- Email: [email protected]
Intrinsic properties of the developing motor cortex in the rat: in vitro axons from the medial somatomotor cortex grow faster than axons from the lateral somatomotor cortex
Developmental Brain Research 122 (2000) 59–66 www.elsevier.com / locate / bres
Research report
Intrinsic properties of the developing motor cortex in the rat: in vitro axons from the medial somatomotor cortex grow faster than axons from the lateral somatomotor cortex J.M.P. Mouveroux a , M. Verkijk a , E.A.J.F. Lakke a , *, E. Marani a,b b
a Neuroregulation Group, Department of Neurosurgery, Leiden University Medical Centre, Leiden, The Netherlands Signals and Systems, Department of Electrical Engineering, Biomedical Technology Institute, University of Twente, Twente, The Netherlands
Accepted 9 May 2000
Abstract The axons that originate in the medial somatomotor cortex of the rat depart, during development, after those from the lateral somatomotor cortex, yet they arrive in the cervical spinal cord first. Either the medially originating axons elongate faster, or the laterally originating ones pause along the descent pathway. To investigate the presence of an intrinsic difference of the axonal elongation velocity between the lateral and medial somatomotor cortical areas, we cultured explants taken from these areas for 2 days, and measured the length of the outgrowth. After 2 days the explants were surrounded by a radiate corona of axons of which the longest measured 1.95 mm. A significant difference was detected between the medial and lateral somatomotor cortical areas in vitro. Axons originating from explants taken from the medial somatomotor cortical area are, after 2 days in culture, on average 0.16 mm longer than those from the lateral somatomotor cortical area. Though the observed difference is not large enough to allow for the overtaking observed in vivo, it does indicate that intrinsic differences exist within the developing rat somatomotor cortex. This in turn indicates that intrinsic cortical traits not only influence regionalization and targeting behavior of cortical projection neurons, but also their axonal elongation speed. 2000 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Process outgrowth, growth cones and sprouting Keywords: Rat; Somatomotor cortex; Development; Culture; Axonal outgrowth; Arealization; Regionalization
1. Introduction The differentiation of areas within the cerebral cortex has been a contested issue. Regionalization has been proposed to be controlled by patterning mechanisms intrinsic to the cortex [29], or by extrinsic influences such as the thalamocortical afferents [26]. At present the evidence strongly supports the presence of intrinsic mechanisms governing cortical regionalization (see Rubenstein and Rakic [31] for a review). The same mechanisms that govern cortical regionalization also influence targeting behavior of the cortical projection neurons, presumably by *Corresponding author. Department of Physiology, P.O. Box 9604, NL-2300 RC Leiden, The Netherlands. Tel.: 131-71-527-6759. E-mail address: [email protected] (E.A.J.F. Lakke)
influencing their sensitivity for axonal guidance molecules [4]. We will argue that axonal elongation speed is another trait of the cortical projection neurons that is under control of these intrinsic cortical mechanisms. Cortical projection axons do not arrive at the spinal cord synchronously. The first corticospinal axons arrive at cervical levels 3 days after birth. These axons originate from a small area located medially in the dorsal parietal cortex, and these axons are destined for the lumbar spinal cord [33]. In adulthood the same area of dorsal parietal cortex represents the hindlimb area (HL), where the motor area I (MI) and the sensory area I (SI) overlap [8,11]. During the first week the area of cortical neurons that reach down to the cervical spinal cord expands to a continuous sheet that covers most of the frontal parietal and cingulate cortex. This area includes the forelimb area
0165-3806 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0165-3806( 00 )00056-0
60
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
(FL), located in the lateral dorsal parietal cortex [33]. Later during postnatal development many of the cortical projection neurons lose their spinal axon, resulting in the discontinuous distribution of spinal projection neurons as observed in adult rats [32]. In short, the axons from the medially located HL area arrive in the cervical spinal cord before the axons from the laterally located FL area. The neurogenetic gradient of the cortex is like a wavepattern emerging from a center located anterolaterally. The wave front travels rostrad, mediad and caudad simultaneously, but the general direction is from rostral to caudal and from lateral to medial [13,34]. Thus the laterally located FL area neurons are generated before those of the medially located HL area. The axogenetic gradient of the cortex is similarly oriented, the first projection axons that appear in the intern capsule originate in the lateral cortex, and in time axons from progressively more medial origins are added [3,9,22,27]. Thus, axons from the laterally located FL area will enter the internal capsule before those of the medially located HL area; the medial area axons are generated later, and have to travel a longer distance to reach the internal capsule. The nervous system itself increases in size during development, and so the distance between the cortex and the spinal cord increases continuously. Thus a later departure from the cortex towards the spinal cord implies a longer distance to be covered. Even though the axons from the HL area of the cortex are initially in arrear, they arrive first at the cervical spinal cord. Their growth cones must thus have overtaken the growth cones that originated in more lateral cortical areas. Either those growth cones have stopped along the way, or the medial growth cones have traveled faster. Medial growth cones could travel faster because of an intrinsic property of their parent neurons in the HL somatomotor area, or because of a stimulating influence along the way. To investigate whether an intrinsic property of the neurons in the HL somatomotor cortical area is responsible for the faster elongation of its descending axons we cultured explants from the lateral and medial areas of the dorsal frontoparietal cortex, corresponding to, respectively, the forelimb and hindlimb somatomotor cortical areas (Fig. 1). The explants were embedded in a collagen matrix, and cultured for 2 days in a chemically defined medium. In our experience explant culture embedded in a three-dimensional matrix results in more profuse outgrowth as compared to outgrowth on a two-dimensional matrix. After 2 days the length of the outgrowth was measured and compared. The culture period was limited to 2 days by the size of the collagen drop in which the explants were embedded. After culture for longer periods the outgrowing axons would reach the periphery of the collagen drop and stop. Cortical projection axons originate in lamina V pyramidal cells. In the rat, lamina V neuroblasts of the cerebral cortex are generated from gestation day 15 (G15) through
G18. We used fetuses and neonates ranging in age from G17 to G25. After 2 days the explants were surrounded by a radiate corona of axons of which the longest measured 1.95 mm. A significant difference was detected between outgrowth from the medial and lateral somatomotor cortical areas in vitro. Axons originating from explants taken from the medial somatomotor cortical area are, after 2 days in culture, on average 0.16 mm longer than those from the lateral somatomotor cortical area. The results demonstrate the existence of an intrinsically determined difference of the axonal growth velocity between cultured explants taken from medial and lateral areas of the rat motor cortex. This indicates that intrinsic cortical traits not only influence regionalization and targeting behavior of cortical projection neurons [4], but also their axonal elongation speed.
