Research
Cell Calcium (1997) 22(5), 367-372 0 Harcoutl Brace & Company Ltd 1997
Axotomv induces transient - calbindin _ _~--_ --_---D28K ikmunoreactivity in hypoglossal motoneurons in vivo Claudia Krebs1y3,Wolfram F. Neissl, Michael Streppel*, Orlando Guntinas-Lichius*, Donald Dassesse3, Eberhard Stennert*, Roland Pochet3 ‘Institut
I fur Anatomie
3Laboratory
and *Klinik fur Hals-,
of Histology,
Faculte
Nasen-
de Medecine,
und Ohrenheilkunde,
Universite
Universitat
Libre de Bruxelles.
zu Kbln, Cologne,
Bruxelles,
Germany
Belgium
Calbindin D28K, an intracellular calcium-binding protein, acts as Ca*+ buffering system in the cytoplasm. By means of this property, calbindin may protect neurons against large fluctuations in free intracellular Ca2+and, hence, may prevent cell death. Although axotomy causes a massive influx of calcium into the lesioned neurons, resection of the hypoglossal nerve does not induce extensive neuronal cell death in rats. Even several weeks after axotomy, about 70% of the motoneurons survive despite permanent target deprivation. The mechanisms responsible for this remarkable survival rate are unknown. In this study, we have looked at the modification of calbindin immunoreactivity in axotomized hypoglossal motoneurons. In non-axotomized motoneurons, no calbindin is detectable by immunocytochemistry. Axotomy induced an increase of calbindin immunoreactivity in lesioned motoneurons. This increase, visualised by the number of calbindin-immunoreactive neurons extended from 1 day to 28 days. At this time most, but not all, motoneurons located on the side of the lesion were calbindin-positive as shown by retrograde labeling and immunoquenching. From 14 days post operation, calbindin immunoreactivity decreased and reached its basal value after 35 days post operation. At that time, only fibres were still calbindin immunoreactive. Interestingly, calbindin-immunoreactivity was also increased in almost all cell nuclei, compatible with a nuclear regulation. These data are consistent with the hypothesis that, as a reaction to axotomy, motoneurons trigger an increase in calbindin expression which acts as a compensatory Ca*+buffering system, enabling neurons to maintain Ca*+ homeostasis and the survival of many motoneurons after axotomy.
Summary
INTRODUCTION
In vivo axotomy and removal of part of the length of a peripheral motor nerve irreversibly separates the lesioned motoneurons from their target muscles and, hence, from the trophic factors supplied by them. This permanent target deprivation causes the death of 25-35% [l-3] of the
Received 10 January Revised 8 September
1997 1997
Accepted
1997
IO October
Correspondence
to: Dr R. Pochet.
Laboratory
of Histology,
Medecine, Universite Libre de Bruxetles, Campus route de Lennik, B-l 070, Brussels, Belgium Fax: +32 2 555 6285:
E-mail:
[email protected]
Erasme,
Faculte
de
CP 620, 808
hypoglossal motoneurons 16 weeks after interruption of the hypoglossal nerve in the adult rat. Remarkably, however, 65-75% of the hypoglossal motoneurons survive despite this permanent target deprivation. The mechanisms preventing the death of so many neurons after hypoglossal axotomy are unknown, but regulation of Caz+-homeostasis might play an important role. Indeed, it is a well established fact that axotomy causes a massive influx of Ca2+ into the damaged neurons [4,5] and the vulnerability of motoneurons seems to be due to neurotoxic concentrations of intracellular free Caz+ [6-lo]. Several recent studies suggest that the presence of calcium-binding proteins, such as calbindin, might prevent neuronal death after injury [ 1 l- 131. Therefore, it was tempting to look if calbindin, not detected in mature motoneurons, could be expressed 367
368
C Krebs, W F Neiss, M Streppel et al
or re-expressed in adult rat hypoglossal motoneurons after axotomy with resection of the peripheral nerve, i.e. after permanent target deprivation of the nerve cells.
