DEVELOPMENTALBIOLOGY 141,84-92(1990)

Increased Expression of Tal a-Tubulin mRNA during Collateral and NGF-Induced Sprouting of Sympathetic Neurons T. C. MATHEW AND F. D. MILLERS Department

of Anatomy

and Cell Biology, University Accepted April

of Alberta, Edmonton,

Canada

27, 1990

We have examined expression of To1 cu-tubulin mRNA in the rat superior cervical ganglion (SCG) to determine whether changes in gene expression accompany neuronal sprouting and to investigate factors that regulate growth-associated genes in intact neurons. Northern blot analysis demonstrates that levels of Tal a-tubulin mRNA increase in the uninjured SCG following transection of contralateral neurons that project to bilaterally innervated, but not unilaterally innervated target organs. The observed increase in uninjured neurons is associated with collateral sprouting, as measured by increased tyrosine hydroxylase immunoreaetivity within the pineal gland. These data suggest that targetderived factors may regulate Tal mRNA in sprouting neurons. Consistent with this hypothesis, systemic NGF treatment of neonatal animals over a developmental interval when Tal c*-tubulin mRNA normally decreases led to a 5- to lo-fold increase in Tal mRNA levels in developing sympathetic neurons. In addition, deafferentation of the SCG, which promotes neuronal sprouting in the ganglion, increases Tal mRNA in ganglia on the ipsilateral and contralateral sides. Together, these data demonstrate that Tal cu-tubulin mRNA elevates as a function of neuronal sprouting, and that Tal mRNA expression in intact neurons can be regulated by extrinsic cues, including NGF and changes in conneetivity.

0 1990 Academic

Press, Ine

at least six different a-tubulin genes (Villasante et al, 1986), and five different /3-tubulin genes (Wang et a& 1986) are expressed in neural and nonneural tissues at various times during development. Of these, two distinct a-tubulin mRNAs termed Tal and T26, are known to be expressed in the developing and mature rat brain (Lemischka et ab, 1981; Ginzburg et a& 1981; Miller et uZ., 1987). Tal cY-tubulin mRNA is abundantly expressed during periods of process outgrowth in developing and regenerating neurons, whereas expression of T26 mRNA is constitutive, and unchanged as a function of cell growth (Miller et uL, 1987, 1989b). These studies, in conjunction with similar studies examining the expression of GAP-43 mRNA (Neve et a& 1987; Hoffman, 1989) and class II 8-tubulin mRNA (Hoffman and Cleveland, 1988) have led to the hypothesis that axonal regeneration involves reexpression of the same genes necessary for neuronal growth during development. The experiments reported in this paper extend these findings by demonstrating that increased expression of Tal cY-tubulin mRNA is associated with the sprouting of intact sympathetic neurons as induced by several different environmental cues. Expression of this growthassociated mRNA in intact neurons is increased by systemic nerve growth factor, and by changes in connectivity. These studies provide strong evidence for modulation of gene expression during neuronal sprouting, and suggest that Tal mRNA may be a useful

INTRODUCTION

The growth and morphological differentiation of mammalian neurons involves a complex interplay between extrinsic influences and genetic mechanisms intrinsic to the neuron. Developing neurons express genes that are involved, sequentially, in commitment, migration, process outgrowth, and synaptogenesis. In the mature nervous system, with a few notable exceptions such as the olfactory system (Graziadei et aZ.,1980), neurons neither develop de nova nor migrate to new positions. However, new process outgrowth in the form of axonal regeneration or sprouting occurs in response to neural trauma or pathology (for reviews, see Brown, 1984; Seil, 1988) and may be ongoing at a low level even in the normal animal (Purves et al, 1986a, 198’7;Lichtman et ak, 1987;Harris and Purves, 1989). The molecular mechanisms underlying this structural plasticity, and the extraneuronal cues that regulate it remain largely undefined. Microtubules, which are assembled from (Y-and p-tubulins, are integral components of growing neurites (Daniels, 1972) and, thus, essential building blocks during neuronal regeneration and sprouting. In mammals, 1To whom correspondence should be addressed at Department of Anatomy and Cell Biology, 5-14 Medical Sciences Building, University of Alberta, Edmonton, Canada, T6G 2H7. 0012-1606/90 $3.00 Copyright All rights

Q 1990 by Academic Press. Inc. of reproduction in any form reserved.

84

MATHEW AND MIUER

T&-Tub&in

FIG. 1. The efferent fibers arising from the postganglionic neurons of the superior cervical ganglion course primarily in two major nerve branches: the internal and external carotid nerves. The preganglionic input from the sympathetic preganglionic neurons in the spinal cord arrives via the cervical sympathetic trunk. The postganglionic sympathetic axons of the internal carotid nerve bilaterally innervate a number of shared target organs including the cerebral vasculature and the pineal gland. Fibers arising from the two SCG mingle at some of these target organs (Tamamaki and Nojyo, 198’7;Lingappa and Zigmond, 1987). In contrast, some of the principal neurons of the SCG unilaterally innervate the ipsilateral iris via the internal carotid nerve. Three different types of axotomy were performed. A “short cut” in which the internal carotid nerve was unilaterally transected 2-3 mm from the SCG. A “long cut” involved unilateral transection of those principal neurons that project to the iris and other structures of the eye by ipsilateral enucleation. “Decentralization” involved unilateral transection of the cervical sympathetic trunk 4 mm from the SCG.

