Nerve Growth Factor Regulates the Expression of Brain-Derived Neurotrophic Factor mRNA in the Peripheral Nervous System

MCN

Molecular and Cellular Neuroscience 7, 134– 142 (1996) Article No. 0010

Nerve Growth Factor Regulates the Expression of Brain-Derived Neurotrophic Factor mRNA in the Peripheral Nervous System Stuart C. Apfel,* Douglas E. Wright,† Andrea M. Wiideman,* Christine Dormia,* William D. Snider,† and John A. Kessler* *Departments of Neurology and Neuroscience, Albert Einstein College of Medicine, Bronx, New York 10461; and †Department of Neurology, Washington University, St. Louis, Missouri

Neurotrophins are profound regulators of neuronal survival in the developing peripheral nervous system and are synthesized by peripheral neurons themselves both during development and in maturity. Neuronal neurotrophin expression may be importantly related to survival of mature neurons, both in normal and pathological states. We show here that brain-derived neurotrophic factor (BDNF) gene expression in dorsal root ganglia is strongly stimulated in vivo by another neurotrophin, nerve growth factor (NGF). Furthermore, colocalization studies show that many BDNF-expressing sensory neurons also express trk A, the high-affinity NGF receptor. These results demonstrate a novel regulatory mechanism for neurotrophin gene expression and suggest a paracrine function for neurotrophins in mature animals.

INTRODUCTION The neurotrophin gene family of neurotrophic factors consists of at least five proteins which share significant sequence homologies, but which differ in their specific actions and sites of activity. For example, among peripheral sensory neurons NGF is primarily trophic for small fiber sensory neurons which mediate pain and temperature sensation (Johnson et al., 1980; Crowley et al., 1994; Smeyne et al., 1994). BDNF and NT-4/5 support the survival of larger, but overlapping, populations of neurons which may mediate mechanical sensation, although the function of BDNF and NT-4/5 responsive neurons are not yet fully defined (Klein et al., 1993; Ernfors et al., 1994a; Jones et al., 1994). NT-3 promotes the survival of predominantly large fiber proprioceptive neurons (HoryLee et al., 1993; Klein et al., 1994; Ernfors et al., 1994b).

The major trophic activities of the neurotrophins are mediated through high-affinity binding to specific members of the trk tyrosine kinase gene family. NGF acts through trk A, BDNF and NT-4/5 through trk B, and NT-3 through trk C (See Eide et al., 1993, for review). Subpopulations of sensory neurons differ with regard to which members of the trk family they express. Predominantly small-diameter sensory neurons express trk A, larger neurons express trk B, and the largest express trk C (Mu et al., 1993). There is considerable overlap in size among the different subpopulations, however, and some neurons express more than one type of trk receptor, while others express none (Wright and Snider, 1995). Each of these factors also binds with lower affinity to a common receptor called p75 (Rodriguez-Tebar et al., 1990, 1992). It is not clear if any biological effect is mediated through the p75 receptor, although several actions have recently been attributed to it (Herrmann et al., 1993; Palmer et al., 1993; Barrett and Bartlett, 1994; Verdi et al., 1994; Curtis et al., 1995). The target tissues for the different sensory subpopulations also differ as to the neurotrophins that they express (Schecterson and Bothwell, 1992; Mu et al., 1993). NGF mRNA is found principally in the superficial epidermis where nociceptive afferent terminals rest. BDNF mRNA is expressed in the deeper dermal layers of the skin where low-threshold mechanoreceptors terminate. NT-3 mRNA is found in muscle fibers where proprioceptive neurons project to muscle spindles. Target-derived neurotrophic factors appear to play an important developmental role in selecting neurons for survival during periods of naturally occurring cell death. At such times neurons growing toward the target must compete for limited 1044-7431/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