50 adult female Wistar rats (175-200 g; strain HsdCpb:WU; Harlan-Winkelmann, Borchen, Germany) were used for resection of the hypoglossal nerve. All animals were kept on standard laboratory chow (Fa. Ssniff, Soest, Germany) and tap water ad Zibitum with an artificial light-dark cycle of 12hon, 12hoff.
l/20), then incubated overnight at room temperature with the rabbit anti-calbindin polyclonal antibody (Swam, Bellinzona, Switzerland) diluted at l/5000. The sections were then treated 10 min with NSS, 10 min with anti-rabbit-IgG (l/100) (DAKO), 10 min with PAPcomplex (l/300) (DAKO). Between each incubation, the sections were rinsed twice for 5 min in PBS. The further steps using the nickel-enhanced diaminobenzidine precipitation were made in O.lM Tris-HCl pH 72 and according to Hsu and Soban [ 161: 10 min in Tris-nickel @g/l ammonium nickel sulfate; Fluka) containing 0.5 mg/ml nickel-enhanced diaminobenzidine (Sigma). The precipitation reaction was performed in the same Trisnickel containing H,O, (0.03%).
Application
RESULTS
MATERIALS
AND
METHODS
Animals
of Fluorogold
Under deep ether narcosis, 1 mg Fluorogold (FG; Fluorochrome Inc., Englewood, CO, USA) in 100 ~1 distilled water was injected into the midline of the tongue in 18 animals, 5 days prior to surgery. Surgery
The rats were anaesthetised by an intraperitoneal injection of 1.4 ml Avertin (2.0 g tribrom-ethanol, 1.O ml 3-pentanol, 8.0 ml absolute ethanol plus 90.0 ml 0.9% saline). The hypoglossal nerve was transected and a piece of 8-10 mm nerve length was removed in order to prevent regeneration [2,14]. Three animals were not subjected to this hypoglossal nerve resection and served as controls. Fixation
At 1, 2, 3, 4, 8, 14, 2 1, 28, 35 and 49 days post operation, rats were deeply anaesthetised with ether and intracardially perfused with 0.9% NaCl for 45 s, followed by 4% paraformaldehyde in 0.1 M phosphate buffer @H Z4) for 30 min. The animals that had been prelabeled with FG were perfused with the periodate-lysine-paraformaldehyde fixative [15] for 45 min. The brainstem was dissected, postfixed overnight in 4% parafonnaldehyde (without lysine-periodate) and cryoprotected in sucrose (30%). lmmunostaining
Free-floating cryostat sections (40 pm) taken every 200 pm throughout the entire hypoglossal nucleus were incubated in PBS and further processed to detect calbindin immunoreactivity. Sections were pretreated at room temperature 30 min in 0.3% Triton X-100,20 min in 1% H202, 15 min in normal swine serum (NSS) (diluted Cell Calcium
(1997)
22(5),
367-372
In normal rats, no calbindin-immunoreactivity could be detected in the neurons of the hypoglossal nucleus, nor were any fibres immunoreactive, whereas ependymocytes were intensively labeled. Injection of FG into the midline of the tongue 5 days prior to the killing of the rats caused a retrograde labeling of all hypoglossal motoneurons on both sides of the brainstem. After axotomy and resection of the hypoglossal nerve, calbindin-immunoreactivity was detectable in the perikarya and the cell nuclei of the motoneurons located on the side of the lesion (Fig. 1). This upregulation of calbindin in the lesioned motoneurons was transient. Already 24 h after the lesion, some calbindin-positive neurons could be found in the nucleus. At 2 days post operation, several fibres and neurons located in the dorsal part of the nucleus (Fig. l), which supplies the retractor muscles of tongue movement [ 17,181, were calbindin positive. At 3 and 4 days after resection, intensively calbindin-positive neurons (somata and processes) were scattered throughout the dorsal and ventral part of the hypoglossal nucleus. The cell nuclei, but not the nucleoli, of most of the neurons were also calbindin-immunoreactive and, in some neurons, the cell nuclei were even more intensively labeled than the perikarya. The maximum number of calbindin-positive fibres and motoneurons scattered throughout the entire ipsilateral hypoglossal nucleus was reached at 8 days post operation and remained at a very high level until 14 days post operation. At 21 days post operation, fewer neurons were calbindin-positive, their number and the intensity of staining then decreased and, at 28 days post operation, very few weakly immunoreactive motoneurons were still detectable (Fig. 1). At 35 days post operation, only the fibres on the side of the lesion remained calbindin-positive. Further analysis throughout the whole nucleus did not reveal preferential labeling in any particular hypoglossal subnucleus. 0 Harcourt
Brace
& Company
Ltd 1997
Axotomy induces transient calbindin D28K immunoreactivity
369
Fig. 2. Fluorogold
labeling of motoneurons in the left and right hypoglossal nucleus 8 days after resection of the right hypoglossal nerve. All motoneurons acquired FG, also on the side of axotomy (right), as the retrograde tracer was injected into the midline of the tongue 5 days prior to axotomy. This figure shows the same section and field of view as Figure 1, 8 days post operation. The dark calbindin-staining with nickel-enhanced diaminobenzidine (Fig. 1) quenches the fluorescence of FG; i.e. all fluorescent neurons visible in Figure 2 appear calbindin-negative in Figure 1. FG in vivo, UV-fluorescence (Zeiss filter Ol), 1:5000 anti-calbindin, PAP, nickel-enhanced diaminobenzidine; scale bar = 120 lrn.