marker for defining the cellular cues responsible for mediating morphological plasticity of mature mammalian neurons.

mRNA duriw Neuronal Sprouting

85

lowing axotomy, animals were sacrificed under deep anesthesia. Superior cervical ganglia were removed and processed for Northern blot analysis. Two animals were analyzed for each timepoint in six separate experiments (a total of 12 animals per timepoint). For the decentralization experiments, the cervical sympathetic trunk was unilaterally transected 3-4 mm caudal to the SCG. Ten animals were examined (two animals in each of five separate experiments). For immunohistochemical studies, animals were anesthetized with sodium pentobarbital, and the internal carotid nerve was unilaterally transected. Two, 3, or 7 days following axotomy, the animals were again anesthetized and transcardially perfused with 4% paraformaldehyde. The pineal gland was removed and processed for tyrosine hydroxylase immunocytochemistry. Two animals were studied for each timepoint. (c) NGF-treatment studies. Neonates were injected daily from Postnatal Days 2 to 11 with 2.5s NGF (generously provided by Dr. Richard Murphy) dissolved in saline. Two animals were injected with 5 mg/kg and one with 10 mg/kg 2.5s NGF. Control littermates were injected daily with a similar volume of saline solution. Animals were subsequently sacrificed at P12 under deep anesthesia (35 mg/kg sodium pentobarbital), and the superior cervical ganglion was processed for RNA isolation. Alternatively, animals were anesthetized with sodium pentobarbital, transcardially perfused with 4% paraformaldehyde, and the superior cervical ganglion was removed and processed for in situ hybridization. RNA Isolation and Analysis

Total cytoplasmic RNA was prepared from ganglia by a modification of the phenol/chloroform/isoamyl alcohol technique (Schibler et aL, 1980). Total RNA (l-3 pg) MATERIALS AND METHODS was fractionated by electrophoresis on 1.2% agarose gels in the presence of 1 M formaldehyde (Rave et al, Animals and Surgical Procedures 1979) and transferred to nitrocellulose (Thomas, 1980). (a) Developmental studies. Timed pregnant Sprague- Anti-sense Tal a-tubulin RNA probes were prepared as Dawley rats provided a source of neonatal animals of previously described (Miller et al, 1987,1989b) with SP6 RNA polymerase (BRL) and 32P-CTP (New England precise chronological age. Neonates were sacrificed Nuclear, Boston, MA; 800 Wmmol) under conditions under deep anaesthesia (35 mg/kg sodium pentobarbital), and the superior cervical ganglia were removed and described by Melton et al. (1984). Anti-sense RNA probes were hybridized to the immunobilized RNA as processed for RNA isolation. (3) Regeneration/sprouting studies. Female Sprague- previously described for probes prepared by nick transDawley rats (150-200 g) were anesthetized with sodium lation (Lenoir et aL, 1986), except that hybridizations were performed at 65”C, and blots were washed to a pentobarbital (35 mg/kg), and the internal carotid nerve was unilaterally transected 2-3 mm from the supe- stringency of 0.05~ SSC at 65°C. Nitrocellulose filters rior cervical ganglion (short cut) (refer to Fig. 1 for a were subsequently exposed to XAR or XRP X-ray film (Kodak) for 2 hr-3 days. To confirm that equivalent schematic representation of these procedures). Alternatively, unilateral enucleation of the eye was performed amounts of RNA were loaded in each lane, ethidium to transect sympathetic axons that project to the eye bromide was added to the sample buffer prior to electronear their terminations (long cut). At various times fol- phoresis, and gels were photographed under ultraviolet

86

DEVELOPMENTALBIOLOGY

illumination. The nitrocellulose was also stained with methylene blue (Monroy, 1988) subsequent to hybridization.