2. Materials and methods
2.1. Explants Wistar Albino Glaxo rats in oestro were mated between 10:00 and 11:00 h [36]. The end of this period was taken as the start of G0 [28]. In our colony, pregnancy lasts 23 days, i.e. the pups are born early on G23. Rat fetuses and pups of ages ranging from G17 to G25 were used. Fetuses were harvested at the required duration of gestation, after sacrificing the pregnant dam. All procedures for the preparation of cortical explants took place under sterile conditions. Immediately before the rats were decapitated, they were introduced into the flowcabinet, submerged into 70% ethanol and rinsed in cold normal saline. Skin, parietal and frontal bones were then quickly resected, to reveal the cortical surface. In the youngest fetuses (G17, G18) the cortical surface was moistened with a few drops of normal saline. One longitudinal strip of dorsal fronto-parietal cerebral cortex was then excised from each hemisphere, each strip measuring |632 mm. On one side the strip was excised from HL, and on the other side from FL (Fig. 1). Excision sequence (FL or HL first) and side (from left or right hemisphere first) were alternated between rats. The strips were collected in a chemically defined, serum free medium (R12) [30], and chopped immediately after excision into 250-mm slices. These slices were collected in the same medium, separated by gentle aspiration into a Pasteur pipette, and subsequently stored in a CO 2 incubator (378C, 5.5% CO 2 ). Processing time, from decapitation to storage never exceeded 13 min. Individual rats were processed sequentially, and all batches of slices were stored in separate Petri dishes. Under a dissection microscope, those slices of each batch whose cortical surface appeared to be undamaged, and that were of uniform thickness were selected for trimming. With a fine scalpel a wedge-shaped section was
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
61
Fig. 1. (A) Location of the cortical strips. Dorsal view of the rat brain. The shaded area indicates the position of the somatomotor cortex, shaped like a rattunculus [25]. The rattunculus is repeated above, to indicate the position of the HL (backward-hatched) and FL (forward-hatched) areas within the somatomotor cortex. Rectangles on the cortex indicate the excision areas of the cortical strips. Excision sequence (FL or HL first) and side (from left or right hemisphere first) were alternated between rats. (B) Preparation of the explants. The cortical strips were chopped into 250-mm slices. A wedge-shaped section was excised out of each slice, to obtain an equilateral triangle |1 mm wide, yet with a recognizable pial surface. Care was taken to exclude white matter (shaded gray) from the resulting explants. Each explant was cultured in a separate 10 culture dish. Each dish was marked with a number tracing it back to the litter from which the culture originated, and a serial number. A separate list was kept to link these numbers to the origin of the explant (HL or FL). Only after completion of the measurements were the measured values correlated to the separately listed data on their site of origin (FL or HL) or age.
excised out of each slice, to obtain an equilateral triangle of |1 mm width, yet with a recognizable pial surface (Fig. 1). Care was taken to exclude white matter from the resulting explants, since it may interfere with axonal outgrowth [6]. Only the most regularly shaped explants that furthermore had smooth edges were used for culturing (612 per cortical strip).
2.2. Culture Explants were cultured in a 30-ml drop of collagen gel, on a coverslip in a 10 culture dish. The collagen solution was prepared extemporaneously from 800 ml Vitrogen (3 mg / ml bovine collagen solution; Collagen Corp), 100 ml
carbonate buffer (0.1 N NaOH and 25 mg / ml NaHCO 3 ), 100 ml 103 DMEM (135 mg / ml DMEM; ICN Biomedicals), and 10 ml 1 N HCl. The collagen solution was stored in melting ice. Two 30-ml drops of collagen solution were placed on the bottom of a culture dish. The explant was washed through one of these drops immediately before introduction into the center of the other drop. Using a fine blunt needle the explant was then gently coaxed into a central and horizontal position. The culture dishes were placed in the CO 2 incubator for 2 h, for gelation of the collagen. Thereafter 1 ml of R12 medium was added to each culture dish [30], and the dishes were maintained in a CO 2 incubator (378C, 5.5% CO 2 ) for exactly 48 h. Each dish was marked with a number tracing it back to the litter from which the culture originated, and a serial number. A
62
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
separate list was kept to link these numbers to the origin of the explant (HL or FL).
2.3. Staining After the culture period (48 h) the cultures were immersed for 18 h in a modified Karnovsky fixative, consisting of 1% paraformaldehyde and 1.25% glutardialdehyde in phosphate buffered saline (PBS: 0.1 M, pH 7.2). The staining procedure used to visualize the explants and their outgrowing axons is a modification of a previously described method [2,5,7,10,12,19,20]. The cultures were washed for 3310 min in PBS, osmicated in a large drop of osmiumtetroxide (1% OsO 4 in PBS) for 2 h [5], washed for 3320 min in distilled water, and rinsed in 70% ethanol for 10 min. To intensify osmium staining, cultures were incubated with 1% orthophenylenediamine in 70% ethanol for 20 min [1,10,20]. The cultures were then washed and dehydrated in ascending concentrations of ethanol (70, 80, 90, 96 and 100%, 10 min each), cleared with xylene (10 min), and mounted with Entellan (Merck).
2.4. Measurements Axonal outgrowth was assessed by measuring the width of the (more or less circular) axonal corona surrounding the explant. Since explants were located eccentrically (the circle centered on the apex; see Results), there would always be one place where the corona was at its widest. At this place the width of the corona was measured. Through a microscope, at low magnification an imaginary line was drawn connecting the outermost growth cones of the axonal corona, and the distance between the most remote growth cone on this line and the nearest surface of the explant was measured by means of a gauged ocular grid. Each imaginary line was defined by at least 25 growth cones. The distance between the tip of the axon (visible as the growth cone), and the nearest surface of the explant was designated as the ‘length’ of the axon (Fig. 2). All 280 explants were measured three times by two independent observers. The means of these three measurements were used for further statistics. When the standard deviation of these three measurements was more than 0.2 mm all data from the explant were excluded from further analysis (n 5 16). Errors generally occurred because of an atypical distribution of the axonal length, i.e. one single axon being much longer than the general corona of axons around the explant. After completion of the measurements the values were correlated to the separately listed data on their site of origin (FL or HL) or age. The SSPS software package was used for statistical analysis of the numerical data. The data set consisted of 264 measurements. Data from the same rat (n518) and the same area (n52) were aggregated out, resulting in 36 means of length (Table 1). The influence of the area of origin (HL or FL) was then assayed with a paired t-test.
Fig. 2. Measurement of the axonal corona: an imaginary line was constructed by connecting the outermost growth cones of the axonal corona. Since neighboring axons tended to be of the same size, this would result in a smooth, approximately circular line, in which the explant was located eccentrically. The linear distance between the most remote growth cone on this line and the nearest surface of the explant was measured. This value was taken to represent the length of the longest axon.
This test was followed by an analysis of variance (ANOVA) to test whether age (five different ages were used) had an influence on the average length per area, and
Table 1 Mean length in mm of the longest axons per animal a Rat
Age
1 G17 2 G18 3 G18 4 G18 5 G18 6 G18 7 G18 8 G19 9 G19 10 G24 11 G24 12 G24 13 G24 14 G24 15 G24 16 G25 17 G25 18 G25 Total n cultures Overall average / area a
n HL
HL
n FL
FL
10 8 3 4 6 5 8 9 5 12 10 11 12 6 5 8 8 9 139
0.7360.20 1.0260.25 1.1060.33 1.2360.31 1.5060.25 1.4860.50 1.1260.36 0.9060.22 0.7860.20 0.5660.14 0.5960.14 1.7960.14 1.1560.21 1.6360.16 1.3860.52 1.3760.38 1.2760.25 1.5860.21
7 3 5 8 3 10 5 6 6 6 10 9 11 5 6 7 8 10 125
1.0360.15 0.8260.46 0.9760.44 0.8760.39 1.2460.54 1.1560.56 1.3660.15 0.5360.15 0.7260.18 0.7960.36 0.7460.20 0.9060.52 0.6760.25 1.2560.27 1.0360.22 1.5060.16 1.3760.21 1.4060.41
1.1860.36
1.0260.29
Age, the number of explants per pup per region (n HL , n FL ) and the mean length6S.D. are listed for each pup. Below the total number of explants per area, and the overall mean length6S.D. per area are indicated. Overall mean length per area was averaged from the 18 values listed in the column above.