134

AID MCN 0538

/

6p07$$$101

04-22-96 10:20:17

mcna

AP: MCN

135

NGF Increases BDNF mRNA in Sensory Ganglia

supplies of the available factor, and large percentages of unsuccessful neurons die. Postnatally, target-derived neurotrophins such as NGF also regulate physiologically important functions such as neurotransmitter synthesis in their responsive neuronal populations (Kessler and Black, 1980; Lindsay and Harmar, 1989). Peripheral target fields are not the only sources of neurotrophins for peripheral sensory neurons. NGF, BDNF, and NT-3 are all synthesized in other tissues, both during development and in maturity (Shelton and Reichardt, 1984; Ernfors et al., 1990, 1992). All three of these neurotrophins have been localized to the DRG, but BDNF expression in particular has been demonstrated in sensory neurons. The localization of BDNF mRNA in DRG neurons, many of which may be responsive to BDNF, has led to the suggestion that BDNF in the DRG may serve an autocrine function. In fact it has been shown that mature sensory neurons are supported in vitro by BDNF via an autocrine mechanism (Acheson et al., 1995). In that study it was demonstrated that suppression of BDNF expression with antisense oligonucleotides decreased the survival of adult sensory neurons in vitro, and exogenous BDNF prevented this loss. Thus, regulation of neuronal neurotrophin expression may be importantly related to the survival of mature neurons, both in normal and pathological states. The mechanisms which regulate expression of the different neurotrophins are just beginning to become known. In the central nervous system, the level of expression of some members of this gene family correlates with the degree of neuronal activity. For example, rats exposed to light (resulting in retinal ganglion cell activity) have higher levels of BDNF mRNA in visual cortex than those reared in the dark (Castren et al., 1992). Kindling models of epilepsy also result in elevated levels of both BDNF and NGF mRNA in hippocampal and various cortical regions (Ernfors et al., 1991). Likewise, application of the excitatory neurotransmitter glutamate or its agonists stimulates the expression of BDNF mRNA in hippocampal cultures, while inhibitory GABAergic activity decreases it (Zafra et al., 1991). In the peripheral nervous system, there is little evidence that depolarizing signals regulate neurotrophin expression. Hematopoietic cytokines, however, appear to play a prominent regulatory role. Most notably, interleukin-1b (IL-1b) stimulates the expression of NGF in the peripheral nervous system (Lindholm et al., 1987), but no effects on the expression of the other neurotrophins have been reported. While it is possible that other cytokines may play an important regulatory role with regard to NGF or the other neurotrophins, this has not yet been demonstrated. We have investigated the possi-

AID MCN 0538

/

6p07$$$102

04-22-96 10:20:17

bility that members of the neurotrophin gene family may regulate their own expression in the peripheral nervous system, either via paracrine or autocrine interactions. We report here that systemic administration of recombinant human NGF (rhNGF) stimulates the expression of BDNF mRNA in the dorsal root ganglion, suggesting the presence of a novel regulatory pathway.

RESULTS Exogenous Administration of rhNGF Stimulates the Expression of BDNF in Adult Rat DRG rhNGF was administered to adult rats as a single injection at a dose of 1 mg/kg subcutaneously. Twenty-four hours later, the rats were sacrificed, and cervical dorsal root ganglia were removed and pooled for RNA analysis by nuclease protection assay. rhNGF administration resulted in a fourfold increase in BDNF mRNA levels (Fig. 1A). By contrast, systemic injections of NT-3 at the same dosage failed to significantly alter levels of BDNF mRNA expression (Fig. 1B). BDNF mRNA expression was maximally stimulated as early as 12 h after rhNGF injection, where levels reached nearly 30 times that of control ganglia (Figs. 2A and 2B). Although levels of expression began to decrease by 24 h, they remained significantly higher than control until the second day, at which time the effect was no longer detectable. Interestingly, when we examined ganglia of rats treated chronically with rhNGF administration at a dose of 1 mg/kg, three times a week for 2 consecutive weeks, we failed to see BDNF mRNA stimulation in the dorsal root ganglion (data not shown). This suggests that the stimulatory effect may be an acute or subacute reaction to the rhNGF administration. Dose response studies in adult rats demonstrated a significant stimulatory effect at doses at and above 0.1 mg/kg (Fig. 3). A similar experiment was performed on neonatal rat pups, to determine whether this phenomenon was restricted to the mature peripheral nervous system. Comparable stimulation of BDNF mRNA expression was observed in this setting as well, 16 h following a single injection of 1 mg/kg rhNGF, suggesting that it was not restricted to mature rats (Fig. 4).

Anatomical Localization of BDNF mRNA Stimulation To identify the cell types in the DRG which respond to NGF with increased synthesis of BDNF mRNA, in situ hybridization analysis was performed on DRGs from

mcna

AP: MCN

136

Apfel et al.

A-expressing neurons coexpressed BDNF mRNA (Figs. 5C, 6B). This coexpression therefore links NGF-responsive neurons to BDNF synthesis within the ganglia. Within the trk A population, the percentage of trk A/ BDNF-expressing neurons may have been underestimated due to the fact that some silver grains are obscured by the alkaline phosphatase reaction product. Indeed, following NGF treatment, the percentage of trk A neurons that coexpress BDNF increased (control, 35.7 { 3.92; rhNGF-treated, 61 { 6.38; P õ 0.05; Fig. 6B). We suspect that this increase is due at least in part to the ability to identify double-labeled neurons resulting from the increase in BDNF message. Following NGF administration, levels of BDNF mRNA appeared to be increased in trk A/BDNF double-labeled cells. We were unable to quantify this increase because attempts to measure grain density over trk A/BDNF (/) neurons were compromised due to the intracellular alka-