VOLUME141, 1330

a

b

In Situ Hybridization Ganglia from perfused animals were cryoprotected by FIG. 2. Northern blot analysis of Tal a-tuhulin mRNA in the supeimmersion in graded sucrose solutions and sectioned rior cervical ganglion following a unilateral (a) short cut, and (h) long onto chrom-alum subbed slides. Anti-sense Tal RNA cut. (a) Equal amounts of total cytoplasmic RNA from the superior cervical ganglion of (1) control adult rats, (2) the contralateral and (3) probes were prepared as above with 35S-CTP (New En- ipsilateral ganglia at 1 day following a short cut, and (4) the contralatgland Nuclear, Boston, MA; 800 Ci/mmol) rather than era1 and (5) ipsilateral ganglia at 3 days following a short cut. (h) 32P-CTP. In situ hybridization for both Tal and T26 Equal amounts of total cytoplasmic RNA from (1) control SCG, and a-tubulin mRNAs was performed as previously de- (2) the contralateral and (3) ipsilateral SCG 3 days following a long cut. scribed (Miller et al, 1989a,b). After hybridization, slides were air-dried and apposed to Kodak XRP film for 12-24 hr to obtain X-ray images. The slides were subsequently dipped in Kodak NTB-2 emulsion, and exposed ing incubation with Texas-red steptavidin the sections for l-3 days prior to development. Controls were done to were washed three times with HBS and mounted with Mowiol (Osborne and Weber, 1982), sealed with nail polensure specificity of hybridization, including hybridization with a sense probe, or with a variety of other heter- ish, and stored at 4°C. Control sections were treated ologous probes for mRNAs that did not change with the identically, with the exception that no primary antibody experimental manipulations. To ascertain cellular local- was used. ization, hybridized tissue sections were counterstained with hematoxylin and eosin, and alternate tissue sec- Analysis and Quantification tions were stained with cresyl violet. Northern blot results were quantitated using an LKB Ultrascan XL scanning laser densitometer. RepresentaTyrosine Hydroxylase Immunocytochemtitry tive Northern blots from different experiments were Animals were anaesthetized with sodium pentobarbichosen for quantitation, after ensuring that the tal and perfused transcardially with a brief wash of PBS amounts of total RNA in the pertinent lanes were iden(pH 7.4) followed by 4% paraformaldehyde in phos- tical. Several different film exposures of the same data phate buffer at pH 7.4. Following perfusion the pineal were analyzed. Results are represented as an approxiglands were removed and postfixed in the same fixative mate value, or as a range of values, based on the asovernight. The tissues were subsequently cryoprotected sumption that this type of analysis is semiquantitative. by immersion in cold graded sucrose solutions (12, 16, RESULTS 18% ) and frozen, and cryostat sections 10 pm thick were cut and mounted on chrom-alum subbed slides. Sections Td cr-Tubulin mRNA Increases during Collateral were further treated with 4% paraformaldehyde in Sprouting of Mature, Uninjured, Sympathetic Neurons phosphate buffer for 5 min at room temperature. The Following unilateral transection of the major effersections were then permeabilized with Hepes-buffered saline (HBS) containing 0.1% Triton Xl00 for 5 min. ent branches of the SCG, contralateral, uninjured symFollowing nonspecific blocking with 3% fetal calf serum pathetic neurons reinnervate partially deafferented for 1 hr at room temperature, the sections were incu- shared target organs such as the pineal gland (Dornay et bated overnight at 4°C with a monoclonal antibody to aL, 1985; Lingappa and Zigmond, 1987). To determine tyrosine hydroxylase (Rohrer et aZ., 1986) (generously whether uninjured, sprouting neurons express inprovided by Dr. Ann Acheson) at a concentration of creased levels of Tal a-tubulin mRNA, we unilaterally 1:lOOO.The monoclonal antibody was raised in female transected the internal carotid nerve 2-3 mm from the SCG (short cut; refer to Fig. l), and, at various times BALB/C mice that had been immunized with purified tyrosine hydroxylase from a rat pheochromocytoma tu- later, isolated RNA from the ipsilateral and contralatera1 ganglia (Fig. 2a) Northern blot analysis revealed mor. Sections were then incubated with biotinylated that Tal cY-tubulin mRNA increased at least two-fold in anti-mouse IgG (Vector; 7.5 pg/ml) in HBS eontaining 5% each of goat serum and horse serum. Between incu- the operated, regenerating ganglia by 1 day and reached levels three- to four-fold higher than controls 3 days bations, sections were washed twice with HBS. Finally, postaxotomy, as determined by laser densitometry of the sections were treated with Texas-red streptavidin (Amersham; 4 hi/ml) in HBS containing 1.5% fetal calf several representative experiments. This axotomy reserum for 1 hr at room temperature in the dark. Follow- sponse was similar to that reported previously in regen-

MATHEW AND MILLER

Td-Tub&n

mRNA during Neuronal Spwting

FIG. 3. Immunocytochemical analysis of tyrosine hydroxylase-like immunoreactivity in the pineal gland of a control animal (a), and at 3 days (b) and ‘7days (c) following a unilateral short cut transection. Tyrosine hydroxylase-like immunoreactivity is significantly decreased at 3 days postaxotomy, and subsequently increases at 7 days. Calibration bar, 10 pm.

erating motor neurons (Miller et ah, 198913).Tal cr-tubulin mRNA expression in the contralateral, sprouting SCG was not affected 1 day following the short cut, but did increase approximately threefold 3 days postaxotomy (Fig. 2a). In situ hybridization confirmed that the increase in the contralateral ganglion occurred in neurons (data not shown). To confirm that the increase in Tal a-tubulin mRNA in the uninjured contralateral neurons was associated with neuronal sprouting, we monitored immunoreactivity for tyrosine hydroxylase in the pineal gland, a target organ bilaterally innervated by fibers in the internal carotid nerves (Bowers et a& 1984). Three days following unilateral resection of the nerve, when Tal-tubulin mRNA was elevated, the intensity of tyrosine-hydroxylase immunoreactivity in the gland was diminished (Fig. 3). By 7 days, tyrosine hydroxylase immunostaining was increased relative to the 3-day timepoint. Thus, the increase in Tc~l mRNA coincides temporally with the collateral sprouting of uninjured neurons. To determine whether partially denervated shared target organs could play a role in the contralateral increase in Tal a-tubulin mRNA, sympathetic axons that project to a unilaterally innervated target, the eye, were transected by enucleation (long cut; see Fig. 1). Northern blot analysis demonstrated that Tal cy-tubulin mRNA increased in the axotomized ipsilateral ganglia to a peak of approximately twofold at 3 days post-transection, but did not increase in the uninjured, contralatera1 ganglion (Fig. 2b). The observed contralateral in-