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
whether it modifies the variation of length between the two areas. A probability of P,0.05 was considered significant.
3. Results Observations at arbitrary intervals during the 48-h culture period revealed a tendency for the first axons to emerge from the sides of the deep apex. Eventually axons would emerge from all sides of the explant. After 48 h the explants were surrounded by a radiate corona of axons (Fig. 3). The axonal corona was generally wider along the sides of the deep apex. Growth cones could be observed throughout the width of the axonal corona, indicating continuous emergence of axons from the explant during the culture period. Most axons pointed away from the explant, along gentle curves to either side. In cultures of explants harvested at G17 and G18 thick bundles of axons were commonly observed. Close to the explant surface the density of visible structures (tubular axonal segments, varicose axonal segments, and growth cones) in each plane of vision made it impossible to follow individual axons. In general the distribution of the lengths of the axons defining the perimeter of any specific corona was uniform, i.e. neighboring axons tended to be of the same size, though the whole corona was generally longer along the deep apices of the explant. In some cases one or a few axons stood out from the corona. These cases were excluded from the analysis (see Discussion). The longest axon observed was 1.95 mm long. The average length of the longest axon per explant was 1.1060.11 mm. Between the various explants the size of the longest axon defining the corona was quite variable (Table 1). The mean length of the axons from the explants taken out of HL and FL was respectively 1.1860.36 and 1.0260.29 mm (Fig. 4 and Table 1). The paired t-test performed to compare the two populations demonstrated that the observed difference of 0.16 mm is significant (P50.044). Using ANOVA analysis no significant effect of age could be demonstrated on the average axonal length within the FL or HL areas (P50.068), nor could a significant effect of age on the difference of the average axonal length between the FL and HL areas be demonstrated (P50.405).
4. Discussion The observed difference of axonal length after 2 days culture is 0.16 mm, i.e. 14% of the mean length of the longest axon. This difference is not sufficient to explain arrival of descending axons from HL at the cervical spinal cord before those from FL in vivo. Making up for lost time and overtaking should occur after G16 (departure from medial cortex [9]), and before G26 [33]. In 10 days the HL originating axons would elongate 0.8 mm more than the
63
FL originating axons, not even enough to overcome the longer distance they have to travel due to the growth of the nervous system itself. The actual time-window available is even shorter, since at the level of the pons the laterally originating cortical fibers are still in advance of the medially originating cortical fibers, necessitating an even larger difference in elongation velocity [21]. However, the circumstances in vitro may diminish the difference in axonal elongation speed as it exists in vivo; it has been shown, for instance, that cortical plate membranes enhance cortical neurite growth in a maturation-dependent fashion [35]. Furthermore, axons extending from cortical explants in vitro are derived from all layers of the cortex [15], while spinal projecting cortical axons arise from cortical lamina V only. Faster growth is not necessarily a trait restricted to lamina V pyramidal neurons, but could be a general trait of medial cortical neurons. The axons that defined the perimeter of the corona could originate in other layers, and obscure a more pronounced difference within the corona between FL and HL spinal projecting axons. Finally, because axons were measured linearly from the growth cone to the surface of the explant, the actual length of the curvilinear axons will be underestimated, possibly diminishing the size of the difference between HL and FL outgrowth. The amount of outgrowth observed from the individual explants was quite variable, in respect to both the width and the density of the corona. In a pilot series we determined that it was important to process the tissue as fast and as gentle as possible, but even under ideal circumstances variability persisted. Despite the use of Wistar Glaxo albino rats that were bred under constant conditions, different pups or pups from different litters might still display different abilities to grow processes. Though special care was taken during trimming of the explants to avoid stretching, to remove all the white matter and to obtain identically shaped and sized explants small variations will persist. We do not know whether the observed difference is the result of a discretely or a gradually divided property of the cortical neurons. Aptitude for growth may be variable across a small distance, and small variations in the location of the excised strip may again result in variation of the resulting outgrowth. Notwithstanding these difficulties it was still possible to demonstrate a small, but significant difference in the average width of the axonal corona of explants from the FL and HL somatomotor cortical areas. Since the difference existed independently of age, and in culture, we conclude that the difference represents an intrinsic property of the somatomotor cortex: axons emerging from the medial somatomotor cortex (HL) elongate faster than axons emerging from the lateral somatomotor cortex (FL). Regionalization itself is an intrinsic property of the developing cortex, independent from the presence of thalamocortical afferents [23,24], and independent from the synaptic activity of these afferents or even from the
64
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
Fig. 3. Microphotograph of a representative explant. The axonal corona surrounds the explant entirely. The imaginary line connecting the outermost growth cone of the axonal corona is drawn. Each imaginary line was defined by at least 25 growth cones.
synaptic activity within the developing cortex [37]. In mutant mice lacking the Emx2 homeodomain transcription factor rostral and lateral cortical areas expand, whereas caudal and medial areas contract [4]. Moreover, in these mice the corticothalamic and thalamocortical projection
patterns change accordingly, indicating that regional identity of cortical projection neurons governs their projective behavior. In relation to the present results this would indicate the presence of intrinsically determined regions within the somatomotor cortex.
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
65
References
Fig. 4. Mean length6S.D. in mm of the longest axon per age and of all cases (All), from the HL (light gray) and FL (dark gray) cortices. Below the abscissa both the age and the number of rats used are indicated. The mean length of the axons from the explants taken out of HL and FL was, respectively, 1.1860.36 and 1.0260.29 mm. The difference of 0.16 mm is significant (P50.044).
The sequence of arrival of axons at the cervical spinal cord, their subsequent descent along the cord, and the sequence of their ingrowth into the spinal cord is identical for the corticospinal and rubrospinal projections [14,18,33]. In both cases the lumbar projecting fibers arrive in the spinal cord in advance of the cervical projecting fibers, and ingrowth occurs along a rostrocaudal gradient; the cervical fibers grow into the gray matter of the cord even before the lumbar projecting fibers have attained their projection level. Both projections are somatotopically organized, and the sequence of arrival of their fibers at the spinal cord is exploited to organize their projection pattern during development [16,17]. Since this developmental organization stratagem is phylogenetically older than the corticospinal projection (or even the cortex), since the corticospinal projection had to employ this stratagem to organize its projection during development, and since the corticospinal projection map had to fit to other cortical projection maps (cortico-ponto-cerebellar), evolutionary pressure must have resulted in the necessary increase in axonal elongation speed for axons from the HL somatomotor cortical area. The present findings thus indicate that intrinsic cortical traits that govern regionalization may not only influence targeting behavior of cortical projection neurons by influencing their sensitivity for axonal guidance molecules [4], but also by influencing their axonal elongation speed.