FIG. 1. (A) Representative ribonuclease protection assay for BDNF mRNA in DRG following systemic administration of rhNGF (1 mg/ kg) to mature rats (N Å 10) 24 h prior to sacrifice. Control rats (N Å 10) were treated with physiologic saline. Relative values were obtained by establishing a ratio between the intensity of the BDNF mRNA band and that for cyclophilin mRNA (1B15) which is constitutively expressed and serves as an internal standard. BDNF mRNA levels were increased in the rhNGF-treated animals to four times that of the control DRG. Each lane was loaded with total RNA (20 mg) pooled from the cervical DRG of five rats. (B) By contrast systemic injections of NT-3 (1 mg/kg) failed to have any effect on levels of BDNF mRNA expression in DRG at 24 h following injection. Similar results were obtained at earlier time points. Each lane was loaded with total RNA (lanes 1 and 3, 20 mg; lanes 2 and 4, 5 mg) pooled from the cervical DRG of five rats.

control and mature rats treated with a single dose of 1 mg/kg rhNGF. In control rats, nearly 50% of DRG neurons expressed BDNF mRNA, with some cells showing quite high levels of expression (Figs. 5A and 5C). Following NGF treatment, levels of BDNF mRNA were greatly increased in neurons (Figs. 5B and 5D). Interestingly, the overall percentage of BDNF-expressing neurons was only slightly increased in the treated group (control, 46.4 { 6.9; NGF-treated, 57.2 { 2.5, P õ 0.05; Fig. 6A. Previous studies have demonstrated that in the adult rat, approximately 45% of DRG neurons express the functional NGF receptor, trk A (Verge et al., 1992; Ernfors et al., 1993; Molliver et al., 1995). To determine whether sensory neurons that express trk A also express BDNF mRNA, dual isotopic and calorimetric in situ hybridization was performed. In untreated animals, 35% of all trk

AID MCN 0538

/

6p07$$$102

04-22-96 10:20:17

FIG. 2. (A) A ribonuclease protection assay for BDNF mRNA in DRG at varying times after administration of rhNGF (1 mg/kg). BDNF mRNA was maximally stimulated (30-fold) as early as 12 h after NGF administration. Levels remained elevated at 24 h, but were no longer elevated at 48 h. Each lane contains total RNA pooled from the cervical DRG of five rats. (B) Graph representing relative levels of BDNF mRNA normalized with respect to 1B15. Each point represents the ratio of BDNF mRNA band density from the corresponding lane in A and the associated 1B15 band.

mcna

AP: MCN

137

NGF Increases BDNF mRNA in Sensory Ganglia

FIG. 3. Ribonuclease protection assay for BDNF in DRG after injection of different doses of rhNGF. Stimulatory effects were seen with doses as low as 0.1 mg/kg of rhNGF. *Signifies that these groups differed significantly from the control group with a P õ 0.001, and ** signifies that this group differed from the control group with a P õ 0.01, by ANOVA with a Student– Newman–Keuls post hoc test. (N Å 3 RNA was pooled from all cervical DRG removed from five rats in each group.)

line phosphatase reaction product. An increase in silver grains over BDNF (/) cells not coexpressing trkA was also observed. Grain density measurements from this population revealed that following NGF treatment, the mean density of silver grains over BDNF-non trk A neurons increased almost twofold (control, 25.6 { 1.5; NGFtreated, 48.5 { 2.07; P õ 0.05; Fig. 7A). Thus administration of NGF increases BDNF message not only in trk A (/) cells, but also in non-trk A neurons. Histograms plotting the average grain density measured over BDNFexpressing neurons show that the distribution shifts to the right following NGF treatment, most noticeably in the population of neurons that express BDNF at low levels (Figs. 7B and 7C). These findings raise the possibility that NGF administration may result in paracrine effects on non-trk A neurons.