28dpo
.
-
35*
Fig. 1 Calbindin immunostaining appears in the right hypoglossal nucleus only after resection of the right hypoglossal nerve. The maximum of immunoreactivity is observed at 8 to 14 days post operation. All sections are from the same level at the caudal third of the hypoglossal nucleus. 1:5000 anti-calbindin, PAP, nickelenhanced diaminobenzidine; scale bar = 120 pm.
The contralateral hypoglossal nucleus of operated rats contained very few, if any, calbindin-positive cells (Fig. 1) and thus served as an internal negative control. In the same sections, the tractus solitarius constitutively displayed an intensive imrnunostaining on both sides of the brainstem and, thereby, provided an internal positive control (not shown). These two extremes on the same section were helpful for semi-quantitative assessment of immunoreactivity intensity. Fluorogold
prior to axotomy
FG was injected into the midline of the tongue 5 days before the hypoglossal nerve was resected. According to our observations in normal rats (see above) and previous experience of our laboratory with facial motoneurons in the rat, 5 days is ample time for uptake and transport of this stable retrograde tracer to the perikarya of the 0 Harcourt
Brace
& Company
Ltd 1997
motoneurons (see [19,20]). Hence all motoneurons were labeled at the time of lesion. As shown in previous studies [ 19-2 11, FG is a stable tracer that is not destroyed or extracted from neurons and other cells in the fixed and natural tissue by standard methods of bright field immunocytochemistry with PAP-reaction and H,O,diaminobenzidine as detector system. III all immunoreactive cells, however, the dark-coloured nickelenhanced diaminobenzidine reaction product blocks all emission of the fluorochrome (immunoquenching of fluorescence, see [19-211). In our experiments, the immunostaining of calbindin blocked the visibility of FG on the side of the lesion. Hypoglossal motoneurons that in bright field illumination showed a moderate to strong immunoreactivity for calbindin (8 days post operation in Fig. I), were invisible with the fluorescent illumination (Fig. 2). Other neurons with bright FG labeling in the fluorescent mode, showed very faint or no calbindin reaction product in the bright field mode (compare 8 days post operation in Figs 1 & 2). At no time point after axotomy did calbindin immunocytochemistry quench the fluorescence of all hypoglossal motoneurons. Even at 8 or 14 days post operation, at the maximum of calbindin expression, some motoneurons were fluorescent and did not contain calbindin reaction product in the bright field mode (Fig. 2, right side) These data show that the majority, but not all, motoneurons upregulate calbindin after axotomy. Cell Calcium
(1997)
22(5),
367-372
370
C Krebs, W F Neiss, M Streppel et al
DISCUSSION
Research on calcium-binding proteins in neurodegenerative disorders has brought forth contradictory results (for a review, see [22]). Each brain system or region seems to have its own mechanism for responding to calciummediated neurodegeneration. As new data, we report here that calbindin, an intracellular calcium-binding protein largely distributed in many (but not all) brain nuclei, is increased in hypoglossal motoneurons of adult rats after axotomy. Induced calbindin immunoreactivity was apparent as early as 24 hours post operation, was transitory and disappeared after 6-7 weeks. This result is noteworthy, as up to now calbindin appeared to be a rather stable protein in neurons of adult animals and, thus far, there are but scarce data indicating that this protein can be upregulated in animal experiments. To our knowledge, only three recent publications reported such modulations. The first showed that expression of calbindin was increased in the hippocampus in response to stimulation of the perforant path [23]. The second indicated that chronic treatment with morphine could induce calbindin expression in rat striosomes [24] and the third showed that axotomy of the rat superior cervical ganglion increased the number of calbindin-immunoreactive neurons in this ganglion [9]. Recently Ziv and Spira [5] proved that axotomy causes an increase of intracellular calcium to the millimolar range and that neurons can compensate this within 2-3 min after resealing of the nerve end, this resealing occurs within 30 min after axotomy [24]. This short time scale makes it difficult to prove that the newly established increase in calbindin could be the