Acknowledgements We thank Dr R. Brand of the Medical Statistics Group (LUMC, Leiden) for his advice on statistical evaluation.
[1] G.F. Bahr, Osmium tetroxide and ruthenium tetroxide and their reactions with biologically important substances, Exp. Cell Res. 7 (1954) 457–479. [2] A.K. Bal, Localization of plant lipids for light microscopy using p-phenylenediamine in tissues of Arachis hypogaea L, Stain Technol. 65 (1990) 91–94. [3] S.A. Bayer, J. Altman, Directions in neurogenetic gradients and patterns of anatomical connections in the telencephalon, Prog. Neurobiol. 29 (1987) 57–106. [4] K.M. Bishop, G. Goudreau, D.D. O’Leary, Regulation of area identity in the mammalian neocortex by Emx2 and Pax6, Science 288 (2000) 334–349. [5] D.P. Boshier, H. Holloway, L.F. Kitchin, A comparison of standard lipid staining techniques used in electron microscopic studies of mammalian tissues, Stain Technol. 59 (1984) 83–89. [6] P. Caroni, M.E. Schwab, Two membrane protein fractions from rat central myelin with inhibitory properties for neurite growth and fibroblast spreading, J. Cell Biol. 106 (1988) 1281–1288. [7] O. Colman, J.C. Stockert, Observations on the osmium tetroxide-pphenylenediamine staining method for Epon semithin sections of nervous tissue, Acta Histochem. 72 (1983) 123–125. [8] J.P. Donoghue, S.P. Wise, The motor cortex of the rat: cytoarchitecture and microstimulation mapping, J. Comp. Neurol. 212 (1982) 76–88. [9] R.S. Erzurumlu, S. Jhaveri, Emergence of connectivity in the embryonic rat parietal cortex, Cereb. Cortex 2 (1992) 336–352. [10] J.F. Estable-Puig, W.C. Bauer, J.M. Blumberg, Paraphenylenediamine staining of osmium-fixed plastic-embedded tissue for light and phase microscopy, J. Neuropathol. Exp. Neurol. 24 (1965) 531–535. [11] R.D. Hall, E.P. Lindholm, Organisation of motor and somatosensory neocortex in the albino rat, Brain Res. 66 (1974) 23–38. [12] W.A. Harris, C.E. Holt, T.A. Smith, N. Gallenson, Growth cones of developing retinal cells in vivo, on culture surfaces, and in collagen matrices, J. Neurosci. Res. 13 (1985) 101–122. [13] S.P. Hicks, C.J. D’Amato, Cell migrations to the isocortex in the rat, Anat. Rec. 160 (1968) 619–634. [14] E.A. Joosten, A.A. Gribnau, P.J. Dederen, An anterograde tracer study of the developing corticospinal tract in the rat: three components, Brain Res. 433 (1987) 121–130. [15] E.A.J. Joosten, W.H. Gispen, P.R. Bar, Tropism and corticospinal target selection in the rat, Neuroscience 59 (1994) 33–41. [16] E.A.J.F. Lakke, Neurotropic attraction: facts and fiction, Biomed. Rev. 4 (1995) 95–102. [17] E.A.J.F. Lakke, The projections to the spinal cord of the rat during development: a timetable of descent, Adv. Anat. Embryol. Cell Biol. 135 (1997) I143. [18] E.A.J.F. Lakke, E. Marani, Prenatal descent of rubrospinal fibers through the spinal cord of the rat, J. Comp. Neurol. 314 (1991) 67–78. [19] J.M. Ledingham, F.O. Simpson, Intensification of osmium staining by p-phenylenediamine: paraffin and epon embedding; lipid granules in renal medulla, Stain Technol. 45 (1970) 255–260. [20] J.M. Ledingham, F.O. Simpson, The use of p-phenylenediamine in the block to enhance osmium staining for electron microscopy, Stain Technol. 47 (1972) 239–243. [21] T.B. Leergaard, E.A.J.F. Lakke, J.G. Bjaalie, Topographical organization in the early postnatal corticopontine projection: a carbocyanine dye and 3-D computer reconstruction study in the rat, J. Comp. Neurol. 361 (1995) 77–94. [22] M.W. Miller, Effect of prenatal exposure to alcohol on the distribution and time of origin of corticospinal neurons in the rat, J. Comp. Neurol. 257 (1987) 372–382. [23] E.M. Miyashita-Lin, R. Hevner, K.M. Wassarman, S. Martinez, J.L.
66
[24]
[25]
[26] [27]
[28] [29] [30]
[31]
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66 Rubenstein, Early neocortical regionalization in the absence of thalamic innervation, Science 285 (1999) 906–909. Y. Nakagawa, J.E. Johnson, D.D. O’Leary, Graded and areal expression patterns of regulatory genes and cadherins in embryonic neocortex independent of thalamocortical input, J. Neurosci. 19 (1999) 10877–10885. E.J. Neafsey, E.L. Bold, G. Haas, K.M. Hurley-Gius, G. Quirk, C.F. Sievert, R.R. Terreberry, The organization of the rat motor cortex: a microstimulation mapping study, Brain Res. 396 (1986) 77–96. D.D. O’Leary, Do cortical areas emerge from a protocortex?, Trends Neurosci. 12 (1989) 400–406. D.D. O’Leary, T. Terashima, Cortical axons branch to multiple subcortical targets by interstitial axon budding: implications for target recognition and ‘waiting periods’, Neuron 1 (1988) 901–910. G. Paxinos, Atlas of the Developing Rat Brain, Academic Press, Sydney, 1991. P. Rakic, Specification of cerebral cortical areas, Science 241 (1988) 170–176. H.J. Romijn, F. van-Huizen, P.S. Wolters, Towards an improved serum-free, chemically defined medium for long-term culturing of cerebral cortex tissue, Neurosci. Biobehav. Rev. 8 (1984) 301–334. J.L. Rubenstein, P. Rakic, Genetic control of cortical development, Cereb. Cortex 9 (1999) 521–523.
[32] D. Schreyer, E.G. Jones, A quantitative electron microscopic study of axon elimination in the growing corticospinal albino rat tract of the rat, Dev. Brain Res. 38 (1988) 103–119. [33] D.J. Schreyer, E.H.G. Jones, Topographic sequence of outgrowths of corticospinal axons in the rat: a study using retrograde labelling with Fast blue, Dev. Brain Res. 38 (1988) 89–101. [34] I.H. Smart, Histogenesis of the mesocortical area of the mouse telencephalon, J. Anat. 138 (Pt 3) (1984) 537–552. [35] R. Tuttle, B.L. Schlaggar, J.E. Braisted, D.D. O’Leary, Maturationdependent upregulation of growth-promoting molecules in developing cortical plate controls thalamic and cortical neurite growth, J. Neurosci. 15 (1995) 3039–3052. [36] J. Van der Zwet, P.A. Van Tilburg, R.E. Poelmann, Oren wapperen, een betrouwbare manier om de bronst bij ratten te bepalen, Biotechniek 3 (1986) 48–49. [37] M. Verhage, A.S. Maia, J.J. Plomp, A.B. Brussaard, J.H. Heeroma, H. Vermeer, R.F. Toonen, R.E. Hammer, T.K. van den Berg, M. Missler, H.J. Geuze, T.C. Sudhof, Synaptic assembly of the brain in the absence of neurotransmitter secretion, Science 287 (2000) 864– 869.