ploying cultured cerebellar granule neurons demonstrated that BDNF administration increased the expression of NT-3 mRNA; however, whether such regulation occurs in vivo is unknown (Leingartner et al., 1994). In that study the increase in expression of NT-3 was eliminated by addition of the protein kinase inhibitor K252a, suggesting that the effect was probably trk mediated. Here we have used double labeling in situ techniques to demonstrate that many neurons which express BDNF mRNA also express trk A mRNA, suggesting that the effect may be mediated, at least in part, by direct NGF binding to its high-affinity receptor. This mechanism may not be sufficient to account for all of our observations, however, since there were also many neurons with high levels of BDNF expression which did not express trk A mRNA. Possible explanations include the following: NGF may stimulate BDNF through indirect pathways, NGF may stimulate BDNF through the low-affinity p75 receptor, or trk A may be expressed in these neurons, but at levels of expression below the sensitivity of our double label in situ assay. In earlier reports of BDNF mRNA expression in peripheral ganglia, it was postulated that BDNF may mediate either an autocrine or a paracrine function (Ernfors et al., 1992; Schecterson and Bothwell, 1992; Wetmore and Olson, 1995). In vitro studies have shown that mature sensory neurons cultured singly express BDNF and depend on this BDNF synthesis for their survival, providing evidence for an autocrine loop (Acheson et al., 1995). Our results provide evidence for a BDNF paracrine loop regulated by NGF. Whether the BDNF produced by NGF

DISCUSSION This report demonstrates that administration of a neurotrophin regulates the gene expression of another neurotrophin in vivo. A single dose of rhNGF administered systemically rapidly and dramatically increased levels of expression of BDNF mRNA in DRG. The effect was observed in adult as well as neonatal rats, and was seen with doses as low as 0.1 mg/kg. A recent study em-

AID MCN 0538

/

6p07$$$103

04-22-96 10:20:17

FIG. 4. Effects in neonates. Comparable stimulation of BDNF mRNA expression (normalized to 1B15) was seen in neonatal rat pups 16 h following rhNGF administration (1 mg/kg) on Postnatal Day 1. *P õ 0.01 by Students t test. (N Å 3 RNA was pooled from all cervical DRG removed from 10 rats in each group.)

mcna

AP: MCN

138

Apfel et al.

FIG. 5. Sensory neurons in the DRG express BDNF mRNA and are responsive to NGF administration. Adult DRGs were hybridized with 33 Plabeled BDNF antisense riboprobes from control (A) and NGF-treated (B) animals. (A) In control DRGs, many sensory neurons express BDNF mRNA at low levels, while a few express BDNF at high levels. (B) Following a single injection of rhNGF, sensory neurons appear to increase levels of BDNF mRNA expression. Sensory neurons that synthesize BDNF mRNA also express the high-affinity NGF receptor trk A. Adult DRGs were hybridized using dual in situ hybridization with 33P-labeled BDNF antisense riboprobes and digoxigenin-labeled trk A antisense riboprobes from control (C) and NGF-treated (D) animals. (C) In control DRGs sensory neurons that express BDNF mRNA (silver grains) but not trk A mRNA (purple cells) were detected. Individual neurons which coexpress BDNF mRNA and trk A mRNA (arrows) were detected. Some trk A mRNA expressing cells do not also express BDNF mRNA. (D) In NGF-treated animals, levels of BDNF mRNA are increased in sensory neurons that coexpress trkA (arrows) as well as in some that do not. Some trk A expressing neurons that do not coexpress BDNF mRNA are still apparent. Scale bars, 200 (A and B) and 100 mm (C and D).

stimulation is released locally to impact on neighboring trk B expressing neurons, or is transported elsewhere via central or peripheral nerve processes is not yet known. If the newly expressed BDNF is active locally, it would be an example of a non-target-derived source of trophic support for a neuronal population. In the companion paper, Robinson et al. demonstrate that during develop-

AID MCN 0538

/

6p07$$$103

04-22-96 10:20:17

ment, NGF-dependent cranial sensory neurons express BDNF mRNA. When cocultured with separate BDNFdependent cranial neurons, BDNF synthesis by NGF responsive neurons acts in a paracrine fashion to increase the survival of BDNF-dependent neurons. Thus paracrine functions regulated by NGF may be important both during development and in maturity.

mcna

AP: MCN

139

NGF Increases BDNF mRNA in Sensory Ganglia

Current efforts to develop clinical applications for the neurotrophins have largely been directed by the longstanding perception that selective subpopulations of neurons respond to different neurotrophins. Nevertheless, recent in vivo studies have demonstrated that some of these factors may have broader effects than would be predicted based upon their classical biology (Johnson et al., 1986; Apfel et al., 1992). For example, NGF can prevent neuropathies involving large fiber, non-trk A containing neurons (Apfel et al., 1992). Our findings that paracrine regulatory interactions may exist between different members of the neurotrophin gene family could offer an explanation for these unexpected effects. The regulatory interactions demonstrated here suggest that the use of NGF as a treatment for peripheral neuropathies may have substantially broader effects than previously believed. Fully defining paracrine interactions among the different neurotrophin family members will be important in the design of rational drug therapy employing these factors.

