor nerves. An increased incidence of peptide-immunoreactive ventral horn cells were also seen in murine muscular dystrophy in mice [7] and in rats treated with IDPN [15], a neurotoxicant which targets motoneurones Furthermore we have shown increases in/3-EP and a - M S H immunoreactivity in intra-
47
muscular nerves in many conditions where motor neuropathy may be present. Edwards et al. [5] also demonstrated the presence of an a-MSH-like peptide in crushed motoneurones but they speculated that it was due to a degeneration product of neurofilament protein which shares a common amino acid sequence with a-MSH, enabling it to cross-react with antibodies to the peptide [3]. Oren et al. [18] demonstrated a-MSHlike immunoreactivity associated with neurofilaments in frog motor nerve axons. However, immunoreactivity for /3-EP which is formed from a different region of POMC is also expressed in the ventral horn cells [7]. It seemed more likely therefore that processing of POMC occurs in these neurones and that A C T H as well as /~-EP peptides are produced via processing of the prohormone in the nerves, rather than from neurofilament protein. In this study we used in situ hybridization to demonstrate POMC m R N A and to demonstrate its localisation in the ventral horn motoneurones thereby indicating that POMC-derived peptides can be synthesised in the motoneurones. We used oligoprobes complementary to two different regions of POMC m R N A to demonstrate the presence of POMC transcript. It seems likely therefore that it was indeed POMC m R N A which was detected and not the m R N A for some other protein which has a common amino acid sequence. Plantinga et al. [19], using polymerase chain reaction and northern blot analysis demonstrated the presence of low levels of POMC transcript in the T13 to L2 region of the rat spinal cord. They could not demonstrate increased synthesis of POMC m R N A in the ventral spinal cord after sciatic nerve crush. It is possible that the expression of the peptides in the rat ventral horn is less marked than that in the mouse. Indeed in earlier work where we investigated peptide immunoreactivity in intramuscular nerves, we observed a markedly lower incidence of immunoreactive nerves after nerve section in the rat [11] than in the mouse [10]. There are however, other reasons why the findings of their study and ours might be different. For example nerve crush is a relatively mild insult to the neurone compared to nerve transection and the pathological changes are less marked and more short-lived. In addition several biochemical procedures were necessary in their study and incomplete recovery of the nucleotide hybrids may therefore have resulted. It is incidently important to note that the sensitive histochemical method used in the present study enabled the m R N A to be precisely localised to the perikarya so that the motoneurone could be positively identified as the site of synthesis of POMC-derived peptides. The incidence of both peptide-immunoreactive cells and mRNA-positive cells showed parallel increases up to 24 h after nerve section and the incidence of both declined thereafter. The time course of the m R N A expression and of the peptide expression was initially
48
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
similar, the time interval between maximum transcription and maximum peptide synthesis indicating a difference of a few hours at most between transcription and synthesis of the peptides. At 7 days after nerve section the proportion of cells expressing the m R N A had returned to the levels in the unoperated controls. Interestingly, however, at this time after nerve section, the proportion of immunoreactive cells was still appreciably higher than in the control mice. The apparent delay in the time course of the expression of the immunoreactivity compared to that of the synthesis of the transcript at later times may be due to the slow rate of transport of the peptides from the perikarya to the axons after their synthesis in the motoneurones. It is not clear why increased peptide expression occurs in motoneurones of the contralateral ventral horn after nerve section. However previous work has shown that immunoreactivity for the peptides could be detected in contralateral intramuscular nerves after nerve section. Evidence from experiments where the nerve section was located at different distances from the spinal cord indicated that the contralateral effect was due to a transneuronal signal originating at the site of injury and travelling to