THE JOURNAL OF COMPARATIVE NEUROLOGY 32691-100 (1992)

Antibody to NGF Inhibits Collateral Sprouting of Septohippocampal Fibers Following Entorhinal Cortex Lesion in Adult Rats CATHARINA E.E.M. VAN DER ZEE, JIM FAWCETT, AND JACK DIAMOND Department of Biomedical Sciences, McMaster University, Hamilton, Ontario, Canada L8N 325

ABSTRACT We have used an antiserum raised against mouse 2.5s NGF to examine the involvement of endogenous neurotrophins in the collateral sprouting of septohippocampal fibers in the adult rat brain. The antiserum was administered intraventricularly. Immunocytochemical techniques indicated that the injected antibodies penetrated into brain tissue that included the basal forebrain, cortex, striatum, corpus callosum, and hippocampus. Unilateral lesioning of the entorhinal cortex was done to evoke the sprouting of the cholinergic septohippocampal fibers. At 8 days postlesion, the sprouting was much advanced, as evidenced by an increase in density of the acetylcholinesterase (AChE) staining in the outer molecular layer (OML) of the dentate gyrus and by the associated increase in the absolute number of AChE-positive fibers in the OML. As well, there was a widening of the inner molecular layer (IML), interpreted as being due to sprouting of noncholinergic axons in that region. In rats injected daily with anti-NGF or anti-NGF Fab fragments, no increase in AChE density, or in the population of AChE-positive fibers, was observed in the OML. In contrast, the widening of the IML seemed to be unaffected by the anti-NGF treatment. No changes were observed in the AChE related parameters in the dentate gyrus of nonlesioned animals treated similarly for 8 days with anti-NGF; there was, however, a decrease of choline acetyltransferase ( C U T ) immunostaining in the CUT-positive cells of the basal forebrain. Our findings and the confirmation that our polyclonal anti-NGF also recognizes other members of the NGF neurotrophin family, specifically brain-derived neurotrophic factor and neurotrophin-3, indicate that at least one of these neurotrophins plays a key role in the collateral sprouting of the cholinergic septohippocampal fibers (but not that presumed to occur within the IML) following an entorhinal cortex lesion. o 1992 Wiley-Liss, Inc. Key words: anti-NGF,CNS sprouting, dentate gyrus, antibody penetration, cholinergicprojections

Collateral sprouting is a growth response of undamaged axons, occurring almost invariably at or within their target tissue; in the mature organism such sprouting is usually observed following partial denervation of the tissue (Diamond et al., '76). In the peripheral nervous system (PNS), collateral sprouting of spared axons can often restore a functionally useful innervation to the deprived tissue, e.g., muscle (Brown, '81) or skin (Diamond et al., '91).Collateral sprouting has also been well characterised within the central nervous system (CNS),particularly in the hippocampus (e.g., Scheff et al., '80), but also in a variety of other brain regions (Tsukahara, '81; Cotman and Anderson, '88). Considerable interest attaches to such sprouting as a possible basis for the recovery of functions compromised by damage or disease of the brain or spinal cord. This is an especially attractive possibility given the present uncertain

o 1992 WILEY-LISS, INC.

status of axonal regeneration in the adult CNS (Aguayo et al., '91). Until recently, none of the putative growth factors had been unambiguously identified as responsible for the spontaneous occurrence of either axonal regeneration or collateral sprouting following lesions of the adult nervous system. Whereas the use of antibodies clearly established endogenous NGF as an essential survival factor for neuronal populations in the developing PNS, its possible role in axonal growth could not be distinguished (Angeletti and Bradshaw, '71; Levi-Montalcini, '87). In the adult animal, however, the antibody approach has allowed the direct Accepted August 3,1992. Address reprint requests to Dr. Jack Diamond, Department of Biomedical Sciences, McMaster University, 1200 Main Street West, Hamilton, Ontario, Canada L8N 325.

C.E.E.M. VAN DER ZEE ET AL.

92 demonstration that endogenous NGF (or a related molecule recognised by the same antibody) is indeed a growth factor in the nervous system, in that it is essential for the collateral sprouting of cutaneous nociceptive nerves (Diamond et al., '87, '92) and sympathetic axons (Springer and Loy, '85; Gloster and Diamond, '89). Within the CNS the search for endogenous growth factors has largely focused on the expression in various brain regions of mRNAs for the various members of the NGF family of neurotrophins (Korsching et al., '85; Whittemore et al., '86; Ernfors et al., '90a,b; Hofer et al., '90; Maisonpierre et al., '90) and their receptors (Richardson et al., '86; Taniuchi et al., '86; Springer et al., '87), on the identification of the receptors themselves (Kaplan et al., '91; Klein et al., '91a,b; Lamballe et al., '91; Radeke and Feinstein, '911, and to a much lesser extent on the identification of the neurotrophins (Alderson et al., '90; Ernfors et al., '90a; Hohn et al., '90). However, there are now numerous findings, based on the exogenous administration of NGF or antibodies to NGF, which implicate NGF (or a related molecule) in the maintenance of the normal character of the cholinergic neurons in the basal forebrain (Gnahn et al., '83; Hefti et al., '84; Hefti, '86; Fusco et al., '89, '90; Vantini et al., '89). Exogenous NGF has also been reported to have evoked axonal growth within the lesioned septohippocampal pathway (Hagg et al., '90, '91; Junard et al., '90). Given these NGF-related findings and the well-established observation that the cholinergic neurons of the basal forebrain sprout in the hippocampus following lesions of the entorhinal cortex (Cotman et al., '73; Cotman and Anderson, '881, it seemed appropriate to examine the effects of anti-NGF treatment. The results indicate that this cholinergic sprouting is indeed dependent on endogenous NGF or a related neurotrophin. In contrast, the signs of an associated sprouting of noncholinergic fibers in the hippocampus seem not to be affected by the anti-NGF treatment. Preliminary reports of these findings have been published (Van der Zee et al., '91a,b).

MATERIALS AND METHODS Animals All studies were carried out on female Wistar rats, weighing 200-250 g.