crease in TcJ cu-tubulin mRNA following transection of fibers that project to bilaterally innervated, but not unilaterally innervated targets, is consistent with a role for a target-derived factor(s) in increasing Tal cy-tubulin mRNA in sprouting neurons. Systemic NGF Increases Tal a-T&n&n mRNA in Intact Sympathetic Neurcms One target-derived factor which could be involved in collateral sprouting of sympathetic neurons is nerve growth factor (NGF) (Edwards et a& 1989; Campenot, 1982; Levi-Montalcini and Angeletti, 1968). To directly assess the effects of NGF on the expression of Tc~l a-tubulin mRNA in intact sympathetic neurons, we injected neonatal animals daily with 2.5s NGF from Postnatal Days 2 to 11, a procedure that leads to sympathetic terminal sprouting and dendritic growth (Levi-Montalcini and Angeletti, 1968;Snider, 1988). Over this developmental interval, in control animals, Tar1 cu-tubulin mRNA decreases, as demonstrated by Northern blot analysis (Fig. 4a). Tal a-tubulin mRNA is abundant at Postnatal Day 5 (PS), decreases at least fivefold between P5 and PlO, and decreases another 5- to lo-fold from PlO to P30 (Fig. 4a). Treatment of developing animals with systemic NGF maintained Tal ar-tubulin mRNA at levels that were similar to those in P5 ganglia, and that were at least 5-fold higher than P12 control SCG (Figs. 4b, 4c and 4d). In contrast, levels of T26 a-tubulin mRNA, which is not regulated as a function of neuronal growth

88

DEVELOPMENTALBIOLOGY

a

b

C

d

FIG. 4. Northern blot analysis of Tcvl cr-tubulin mRNA in the superior cervical ganglion (a) during postnatal development, (b), (c), and (d) following systemic NGF treatment for 10 days. (a) Total cytoplasmic RNA from the superior cervical ganglia of developing rats at (1) Postnatal Day 5, (2) Postnatal Day 10, (3) Postnatal Day 30, and (4) Postnatal Day 45. (b) and (c) Total cytoplasmic RNA from the SCG of two different Postnatal Day 12 rats that were treated with NGF from P2 to Pll (lane 2 in each of b and c) versus their age-matched control littermates (lane 1 in each of b and c). (d) Equal amounts of total cytoplasmic RNA from the SCG of (1) a control P12 animal, (2) a NGF-treated P12 animal, and (3) a control P5 animal.

(Miller et al, 1987; 1989b), were similar in control and NGF-treated animals, as determined by in situ hybridization (data not shown). The specific increase in Tal mRNA can be only partially explained by NGF-mediated rescue of sympathetic neurons since NGF-treatment permits only 30% more SCG neurons to survive (Hendry and Campbell, 1976; Hendry, 1977). Although Tal a-tubulin mRNA is expressed predominantly in neurons of the developing nervous system, it is possible that systemic NGF induces Tal cY-tubulin mRNA in nonneuronal cells of the ganglia. To determine the cellular localization of the increased Tal a-tubulin mRNA levels, the SCG of control and NGF-treated animals were analyzed by in situ hybridization. The SCG were enlarged in all of the NGF-treated animals, as previously reported (Levi-Montalcini and Booker, 1960; Thoenen et aZ.,1971). However, the pattern of hybridization of probes for Tal cY-tubulin mRNA was similar in control and NGF-treated animals (Fig. 5), with silver grains being predominantly localized over neurons (Fig. 5b). There was no significant hybridization of the probe either to the epineuraVperineuria1 connective tissue or to nonneuronal cells within the ganglia. Thus, Tc~l a-tubulin mRNA increases were specific to intact, sympathetic neurons that were induced to sprout with systemic NGF. Regulation of Toll a-Tub&in in Connectivity

rnRNA following

Changes

To determine whether Tal cY-tubulin mRNA is induced in uninjured neurons by stimuli other than trophic factors, the cervical sympathetic trunk (source of the afferent input for the SCG) was unilaterally transected 4 mm from the ganglion (decentralization, refer to Fig. l), and RNA was isolated from the ipsilateral and contralateral ganglia 3 days later. Northern blot analysis (Fig. 6a) demonstrated that a unilateral lesion of this

VOLUME141,1%0

type, which leads to intrinsic ganglionic sprouting (Ramsay and Matthews, 1985; Purves, 1976), increases Tal a-tubulin mRNA two- to threefold in both ipsilatera1 and contralateral ganglia. The ipsilateral increase could be due either to factors (like NGF) produced locally at the transection site, or to physical deafferentation of the sympathetic neurons. Although 1% of the principal neurons of the SCG project into the cervical sympathetic trunk (Bowers and Zigmond, 1979) (see Fig. l), axotomy of these neurons alone is unlikely to account for the ipsilateral response. In contrast, the contralatera1 increase in Tal mRNA must either reflect transynaptic regulation via the sympathetic preganglionic neurons in the spinal cord, or the partial functional denervation of shared target organs. Increases in Tal a-tubulin mRNA in sprouting neurons following a short cut or decentralization could also be mediated by less-localized, systemic “injury factors.” To address this possibility we performed a unilateral short cut, and isolated RNA from the sympathetic stellate ganglion. Northern blot analysis demonstrated no detectable increase in either the ipsilateral or contralatera1 stellate ganglia following axotomy of the SCG (Fig. 6b). Thus, it is unlikely that increased Tal mRNA in the SCG following a short cut or deafferentation is mediated systemically. DISCUSSION