Research report
Intrinsic properties of the developing motor cortex in the rat: in vitro axons from the medial somatomotor cortex grow faster than axons from the lateral somatomotor cortex J.M.P. Mouveroux a , M. Verkijk a , E.A.J.F. Lakke a , *, E. Marani a,b b
a Neuroregulation Group, Department of Neurosurgery, Leiden University Medical Centre, Leiden, The Netherlands Signals and Systems, Department of Electrical Engineering, Biomedical Technology Institute, University of Twente, Twente, The Netherlands
Accepted 9 May 2000
Abstract The axons that originate in the medial somatomotor cortex of the rat depart, during development, after those from the lateral somatomotor cortex, yet they arrive in the cervical spinal cord first. Either the medially originating axons elongate faster, or the laterally originating ones pause along the descent pathway. To investigate the presence of an intrinsic difference of the axonal elongation velocity between the lateral and medial somatomotor cortical areas, we cultured explants taken from these areas for 2 days, and measured the length of the outgrowth. After 2 days the explants were surrounded by a radiate corona of axons of which the longest measured 1.95 mm. A significant difference was detected between the medial and lateral somatomotor cortical areas in vitro. Axons originating from explants taken from the medial somatomotor cortical area are, after 2 days in culture, on average 0.16 mm longer than those from the lateral somatomotor cortical area. Though the observed difference is not large enough to allow for the overtaking observed in vivo, it does indicate that intrinsic differences exist within the developing rat somatomotor cortex. This in turn indicates that intrinsic cortical traits not only influence regionalization and targeting behavior of cortical projection neurons, but also their axonal elongation speed. 2000 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Process outgrowth, growth cones and sprouting Keywords: Rat; Somatomotor cortex; Development; Culture; Axonal outgrowth; Arealization; Regionalization
1. Introduction The differentiation of areas within the cerebral cortex has been a contested issue. Regionalization has been proposed to be controlled by patterning mechanisms intrinsic to the cortex [29], or by extrinsic influences such as the thalamocortical afferents [26]. At present the evidence strongly supports the presence of intrinsic mechanisms governing cortical regionalization (see Rubenstein and Rakic [31] for a review). The same mechanisms that govern cortical regionalization also influence targeting behavior of the cortical projection neurons, presumably by *Corresponding author. Department of Physiology, P.O. Box 9604, NL-2300 RC Leiden, The Netherlands. Tel.: 131-71-527-6759. E-mail address: [email protected] (E.A.J.F. Lakke)
influencing their sensitivity for axonal guidance molecules [4]. We will argue that axonal elongation speed is another trait of the cortical projection neurons that is under control of these intrinsic cortical mechanisms. Cortical projection axons do not arrive at the spinal cord synchronously. The first corticospinal axons arrive at cervical levels 3 days after birth. These axons originate from a small area located medially in the dorsal parietal cortex, and these axons are destined for the lumbar spinal cord [33]. In adulthood the same area of dorsal parietal cortex represents the hindlimb area (HL), where the motor area I (MI) and the sensory area I (SI) overlap [8,11]. During the first week the area of cortical neurons that reach down to the cervical spinal cord expands to a continuous sheet that covers most of the frontal parietal and cingulate cortex. This area includes the forelimb area
0165-3806 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0165-3806( 00 )00056-0
60
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
(FL), located in the lateral dorsal parietal cortex [33]. Later during postnatal development many of the cortical projection neurons lose their spinal axon, resulting in the discontinuous distribution of spinal projection neurons as observed in adult rats [32]. In short, the axons from the medially located HL area arrive in the cervical spinal cord before the axons from the laterally located FL area. The neurogenetic gradient of the cortex is like a wavepattern emerging from a center located anterolaterally. The wave front travels rostrad, mediad and caudad simultaneously, but the general direction is from rostral to caudal and from lateral to medial [13,34]. Thus the laterally located FL area neurons are generated before those of the medially located HL area. The axogenetic gradient of the cortex is similarly oriented, the first projection axons that appear in the intern capsule originate in the lateral cortex, and in time axons from progressively more medial origins are added [3,9,22,27]. Thus, axons from the laterally located FL area will enter the internal capsule before those of the medially located HL area; the medial area axons are generated later, and have to travel a longer distance to reach the internal capsule. The nervous system itself increases in size during development, and so the distance between the cortex and the spinal cord increases continuously. Thus a later departure from the cortex towards the spinal cord implies a longer distance to be covered. Even though the axons from the HL area of the cortex are initially in arrear, they arrive first at the cervical spinal cord. Their growth cones must thus have overtaken the growth cones that originated in more lateral cortical areas. Either those growth cones have stopped along the way, or the medial growth cones have traveled faster. Medial growth cones could travel faster because of an intrinsic property of their parent neurons in the HL somatomotor area, or because of a stimulating influence along the way. To investigate whether an intrinsic property of the neurons in the HL somatomotor cortical area is responsible for the faster elongation of its descending axons we cultured explants from the lateral and medial areas of the dorsal frontoparietal cortex, corresponding to, respectively, the forelimb and hindlimb somatomotor cortical areas (Fig. 1). The explants were embedded in a collagen matrix, and cultured for 2 days in a chemically defined medium. In our experience explant culture embedded in a three-dimensional matrix results in more profuse outgrowth as compared to outgrowth on a two-dimensional matrix. After 2 days the length of the outgrowth was measured and compared. The culture period was limited to 2 days by the size of the collagen drop in which the explants were embedded. After culture for longer periods the outgrowing axons would reach the periphery of the collagen drop and stop. Cortical projection axons originate in lamina V pyramidal cells. In the rat, lamina V neuroblasts of the cerebral cortex are generated from gestation day 15 (G15) through
G18. We used fetuses and neonates ranging in age from G17 to G25. After 2 days the explants were surrounded by a radiate corona of axons of which the longest measured 1.95 mm. A significant difference was detected between outgrowth from the medial and lateral somatomotor cortical areas in vitro. Axons originating from explants taken from the medial somatomotor cortical area are, after 2 days in culture, on average 0.16 mm longer than those from the lateral somatomotor cortical area. The results demonstrate the existence of an intrinsically determined difference of the axonal growth velocity between cultured explants taken from medial and lateral areas of the rat motor cortex. This indicates that intrinsic cortical traits not only influence regionalization and targeting behavior of cortical projection neurons [4], but also their axonal elongation speed.