the contralateral motoneurones via the spinal cord [11]. The results of the present study showing staining apparently in motoneurones in different locations in the two horns indicate that if this is the case then the 'crosstalk' between neurones in the two ventral horns of the spinal cord, at least at the same segmental level, may involve neurones of different muscle pools in the two sides. It is not known whether there are direct connections between motoneurones in the two ventral horns at the same segmental level and the nature of the message transferred is also unclear. The POMC-derived peptides probably act in concert to exert a t r o p h i c control of a number of different processes at the developing or regenerating neuromuscular junction. These processes include nerve growth [23,4], maturation of the endplate [6], production of the different acetylcholinesterase molecular forms [9], development of muscle strength [22] and neuromuscular transmission [2]. We have shown that at least one POMC-derived peptide,/3-EP, is released upon electrical stimulation of the motor nerve in the developing rat [8]. Furthermore receptors for these peptides were demonstrated in skeletal muscles [21]. It seems possible therefore that the trophic effects are mediated by the neuronal peptides, probably in conjunction with the circulating peptides. The studies of Strand and Kung [23] and Edwards et al. [4] have shown that exogenous ACTH-derived peptides can increase the rate of nerve regeneration after sciatic nerve transection. The time course of the upregulation of the P O M C gene shown in the present study is very much shorter than that for re-innervation after
motor nerve section, having returned to normal levels by 7 days after the section. Thus the suppression of P O M C transcription in the motoneurones is not controlled solely by the target muscle. The brevity of the time course is interesting in the light of the findings of others [20,24] that the susceptibility of nerve regeneration to the synthetic A C T H derivative, O R G 2766, is restricted to the first few days after nerve injury. However, the actions of peptides on nerve growth at late or early times after nerve injury may depend partly on their uptake by the nerve terminals from the blood or the medium surrounding the nerve and administered exogenous peptides could gain access to the nerves from the blood. In previous work increases in P O M C peptide immunoreactivity were seen in the distal stump or 'ghost' of the transected sciatic nerve after the axonal material (investigated using a silver staining technique) had disappeared, in both mice and rats [10,11]. Furthermore, in the rat the time course of the expression of the peptides in the intramuscular nerve terminals of both the transected and contralateral sides was more prolonged than that of the time course of the expression of P O M C m R N A in the mouse perikarya in the present study. Thus there was an initial increase up to approximately 24 h followed by a slight decrease to a plateau which remained high for at least 12 days [11]. In this case the immunoreactivity at later times could reflect a specis difference or it could have been due partly to uptake of the peptides from the blood. On the other hand the ventral horn cell may not be the only relevant site where increased synthesis of P O M C occurs after nerve damage. It is also interesting in this context that in the normal unoperated mouse immunoreactivity was present in a higher proportion of the nerve terminals than the proportion seen in the ventral horn cells of unoperated mice in the present study. It is possible therefore that increased expression of P O M C also occurs in extraneuronal elements such as Schwann cells or macrophages. POMC-derived peptides synthesised in elements at the neuromuscular junction (or circulating in the blood) may take over the functions of the neuronally synthesised peptides after they are no longer synthesised in the nerves. Our results indicate that damage to the motor nerve can result in increased synthesis of the P O M C precursor and the POMC-derived peptides. It seems likely that the peptides are synthesised at these locations whenever their actions are required to facilitate the repair processes or to augment neuromuscular transmission.
Acknowledgements We are grateful to the Wellcome Trust and the Motor Neurone Disease Association for supporting this work.
S. Hughes, M.E. Smith/Molecular Brain Research 25 (1994) 41-49