These results demonstrate that increased expression of the major embryonic cu-tubulin mRNA, Tal, is associated with collateral sprouting of sympathetic neurons of the KG, as induced by partial denervation of bilaterally innervated target organs. Systemic NGF and deafferentation, which also lead to neuronal sprouting (Ramsay and Matthews, 1985; Purves, 1976), produce a similar increase in Tc~l mRNA. These results indicate that sprouting involves modulation of gene expression and suggest that genes associated with structural growth are regulated by a number of extrinsic cues, including target-derived or locally produced trophic factors and/or changes in connectivity. Unilateral transection of the internal and external carotid nerves leads to a regenerative response in the ipsilateral SCG, and indirectly affects the physiology of the contralateral SCG. Many target organs are bilaterally innervated by the SCG, and, in the cerebral vasculature and the pineal gland, the fibers of the contralateral neurons intermingle (Tamamaki and Nojyo, 1987; Lingappa and Zigmond, 1987). Several investigators (Dornay et aL, 1985; Lingappa and Zigmond, 1987) have demonstrated that the contralateral, uninjured neurons innervating the pineal gland sprout following unilateral ganglionectomy or transection of the internal and external carotid nerves. The timecourse of collateral

MATHEW AND MILLER

Tal-Tub&n

mRNA during Neuronal Sprouting

89

CON

FIG. 5. Expression of Tal a-tubulin mRNA in sympathetic neurons of Postnatal Day 12 animals that were injected daily with NGF for 10 days. Sections of the SCG of control (CON) and NGF-treated (NGF) animals were hybridized with a probe specific for Tal mRNA, coated with emulsion for autoradiography, developed, counterstained with hematoxylin and eosin, and visualized under darkfield (a, b) or brightfield (c, d) illumination. Note the clustering of grains over the large, pale-staining neurons in (c) and (d), as denoted by arrowheads, and the relative lack of signal over the smaller, nonneuronal cells. Calibration bar for (a, b) = 10 pm, and for (c, d) = 5 pm.

as confirmed in this paper, is temporally correlated with increased TCX~cu-tubulin mRNA. The increase in Tal a-tubulin mRNA in sprouting sympathetic neurons as well as in developing and regenerating neurons (Miller et aL, 1987,1989b) supports the hypothesis that induction of Tcvl mRNA is a reflection of a generalized neuronal growth response, and that the sprouting,

growth response, once initiated, may be similar in developing and mature neurons. It is likely that the common subset of neuronal growth-associated mRNAs plays a role in the morphological differentiation of mammalian neurons. Growth-associated mRNAs like Tal could therefore provide reliable genetic markers for monitoring neuronal growth and plasticity. Interestingly, Tal