Cannulation of the lateral ventricle: Intraventricular injections Rats were anesthetised with sodium pentobarbital (65 mgikg), placed in a head holder (skull surface horizontal), and a hole was drilled at 0.8 mm caudal and 1.0 mm lateral to bregma for implantation of the cannula in the lateral ventricle. The cannula, made from polyethylene tubing (internal diameter 0.58 mm; external diameter 0.965 mm) according to Brakkee et al. ('791, was firmly attached to the skull with dental cement, anchored with three stainless steel screws in the skull. The location of the cannula tip in the lateral ventricle was always checked in the subsequent histological examination (see below). Intraventricular injections, given daily to nonrestrained, conscious animals starting on the day of operation, were 5 pl (concentration 20 pg/ pl) of anti-NGF (affinity-purified IgGs); anti-NGF Fab fragments; control serum IgGs; or control Fab fragments.

Entorhinal cortex lesion Unilateral electrolytic lesions of the entorhinal cortex were performed according to Scheff et al. ('80). Briefly, after a small craniotomy and exposure of the underlying cortex, nine stereotactically determined points were lesioned by passing a 1 mA current of 45-second duration through a tungsten electrode, which was tilted at 10" to the vertical from medial to lateral. The electrode was positioned, with respect to the midline and interaural line, first, 3.3 mm lateral and 0.7 mm anterior, second, at 4.3 mm lateral and 0.7 mm anterior, and third, at 4.3 mrn lateral and 1.4 mm anterior. At each position a lesion was made at 2, 4, and 6 mm from the brain surface. The set of nine lesions spanned the entire medial and lateral portion of the entorhinal cortex and included areas of the parasubiculum (Scheff et al., '80). The extent of the entorhinal cortex lesion was checked in the subsequent histological examination (see below).

Anti-NGF antibody, IgG purification, and Fab fragment preparation NGF (2.5s) was prepared from male mouse salivary glands according to the method of Mobley et al. ('76) and further purified by HPLC according to Petrides and Shooter ('86). To obtain antibodies, sheep were immunized with 0.5 mg of 2.5s NGF intradermally (5-10 sites) in complete Freund's adjuvant initially and in incomplete adjuvant every 4 weeks thereafter. Blood (150 mlianimal) was collected at 10 days after each booster injection. Serum was prepared by clotting the blood at room temperature followed by centrifugation at 1,500 g for 30 minutes, heat inactivation at 56°C for 30 minutes, and sterilization using 0.22 pm filters (Nalgene). To determine serum titers of anti-NGF preparations, serial dilutions of the antiserum in culture medium were combined with equal volumes of medium containing 20 ngiml 7 s NGF and incubated at 37°C for 1 hour. The biological activity was determined using cultures of dissociated superior cervical ganglion neurons (Coughlin and Collins, '85). IgG was purified from serum by differential precipitation by using caprilic acid followed by ammonium sulfate (McKinney and Parkinson, '87). NGF-specific antibody was further purified using affinity chromatography on 2 . 5 s NGF coupled to CN-Br sepharose 4B (Pharmacia). Fab fragments were prepared by papain (linked to CH-Sepharose, Pierce) digestion using a solid phase method as described in the manufacturer's protocol. Neutralizing activity of purified antibody and its fragments was assessed by bioassay as described for serum preparations.

Perfusion, AChE histochemistry Rats were anesthetised with sodium pentobarbital (65 mgikg) and perfused with 4% paraformaldehyde in 0.1 M PO4 (pH 7.3). The brains were removed, postfixed in the same solution overnight at 4"C, and 50-pm coronal sections were cut on a vibratome. AChE staining was performed the next day according to Tag0 et al. ('86) on free-floating sections on a shaker at room temperature. Briefly, the sections were incubated in 0.1% HzOzin 0.1 M PO4 (pH 7.3) for 25 minutes, washed three times in 0.1 M maleate buffer (pH 6.0), then incubated in 18 pM acetylthiocholine iodide, 5 pM K3Fe(CNI6,30 pM CuS04.5H20,and 50 pM sodium citrate.2HzO in the maleate buffer for 35 minutes. After five washes in 50 mM Tris (pH 7.61, the sections were

ANTI-NGF INHIBITS COLLATERAL SPROUTING IN CNS incubated for 5 minutes in 0.04% 3,3 diaminobenzidine (DAB) and 0.3% nickel ammonium sulfate in the Tris buffer. Finally, 0.003% HzOz was added and the sections were incubated for 20 minutes, then rinsed in 5 mM Tris buffer (pH 7.6).

Immunocytochemistry

93

Micro Computer Imaging Device, Brock University, St. Catharines, Ont., Canada). The width of the inner molecular layer (IML) was measured at five points in corresponding locations on the ipsilateral and contralateral sides. ChAT-positive cell bodies in the medial septum and vertical diagonal band of Broca of 12-15 sections of each brain were identified at a magnification of 100 x . The optical density of 30 randomly chosen cell bodies was measured in each of these sections. The number of AChE-positive fibers in the OML was counted microscopically in sections viewed at 400 x magnification, with a square reticle (180 pm/side) with 10 vertical and 10 horizontal lines in the ocular. The AChE positive fibers crossing the 10 vertical lines were counted at three different locations, randomly chosen, in the dorsal part of both the ipsilateral and contralateral OML in each section; 4-6 sections were used from each brain, and each treatment group contained five rats. The reticle spanned the entire width of the normal OML. Since this width is slightly reduced following entorhinal cortex lesion, the areas examined on the lesioned side included a narrow, relatively “empty” zone lying within the IML. This approach thus estimated an absohte value for the number of AChE positive fibers in the OML, rather than the density of these fibers. The differences in optical density values, and in the number of fibers, of the lesioned side were compared to those of the contralateral side and noted as “percentage increase”:

The buffers used for the immunocytochemical staining were phosphate-buffered saline (PBS, pH 7.2) and 0.1 M phosphate buffer (PO4,pH 7.3). For the antibody penetration study, rats were given a single 5-p1 intraventricular injection of either PBS or anti-NGF (20 pgipl), then anesthetised with sodium pentobarbital (65 mgikg), and perfused with 4% paraformaldehyde in 0.1 M PO4at 5,13.5,24 hours and 2 , 3 , 7 days (four rats at each time). The brains were removed, postfixed for 2-3 hours, and cryoprotected in a 15% buffered sucrose solution (pH 7.3) overnight. Coronal 40-pm cryostat sections were saved at every 400 pm and stored in PBS. After rinses in PBS (three washes, 10 minutes each), and incubation for 1 hour in a blocking solution containing 5% heat-inactivated normal horse serum and 0.3% Triton X-100 in 0.1 M PO4, the sections were incubated for 2-3 hours in a solution of biotinylated rabbit antisheep antibody (Sigma) diluted 1:200 in 1.5%heat-inactivated normal horse serum in 0.1 M PO4. After washes (10 minutes each) in PBS for 1.5 hours, the sections were incubated for 1hour in a solution of avidin-biotin HRP (Vector) diluted 1 : l O O in 0.1 M PO4, followed by a 30-minute wash in PBS. The final step was a 5-minute immersion into 0.05% DAB, 0.006% hydrogen peroxide in 0.1 M PO4, then rinsing in PBS. value on lesioned side - value on contralateral side x 100. Camera lucida drawings (magnification 4 0 ~ of ) each secvalue on contralateral side tion were made; the areas of lighter or darker precipitate were indicated respectively by fewer or more dots. Statistical analysis For ChAT immunocytochemistry, rats were perfused in 4% paraformaldehyde, 4% picric acid in 0.1 M PO4. The For the statistical analysis of the data of the optical brains were removed, postfixed for 3 hours in the same density of the OML, the number of AChE positive fibers in fixative, then immersed in 0.1 M PO4 (five changes), and left the OML, and the width of the IML, the Student’s t-test overnight. Coronal 50-pm sections were cut on avibratome. was used to compare the ipsilateral side with the contralatThe ChAT immunostaining was performed on free-floating era1 side. The same test was used to compare the optical sections on a shaker at room temperature. They were density of CUT-stained basal forebrain cell bodies in the incubated for 1 hour in a blocker solution containing 5% control serum treated, and the anti-NGF treated, groups of heat-inactivated normal horse serum and 0.3% Triton animals. Statistical significance at p < 0.05. X-100 in 0.1 M PO4,then incubated overnight in a solution of the primary antibody (anti-ChAT, 143-Chemicon) diRESULTS luted 1:600 in 1.5%heat-inactivated normal horse serum in Antibody penetration 0.1 M PO4. The following day the sections were washed repeatedly in PBS for 4 hours (every 15 minutes), then In rats injected with anti-NGF, the identifying brown incubated for 3 hours in a secondary antibody (biotinylated precipitate was clearly visible in the areas surrounding the goat-antirabbit, Sigma) solution diluted 1:200 in 1.5% lateral ventricles and in structures such as the basal heat-inactivated normal horse serum in 0.1 M PO4. The forebrain, hippocampus, cortex, striatum, and corpus callosections were washed for 1hour in PBS, then incubated for sum. Control rats, injected with PBS, showed only a pale 1 hour in a avidin-biotin HRP (Vector) solution diluted background staining. In the anti-NGF brains, the precipi100 in 0.1 M PO4, followed by a 1-hour washing in PBS. tate was evident at 5 , 13.5, and 24 hours (Fig. 11, but was nally, the sections were immersed for 5 minutes in 0.05% absent in brains examined at 2, 3, and 7 days (not shown). DAB, 0.006% hydrogen peroxide in 0.1 M PO4 and rinsed in The precipitate in the basal forebrain and hippocampal sections was most dense and most extensive at 5 hours (Fig. PBS. la,b), but by 13.5 hours, and even more so at 24 hours, the Histochemical, immunocytochemical,and precipitate was less concentrated and covered a smaller anatomical measurements area (Fig. lc-fl. The dentate gyrus was analyzed in six sections per rat. Collateral sprouting in the dentate gyrus Sections were taken from 2.3 to 3.8 mm posterior to following unilateral entorhinal cortex lesion Bregma along the septotemporal axis. The OML of the AChE density. The AChE staining density of the OML ipsilateral and contralateral dentate gyrus of the digitized image of each brain section was outlined and its optical of the dentate gyrus of the hippocampus was measured at 4, density measured with an image-analysis system (MCID: 6, 8, and 14 days following unilateral entorhinal cortex

C.E.E.M. VAN DER ZEE ET AL.

94

Basal Forebrain

Hippocampus

Fig. 1. Antibody penetration. Typical examples from a total of four animals per time point of camera lucida drawings of coronal brain sections of the basal forebrain (a,c,e) and hippocampal (b,d,f)area. The dots indicate the presence of anti-NGF at 5 hours (a,b), 13.5 hours (c,d), and 24 hours (e,f) following injection in the lateral ventricle. Scale bar = 1 mm.

ANTI-NGF INHIBITS COLLATERAL SPROUTING IN CNS lesion. The AChE density of the OML on the lesioned side was significantly increased when compared to the contralatera1 side at day 6 (18%),day 8 (21%),and day 14 (23%) (Fig. 2; p < 0.01). In contrast, the AChE staining density was not changed in the supragranular band (data not shown), another zone where septa1 fibers terminate, but one that is not de-afferented by entorhinal cortex (Cotman and Anderson, '88). Number of AChE positive fibers. The number of AChE positive fibers was counted in both the ipsilateral and contralateral OMLs of the dentate gyrus in lesioned rats treated with anti-NGF or control serum. By using a reticle that spanned the entire width of the (contralateral) OML, an absolute value for the AChE positive fibers was obtained, independent of OML shrinkage (see Methods). Values of AChE fiber counts utilising this technique were 83 & 3 in the OMLs of unlesioned rats, 85 5 5 in the contralateral OMLs of brains following entorhinal cortex lesion, and 108 f 5 in the OMLs ipsilateral to the lesion.