EXPERIMENTAL METHODS Animals A minimum of five adult female Sprague –Dawley rats (Charles River Laboratories, Wilmington, MA) were used per group in each experiment. The rats were injected with a single dose of rhNGF generously provided by Dr. Gene Burton at Genentech, Inc. (South San Francisco, CA). They were euthanized painlessly with CO2 in accordance with the recommendations of the AVMA Panel on Euthanasia. The specific method was approved by our attending veterinarian and ACUC.

RNA Isolation Total RNA was isolated from rat DRG immediately after they were sacrificed by guanidinium thiocyanate lysis (Chomczynski and Sacchi, 1987). Concentration of RNA was determined by absorbance at 260 nm.

Solution Hybridization/Nuclease Protection Assay 32

P-labeled antisense riboprobe and sense cRNA standard construction. Riboprobes and standards were synthesized from cDNAs (rat BDNF cDNA and NT-3 cDNA were generously provided by Dr. Karoly Nicholics of Genentech, Inc., South San Francisco) subcloned into ri-

AID MCN 0538

/

6p07$$$103

04-22-96 10:20:17

FIG. 6. (A) Graph illustrating the percentage of neurons expressing BDNF mRNA in control and rhNGF-treated rats. Overall, rhNGF treatment slightly increased the number of BDNF (/) neurons. (B) Graph illustrating the percentage of trk A neurons that coexpress BDNF in control and NGF-treated rats. NGF treatment increased the percentage of trk A neurons that coexpress BDNF (* signifies that this group differed with a P õ 0.05).

boprobe vectors. Plasmids were linearized and in vitro transcription was initiated with 15-20 U of either T7, T3 (Promega), or SP6 (Gibco) RNA polymerases. In vitro transcription reactions were performed in a 25-ml mixture of 0.5 mg linearized template DNA, RNA polymerase transcription buffer, 4 mM DTT, 1 U RNasin (Promega), and 0.4 mM each of ATP, CTP, and GTP (Promega). For antisense riboprobe construction 8 ml of [32P]UTP (3000 Ci/mmol, NEN) and 0.02 mM UTP were added, or for cRNA construction 4 mM UTP was added to the mixture. After 60 min incubation at 377C and 10 min DNase 1 digestion the probes were purified on an 8 M urea-denatured 5% polyacrylamide gel. The full-length transcripts were excised from the gel and extracted with the RNaid kit (BIO 101) following the manufacturer’s instructions.

mcna

AP: MCN

140

Apfel et al.

Probes were used within 48 h of construction, and 200,000 cpm of each probe was used per sample. Solution hybridization and RNase digestion. Samples of total cellular RNA (25 –40 mg) or cRNA standards (0–40 pg, plus 25 mg yeast tRNA) were added to 200,000 cpm of riboprobe and coprecipitated in ethanol. Pellet was resuspended in 10 ml of hybridization mixture [40 mM Pipes (pH 6.4), 400 mM NaCl, 1 mM EDTA, and 80% (v/v) formamide]. The mixture was heated to 957C for 5 min and then incubated for 16 –20 h at 457C. Following hybridization, nonhybridized probe and sample were digested in digestion buffer [0.3 M NaCl, 10 mM Tris– HCl (pH 7.4), 10 mM DTT, and 5 mM EDTA], RNase A (Sigma) and RNase T1 (Boehringer Mannheim), followed by digestion with 0.5% SDS and proteinase K. Following extraction with phenol–chloroform (1:1), EtOH precipitation, and resuspension in water, the samples were resolved on a 5% denaturing acrylamide gel (8 M urea). Levels of 1B15 (a constitutively expressed message) was determined to standardize the amount of RNA in each sample. Band densities were determined by a computing densitometer (Molecular Dynamics).

In Situ Hybridization Dual in situ hybridization protocols were used as previously described (Wright and Snider, 1995). Briefly, sections were pretreated, air dried, and hybridized overnight with both [33P]UTP-labeled riboprobes (1 1 106 cpm/slide) and digoxigenin-labeled riboprobes (500 ng probe/slide) diluted in a hybridization mixture containing 50% deionized formamide, 10% dextran sulfate, 11 Denhardt’s, 41 SSC, 10 mM dithiothreital, 1 mg/ml yeast tRNA, and 1 mg/ml denatured salmon sperm DNA. The following day, slides were rinsed with successive washes of SSC as well as being RNase treated for 30 min at 377C with RNase A (20 mg/ml) in 10 mM Tris (pH 7.4), 0.5 M NaCl, and 100 mM EDTA. Sections were then transfered to a blocking solution composed of 21 SSC, 0.05% Triton X-100, and 2% normal lamb serum for 2 h. Following two 10-min washes in 100 mM Tris, 150 mM NaCl, pH 7.5, sections were incubated overnight at room temperature in anti-digoxigenin Fab antibody conjugated to alkaline phosphatase (1:1000, Boehringer Mannheim) diluted in the same solution containing 1% normal lamb serum and 0.3% Triton X-100. Sections were washed twice (5 min each) with the above buffer, followed by a 10-min wash in 100 mM Tris, 100 mM NaCl, 50 mM MgCl2 , pH 9.5. The reaction product was visualized by immersion in the dark for 2 – 24 h in the following: 45 ml nitro blue tetrazolium salt (75 mg/ml in 70% dimethylformamide), 35 ml 5-bromo-4-chloro-3-indolyl