2. Materials and methods
2.1. Explants Wistar Albino Glaxo rats in oestro were mated between 10:00 and 11:00 h [36]. The end of this period was taken as the start of G0 [28]. In our colony, pregnancy lasts 23 days, i.e. the pups are born early on G23. Rat fetuses and pups of ages ranging from G17 to G25 were used. Fetuses were harvested at the required duration of gestation, after sacrificing the pregnant dam. All procedures for the preparation of cortical explants took place under sterile conditions. Immediately before the rats were decapitated, they were introduced into the flowcabinet, submerged into 70% ethanol and rinsed in cold normal saline. Skin, parietal and frontal bones were then quickly resected, to reveal the cortical surface. In the youngest fetuses (G17, G18) the cortical surface was moistened with a few drops of normal saline. One longitudinal strip of dorsal fronto-parietal cerebral cortex was then excised from each hemisphere, each strip measuring |632 mm. On one side the strip was excised from HL, and on the other side from FL (Fig. 1). Excision sequence (FL or HL first) and side (from left or right hemisphere first) were alternated between rats. The strips were collected in a chemically defined, serum free medium (R12) [30], and chopped immediately after excision into 250-mm slices. These slices were collected in the same medium, separated by gentle aspiration into a Pasteur pipette, and subsequently stored in a CO 2 incubator (378C, 5.5% CO 2 ). Processing time, from decapitation to storage never exceeded 13 min. Individual rats were processed sequentially, and all batches of slices were stored in separate Petri dishes. Under a dissection microscope, those slices of each batch whose cortical surface appeared to be undamaged, and that were of uniform thickness were selected for trimming. With a fine scalpel a wedge-shaped section was
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
61
Fig. 1. (A) Location of the cortical strips. Dorsal view of the rat brain. The shaded area indicates the position of the somatomotor cortex, shaped like a rattunculus [25]. The rattunculus is repeated above, to indicate the position of the HL (backward-hatched) and FL (forward-hatched) areas within the somatomotor cortex. Rectangles on the cortex indicate the excision areas of the cortical strips. Excision sequence (FL or HL first) and side (from left or right hemisphere first) were alternated between rats. (B) Preparation of the explants. The cortical strips were chopped into 250-mm slices. A wedge-shaped section was excised out of each slice, to obtain an equilateral triangle |1 mm wide, yet with a recognizable pial surface. Care was taken to exclude white matter (shaded gray) from the resulting explants. Each explant was cultured in a separate 10 culture dish. Each dish was marked with a number tracing it back to the litter from which the culture originated, and a serial number. A separate list was kept to link these numbers to the origin of the explant (HL or FL). Only after completion of the measurements were the measured values correlated to the separately listed data on their site of origin (FL or HL) or age.
excised out of each slice, to obtain an equilateral triangle of |1 mm width, yet with a recognizable pial surface (Fig. 1). Care was taken to exclude white matter from the resulting explants, since it may interfere with axonal outgrowth [6]. Only the most regularly shaped explants that furthermore had smooth edges were used for culturing (612 per cortical strip).
2.2. Culture Explants were cultured in a 30-ml drop of collagen gel, on a coverslip in a 10 culture dish. The collagen solution was prepared extemporaneously from 800 ml Vitrogen (3 mg / ml bovine collagen solution; Collagen Corp), 100 ml
carbonate buffer (0.1 N NaOH and 25 mg / ml NaHCO 3 ), 100 ml 103 DMEM (135 mg / ml DMEM; ICN Biomedicals), and 10 ml 1 N HCl. The collagen solution was stored in melting ice. Two 30-ml drops of collagen solution were placed on the bottom of a culture dish. The explant was washed through one of these drops immediately before introduction into the center of the other drop. Using a fine blunt needle the explant was then gently coaxed into a central and horizontal position. The culture dishes were placed in the CO 2 incubator for 2 h, for gelation of the collagen. Thereafter 1 ml of R12 medium was added to each culture dish [30], and the dishes were maintained in a CO 2 incubator (378C, 5.5% CO 2 ) for exactly 48 h. Each dish was marked with a number tracing it back to the litter from which the culture originated, and a serial number. A
62
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
separate list was kept to link these numbers to the origin of the explant (HL or FL).
2.3. Staining After the culture period (48 h) the cultures were immersed for 18 h in a modified Karnovsky fixative, consisting of 1% paraformaldehyde and 1.25% glutardialdehyde in phosphate buffered saline (PBS: 0.1 M, pH 7.2). The staining procedure used to visualize the explants and their outgrowing axons is a modification of a previously described method [2,5,7,10,12,19,20]. The cultures were washed for 3310 min in PBS, osmicated in a large drop of osmiumtetroxide (1% OsO 4 in PBS) for 2 h [5], washed for 3320 min in distilled water, and rinsed in 70% ethanol for 10 min. To intensify osmium staining, cultures were incubated with 1% orthophenylenediamine in 70% ethanol for 20 min [1,10,20]. The cultures were then washed and dehydrated in ascending concentrations of ethanol (70, 80, 90, 96 and 100%, 10 min each), cleared with xylene (10 min), and mounted with Entellan (Merck).
2.4. Measurements Axonal outgrowth was assessed by measuring the width of the (more or less circular) axonal corona surrounding the explant. Since explants were located eccentrically (the circle centered on the apex; see Results), there would always be one place where the corona was at its widest. At this place the width of the corona was measured. Through a microscope, at low magnification an imaginary line was drawn connecting the outermost growth cones of the axonal corona, and the distance between the most remote growth cone on this line and the nearest surface of the explant was measured by means of a gauged ocular grid. Each imaginary line was defined by at least 25 growth cones. The distance between the tip of the axon (visible as the growth cone), and the nearest surface of the explant was designated as the ‘length’ of the axon (Fig. 2). All 280 explants were measured three times by two independent observers. The means of these three measurements were used for further statistics. When the standard deviation of these three measurements was more than 0.2 mm all data from the explant were excluded from further analysis (n 5 16). Errors generally occurred because of an atypical distribution of the axonal length, i.e. one single axon being much longer than the general corona of axons around the explant. After completion of the measurements the values were correlated to the separately listed data on their site of origin (FL or HL) or age. The SSPS software package was used for statistical analysis of the numerical data. The data set consisted of 264 measurements. Data from the same rat (n518) and the same area (n52) were aggregated out, resulting in 36 means of length (Table 1). The influence of the area of origin (HL or FL) was then assayed with a paired t-test.
Fig. 2. Measurement of the axonal corona: an imaginary line was constructed by connecting the outermost growth cones of the axonal corona. Since neighboring axons tended to be of the same size, this would result in a smooth, approximately circular line, in which the explant was located eccentrically. The linear distance between the most remote growth cone on this line and the nearest surface of the explant was measured. This value was taken to represent the length of the longest axon.
This test was followed by an analysis of variance (ANOVA) to test whether age (five different ages were used) had an influence on the average length per area, and
Table 1 Mean length in mm of the longest axons per animal a Rat
Age
1 G17 2 G18 3 G18 4 G18 5 G18 6 G18 7 G18 8 G19 9 G19 10 G24 11 G24 12 G24 13 G24 14 G24 15 G24 16 G25 17 G25 18 G25 Total n cultures Overall average / area a
n HL
HL
n FL
FL
10 8 3 4 6 5 8 9 5 12 10 11 12 6 5 8 8 9 139
0.7360.20 1.0260.25 1.1060.33 1.2360.31 1.5060.25 1.4860.50 1.1260.36 0.9060.22 0.7860.20 0.5660.14 0.5960.14 1.7960.14 1.1560.21 1.6360.16 1.3860.52 1.3760.38 1.2760.25 1.5860.21
7 3 5 8 3 10 5 6 6 6 10 9 11 5 6 7 8 10 125
1.0360.15 0.8260.46 0.9760.44 0.8760.39 1.2460.54 1.1560.56 1.3660.15 0.5360.15 0.7260.18 0.7960.36 0.7460.20 0.9060.52 0.6760.25 1.2560.27 1.0360.22 1.5060.16 1.3760.21 1.4060.41
1.1860.36
1.0260.29
Age, the number of explants per pup per region (n HL , n FL ) and the mean length6S.D. are listed for each pup. Below the total number of explants per area, and the overall mean length6S.D. per area are indicated. Overall mean length per area was averaged from the 18 values listed in the column above.