Effect of anti-NGF on collateral sprouting Four groups of rats were injected daily, beginning at the time of lesion, with 5 pl of one of four test solutions: (1) anti-NGF, (2) control serum, (3) anti-NGF Fab fragments, or (4) control Fab fragments. They were perfused after 8 days. Figure 3 shows photomicrographs of the dentate gyrus on the lesioned (Fig. 3a) and contralateral (Fig. 3b) side of an animal treated with control serum, and the lesioned (Fig. 3c) and contralateral (Fig. 3d) side following anti-NGF treatment. AChE density. The AChE density of the OML of the lesioned side was compared with that of the contralateral side and expressed as percentage increase (see Methods). Rats treated with control serum showed a significant increase in density of the OML on the lesioned side (Fig. 4a; p < 0.01). In contrast, animals injected with anti-NGF did not show this increase (Fig. 4b; not significant). With

30 35

1

25 -

*

T

*

95

control Fab fragments, like control serum, there was a significant increase in AChE density on the lesioned side (Fig. 4c; p < 0.021, whereas with anti-NGF Fab fragments administration, this increase was absent (Fig. 4d, n.s.). In nonlesioned, untreated rats, no differences were observed between the AChE densities of the OMLs on either side (Fig. 4e, n.s.1. A critical finding came from another group of nonlesioned rats that had been treated with anti-NGF for 8 days to evaluate the possibility that the AChE marker, used to identify sprouting, would be downregulated by the anti-NGF treatment. Not only were the densities of the two OMLs similar on the two sides in these animals (Fig. 4f, n.s.1, but their values were not significantly different from those in untreated animals. Number o f AChE positive fibers. Figure 5 shows photomicrographs of the AChE stained fibers in the OML at the lesioned side of a rat treated with control serum (Fig. 5a) or anti-NGF (Fig. 5b). Rats treated with control serum showed a significant increase (26.4 ? 3.5%; p < 0.01) in the number of AChE positive fibers on the lesioned side (Fig. 6a). In animals treated with anti-NGF, this increase was absent (Fig. 6b, ns.).

Continued inhibition of collateral sprouting by anti-NGF We examined whether the anti-NGF block of collateral sprouting of the basal forebrain cholinergic neurons persisted during a prolonged period of treatment, or whether this only delayed the onset of sprouting. Two groups of rats received an entorhinal cortex lesion; for the next 14 days one group received anti-NGF treatment, the other control serum. In this study we measured only AChE density of the OML, the lesioned side being compared to the contralateral side, as in the preceding study. The increase in density of the OML observed on the lesioned side in the animals that received the control serum, was even greater at 14 days than in those animals examined at 8 days (Fig. 7a,b; p < 0.01). However, the inhibition of sprouting in the anti-NGF treated animals was no different at 14 days than at 8 days (Fig. 7c,d; ns.).

Width of the inner molecular layer

Silver staining has revealed that following entorhinal cortex lesion, the commissuraliassociational (CIA) afferL ents, which derive from CA4 neurons and normally termi0 nate in the IML of the dentate gyrus, expand into the partially denervated OML (Lynch et al., '77). This sprouting of CIA fibers can also be evaluated by measuring the wideningof the clear zone of the IML in the sections stained for AChE. Eight days following the lesion, the width of the IML of the dentate gyrus of the lesioned side was significantly increased compared to the contralateral side (142%, when calculated according to the approach used by Lynch et I al., '77), with the contralateral side taken as 100%; p < 14 4 6 0 0.01; calculated as described in Methods, the increase was days 42%; p < 0.01; Table 1).Anti-NGF treatment did not affect Fig. 2. Time course of collateral sprouting of the cholinergic this lesion-induced widening of the IML (Table 1). The septohippocampal fibers. The acetylcholinesterase (AChE) density of width of the contralateral IML was constant in all the the outer molecular layer (OML) of the dentate gyrus on the lesioned animal groups studied and was not significantly different side, as compared to the contralateral side, is expressed as % increase; day 4 (n = 51, day 6 (n = 51, day 8 (n = 61, and day 14 (n = 11). from that of the IMLs in nonlesioned treated or untreated rats (Table 1,n.s.1. *significant compared to the contralateral side (p < 0.01).

C.E.E.M. VAN DER ZEE ET AL.

96

Fig. 3. Typical photomicrographs of the dentate gyrus showing the lesioned (a) and contralateral (b) side of a rat treated with control serum, and the lesioned ( c ) and contralateral (d)side of a rat treated with anti-NGF. The outer (OML) and inner (IML) molecular layer are indicated. Scale bar = 100 km.

Effect of anti-NGF on ChAT immunostaining ChAT immunostaining was performed on brain sections from six nonlesioned rats treated for 8 days with either anti-NGF (n = 3 ) or control serum (n = 3 ) . The cholinergic neurons in the medial septum and vertical diagonal band of Broca in rats treated with anti-NGF showed a decrease in ChAT immunostaining density, which ranged from 43-47% when compared to those receiving control serum. This decrease was highly significant (45 ? 2%; p < 0.001).

DISCUSSION Antibody penetration The immunocytochemical findings are interpreted by us as indicating that antibodies to NGF do penetrate well into the adult rat brain following their injection into the lateral ventricle. In addition to the apparent inhibition of collateral

sprouting in the OML, intraventricular injections of antibodies to NGF resulted in a 45% decrease of ChAT immunostaining in the cholinergic neurons of the basal forebrain. As well, the effects of anti-NGF Fab fragments, which had the same biological activity as the anti-NGF I g G by in vitro assay, though differing in molecular weight (50 kD instead of 150 kD), were comparable to those of whole antibody in the in vivo experiment, indicating that the size of the anti-NGF IgG molecule was not a significant impediment per se to its diffusion within the brain. It is interesting that the direct intrahippocampal injection of antiserum to NGF was found by Springer and Loy ('85) to reduce the sprouting of sympathetic fibers in the hippocampus after fimbria fornix lesion only at the injection site. Although earlier measurements of ChAT activity indicated poor penetration of anti-NGF antibodies in the neonatal rat basal forebrain following their intraventricular or intracortical injection (Gnahn et al., '83; Thoenen et al., '871, Vantini et al. ('89)

ANTI-NGF INHIBITS COLLATERAL SPROUTING IN CNS

97

20 Q)

In cll 0)