main cause for this compensation as we have seen only very few neurons becoming calbindin immunoreactive 24 h after axotomy. However, it can be hypothesized that massive inilux of calcium triggers calbindin mRNA synthesis and provides the cell with an increase in intracellular calcium buffer protecting it from any further Ca2+influx. Such a role for calcium in the regulation of calbindin expression has already oeen shown in chick kidney cells 1251 and in the uterus of laying hens [26]. Furthermore, Umemiya et al 1271documented that axotomy of rat facial motoneurons causes an enlargement of the action potential, which, in turn, may increase calcium influx during electrical activity. When calcium is bound to calbindin the concentration of free intracellular calcium is decreased, the total amount of intracellular calcium, however, remains increased; by continuously releasing calcium ions a sustained increase of intracellular ‘Ca2+might be achieved, as suggested by Collins et al to be important for neuronal survival 1281.On the other hand, prolonged elevation of intracellular Caz+ has shown to lead to cell death, and aberrant Ca2+ Cell Calcium
(1997)
22(5),
367-372
handling by motoneurons is thought to contribute to motoneuron disease 131. Interestingly, hypoglossal motoneurons of adult rats resist rather well to cell death after peripheral nerve resection, i.e. axotomy plus permanent target deprivation [l-3], and appear quite well able to maintain non toxic concentrations of free intracellular Ca2+ in the face of increased Ca2+influx. Our results are compatible with a mechanism involving the increased Ca2+buffering capacity of the injured neurons expressing calbindin. Such a role for calbindin has already been suggested by several authors studying different brain regions following injury [l l-13,16,29-31]. Recently, it was shown that an additional mechanism for calbindin to protect Ca2+-mediated excitotoxicity in neurons could be to trigger protein by affecting Ca2+-regulating proteins, such as Ca2+channels 1321. We have previously shown that the cytosolic enzyme neuron specific enolase (NSE) is transiently expressed in the nuclei of axotomized neurons 1331, and the intranuclear calbindin expression reported here runs strikingly parallel to the intranuclear NSE localisation. A possible role of NSE in genome regulation has been suggested. Whether calbindin plays a role in this mechanism remains to be studied; such a role has already been suggested for the calmodulin-dependent phosphatase calcineurin. Calbindin might be important for regulating the intranuclear calcium in the event that the ionic pumps of the nuclear membrane fail to function because the ionic milieu in the cytoplasm is disturbed too much (for a review, see [34]). Considering the fact that only 25-33% of the motoneurons die following axotomy with permanent target deprivation and considering the low number of neurons not expressing calbindin at 8 or 14 days post operation, we dare to hypothesize that those expressing calbindin have a higher probability of survival. An argument in favour of calbindin being important in modulating neuronal Ca2+ homeostasis, is its transient presence during neural development [35-371 in which calcium ions have a dominant influence, especially contemporarily to growth cones [38]. Therefore, to sub stantiate this hypothesis further, it would be of interest to look at calbindin expression in the hypoglossal nucleus during development. Another approach to demonstrate that calbindin acts as a survival-promoting factor, would be to perform hypoglossal axotomy on calbindin knockout gene animals. CONCLUSIONS
In the adult rat, two-thirds of all hypoglossal motoneurons persist after axotomy and resection of the peripheral nerve, i.e. they survive permanent target deprivation despite of the massive increase of intracellular free Ca2+. We have now shown in vivo that 0 Harcourt
Brace
& Company
Ltd 1997
Axotomy
induces
transient
calbindin
D28K immunoreactivity
371
axotomy of the hypoglossal nerve induced an increase of calbindin in motoneurons of the hypoglossal nucleus located on the side of the lesion. This transient expression was first observed as early as 1 day after axotomy, culminated at 8-14 days and completely disappeared 6-7 weeks after the lesion. These results provide an anatomical substrate for the initiation of a possible compensatory mechanism in the neurons to buffer the neurotoxic increase of intracellular free Ca2+.
