Estrogen receptors colocalize with low-affinity nerve growth factor ...

Proc. Nadl. Acad. Sci. USA Vol. 89, pp. 4668-4672, May 1992 Neurobiology

Estrogen receptors colocalize with low-affinity nerve growth factor receptors in cholinergic neurons of the basal forebrain (steroid autoradography/nonlsotopIc iu sift hybrdatlou/mo

tohe

sty/c t/aglng ahdAle

r diease)

C. DOMINIQUE TORAN-ALLERAND*t, RAJESH C. MIRANDA*, WAYNE D. L. BENTHAM*, FARIDA SOHRABJI*, THEODORE J. BROWNS, RICHARD B. HOCHBERG§, AND NEIL J. MACLUSKY* *Department of Anatomy and Cell Biology and Center for Reproductive Sciences, Columbia University College of Physicians and Surgeons, New York, NY

10032; tDivision of Reproductive Science, University of Toronto, Toronto, Canada M5G 1L4; and §Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, CT 06510

Communicated by Patricia S. Goldman-Rakic, February 28, 1992

effects of estrogen in the brain is whether these actions are exerted directly. Alternatively, the steroid may potentiate synthesis of endogenous growth and trophic factors and their receptors (10), as in extraneural targets (11, 12), or it may interact with them in a stimulatory or inhibitory fashion to promote the growth and differentiation of specific neuromodulatory or neurotransmitter systems by autocrine or paracrine mechanisms, which may lead to shifts in the developmental patterns of resulting neural networks. To test the hypothesis that estrogen and the neurotrophins may act on the same basal forebrain neurons, we combined autoradiography with 1251-labeled estrogen ("2'I-estrogen) and nonisotopic in situ hybridization histochemistry or immunohistochemistry to identify the mRNA and encoded protein for the low-affinity (p75NOFR) NGF receptor and for choline acetyltransferase (ChAT; acetyl-CoA: choline O-acetyltransferase, EC 2.3.1.6), the acetylcholine-synthesizing enzyme, a marker of cholinergic neurons. We report here the colocalization of estrogen binding sites with the low-affinity NGF receptor p75NGFR mRNA and immunoreactive protein in neuronal subsets of the rodent medial septum, nuclei of the diagonal band of Broca, and in the continuum of neurons of the ventral pallidum, substantia innominata, and nucleus basalis of Meynert. Moreover, we also provide definitive evidence that cholinergic basal forebrain neurons bind estrogen as well.

The rodent and primate basal forebrain is a ABSTRACT target of a family of endogenous peptide signaling molecules, the neurotrophins-nerve growth factor, brain-derived neurotrophic factor, and neurotrophin 3-and of the gonadal steroid hormone estrogen, both of which have been implated in cholinergic function. To investigate whether or not these ligands may act on the same neurons in the developing and adult rodent basal forebrain, we combined autoradlography with '2I-labeled estrogen and either nonisotopic in situ hybridization histochemistry or Im nohistochemistry. We now report colocalization of intranuclear estrogen binding sites with the mRNA and immunoreactive protein for the low-affinity nerve growth factor receptor, which binds all three neurotrophins, and for the cholinergic marker enzyme choline acetyltransferase (acetyl-CoA:choline O-acetyltransferase, EC 2.3.1.6). Colocalization of estrogen and low-affinity nerve growth factor receptors implies that their ligands may act on the same neuron, perhaps synergistically, to regulate the expression of specific genes or gene networks that may influence neuronal survival, differentiation, regeneration, and plasticity. That cholinergic neurons in brain regions subserving cognitive functions may be regulated not only by the neurotrophins but also by estrogen may have considerable relevance for the development and maintenance of neural substrates of cognition. If estrogen-neurotrophin interactions are important for survival of target neurons, then clinical conditlons associated with estrogen deficiency could contribute to the atrophy or death of these neurons. These findgs have implications for the subsequent decline in those differentiated neural functions associated with aging and Alzheimer disease.

MATERIALS AND METHODS Animals. Postnatal female Sprague-Dawley rats (Charles River Breeding Laboratories) 10 and 12 days old (n = 15) and adult female mice (RIII) from our breeding colony, ovariectomized under methoxyflurane (Metofane) for 1 week before use (n = 5), were injected s.c. or i.v. in the jugular [1 1ACi per g of body weight (1 Ci = 37 GBq); =0.2 ng per g of body weight] with ll(-methoxy[16a-125I]iodoestradiol ([125I]MIE2) synthesized by R.B.H. as described (13) (100 1.d; 506% propylene glycol/50% saline) and killed 1 h later by transcardial perfusion under deep ether anesthesia with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.6) containing 2.5% (vol/vol) dimethyl sulfoxide (DMSO) and 0.1% of the RNase inhibitor diethylpyrocarbonate. Specific binding of this iodinated estrogen is completely abolished by concurrent exposure to 100-fold molar excess of the unlabeled nonsteroidal estrogen diethylstilbestrol (14). Brains were postfixed for 2 h at 40C in the same fixative, without DMSO, and then equilibrated overnight at 40C in 15% buffered sucrose before

The basal forebrain is a region that, in the human, is important for learning, memory, and other cognitive functions and in which the cholinergic neurons are affected early in Alzheimer disease and age-related cognitive impairment (1, 2). Developing and adult basal forebrain neurons are targets not only of the family of neurotrophins-nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophin 3 (NT-3)-but of the gonadal steroid estrogen as well. Basal forebrain cholinergic neurons of the developing rodent and primate have been shown to express the mRNA (3, 4) or encoded protein (5, 6) for the low-affinity form of the NGF receptor, which binds the neurotrophins (7, 8). Overlapping basal forebrain regions have also been shown to exhibit estrogen-receptor mRNA, estrogen binding sites, and high levels of the estrogen synthesizing enzyme cytochrome P450-dependent aromatase (for refs. see ref. 9), although the neurotransmitter phenotype(s) of these neurons has not been reported. A critical question concerning the developmental