.-

15

u 10

5

a b c d control anti Fab Fab serum NGF control anti NGF

e f no anti treat NGF ment

Fig. 4. Effect of anti-NGF and anti-NGF Fab fragments on collateral sprouting at 8 days following entorhinal cortex lesion and of anti-NGF in nonlesioned controls. The AChE density of the OML at the ipsilateral (lesioned, a-d) side, as compared to the contralateral side, is expressed as % increase. (a) control serum (n = 5); (b) anti-NGF (n = 5 ) ;( c )control Fab fragments (n = 3); (d)anti-NGF Fab fragments (n = 3); 2 nonlesioned control groups: ( e ) no treatment (n = 5 ) ; (f) anti-NGF (n = 4). *significant compared to the contralateral side (a, p < O.Ol;c,p < 0.02).

found such injections to produce a decrease of ChAT activity and immunostaining in a variety of brain regions. Other evidence of antibody penetration into the adult rat brain came from the findings of Funabashi et al. ('88),who demonstrated that intraventricularly injected anti-NGF delayed the development of amygdaloid kindling. Given the variability of techniques and findings attached to these earlier observations, our present findings can be taken to confirm the effectiveness of intraventricular administration as a way of getting significant antibody penetration into adult brain tissue.

Fig. 5. Typical photomicrographs of AChE positive fibers in the OML of the dentate gyrus of the lesioned side: a rat treated with control serum (a), or with anti-NGF (b).Scale bar = 50 wm.

Collateral sprouting in the dentate gyrus The occurrence of collateral sprouting following entorhinal cortex lesion was indicated particularly by the progressive increase in the AChE staining density of the OML of the ipsilateral dentate gyrus, as described by Cotman et al. ('73), Lynch et al. ('77), Scheff et al. ('80), Gomez-Pinilla et al. ('87), and Cotman and Anderson ('88). Supportive evidence for the explanation that this increased AChE staining represented an increase in the septohippocampal innervation of the dentate gyrus has come from biochemical measurements by Nadler et al. ('731, which showed that both AChE and ChAT activity were elevated in the OML of the dentate gyrus after entorhinal cortex lesion. In the present experiments, the axon counts revealed a significant increase in the absolute numbers of AChE positive fibers; in the absence of any evidence for the occurrence of AChE negative cholinergic fibers, we regard this result as ruling out an explanation based upon an increased AChE density within an unchanged population of cholinergic terminals. The conclusion that the findings represent collateral sprouting of cholinergic septohippocam-

pal fibers within the OML is further supported by the absence of increased AChE staining in the supragranular zone, which is also a target for septa1 fibers but is not de-afferented by the entorhinal cortex lesion (Cotman and Anderson, '88). The IML, a palely stained zone when AChE staining is employed, was expanded (see Table 1). This expansion reflects the collateral sprouting of the CIA fibers (Lynch et al., '77; Gomez-Pinilla et al., '87).

Anti-NGF administration blocks collateral sprouting of the cholinergic, but not the noncholinergic, fibers Cholinergic axons. The most important finding was that daily injections of polyclonal antibodies to NGF blocked the collateral sprouting of the cholinergic fibers evoked by a lesion of the entorhinal cortex. That the lack of increase in the density of AChE staining in the ipsilateral dentate gyms was due to inhibition of sprouting and not to a down-regulation of the AChE marker was shown by the

C.E.E.M. VAN DER ZEE ET AL.

98

TABLE 1. Width of the Inner Molecular Layer (IML) of the Dentate Gyrus (Fm)

30 351 Entorhind lesion * 8 days control sernm * 8 days anti-NGF No lesion *no treatment *8 days anti-NGF

Lesioned side

Contralateral side

68 ? 4l 67 i- Z1

48 t 2 50 5 6

( n = 5) In = 5)

44 5 4 42 i 5

46 f 4 44 ? 3

In in

= 5) = 41

'Eight days following the entorhinal cortex lesion, the width of the IML of the dentate gyms of the lesioned side was significantly (p < 0.01) increased compared to the contralateral side. Anti-NGF treatment did not affect this lesion-induced widening of the IML. The width of the contralateral IML was not significantly different from the IMLs in nonlesioned treated or untreated rats.

a control serum

b anti NGF

Fig. 6. Number of AChE positive fibers in the ipsilateral OML 8 days after entorhinal cortex lesion. The increased number of fibers at the lesioned side, as compared to the contralateral side, is expressed as ?u increase: (a)rats treated with control serum (n = 5); (b) or with anti-NGF (n = 5). *significant compared to the contralateral side (p < 0.01).

40 35'1

*

T

a b 8d 14d control serum

c d 8d 14d anti-NGF

Fig. 7. Anti-NGF continues to block collateral sprouting at 14 days following entorhinal cortex lesion. The AChE density of the OML at the ipsilateral (lesioned) side, as compared to the contralateral side, is expressed as 5% zncreuse. Treatment with control serum (a)8 days (n = 5 ) ;(b)14 days (n = 6);or with anti-NGF (c) 8 days (n = 5); (d) 14 days (n = 3). *significant compared to the contralateral side (p < 0.01).

findings from nonlesioned control rats. In these animals anti-NGF injections produced no differences in density on either side, their values being no different from those of OMLs from nonlesioned rats that received no anti-NGF treatment. We conclude from these findings that the sprouting of the cholinergic fibers evoked within the OML by an entorhinal cortex lesion is attributable to endogenous NGF or a related