AID MCN 0538

/

6p07$$$103

04-22-96 10:20:17

FIG. 7. (A) Graph illustrating the effect of rhNGF treatment on neurons that express BDNF mRNA but not trkA mRNA. NGF-treatment results in a twofold increase in the percentage of some area covered by silver grains (grain density) in BDNF (/) neurons. (B and C) Histograms plotting the frequency of neurons grouped by grain density. In control rats (B), two populations of BDNF ( /) cells were apparent. Most neurons express BDNF mRNA at low levels (left), while a smaller population expresses at high levels (right). Following NGF treatment (C), BDNF ( /) neurons that previously expressed BDNF mRNA at low levels dramatically upregulate BDNF message.

mcna

AP: MCN

141

NGF Increases BDNF mRNA in Sensory Ganglia

phosphate (50 mg/ml in dimethylformamide), and 2.4 mg levamisole diluted in 10 ml of the alkaline buffer. The reaction was terminated by washing sections twice (30 min each) in 10 mM Tris, 1 mM EDTA, pH 8.0, rapidly dehydrated in 70 and 95% ethanol, and air dried. To prevent positive chemography resulting from the alkaline phosphatase reaction, sections were dipped briefly in a 3% solution of parlodion dissolved in isoamyl acetate and allowed to air dry for at least 2 h. Sections were then dipped in Kodak NTB-2 liquid emulsion and stored in dessicated light-tight boxes at 47C for 10 – 15 days. Slides were developed in Kodak D19 and fixed in Kodak Fixer, rinsed in distilled water, and coverslipped with 50% glycerol.

Hybridization Controls Control experiments were performed to assess both the dual hybridization method and the specificity of the riboprobes. Sections were incubated with individual sense strand 35S-labeled riboprobes or were pretreated with RNase (Boehringer Mannheim, 20 mg/ml for 30 min at 377C) followed by hybridization with individual antisense riboprobes. In each case, control hybridizations resulted in complete loss of hybridization signal.

Analysis To determine the extent of colocalization, every third section was drawn using camera lucida using a 101 objective. Neurons labeled using nonisotopic probes were drawn under bright-field and neurons labeled using isotopic probes were drawn under dark-field conditions. Positive neurons were identified based on the presence of blue to purple precipitate (nonisotopic riboprobes), whereas isotopically labeled neurons were identified by the presence of silver grains (5 1 background) dispersed in a tight circular profile over a neuron. To quantify levels of mRNA expression in identified neurons, random sections from control (n Å 3) and NGFtreated (n Å 3) animals were photographed under darkfield conditions using a Nikon Microphot FXA microscope and a Kodak DCS 200 digital camera. Images were acquired using Adobe Photoshop 2.5.1 software for Macintosh. Each image typically included 10 –20 sensory neurons. The images were then quantified using the PCbased Metamorph image analysis system (Universal Imaging). The profiles of labeled sensory neurons were encircled and a measurement of the soma area was performed. The image was then assigned a threshold at a level which ensured that only the reflective silver grains were recognized by the computer. The number of pixels

AID MCN 0538

/

6p07$$$103

04-22-96 10:20:17

which were at or above the desired threshold value and thus occupied by silver grains was determined. To obtain an estimate of the grain density, the percentage of pixels above threshold over the total number of pixels outlined per neuron was then calculated. At least 90 neurons were quantitated for each animal. Comparisons of means between groups for cell counts and grain density measurements were performed using an unpaired Student’s t test (P õ 0.05).