90

DEVELOPMENTALBIOLOGYV0~~~~141.1990

a

b

total RNA) is increased approximately five-fold by NGF treatment (Hendry and Campbell, 1976). The increase in Tal cY-tubulin mRNA observed upon systemic NGF-treatment could be mediated either directly or indirectly as a consequence of increased strucFIG. 6. Expression of TLII o-tubulin mRNA in sympathetic neurons tural growth. However, in PC12 cells, Tal mRNA is up(a) following decentralization and (b) in the stellate ganglia following unilateral short cut axotomy. Decentralization alone induces Toll (Y- regulated within 6 hr of added NGF (Miller et aL, 1987b) an effect that significantly precedes neurite extension tubulin mRNA, as demonstrated in (a). Equal amounts of total cytoplasmic RNA from (3) control SCG, and (1) the contralateral and (2) (Greene and Tischler, 1976). Thus, Tal a-tubulin gene ipsilateral SCG 3 days following transection of the CST. Tel or-tubulin expression is either directly regulated by NGF itself, or mRNA does not increase in all sympathetic ganglia following a short is closely tied to early events that precede process outcut axotomy, as shown in (b). Total cytoplasmic RNA was isolated from the stellate ganglia 3 days following a unilateral short cut of the growth. The hypothesis that an increased available targetinternal carotid nerve. Equal amounts of RNA from (1) control stellate ganglia, and (2) the contralateral and (3) ipsilateral stellate gan- derived factor(s) is responsible for increased Tal glia of a short cut animal. mRNA during collateral sprouting is based upon the assumption that genes like Tcrl respond to these factors in a dose-dependent fashion. This would provide a mechamRNA expression is elevated in some populations of nism whereby the supply of proteins required for growth and maintenance of the terminal and dendritic adult mammalian CNS neurons (F. Miller, unpublished observations), which may reflect, at a genetic level, arbors of sympathetic neurons could be tied to the size of the innervated target field (Purves and Lichtman, structural remodeling of mature neurons. One likely candidate for a target-derived factor capa- 1985; Purves et al, 1986b). Direct support for such an ble of up-regulating Tcvl Lu-tubulin mRNA during collat- assumption derives from recent experiments demonstrating that NGF modulates Tal mRNA levels in culeral sprouting of sympathetic neurons is nerve growth tured sympathetic neurons in a dose-dependent manner factor (NGF). Diamond et al. (1987) have previouslydemonstrated that sensory neurons (which, like sympa- (Miller et c& 1989c). Experiments described here also suggest that targetthetic neurons, are responsive to NGF (Levi-Montalcini derived trophic factors may represent only one group of and Angeletti, 1963)) sprout in response to denervation of adjacent sensory fields in the skin, an effect that is extrinsic signals capable of up-regulating Tal a-tubulin inhibited by systemically administered antibodies to mRNA in uninjured neurons. Unilateral decentralizaNGF. Furthermore, increased local concentrations of tion of the SCG (which leads to intrinsic ganglionic NGF at the termini of developing sympathetic neurons sprouting (Ramsay and Matthews, 1985; Purves, 1976)) promote increased neuronal growth both in vivo (Ed- increases Tal mRNA in both the ipsilateral and contralateral ganglia. This increase is probably not due to syswards et al, 1989) and in vitro (Campenot, 1982). temic “injury factors,” since, in other experiments, levIn support of this hypothesis, data reported here demels of Tal mRNA did not change in the stellate ganglion onstrated that systemic NGF substantially increases following short cut axotomy of the SCG. The most plauTal a-tubulin mRNA levels in developing sympathetic neurons. Administration of NGF to neonates causes sible explanations for the observed increases are local both increased terminal sprouting (Levi-Montalcini and production of a trophic factor(s) at the site of the lesion, transynaptic regulation via the sympathetic pregangliAngeletti, 1968) and increased dendritic arborization (Snider, 1988) of sympathetic neurons. Exogenous NGF onic neurons in the spinal cord, and/or partial functional denervation of target organs. It is difficult to difalso leads to increased proliferation of nonneuronal cells in the SCG (Hendry and Campbell, 1976), and res- ferentiate between these possibilities in viva, and furcues the 30% of SCG neurons that are destined for cell ther clarification will likely arise from experiments death (Hendry and Campbell, 1976; Hendry, 19’77). performed on cultured neurons in vitro. On the basis of our observations, we cannot identify NGF-treatment in doses that cause extensive morphological growth maintained T&l mRNA in P12 sympa- the mechanisms underlying induction of Tcvl mRNA thetic neurons at levels similar to those in P5 and at during regeneration of mature sympathetic neurons. It least five-fold higher than in control P12 neurons. The is possible that NGF is responsible for increasing Tal actual increase in Tal cu-tubulin mRNA on a per neuron cu-tubulin mRNA expression during both regeneration basis may actually be greater; the Northern blots used and sprouting since NGF mRNA is produced locally at the site of peripheral nerve lesions (Heumann et al. for quantitation are normalized by analyzing equivalent amounts of ganglia RNA at each timepoint, and the ra- 1987). However, Purves (1975) has suggested that the tio of nonneuronal cells to neurons (and thus, the per- cell body reaction in axotomized sympathetic neurons is centage of nonneuronal cell mRNA in a given amount of due to the loss of target-derived trophic factors, and Nja

MATHEW AND MILLER

Td-Tubulin

mRNA during Neurwnal Sprouting

91

and Purves (1978) have demonstrated that many axotomy-induced changes could be prevented by exogenous NGF. Furthermore, collateral sprouting, but not regeneration, of sensory neurons appears to rely on an NGFdependent mechanism (Diamond et aL, 1987). It is therefore equally feasible that induction of Tal cy-tubulin mRNA during regeneration is the consequence of a different, potentially intrinsic mechanism. In summary, Tal a-tubulin mRNA increases during sprouting of sympathetic neurons, as induced by a number of different extrinsic cues. The fact that the Tc~l cY-tubulin gene is recruited during development, regeneration, and sprouting provides strong support for the hypothesis that there is a subset of neuronal growth-associated mRNAs that are utilized throughout a neuron’s lifetime. However, the intrinsic and extrinsic mechanisms responsible for inducing these genes are likely to be diverse, and may differ for sprouting versus regenerating neurons. Elucidation of these mechanisms at the molecular genetic level will provide fundamental information regarding the control of nerve growth, and may provide insights into some of the factors involved in regulating the ongoing structural plasticity of mature neurons.

GINZBURG, J., BEHAR, L., GIVOL, D., and LITPAUER, U. Z. (1981). The

We thank Lenna Mah, Yanling Ma, Linda Weiner, and Vera Chlumecky for technical advice and assistance, and Ann Acheson, Richard Murphy, Robert Campenot, Richard Smith, Bruce Stevenson, and Warren Gallin for frequent discussions and advice. This work was supported by grants from the Canadian M.R.C. and the Alberta Heritage Foundation for Medical Research to F.M. F.M. is an Alberta Heritage Foundation Scholar and T.M. is supported by a studentship from the Alberta Association for Paraplegic Research.