neurotrophin that is recognised by the polyclonal antibody to NGF. Acheson et al. ('91) reported that polyclonal antibodies raised against NGF recognize BDNF. Recent findings indicate that our polyclonal antibody recognises NT3 and, to a lesser extent, BDNF (R. Murphy, University of Alberta, Edmonton, Canada, personal communication). Since we have no comparable data on the ability of our anti-NGF to block the biological activity of NT-3 and BDNF, at present we cannot say which of the NGF family of neurotrophins is responsible for the observed sprouting. The same antibody to NGF that was used in the present study very effectively blocks the collateral sprouting of sensory axons (Diamond et al., '87, '92) and postganglionic sympathetic axons (Gloster and Diamond, '89) in adult rat skin. The treatment with anti-NGF does not just delay the onset of collateral sprouting but appears to completely block it, since the 14 day anti-NGF treatment inhibited the increase in AChE density in the OML to the same extent as shown after 8 days. Noncholinergic axons. Lynch et al. ('771, using the Holmes silver stain, showed that following entorhinal cortex lesion, there is an expansion to 130-140%, of the normal width of the commissuraliassociational (CIA) fiber plexus. Their interpretation that this represented CIA fiber sprouting was supported by the increase in the IML of receptors for kainic acid, a glutamate analogue (Cotman and Nadler, '811, and more recently by the concomitant increase in the IML of the growth-associated protein B-501 GAP43 (Benowitz et al., '90). In the present study, we assessed the sprouting of the CIA fibers indirectly by measuring the increase in width of the IML in AChEstained brain sections, and our results were comparable with those of Lynch et al. ('77). Treatment with anti-NGF did not alter this lesion-induced response, indicating that NGF does not play a role in the CIA fiber sprouting. This observation is not surprising since the IML, in which glutamatergic CIA fibers terminate, does not show NGF receptor immunoreactivity (Gomez-Pinilla, '87). A role for NGF in the collateral sprouting of sympathetic fibers into the hippocampal formation, which occurs following disruption of the septohippocampal pathway, has been proposed by Crutcher et al. ('79) and Springer and Loy ('85).Although an elevation of hippocampal NGF seems the most likely explanation of such sprouting (Springer and Loy, '85; Gloster and Diamond, '921, the intraventricular infusion of exogenous NGF failed to elicit sympathetic ingrowth into the hippocampal formation (Saffran et al., '89). It is also somewhat paradoxical that sympathetic fibers sprout in the hippocampus after a fimbria fornix lesion, but not after lesions to other hippocampal afferent systems, e.g., entorhinal, commissural, or locus coeruleus

ANTI-NGF INHIBITS COLLATERAL SPROUTING IN CNS (Crutcher et al., '79, '81). One explanation could be that whereas sympathetic fibers respond only to NGF, the cholinergic fibers normally utilise NGF but respond to both NGF and other neurotrophins (like BDNF or NT-3). After entorhinal cortex lesion, BDNF or NT-3 levels may become elevated, evoking sprouting of the cholinergic fibers but not of the sympathetic fibers (see, however, Lapchak and Hefti, '92). It is known that BDNF mRNA is present (although in low levels) in various brain areas, e.g., the entorhinal cortex (Ernfors et al., '90). However, both NGF mRNA and BDNF mRNA increase in the hippocampus after central lesions by kainic acid treatment (Gall et al., '911, ischemia (Bengzon et al., '911, and kindling (Ernfors et al., '91). The question therefore of exactly which neurotrophin of the NGF family is the most important for the cholinergic collateral sprouting in the hippocampus is not yet settled and is a focus of our further research.

CONCLUSIONS Our results show that intraventricularlyinjected antibodies to NGF penetrate into the brain tissue and implicate NGF, or perhaps BDNF or NT-3, as playing a critical role in the collateral sprouting of cholinergic septohippocampal fibers.

ACKNOWLEDGMENTS Funding for this research has been provided by the Network of Neural Regeneration and Functional Recovery, one of the 15 Networks of Centres of Excellence supported by the Government of Canada, and by the NIH (NIA) lROl AG 07732-05. C.E.E.M. V.d.Z. is a MRC PDF. We thank Jolanta Stanisz, Michael Holmes, and Karen Ann Moore for technical assistance, and Dr. Anne Foerster and Dr. Ron Racine for their valuable suggestions.

LITERATURE CITED Acheson, A., P.A. Barker, R.F. Alderson, F.D. Miller, and R.A. Murphy (1991) Detection of brain-derived neurotrophic factor-like activity in fibroblasts and Schwann cells: Inhibition by antibodies to NGF. Neuron 7265-275. Aguayo, A.J., M. Rasminsky, G.M. Bray, S. Carbonetto, L. McKerracher, M.P. Villegas-PQrez,M. Vidal-Sanz, and D.A. Carter (1991) Degenerative and regenerative responses of injured neurons in the central nervous Lond. (Biol.)331:337system of adult mammals. Philos. Trans. R. SOC. 343. Alderson, R.F., A.L. Alterman, Y.-A. Barde, and R.M. Lindsay (1990) Brain-derived neurotrophic factor increases survival and differentiated functions of rat septa1 cholinergic neurons in culture. Neuron 5,297306. Angeletti, R.H., and R.A. Bradshaw (1971) Nerve growth factor from mouse submaxillary gland: Amino acid sequence. Proc. Natl. Acad. Sci. USA 68,2417-2420. Bengzon, J.,Z. Kokaia, P . Ernfors, M.L. Smith, R. Ekman, B.K. Siesjo, H. Persson, and 0. Lindvall (1991) Ischemia, hypoglycemia and kindling increase brain levels of mRNAs for neurotrophic factors. SOC. Neurosci. Abstr. 17:183. Benowitz, L.I., W.R. Rodriguez, and R.L. Neve (1990) The pattern of GAP-43 immunostaining changes in the rat hippocampal formation during reactive synaptogenesis. Mol. Brain Res. 8:17-23. Brakkee, J.H., V.M. Wiegant, and W.H. Gispen (1979) A simple technique for rapid implantation of a permanent cannula into the rat brain ventricular system. Lab. Anirn. Sci. 29:78-81. Brown, M.C., R.L. Holland, and W.G. Hopkins (1981) Motor nerve sprouting. Annu. Rev. Neurosci. 4:17-42. Cotman, C.W., and K.J. Anderson (1988) Synaptic plasticity and functional stabilization in the hippocampal formation: Possible role in Alzheimer's