15. McLean I.W., Nakane P.K. Periodate-lysine-paraformaldehyde fixative a new ftvative for immunoelectron microscopy. J
ACKNOWLEDGMENTS
19. Angelov D.N., Gunkel A., Stennert E., Neiss W.F. Phagocytic microglia during delayed neuronal loss in the facial nucleus of the rat: time course of the neuronofugal migration of brain macrophages. G&z 1995; 13: 113-129. 20. Angelov D.N., Neiss W.F., Streppel M., Walther M., GuntinasLichius O., Stennert E. ED2-positive perivascular cells act as neuronophages during delayed neuronal loss in the facial nucleus of the rat. Glia 1996; 16: 129-139. 21. Rinaman L., Milligan C.E., Levitt P. Persistence of Fluorogold following degeneration of labeled motoneurons is due to phagocytosis by microglia and macrophages. Neuroscience 1991; 44: 765-776. 22. Heizmann C.W., Braun K. cdl&m reguhtiun by calcium-binding proteins in neurodegenerative disorders. Heidelberg: Springer, 1995 23. Garcia M.M., Harlan R.E. Chronic morphine increases calbindin D28k in rat striatum: possible NMDA receptor involvement. Neurorepoti 1993; 5: 65-68. 24. Yawo H., Kuno M. Calcium dependence of membrane sealing at the cut end of the cockroach giant axon. J Neurosci 1985; 5: 1626-1632. 25. Enomoto H., Hendy G.N., Andrews G.K., Clemens T.L. Regulation of avian calbindin-D28K gene expression in primary chick kidney cells: importance of posttranscriptional mechanisms and calcium ion concentration. Endocrinology
Histochem
REFERENCES 1. Aldskogius H., Barron K.D., Regal R. Axon reaction in dorsal motor vagal and hypoglossal neurons of the adult rat. Light microscopy and RNA-cytochemisuy. JComp Neural 1980; 193: 165-177. 2. Neiss W.F., Cuntinas-Iichius O., Angelov D.N., Gunkel A., Stennert E. The hypoglossal-facial anastomosis as model of neuronal plasticity in the rat. Ann Anat 1992; 174: 4 19-433. 3. Snider W.D., Thanedar S. Target dependence of hypoglossal motor neurons during development and in maturity. J Camp New-01
1989;
279: 489-498.
4. Borgens R.B. Voltage gradients and ionic currents in injured and regenerating axons. Adu Neural 1988; 47: 5 l-66. 5. Ziv N.E., Spira M.E. Axotomy induces a transient and localized elevation of the free intracellular calcium concentration to the millimolar range. JNeurophysiol 1995; 74: 2625-2637. 6. George E.B., Glass J.D., Griffin J.W. Axotomy-induced axonal degeneration is mediated by calcium influx through ionspecific channels. JNeurosci 1995; 15: 6445-6452. Z Krieger C., Jones K., Kim S.U., Eisen A.A. The role of intracellular free calcium in motor neuron disease. J Neural Sci 1994; 124: 27-32. 8. Choi D.W. Calcium and excitotoxic neuronal injury. Ann NY Acad
Sci 1994;
747: 162-171.
9. Sanchez-Vives M.V., Valdeolmillos M., Martinez S., Gallego R. Axotomy-induced changes in CaZ+homeostasis in rat sympathetic ganglion cells. Eur J Neurosci 1994; 6: 9- 17 10. Siesjo B.K., Bengtsson F., Grampp W., Theander S. Calcium, excitotoxins, and neuronal death in the brain. Ann NYAcad Sci 1989; 568: 234-25 1. 11. Mattson M.P., Rychlik B., Chu C., Christakos S. Evidence for calcium-reducing and excite-protective roles for the calciumbinding protein calbindin-D28k in cultured hippocampal neurons. Neuron 199 1; 6: 4 1 -5 1. 12. Scharfman H.E., Schwartzkroin P.A. Protection of dentate hilar cells from prolonged stimulation by intracellular calcium chelation. Science 1989; 246: 257-260. 13. Alexianu M.E., Ho B.K., Mohamed A.H., La Bella V., Smith R.G., Appel S.H. The role of calcium-binding proteins in selective motoneuron vulnerability in amyotrophic lateral sclerosis. Ann Neural 1994; 36: 846-858. 14. Angelov D.N., Gunkel A., Stennert E., Neiss W.F. Recovery of original nerve supply after hypoglossal-facial anastomosis causes permanent motor hyperinnervation of the whisker-pad muscles in the rat. J Comp Neural 1993; 338: 2 14-224. 0 Harcourr Brace
& Company
Ltd 1997
1974; 22: 1077-1087.