LEMISCHKA, I. R., FARMER, S., RACANIELLO,V. R., and SHARP, P. A. (1981). Nucleotide sequence and evolution of a mammalian a-tubulin mRNA. J. Mol. Biol. 150,101-120. LENOIR,D., BATTENBERG,E., KIEL, M., BLOOM,F. E., and MILNER, R. J. (1986). The brain-specific gene lB236 is expressed postnatally in the developing rat brain. J. Neurosti 6,522-530. LEVI-M• NTALCINI, R., and ANGELE~I, P. U. (1963). Essential role of the nerve growth factor in the survival and maintenance of dissociated sensory and sympathetic embryonic nerve cells in vitro. Dev.

REFERENCES BOWERS,C. W., and ZIGMOND,R. E. (1979). Localization of neurons in the rat superior cervical ganglion that project into different postganglionic trunks. J. Camp. Neural 185,381-392. BOWERS,C. W., DAHM, L. M., and ZIGMOND,R. E. (1984). The number and distribution of sympathetic neurones that innervate the rat pineal gland. Neuroscience 13,87-96. BROWN,M. C. (1984). Sprouting of motor nerves in adult muscles: a recapitulation of ontogeny. TINS l, lO-14. CAMPENOT,R. (1982). Development of sympathetic neurons in compartmentalized cultures. 1. Local control of neurite growth by nerve growth factor. Dev. Biol 93,1-12. DANIELS, M. P. (1972). Colchicine inhibition of nerve fiber formation in vitro. J. Cell Biol. 53,&i-176. DIAMOND, J., COUGHLIN, M., MACINTYRE, L., HOLMES, M., and VISHEAU, B. (1987). Evidence that endogenous B nerve growth factor is responsible for the collateral sprouting, but not the regeneration, of nociceptive axons in adult rats. Proc. Natl. Acml. Sci USA 84, 6596-6600.

DORNAY, M., GJLAD, V. H., and GILAD, G. M. (1985). Compensatory changes in contralateral sympathetic neurons of the superior cervical ganglion and in their terminals in the pineal gland following unilateral ganglionectomy. J. Neurosti 5,1522-1526. EDWARDS,R. M., RUT~ER, W. J., and HANAHAN, D. (1989). Directed expression of NGF to pancreatic fl cells in transgenic mice leads to selective hyperinnervation of the islets. Cell 58,161-170.

nucleotide sequence of rat cY-tubulin: 3’-end characteristics and evolutionary conservation. Nucleic Acids Res. 9,2691-2697. GRAZIADEI, P. P. C., KARLAN, M. S., MONTI-GRAZIADEI, G. A., and BERNSTEIN,J. J. (1980). Neurogenesis of sensory neurons in the primate olfactory system after section of thej%u dfactoria. Brain Res. 186.289-300.

GREENE,L. A., and TXX!HL.ER,A. S. (1976). Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc Nat1 Acad Sci (Wash.) 73, 2424-2428. HARRIS, L. W., and PURVES,D. (1989). Rapid remodeling of sensory endings in the corneas of living mice. J. Neurosci. 9,2210-2214. HENDRY, I. A. (1977). Cell division in the developing sympathetic nervous system. J. Neuroeytol. 6,299-309. HENDRY, I. A., and CAMPBELL,J. (1976). Morphometric analysis of rat superior cervical ganglion after axotomy and nerve growth factor treatment. J. NeurocytoL 5,351-360. HEUMANN,R., KORSCHING,S., BANDTLOW,C., and THOENEN,H. (1987). Changes of nerve growth factor synthesis in nonneuronal cells in response to sciatic nerve transection. J. Cell Biol 104,1623-1631. HOFFMAN,P. N. (1989). Expression of GAP-43, a rapidly transported growth-associated protein and class II beta tubulin, a slowly transported cytoskeletal protein, are co-ordinated in regenerating neurons. J. Neurosci. 9(3), 893-897. HOFFMAN, P. N., and CLEVELAND,D. W. (1988). Neurofilament and tubulin expression recapitulates the developmental program during axonal regeneration: Induction of a specific beta tubulin isotype. Proc. Natl

Acad Sci USA 85,4530-4533.

Biol. 7,653-659.

LEVI-M• NTALCINI,R., and ANGELETPI,P. U. (1968). Nerve growth factor. Physiol. Rev. 48,534-569. LEVI-M• NTALCINI,R., and BOOKER,B. (1960). Excessive growth of the sympathetic ganglia evoked by a protein isolated from mouse salivary glands. Proc Natl Acad Sci. USA 46,373-3.X LICHTMAN, J. W., MAGRASSI,L., and PURVES,D. (1987). Visualization of neuromuscular junctions over periods of several months in living mice. J. Neurosci. 7,1215-1222. LINGAPPA, J. R., and ZIGMOND,R. E. (1987). A histochemical study of the adrenergic innervation of the rat pineal gland: Evidence for overlap of the innervation from the two superior cervical ganglia and for sprouting following unilateral decentralization. Neurcscience 21,893-902. MELTON,D. A., KRIEG, D. A., RABAPHLATI,M. R., MANIATIS, T., ZINN, K., and GREEN, M. R. (1984). Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res. 12,70357056.

MILLER, F. D., NAUS, C. C. G., DURAND,M., BLOOM,F. E., and MJLNER, R. J. (1987). Isotypes of a-tubulin are differentially regulated during neuronal maturation. J. Cell Bid 105,3065-3073. MILLER, F. D., OZIMEK, G., MILNER, R. J., and BLOOM,F. E. (1989a). Regulation of neuronal oxytocin mRNA by ovarian steroids in the mature and developing hypothalamus. Proc. Natl. Ad Sci. USA 86,2468-2472.