REFERENCES Acheson, A., Conover, J. C., Fandl, J. P., DeChiara, T. M., Russell, M., Thadani, A., Squinto, S. P., Yancopoulos, G. D., and Lindsay, R. M. (1995). A BDNF autocrine loop in adult sensory neurons prevents cell death. Nature 374: 450 –453. Anton, E. S., Weskamp, G., Reichardt, L. F., and Matthew, W. D. (1994). Nerve growth factor and its low-affinity receptor promote Schwann cell migration. Proc. Natl. Acad. Sci. USA 91: 2795– 2799. Apfel, S. C., Arezzo, J. C., Lipson, L., and Kessler, J. A. (1992). Nerve growth factor prevents experimental cisplatin neuropathy. Ann. Neurol. 31: 76–80. Barrett, G. L., and Bartlett, P. F. (1994). The p75 nerve growth factor receptor mediates survival or death depending on the stage of sensory neuron development. Proc. Natl. Acad. Sci. USA 91: 6501– 6505. Castren, E., Zafra, F., Thoenen, H., and Lindholm, D. (1992). Light regulates expression of brain derived neurotrophic factor mRNA in rat visual cortex. Proc. Natl. Acad. Sci. USA 89: 9444 –9448. Chomczynski, P., and Sacchi, N. (1987). Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156– 159. Crowley, C., Spencer, S. D., Nishimura, M. C., Chen, K. S., Pitts-Meek, S., Armanini, M. P., Ling, L. H., McMahon, S. B., Shelton, D. L., Levinson, A. D., and Phillips, H. S. (1994). Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons. Cell 76: 1001 –1011. Curtis, R., Adryan, K. M., Stark, J. L., Park, J. S., Compton, D. L., Weskamp, G., Huber, L. J., Chao, M. V., Jaenisch, R., Lee, K-F., Lindsay, R. M., and DiStefano, P. S. (1995). Differential role of the low affinity neurotrophin receptor (p75) in retrograde axonal transport of the neurotorphins. Neuron 14: 1201 –1211. Eide, F. F., Lowenstein, D. H., and Reichardt, L. F. (1993). Neurotrophins and their receptors: Current concepts and implications for neurologic disease. Exp. Neurol. 121: 200 – 214. Ernfors, P., Wetmore, C., Olson, L., and Persson, H. (1990). Identification of cells in rat brain and peripheral tissues expressing mRNA for members of the nerve growth factor family. Neuron 5: 511 –526. Ernfors, P., Bengzon, J., Kokaia, Z., Persson, H., and Lindvall, O. (1991). Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7: 165 – 176. Ernfors, P., Merlio, J-P., and Persson, H. (1992). Cells expressing mRNA for neurotrophins and their receptors during embryonic rat development. Eur. J. Neurosci. 4: 1140– 1158. Ernfors, P., Rosario, C. M., Merlio, J-P., Grant, G., Aldskogius, H., and Persson, H. (1993). Expression of mRNAs for neurotrophin receptors in the dorsal root ganglion and spinal cord during development and following peripheral or central axotomy. Mol. Brain Res. 17: 217 –223. Ernfors, P., Lee, K-F., and Jaenisch, R. (1994a). Mice lacking brain-de-

mcna

AP: MCN

142

Apfel et al.

rived neurotrophic factor develop with sensory deficits. Nature 368: 147 – 150. Ernfors, P., Lee, K-F., Kucera, J., and Jaenisch, R. (1994b). Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents. Cell 77: 503 –512. Herrmann, J. L., Menter, D. G., Hamada, J-I., Marchetti, D., Nakajima, M., and Nicolson, G. L. (1993). Mediation of NGF-stimulated extracellular matrix invasion by the human melanoma low-affinity p75 neurotrophin receptor: Melanoma p75 functions independently of trk A. Mol. Biol. Cell 4: 1205– 1216. Hory-Lee, F., Russell, M., Lindsay, R. M., and Frank, E. (1993). Neurotrophin-3 supports the survival of developing muscle sensory neurons in culture. Proc. Natl. Acad. Sci. USA 90: 2613 –2617. Johnson, E. M., Gorin, P. D., Brandeis, L. D., and Pearson, J. (1980). Dorsal root ganglion neurons are destroyed by exposure in utero to maternal antibody to nerve growth factor. Science 219: 916 – 918. Johnson, E. M., Rich, K. M., and Yip, H. K. (1986). The role of NGF in sensory neurons. Trends Neurosci. 1: 33– 37. Jones, K. R., Farinas, I., Backus, C., and Reichardt, L. F. (1994). Targeted disruption of the BDNF gene perturbs brain and sensory development but not motor neuron development. Cell 76: 989 –999. Kessler, J. A., and Black, I. B. (1980). Nerve growth factor stimulates the development of substance P in sensory ganglia. Proc. Natl. Acad. Sci. USA 77: 649 –652. Klein, R., Smeyne, R. J., Wurst, W., Long, L. K., Auerbach, B. A., Joyner, A. L., and Barbacid, M. (1993). Targeted disruption of the trk B neurotrophin receptor gene results in nervous system lesions and neonatal death. Cell 75: 113 –122. Klein, R., Silos-Santiago, I., Smeyne, R. J., Lira, S. A., Brambilla, R., Bryant, S., Zhang, L., Snider, W. D., and Barbacid, M. (1994). Disruption of the neurotorphin-3 receptor gene trk C eliminates la muscle afferents and results in abnormal movements. Nature 368: 249 – 251. Leingartner, A., Heisenberg, C. P., Kolbeck, R., Thoenen, H., and Lindholm, D. (1994). Brain-derived neurotrophic factor increases neurotrophin-3 expression in cerebellar granule neurons. J. Biol. Chem. 269: 828 – 830. Lindholm, D., Heumann, R., Meyer, M., and Thoenen, H. (1987). Interleukin-1 regulates synthesis of nerve growth factor in non-neuronal cells of rat sciatic nerve. Nature 330: 658 – 659. Lindsay, R. M., and Harmar, A. J. (1989). Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons. Nature 337: 362 –364. Molliver, D. C., Radeke, M. J., Feinstein, S. C., and Snider, W. D. (1995). Presence or absence of Trk A protein distinguishes subsets of small sensory neurons with unique cytochemical characteristics and dorsal horn projections. J. Comp. Neurol. 361: 404 –416.