1079-1082.
1% Krammer E.B., Rath T., Lischka M.F. Somatotopic organization of the hypoglossal nucleus: a HRP study in the rat. Brain Res 1979;
170: 533-537.
18. Uemura-Sumi M., Itoh M., Mizuno N. The distribution of hypoglossal motoneurons in the dog, rabbit and rat. Anat Embryo1
This research was supported by grants from the FRSM (Belgium) and the Jean-Uhrmacher-Stiftung (OGL, MS and ES).
Cytochem
16. Hsu S.M., Soban E. Color modification of diaminobenzidine (DAB) precipitation by metallic ions and its application for double immunohistochemistry. J Histochem Cytochem 1982; 30:
1992;
1988;
177: 389-394.
130: 3467-3474.
26. Nys Y., Baker K., Lawson D.E.M. Estrogen and a calcium flux dependent factor modulate the calbindin gene expression in the uterus of laying hens. Gen Comp Endocrinoll992; 87: 87-94. 22 Umemiya M., Araki I., Kuno M. Electrophysiological properties of axotomized facial motoneurones that are destined to die in neonatal rats. JPhysiol (Land) 1993; 462: 661-678. 28. Collins F., Schmidt M.F., Guthrie P.B., Kater S.B. Sustained increase in intracellular calcium promotes neuronal survival. J Neurosci 1991; 11: 2582-2582 29. Iacopino A., Christakos S., German D., Sonsalla P.K., Altar CA. Calbindin-D28K-containing neurons in animal models of neurodegeneration: possible protection from excitotoxicity. Brain ResMolBrain Res 1992; 13: 251-261. 30. Ince P., Stout N., Shaw P. et al. Parvalbumin and calbindin D28k in the human motor system and in motor neuron disease. Neuropathol Appl Neurobioll993; 19: 291-299. 31. Heizmann C.W., Braun K. Changes in Ca(2+)-binding proteins in human neurodegenerative disorders. Trends Neurosci 1992; 15: 259-264.
32. Lledo P.M., Somasundaram B., Morton AJ., Emson P.C., Mason W.T. Stable transfection of calbindin-D28k into the GH3 cell line alters calcium currents and intracellular calcium homeostasis. Neuron 1992; 9: 943-954. 33. Angelov D.N., Neiss W.F., Gunkel A., Guntinas-Lichius O., Stennert E. Axotomy induces intranuclear immunolocalization of neuron-specific enolase in facial and hypoglossal neurons of the rat. JNeuroqtol 1994; 23: 218-233. Cell Calcium
(1997)
22(5),
367-372
372
C Krebs, W F Neiss, M Streppel et al
34. Santella L. The cell nucleus: an eldorado to future calcium research? J Met& Biol1996; 153: 83-92. 35. Liu F-C., Graybel A.M. Transient calbindin-D28k-positive systems in the telencephalon: ganglionic eminence, developing striatum and cerebal cortex. J Neurosci 1992; 12: 674-690. 36. Friauf E. Transient appearance of calbindin-D28k-positive neurons in the superior olivary complex of developing rats. J Camp New-01 1993; 334: 59-74.
Cell Calcium
(1997)
Z?(5),
367-372
37 Hogan D., Berman N.EJ. Transient expression of calbindinD28k immunoreactivity in layer V pyramidal neurons during postnatal development of kitten cortical areas. Deu Bruin Res 1993; 74: 177-192. 38. Amato A., Al-Mohamra EA., Bolsover S.R. Spatial organisation of calcium dynamics in growth cones of sensory neurons. Des Bruin Res 1996; 92: 101-110.
0 Harcourt
Brace
& Company
Ltd 1997