Abbreviations: NGF, nerve growth factor; BDNF, brain-derived neurotrophic factor; NT-3, neurotrophin 3; ChAT, choline acetyl-

transferase; [125I]MIE2, 11.8-methoxy[16a-125I]iodoestradiol; ERE, estrogen response element. tTo whom reprint requests should be addressed.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 4668

Neurobiology: Toran-Aflerand et aL embedding in M-1 embedding matrix (Lipshaw Manufacturing, Detroit) and freezing on dry ice/acetone as described

(14).

Steroid Autoradiography. Steroid autoradiography was carried out first. Coronal sections (10-12 ,um) through the basal forebrain region were cut on a cryostat and thawmounted onto NTB-3 (Kodak) emulsion-coated slides and processed for steroid autoradiography combined with either immunohistochemistry (14) or nonisotopic (digoxigenin) in situ hybridization histochemistry as described (9). After photographic development, the autoradiograms were processed for immunohistochemistry with antibodies raised against either p75NGFR [monoclonal antibody IgG 192 (15); 1:25; gift of Eugene M. Johnson, Jr., Washington University, St. Louis] or ChAT (rabbit polyclonal antibody; 1:500; Chemicon) by means of the avidin-biotin-peroxidase complex method with 3,3'-diaminobenzidine as the chromagen (Vectastain ABC Elite kit, Vector Laboratories). Both antibodies have been extensively characterized and both antigens survive autoradiographic development [D170, 2.5% sodium sulfite/0.1% potassium bromide/0.45% 2,4diaminophenol dihydrochloride (Kodak), pH =7 (9, 14)]. In Situ Hybridization Histochemistry. Nonisotopic (digoxigenin) in situ hybridization histochemistry was carried out after steroid autoradiography for p75NG'R and ChAT mRNA as described (9). Briefly, digoxigenin-labeled oligonucleotides were synthesized as described (16, 17). The oligonucleotides were 3'-end-labeled with digoxigenin-labeled deoxyuridine triphosphate (dUTP) (Boehringer Mannheim) by terminal deoxynucleotidyl transferase (BRL). The developed autoradiograms were hybridized overnight at 35°C (NTB-3 emulsion melts off the slide at temperatures >37°C) with 50 ng of probe per ml as well as with 2 pmol of a random composition, unlabeled, 36-base oligonucleotide per ml containing >1020 sequence combinations (DuPont; NEP 550) added to decrease nonspecific hybridization. After hybridization to the target cDNA, the hybrids were detected by enzyme-linked immunohistochemistry with anti-digoxigenin antibodies conjugated to alkaline phosphatase (Fab fragment, 1:500; 48 hr; 4°C) (Boehringer Mannheim) and an enzymecatalyzed, blue-color reaction with 5-bromo-4-chloro-3indolyl phosphate and nitroblue tetrazolium salt (Boehringer Mannheim). After washing at 35°C and at relatively high stringency, sections were cleared in Histo-Clear (National Diagnostics, Manville, NJ) (to avoid fading of the color product) and mounted in Permount. p75NGFR mRNA. Because monoclonal antibody IgG 192 recognizes the low-affinity NGF receptor of the rat only (15), the distribution of immunoreactive p75NGIF could be studied only in autoradiograms of developing rats. To minimize the amounts of [125I]MIE2 needed and to address this question in the adult rodent, nonisotopic in situ hybridization histochemistry was carried out for p75NGFR mRNA in the much smaller adult mouse, as well as in developing rats. The p75NGFR probe was a 46-base synthetic oligonucleotide sequence from the membrane-spanning region of chicken low-affinity NGF receptor cDNA (18), which has been extensively characterized and which shares extensive homology with the low-affinity rat NGF receptor. ChAT mRNA. Two synthetic oligonucleotides, 39 and 48 bases, were synthesized from opposite ends of the cDNA sequence that encodes active rat ChAT (19). The first corresponds to bases 202-243, which is comparable to that used in molecular biological studies of the enzyme (20). The second spans bases 1120-1168, which corresponds to the porcine ChAT sequence used in enzyme induction studies (3). The p75NGFR and two ChAT probes have no significant homology with any of the known nucleotide sequences in the GenBank/EMBL data bases.

Proc. Natl. Acad. Sci. USA 89 (1992)

4669

Controls for the specificity of p75NGFR and ChAT mRNA detection included, as described (9, 17): (i) the demonstration on Northern blot analysis that signal detection by each probe hybridized only to a species of RNA of a size corresponding to that of the mRNA encoding the appropriate protein; (ii) incubation with noncomplementary (sense) digoxigeninlabeled probes; (iii) pretreatment with excess unlabeled 90base oligonucleotides synthesized against sequences that overlapped the experimental probes; and (iv) analyses of thermal stability, which showed a close match (-14C) between theoretical and experimentally calculated melting temperatures, indicating hybridization to a single species of mRNA containing the expected sequence. Neurons expressing p75NGFR or ChAT immunoreactivity or the blue cytoplasmic hybridization signal for their mRNA were characterized with respect to the presence or absence of discrete concentrations of silver grains in the emulsion underlying cell nuclei, indicating binding of the iodinated estrogenic ligand. [12-I]MIE2 binding was considered specific, using a 5 x background labeling criterion, where the probability of false labeling is