Mu, X., Silos-Santiago, I., Carroll, S. L., and Snider, W. D. (1993). Neurotrophin receptor genes are expressed in distinct patterns in developing dorsal root ganglia. J. Neurosci. 13: 4029– 4041. Palmer, M. R., Eriksdotter-Nilsson, M., Henschen, A., Ebendal, T., and Olson, L. (1993). Nerve growth factor-induced excitation of selected neurons in the brain which is blocked by a low affinity receptor antibody. Exp. Brain Res. 93: 226 –230. Robinson, M., Buj-Bello, A., and Davies, A. M. (1996). Paracrine interactions of bdnf involving ngf-dependent embryonic sensory neurons. Mol. Cell. Neurosci. 7: 143 –151. Rodriguez-Tebar, A., Dechant, G., and Barde, Y-A. (1990). Binding of brain derived neurotrophic factor to the nerve growth factor receptor. Neuron 4: 487 –492. Rodriguez-Tebar, A., Dechant, G., Gotz, R., and Barde, Y-A. (1992). Binding of neurotrophin - 3 to its neuronal receptors and interaction with nerve growth factor and brain-derived neurotrophic factor. EMBO J. 11: 917– 922. Schecterson, L. C., and Bothwell, M. (1992). Novel roles for neurotrophins are suggested by BDNF and NT-3 mRNA expression in developing neurons. Neuron 9: 449 – 463. Shelton, D. L., and Reichardt, L. F. (1984). Expression of b-nerve growth factor gene correlates with the density of sympathetic innervation in effector organs. Proc. Natl. Acad. Sci. USA 81: 7951 –7955. Smeyne, R. J., Klein, R., Schnapp, A., Long, L. K., Bryant, S., Lewin, A., and Lira, S. A. (1994). Severe sensory neuropathies in mice carrying a disrupted trk/NGF receptor gene. Nature 368: 246 –249. Verdi, J. M., Birren, S. J., Ibanez, C. F., Persson, H., Kaplan, D. R., Benedetti, M., Chao, M. V., and Anderson, D. J. (1994). p75LNGFR regulates trk signal transduction and NGF-induced neuronal differentiation in MAH cells. Neuron 12: 733– 745. Verge, V. M. K., Merlio, J. P., Grondin, J., Ernfors, P., Persson, H., Riopelle, R. J., Hokfelt, T., and Richardson, P. M. (1992). Colocalization of NGF binding sites, trk mRNA, and low affinity NGF receptor mRNA in primary sensory neurons: Response to injury and infusion of NGF. J. Neurosci. 12: 4011 –4022. Wetmore, C., and Olson, L. (1995). Neuronal and nonneuronal expression of neurotrophins and their receptors in sensory and sympathetic ganglia suggest new intercellular trophic interactions. J. Comp. Neurol. 353: 143 – 159. Wright, D. E., and Snider, W. D. (1995). Neurotrophin receptor mRNA expression defines distinct populations of neurons in rat dorsal root ganglia. J. Comp. Neurol. 351: 329 –338. Zafra, F., Castren, E., Thoenen, H., and Lindholm, D. (1991). Interplay between glutamate and g-aminobutyric acid transmitter systems in the physiological regulation of brain derived neurotrophic factor and nerve growth factor synthesis in hippocampal neurons. Proc. Natl. Acad. Sci. USA 88: 10037 –10041.

Received January 22, 1996 Revised February 19, 1996 Accepted February 19, 1996

AID MCN 0538

/

6p07$$$104

04-22-96 10:20:17

mcna

AP: MCN