MILLER, F. D., TETZLAFF, W., BISBY, M. A., FAWCETT, J. W., and

92

DEVELOPMENTALBIOLOGY

MILNER, R. J. (1989b). Rapid induction of the major embryonic cu-tubulin mRNA, Tal, during nerve regeneration in adult rats. J. Neumsci. 9,1452-1463. MILLER, F., MATHEW, T. C., WILSON, H. C., and CAMPENOT,R. B. (1989c). Regulation of the growth-associated cY-tubulin gene, Tel, in sympathetic neurons in tivo and in vitro. Sot Neurosti Abstr. 844. MONROY,A. F. (1988). Staining immobilized RNA ladder. Focus 10,l. NJA, A., and PURVES,D. (1978). The effects of nerve growth factor and its antiserum on synapses in the superior cervical ganglion of the guinea-pig. J. Physiol 277.53-75. NEVE, R. L., PERROEE-BIZZOZERO, N. I., FINKELSTEIN,S., ZWIERS,H., KUENIT, D. M. and BENOWITZ,L. I. (1987). The neuronal growth associated protein GAP-43 (B-5OFl): Neuronal specificity, developmental regulation and regional distribution of the human and rat mRNAs. Mel Brain Res. 2,177-183. OSBOEE,M., and WEBER,K. (1982). Immunofluorescence and immunocytochemical procedures with affinity purified antibodies: Tubulincontaining structures. Methods Cell Bid 24,97-132. PURVES,D. (1975). Functional and structural changes in mammalian sympathetic neurones following interruption of their axons. J. Physid 252,429-463. PURVES,D. (1976). Competitive and non-competitive re-innervation of mammalian sympathetic neurones by native and foreign fibres. J. Physiol 261,453-475. PURVES, D., and LICHTMAN, J. W. (1985). Geometrical differences among homologous neurons in mammals. Science 228,298-302. PUEVES, D., HADLEY, R. D., and VOYVODIC,J. T. (1986a). Dynamic changes in the dendritic geometry of individual neurons visualized over periods of up to three months in the superior cervical ganglion of living mice. J. Neurosci. 6,1051-1060. PURVES,D., RUBIN, E., SNIDER,W. D., and LICHTMAN,J. (1986b). Relation of animal size to convergence, divergence, and neuronal number in peripheral sympathetic pathways. J. Neuro~ti 6,158-163. PURVES,D., VOYVODIC,J., MAGRASSI,L., and YAWO,H. (1987). Nerve terminal remodeling visualized in living mice by repeated examination of the same neuron. Science 238,1122-1126.

VOLUME141,199O

RAMSAY, D. A., and MA~EWS, M. R. (1985). Denervation-induced formation of adrenergic synapses in the superior cervical sympathetic ganglion of the rat and the enhancement of this effect by postganglionic axotomy. Neuroscience 16,997-1026. RAVE, N., CRKVENJAKOV,R., and BOEDTKER,H. (1979). Identification of procollagen mRNAs transferred to diazobenzyloxymethyl paper from formaldehyde agarose gels. Nucleic Acid-s Res. 6,3559-3567. ROHRER,H., ACHESON,A. L., THIBAULT, J., and THOEEEE, H. (1986). Developmental potential of quail dorsal root ganglion cells analyzed in vitro and in viva J. NIYUTOSC~. 6,2616-2624. SCHIBLER,K., TOSEI,M., PI~TET, A. C., FABIANI, L., and WELLAMES,P. (1980). Tissue-specific expression of mouse a-amylase genes. J. Mel Bid 142,93-116. SEIL, F. J. (1988). Axonal sprouting in response to injury. In “Neural Regeneration and Transplantation,” pp. 123-135. A. R. Liss, New York. SNIDER,W. D. (1988). Nerve growth factor enhances dendritic arborization of sympathetic ganglion cells in developing mammals. J. Neurosci. S(7), 2628-2634. TAMAMAKI, N., and NOJYO,Y. (1987). Intracranial trajectories of sympathetic nerve fibers originating in the superior cervical ganglion in the rat: WGA-HRP anterograde labeling study. Brain Res. 437,387392. THOENEN,H., ANGELETTI, P. V., LEVI-M• NTALANI, R., and K&XTLER, R. (1971). Selective induction by nerve growth factor of tyrosine hydroxylase and dopamine+hydroxylase in the rat superior cervical ganglia. PTOCNat1 Ad Sci USA 68,1598-1602. THOMAS, P. S. (1980). Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Nat1 Acad Sci USA 77,5201-5215. VILLAS-, A., WANG, D., DOBNER, P., DOLPH, P., LEWIS, S. A., and COWAN,N. J. (1986). Six mouse Lu-tubulin mRNAs encode five distinct isotypes: Testis-specific expression of two sister genes. Mel Cell Biol 6,2409-2419. WANG, D., VILLASANTE, A., LEWIS, S. A., and COWAN, N. J. (1986). The mammalian j%tubulin isotype. J. CeUBiol 103,1903-1910.