References [1] Chang, A.C.Y., Cochet, M. and Cohen, S.N., Structural organisation of human genomic DNA encoding the pro-opiomelanocortin peptide, Proc. Natl. Acad. Sci. USA, 77 (1980) 4890-4894. [2] Davies, D.A. and Smith, M.E., ACTH (4-10) increases increases quantal content at the mouse neuromuscular junction, Brain Res., 637 (1994) 328-330. [3] Drager, U.C., Edwards, L.D. and Kleinschmidt, J., Neurofilaments contain a-melanocyte-stimulating hormone (a-MSH)-like immunoreactivity, Proc. Natl. Acad. Sci. USA, 80 (1983) 64086412. [4] Edwards, P.M., Kuiters, R.R.F., Boer, G.J. and Gispen, W.H., Recovery from peripheral nerve transection is accelerated by local application of a-MSH by means of microporous Accurel polypropylene tubes, J. Neurol. Sci., 74 (1986) 171-176. [5] Edwards, P.M., Van der Zee, C.E.E.M., Verhaagen, J., Schotman, P., Jennekens, F.G.I. and Gispen, W.H., Evidence that the neurotrophic actions of ot-MSH may derive from its ability to mimic the actions of a peptide formed in degenerating nerve stumps, J. Neurol. Sci., 64 (1984) 333-340. [6] Frischer, R.E., E1-Kawa, N.M. and Strand, F.L., ACTH peptides as organisers of neuronal patterns in development: maturation of the rat neuromuscular junction as seen by scanning electron microscopy, Peptides, 6, Suppl. 2 (1985) 13-18. [7] Haynes, L.W. and Smith, M.E., Presence of immunoreactive a-melanotropin and/3-endorphin in spinal motoneurones of the dystrophic mouse, Neurosci. Lett., 58 (1985) 13-18. [8] Haynes, L.W., Smith, M.E. and Li, C.H., The regulation by /3-endorphin and related peptides of collagen-tailed acetylcholinesterase forms in the skeletal muscles of vertebrates. In G.B. Stefano (Ed.), CRC Handbook of Comparative Aspects of Opioid and Related Neuropeptides Mechanisms, Vol. II, Ch. 31, CRC Press, San Francisco, 1986, pp. 65-79. [9] Haynes, L.W., Smith, M.E. and Smyth, D.G., Evidence for the neurotrophic regulation of collagen-tailed acetylcholinesterase in immature skeletal muscle by/3-endorphin, J. Neurochem., 42 (1984) 1542-1551. [10] Hughes, S. and Smith, M.E., Effect of nerve section on /3-endorphin and a-melanotropin immunoreactivity in motoneurones of normal and dystrophic mice, Neurosci. Lett., 92 (1988) 1-7. [11] Hughes, S. and Smith, M.E., Pro-opiomelanocortin-derived pep-
49
tides in transected and contralateral motor nerves of the rat, J. Chem. Neuroanat., 2 (1989) 227-237. [12] Hughes, S. and Smith, M.E., Proopiomelanocortin-derived peptides in mice with motoneurone disease, Neurosci. Lett., 103 (1989) 169-173. [13] Hughes, S and Smith, M.E., Expression of the POMC gene in motoneurones of the mouse, J. Physiol., 473 (1993) 40P [14] Hughes, S., Smith, M.E. and Bailey, C.J., POMC-derived peptides in the neuromuscular system of streptozotocin-diabetic mice, Peptides, 13 (1992) 873-877. [15] Hughes, S., Smith, M.E., Simpson, M.G. and Allen, S.L., Effect of IDPN on the expression of POMC-derived peptides in rat motoneurones, Peptides, 13 (1992) 1021-1023. [16] Kramer, H and Windrum, G.M., The metachromatic staining reaction, J. Histochem. Cytochem., 3 (1955) 227-237. [17] McGadey, J., A tetrazolium method for non-specific alkaline phosphatase, Histochemie, 23 (1970) 180-184. [18] Oren, N., Micevych, P.E. and Letinsky, M.S., Presence of amelanocyte-stimulating hormone-like immunoreactivity in the innervation of amphibian skeletal muscle, Z Neurosci. Res., 23 (1989) 225-233. [19] Plantinga, L.C., Verhaagen, J., Edwards, P.M., Schrama, L.H., Burbach, J.P.H. and Gispen, W.H., Expression of the proopiomelanocortin gene in dorsal root ganglia, spinal cord and sciatic nerve after sciatic nerve crush in the rat, Mol. Brain Res., 16 (1992) 135-142. [20] Saint-Come, C. and Strand, F.L., ACTH 4-9 analogue (Org 2766) improves qualitative and quantitative aspects of motor nerve regeneration, Peptides, 8, Suppl. 1 (1988) 215-221. [21] Smith, M.E. and Hughes, S., Pro-opiomelanocortin neuropeptide receptors on developing and dystrophic muscle fibres, MoL Chem. Neuropathol., 19 (1993) 137-145. [22] Smith, C.M. and Strand, F.L., Neuromuscular response of the immature rat to A C T H / M S H 4-10, Peptides, 2 (1981) 197-206. [23] Strand, F.L. and Kung, T.T., ACTH accelerates recovery of neuromuscular function following crushing of peripheral nerve, Peptides, 1 (1980) 135-138. [24] Verhaagen, J., Edwards, P.M.,Jennekens, F.G.I., Schotman, P. and Gispen, W.H., Early effect of an ACTH 4-9 analog (Org 2766) on regenerative sprouting demonstrated by the use of neurofilament binding antibodies isolated from a serum raised by a-MSH immunization, Brain Res., 404 (1987) 147-150.