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
whether it modifies the variation of length between the two areas. A probability of P,0.05 was considered significant.
3. Results Observations at arbitrary intervals during the 48-h culture period revealed a tendency for the first axons to emerge from the sides of the deep apex. Eventually axons would emerge from all sides of the explant. After 48 h the explants were surrounded by a radiate corona of axons (Fig. 3). The axonal corona was generally wider along the sides of the deep apex. Growth cones could be observed throughout the width of the axonal corona, indicating continuous emergence of axons from the explant during the culture period. Most axons pointed away from the explant, along gentle curves to either side. In cultures of explants harvested at G17 and G18 thick bundles of axons were commonly observed. Close to the explant surface the density of visible structures (tubular axonal segments, varicose axonal segments, and growth cones) in each plane of vision made it impossible to follow individual axons. In general the distribution of the lengths of the axons defining the perimeter of any specific corona was uniform, i.e. neighboring axons tended to be of the same size, though the whole corona was generally longer along the deep apices of the explant. In some cases one or a few axons stood out from the corona. These cases were excluded from the analysis (see Discussion). The longest axon observed was 1.95 mm long. The average length of the longest axon per explant was 1.1060.11 mm. Between the various explants the size of the longest axon defining the corona was quite variable (Table 1). The mean length of the axons from the explants taken out of HL and FL was respectively 1.1860.36 and 1.0260.29 mm (Fig. 4 and Table 1). The paired t-test performed to compare the two populations demonstrated that the observed difference of 0.16 mm is significant (P50.044). Using ANOVA analysis no significant effect of age could be demonstrated on the average axonal length within the FL or HL areas (P50.068), nor could a significant effect of age on the difference of the average axonal length between the FL and HL areas be demonstrated (P50.405).
4. Discussion The observed difference of axonal length after 2 days culture is 0.16 mm, i.e. 14% of the mean length of the longest axon. This difference is not sufficient to explain arrival of descending axons from HL at the cervical spinal cord before those from FL in vivo. Making up for lost time and overtaking should occur after G16 (departure from medial cortex [9]), and before G26 [33]. In 10 days the HL originating axons would elongate 0.8 mm more than the
63
FL originating axons, not even enough to overcome the longer distance they have to travel due to the growth of the nervous system itself. The actual time-window available is even shorter, since at the level of the pons the laterally originating cortical fibers are still in advance of the medially originating cortical fibers, necessitating an even larger difference in elongation velocity [21]. However, the circumstances in vitro may diminish the difference in axonal elongation speed as it exists in vivo; it has been shown, for instance, that cortical plate membranes enhance cortical neurite growth in a maturation-dependent fashion [35]. Furthermore, axons extending from cortical explants in vitro are derived from all layers of the cortex [15], while spinal projecting cortical axons arise from cortical lamina V only. Faster growth is not necessarily a trait restricted to lamina V pyramidal neurons, but could be a general trait of medial cortical neurons. The axons that defined the perimeter of the corona could originate in other layers, and obscure a more pronounced difference within the corona between FL and HL spinal projecting axons. Finally, because axons were measured linearly from the growth cone to the surface of the explant, the actual length of the curvilinear axons will be underestimated, possibly diminishing the size of the difference between HL and FL outgrowth. The amount of outgrowth observed from the individual explants was quite variable, in respect to both the width and the density of the corona. In a pilot series we determined that it was important to process the tissue as fast and as gentle as possible, but even under ideal circumstances variability persisted. Despite the use of Wistar Glaxo albino rats that were bred under constant conditions, different pups or pups from different litters might still display different abilities to grow processes. Though special care was taken during trimming of the explants to avoid stretching, to remove all the white matter and to obtain identically shaped and sized explants small variations will persist. We do not know whether the observed difference is the result of a discretely or a gradually divided property of the cortical neurons. Aptitude for growth may be variable across a small distance, and small variations in the location of the excised strip may again result in variation of the resulting outgrowth. Notwithstanding these difficulties it was still possible to demonstrate a small, but significant difference in the average width of the axonal corona of explants from the FL and HL somatomotor cortical areas. Since the difference existed independently of age, and in culture, we conclude that the difference represents an intrinsic property of the somatomotor cortex: axons emerging from the medial somatomotor cortex (HL) elongate faster than axons emerging from the lateral somatomotor cortex (FL). Regionalization itself is an intrinsic property of the developing cortex, independent from the presence of thalamocortical afferents [23,24], and independent from the synaptic activity of these afferents or even from the
64
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
Fig. 3. Microphotograph of a representative explant. The axonal corona surrounds the explant entirely. The imaginary line connecting the outermost growth cone of the axonal corona is drawn. Each imaginary line was defined by at least 25 growth cones.
synaptic activity within the developing cortex [37]. In mutant mice lacking the Emx2 homeodomain transcription factor rostral and lateral cortical areas expand, whereas caudal and medial areas contract [4]. Moreover, in these mice the corticothalamic and thalamocortical projection
patterns change accordingly, indicating that regional identity of cortical projection neurons governs their projective behavior. In relation to the present results this would indicate the presence of intrinsically determined regions within the somatomotor cortex.
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66
65
References
Fig. 4. Mean length6S.D. in mm of the longest axon per age and of all cases (All), from the HL (light gray) and FL (dark gray) cortices. Below the abscissa both the age and the number of rats used are indicated. The mean length of the axons from the explants taken out of HL and FL was, respectively, 1.1860.36 and 1.0260.29 mm. The difference of 0.16 mm is significant (P50.044).
The sequence of arrival of axons at the cervical spinal cord, their subsequent descent along the cord, and the sequence of their ingrowth into the spinal cord is identical for the corticospinal and rubrospinal projections [14,18,33]. In both cases the lumbar projecting fibers arrive in the spinal cord in advance of the cervical projecting fibers, and ingrowth occurs along a rostrocaudal gradient; the cervical fibers grow into the gray matter of the cord even before the lumbar projecting fibers have attained their projection level. Both projections are somatotopically organized, and the sequence of arrival of their fibers at the spinal cord is exploited to organize their projection pattern during development [16,17]. Since this developmental organization stratagem is phylogenetically older than the corticospinal projection (or even the cortex), since the corticospinal projection had to employ this stratagem to organize its projection during development, and since the corticospinal projection map had to fit to other cortical projection maps (cortico-ponto-cerebellar), evolutionary pressure must have resulted in the necessary increase in axonal elongation speed for axons from the HL somatomotor cortical area. The present findings thus indicate that intrinsic cortical traits that govern regionalization may not only influence targeting behavior of cortical projection neurons by influencing their sensitivity for axonal guidance molecules [4], but also by influencing their axonal elongation speed.
Acknowledgements We thank Dr R. Brand of the Medical Statistics Group (LUMC, Leiden) for his advice on statistical evaluation.