99

disease. In S.G. Waxman (ed): Advances in Neurology, Vol47: Functional Recovery. New York: Raven Press, pp. 313-335. Cotman, C.W., and J.V. Nadler (1981)Glutamate and aspartate as hippocampal transmitters: biochemical and pharmacological evidence. In P.J. Roberts, J. Storm-Mathisen, and G.A.R. Johnston (eds): Glutamate: Transmitter in the Central Nervous System. London: John Wiley & Sons, pp. 117-154. Cotman, C.W., D.A. Matthews, D. Taylor, and G. Lynch (1973) Synaptic rearrangement in the dentate g y m s : Histochemical evidence for adjustments after lesions in immature and adult rats. Proc. Natl. Acad. Sci. USA 70:3473-3477. Coughlin, M.D., and M.B. Collins (1985) Nerve growth factor independent development of embryonic mouse sympathetic neurons in dissociated cell culture. Dev. Biol. 110:392-401. Crutcher, K.A., L. Brothers, and J.N. Davis (1979) Sprouting of sympathetic nerves in the absence of afferent input. Exp. Neurol. 66:778-783. Crutcher, K.A., L. Brothers, and J.N. Davis (1981) Sympathetic noradrenergic sprouting in response to central cholinergic denervation: A histochemical study of neuronal sprouting in the rat hippocampal formation. Brain Res. 210:115-128. Diamond, J.,A. Gloster, and P. Kitchener (1992) Regulation of the sensory innervation of the skin: trophic control of collateral sprouting. In S. Scott (ed): Sensory Neurons: Diversity, Development and Plasticity. New York Oxford University Press pp. 309-332. Diamond, J.,M. Coughlin, and M. Holmes (1992) Endogenous NGF and nerve impulses regulate the collateral sprouting of sensory axons in the skin of the adult rat. J. Neurosci. 121454-1466. Diamond, J.,E. Cooper, C. Turner, and L. Macintyre (1976) Trophic regulation of nerve sprouting. Science 193:371-377. Diamond, J.,M. Coughlin, L. Macintyre, M. Holmes, and B. Visheau (1987) Evidence that endogenous p-nerve growth factor is responsible for the collateral sprouting but not the regeneration of nociceptive axons in adult rats. Proc. Natl. Acad. Sci. USA 84:659&6600. Ernfors, P., C. Wetmore, L. Olson, and H. Persson (1990b) Identification of cells in rat brain and peripheral tissues expressing mRNA for members of the nerve growth factor family. Neuron 5511-526. Ernfors, P., C.F. Ibanez, T. Ebendal, L. Olson, and H. Persson (1990a) Molecular cloning and neurotrophic activities of a protein with structural similarities to p-nerve growth factor: A developmental and topographical expression in the brain. Proc. Natl. Acad. Sci. USA 87:54545458. Ernfors, P., J. Bengzon, Z. Kokaia, H. Persson, and 0. Lindvall (1991) Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7: 165-176. Funabashi, T., H. Sasaki, and F. Kimura (1988) Intraventricular injection of antiserum to nerve growth factor delays the development of arnygdaloid kindling. Brain Res. 458,132-136. Fusco, M., M.A. Tria, N. Schiavo, A. Leon, and G. Vantini (1990) Nerve growth factor induces a marked and long-lasting enhancement of choline acetyltransferase activity in septohippocampal neurons of uninjured adult rats. Neurosci. Res. Commun. 7t97-103. Fusco, M., B. Oderfeld-Nowak, G. Vantini, N. Schiavo, M. Gradkowska, M. Zaremba, and A. Leon (1989) Nerve growth factor affects uninjured, adult rat hippocampal cholinergic neurons. Neuroscience 33:47-52. Gall, C., K. Murray, and P.J. Isackson (1991) Kainic acid-induced seizures stimulate increased expression of nerve growth factor mRNA in rat hippocampus. Mol. Brain Res. 9t113-123. Gloster, A., and J. Diamond (1989) Adult sympathetic axons regenerate independently of NGF. Soc. Neurosci. Abstr. 15:333. Gloster, A,, and J. Diamond (1992) Sympathetic nerves in adult rats regenerate normally and restore pilomotor function during an anti-NGF treatment that prevents their collateral sprouting. J. Comp. ,Neural., in press. Gnahn, H., F. Hefti, R. Heumann, M.E. Schwab, and H. Thoenen (1983) NGF-mediated increase of choline acetyltransferase (CMT) in the neonatal rat forebrain: Evidence for a physiological role of NGF in the brain? Dev. Brain Res. 9:45-52. Gomez-Pinilla, F., C.W. Cotman, and M. Nieto-Sampedro (1987) NGF receptor immunoreactivity in rat brain: Topographic distribution and response to entorhinal ablation. Neurosci. Lett. 82:260-266. Hagg, T., H.L. Vahlsing, M. Manthorpe, and S . Varon (1990) Nerve growth factor infusion into the denervated adult rat hippocampal formation promotes its cholinergic reinnervation. J. Neurosci. 10:3087-3092. Hagg, T., A.K. Gulati, M.A. Behzadian, H.L. Vahlsing, S. Varon, and M. Manthorpe (1991) Nerve growth factor promotes CNS cholinergic axonal

100 regeneration into acellular peripheral nerve grafts. Exp. Neurol. 112:7988. Hefti, F. (1986) Nerve growth factor promotes survival of septa1 cholinergic neurons after fimhrial transections. J. Neurosci. 6:2155-2162. Hefti, F., A. Dravid, and J. Hartikka (1984) Chronic intraventricular injections of nerve g o w t h factor elevate hippocampal choline acetyltransferase activity in adult rats with partial septo-hippocampal lesions. Brain Res. 293:305-309. Hofer, M., S.R. Pagliusi, A. Hohn, J. Leibrock, and Y.-A. Barde (1990) Regional distribution of brain-derived neurotrophic factor mRNA in the adult mouse brain. EMBO J. 9.2459-2464. Hohn, A., J. Leibrock, K. Bailey, and Y.-A. Barde (1990) Identification and characterization of a novel member of the nerve growth factor brainderived neurotrophic factor family. Nature 344:339-341. Junard, E.O., C.N. Montero, and F. Hefti (1990) Long-term administration of mouse nerve growth factor to adult rats with partial lesions of the cholinergic septohippocampal pathway. Exp. Neurol. 110:25-38. Kaplan, D.R., B.L. Hempstead, D. Martin-Zanca, M.V. Chao, and L.F. Parada (1991) The trk proto-oncogene product: A signal transducing receptor for nerve growth factor. Science 252554-557. Klein, R., S. Jing, V. Nanduri, E. O’Rourke, and M. Barbacid (1991a) The trk proto-oncogene encodes a receptor for nerve growth factor. Cell 65189197. Klein, R., V. Nanduri, S. Jing, F. Lamballe, P. Tapley, S. Bryant, C. Cordon-Cardo, K.R. Jones, L.F. Reichardt, and M. Barbacid (1991b) The trkB tyrosine protein kinase is a receptor for brain-derived neurotrophic factor and neurotrophin-3. Cell 66:395-403. Korsching, S., G. Auburger, R. Heumann, J. Scott, and H. Thoenen (1985) Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic innervation. EMBO J. 4:1389-1393. Lamballe., F.., R. Klein. and M. Barbacid (1991) trkC. a new member of the trk family of tyrosine protein kinases, is a receptor for neurotrophin-3. Cell 66:967-979. Lapchak, P.A., and F. Hefti (1992) BDNF and NGF treatment in lesioned rats: effects on cholinergic function and weight gain. NeuroReport 3:405-408. Levi-Montalcini, R. (1987) The nerve growth factor: Thirty-five years later. EMBO J. 6:1145-1154. Lynch, G.S., C. Gall, and C.W. Cotman (1977) Temporal parameters of axon “sprouting” in the brain of the adult rat. Exp. Neurol. 54:179-183. Maisonpierre, P.C., L. Belluscio, B. Friedman, R.F. Alderson, S.J. Wiegand, M.E. Furth, R.M. Lindsay, and G.D. Yancopoulos (1990) NT-3, BDNF and NGF in the developing rat nervous system: Parallel as well as reciprocal patterns of expression. Neuron 5501-509. McKinney, M.H., and A. Parkinson (1987) A simple non-chromatographic procedure to purify immunoglobulins from serum and ascites fluid. J. Immunol. Meth. 96271-278. Mobley, W.C., A. Schenker, and E.M. Shooter (1976) Characterization and isolation of proteolytically modified nerve growth factor. Biochemistry 15,5543-5551. Nadler, J.V., C.W. Cotman, and G.S. Lynch (1973) Alered distribution of choline acetyltransferase and acetylcholinesterase activities in the devel-

C.E.E.M. VAN DER ZEE ET AL. oping rat dentate gyrus following entorhinal lesion. Brain Res. 63:215230. Petrides, P.E., and E.M. Shooter (1986) Rapid isolation of the 7s-nerve growth factor complex and its subunits from murine submaxillary glands and saliva. J. Neurochem. 46721-725. Radeke, M.J., and S.C. Feinstein (1991) Analytical purification of the slow, high affinity NGF receptor: identification of a novel 135 kd polypeptide. Neuron 7:141-150. Richardson, P.M., V.M.K. Verge Issa, and R.J. Riopelle (1986) Distribution of neuronal receptors for nerve growth factor in the rat. J. Neurosci. 6.2312-2321. Saffran, B.N., J.E. Woo, W.C. Mobley, and K.A. Crutcher (1989) Intraventricular NGF infusion in the mature rat brain enhances sympathetic innervation of cerebrovascular targets but fails to elicit sympathetic ingrowth. Brain Res. 492245-254. Scheff, S.W., L.S. Benardo, and C.W. Cotman (1980) Decline in reactive fiber growth in the dentate gyrus of aged rats compared to young adult rats following entorhinal cortex removal. Brain Res. 19921-38. Springer, J.E., and R. Loy (1985) lntrahippocampal injections of antiserum to nerve growth factor inhibit sympathohippocampal sprouting. Brain Res. Bull. 15:629-634. Springer, J.E., S. Koh, M.W. Tayrien, and R. Loy (1987) Basal forebrain magnocellular neurons stain for nerve growth factor receptor: Correlation with cholinergic cell bodies and effects of axotomy. J. Neurosci. Res. 17:lll-118. Tago, H., H. Kimura, and T. Maeda (1986) Visualization of detailed acetylcholinesterase fiber and neuron staining in rat brain by a sensitive histochemical procedure. J. Histochem. and Cytochem. 34: 1431-1438. Taniuchi, M., J.B. Schweitzer, and E.J. Johnson (1986) Nerve growth factor receptor molecules in rat brain. Proc. Natl. Acad. Sci. USA 83:19501954. Thoenen, H., C. Bandtlow, and R. Heumann (1987) The physiological function of nerve growth factor in the central nervous system: Comparison with the periphery. Rev. Physiol. Biochem. Pharmacol. 109:145-178. Tsukahara, N. (1981) Synaptic plasticity in the mammalian central nervous system. Annu. Rev. Neurosci. 4:351-379. Van der Zee, C.E.E.M., R. Racine, J. Diamond (1991a) Is collateral sprouting of septo-hippcampal fibers in the dentate gyrus NGF-dependent? Third IBRO World Congress of Neuroscience, Montreal, Canada, Abstr. P4.9. Van der Zee, C.E.E.M., J. Fawcett, J. Stanisz, R. Racine, and J. Diamond (1991b) Anti-NGF treatment blocks the collateral sprouting of cholinergic fibers in the hippocampus. Soc. Neurosci. Abstr. 17:1312. Vantini, G., N. Schiavo, A. Di Martino, P. Polato, C. Triban, L. Callegaro, G. Toffano, and A. Leon (1989) Evidence for a physiological role of nerve growth factor in the central nervous system of neonatal rats. Neuron 3267-273. Whittemoro, S.R., T. Ebendal, L. Lirkfors, L. Olson, A. Seiger, I. Stromberg, and H. Persson (1986) Developmental and regional expression of p-nerve growth factor messenger RNA and protein in the rat central nervous system. Proc. Natl. Acad. Sci. USA 83:817-821.

Antibody to NGF inhibits collateral sprouting of septohippocampal fibers following entorhinal cortex lesion in adult rats.

We have used an antiserum raised against mouse 2.5S NGF to examine the involvement of endogenous neurotrophins in the collateral sprouting of septohip...
1MB Sizes 0 Downloads 0 Views