[1] G.F. Bahr, Osmium tetroxide and ruthenium tetroxide and their reactions with biologically important substances, Exp. Cell Res. 7 (1954) 457–479. [2] A.K. Bal, Localization of plant lipids for light microscopy using p-phenylenediamine in tissues of Arachis hypogaea L, Stain Technol. 65 (1990) 91–94. [3] S.A. Bayer, J. Altman, Directions in neurogenetic gradients and patterns of anatomical connections in the telencephalon, Prog. Neurobiol. 29 (1987) 57–106. [4] K.M. Bishop, G. Goudreau, D.D. O’Leary, Regulation of area identity in the mammalian neocortex by Emx2 and Pax6, Science 288 (2000) 334–349. [5] D.P. Boshier, H. Holloway, L.F. Kitchin, A comparison of standard lipid staining techniques used in electron microscopic studies of mammalian tissues, Stain Technol. 59 (1984) 83–89. [6] P. Caroni, M.E. Schwab, Two membrane protein fractions from rat central myelin with inhibitory properties for neurite growth and fibroblast spreading, J. Cell Biol. 106 (1988) 1281–1288. [7] O. Colman, J.C. Stockert, Observations on the osmium tetroxide-pphenylenediamine staining method for Epon semithin sections of nervous tissue, Acta Histochem. 72 (1983) 123–125. [8] J.P. Donoghue, S.P. Wise, The motor cortex of the rat: cytoarchitecture and microstimulation mapping, J. Comp. Neurol. 212 (1982) 76–88. [9] R.S. Erzurumlu, S. Jhaveri, Emergence of connectivity in the embryonic rat parietal cortex, Cereb. Cortex 2 (1992) 336–352. [10] J.F. Estable-Puig, W.C. Bauer, J.M. Blumberg, Paraphenylenediamine staining of osmium-fixed plastic-embedded tissue for light and phase microscopy, J. Neuropathol. Exp. Neurol. 24 (1965) 531–535. [11] R.D. Hall, E.P. Lindholm, Organisation of motor and somatosensory neocortex in the albino rat, Brain Res. 66 (1974) 23–38. [12] W.A. Harris, C.E. Holt, T.A. Smith, N. Gallenson, Growth cones of developing retinal cells in vivo, on culture surfaces, and in collagen matrices, J. Neurosci. Res. 13 (1985) 101–122. [13] S.P. Hicks, C.J. D’Amato, Cell migrations to the isocortex in the rat, Anat. Rec. 160 (1968) 619–634. [14] E.A. Joosten, A.A. Gribnau, P.J. Dederen, An anterograde tracer study of the developing corticospinal tract in the rat: three components, Brain Res. 433 (1987) 121–130. [15] E.A.J. Joosten, W.H. Gispen, P.R. Bar, Tropism and corticospinal target selection in the rat, Neuroscience 59 (1994) 33–41. [16] E.A.J.F. Lakke, Neurotropic attraction: facts and fiction, Biomed. Rev. 4 (1995) 95–102. [17] E.A.J.F. Lakke, The projections to the spinal cord of the rat during development: a timetable of descent, Adv. Anat. Embryol. Cell Biol. 135 (1997) I143. [18] E.A.J.F. Lakke, E. Marani, Prenatal descent of rubrospinal fibers through the spinal cord of the rat, J. Comp. Neurol. 314 (1991) 67–78. [19] J.M. Ledingham, F.O. Simpson, Intensification of osmium staining by p-phenylenediamine: paraffin and epon embedding; lipid granules in renal medulla, Stain Technol. 45 (1970) 255–260. [20] J.M. Ledingham, F.O. Simpson, The use of p-phenylenediamine in the block to enhance osmium staining for electron microscopy, Stain Technol. 47 (1972) 239–243. [21] T.B. Leergaard, E.A.J.F. Lakke, J.G. Bjaalie, Topographical organization in the early postnatal corticopontine projection: a carbocyanine dye and 3-D computer reconstruction study in the rat, J. Comp. Neurol. 361 (1995) 77–94. [22] M.W. Miller, Effect of prenatal exposure to alcohol on the distribution and time of origin of corticospinal neurons in the rat, J. Comp. Neurol. 257 (1987) 372–382. [23] E.M. Miyashita-Lin, R. Hevner, K.M. Wassarman, S. Martinez, J.L.
66
[24]
[25]
[26] [27]
[28] [29] [30]
[31]
J.M.P. Mouveroux et al. / Developmental Brain Research 122 (2000) 59 – 66 Rubenstein, Early neocortical regionalization in the absence of thalamic innervation, Science 285 (1999) 906–909. Y. Nakagawa, J.E. Johnson, D.D. O’Leary, Graded and areal expression patterns of regulatory genes and cadherins in embryonic neocortex independent of thalamocortical input, J. Neurosci. 19 (1999) 10877–10885. E.J. Neafsey, E.L. Bold, G. Haas, K.M. Hurley-Gius, G. Quirk, C.F. Sievert, R.R. Terreberry, The organization of the rat motor cortex: a microstimulation mapping study, Brain Res. 396 (1986) 77–96. D.D. O’Leary, Do cortical areas emerge from a protocortex?, Trends Neurosci. 12 (1989) 400–406. D.D. O’Leary, T. Terashima, Cortical axons branch to multiple subcortical targets by interstitial axon budding: implications for target recognition and ‘waiting periods’, Neuron 1 (1988) 901–910. G. Paxinos, Atlas of the Developing Rat Brain, Academic Press, Sydney, 1991. P. Rakic, Specification of cerebral cortical areas, Science 241 (1988) 170–176. H.J. Romijn, F. van-Huizen, P.S. Wolters, Towards an improved serum-free, chemically defined medium for long-term culturing of cerebral cortex tissue, Neurosci. Biobehav. Rev. 8 (1984) 301–334. J.L. Rubenstein, P. Rakic, Genetic control of cortical development, Cereb. Cortex 9 (1999) 521–523.
[32] D. Schreyer, E.G. Jones, A quantitative electron microscopic study of axon elimination in the growing corticospinal albino rat tract of the rat, Dev. Brain Res. 38 (1988) 103–119. [33] D.J. Schreyer, E.H.G. Jones, Topographic sequence of outgrowths of corticospinal axons in the rat: a study using retrograde labelling with Fast blue, Dev. Brain Res. 38 (1988) 89–101. [34] I.H. Smart, Histogenesis of the mesocortical area of the mouse telencephalon, J. Anat. 138 (Pt 3) (1984) 537–552. [35] R. Tuttle, B.L. Schlaggar, J.E. Braisted, D.D. O’Leary, Maturationdependent upregulation of growth-promoting molecules in developing cortical plate controls thalamic and cortical neurite growth, J. Neurosci. 15 (1995) 3039–3052. [36] J. Van der Zwet, P.A. Van Tilburg, R.E. Poelmann, Oren wapperen, een betrouwbare manier om de bronst bij ratten te bepalen, Biotechniek 3 (1986) 48–49. [37] M. Verhage, A.S. Maia, J.J. Plomp, A.B. Brussaard, J.H. Heeroma, H. Vermeer, R.F. Toonen, R.E. Hammer, T.K. van den Berg, M. Missler, H.J. Geuze, T.C. Sudhof, Synaptic assembly of the brain in the absence of neurotransmitter secretion, Science 287 (2000) 864– 869.