Generation and phenotypic characterization of a galanin overexpressing mouse

Neuroscience 133 (2005) 59 –77

GENERATION AND PHENOTYPIC CHARACTERIZATION OF A GALANIN OVEREXPRESSING MOUSE K. HOLMBERG,a,i E. KUTEEVA,a P. BRUMOVSKY,a U. KAHL,e H. KARLSTRÖM,b G. A. LUCAS,c J. RODRIGUEZ,c H. WESTERBLAD,d S. HILKE,f E. THEODORSSON,f O.-G. BERGE,g U. LENDAHL,b T. BARTFAIh AND T. HÖKFELTa*

was similar in overexpressing and wild type mice. Axotomy reduced the total number of DRG neurons less in overexpressing than in wild type mice, indicating a modest rescue effect. Aging by itself increased galanin expression in the superior cervical ganglion in wild type and transgenic mice, and in the latter also in preganglionic cholinergic neurons projecting to the superior cervical ganglion. Galanin overexpressing mice showed an attenuated plasma extravasation, an increased pain response in the formalin test, and changes in muscle physiology, but did not differ from wild type mice in sudomotor function. These findings suggest that overexpressed galanin in some tissues of these mice can be released and via a receptor-mediated action influence pathophysiological processes. © 2005 Published by Elsevier Ltd on behalf of IBRO.

a Department of Neuroscience, Karolinska Institutet, Retzius väg 8, B3-4, SE-171 77 Stockholm, Sweden b Department of Cell and Molecular Biology, Karolinska Institutet, 171 77 Stockholm, Sweden c

Department of Medical Biochemistry and Biophysics, Karolinska Institutet, 171 77 Stockholm, Sweden

d Department of Physiology and Pharmacology, Karolinska Institutet, 171 77 Stockholm, Sweden e Department of Neurochemistry and Neurotoxicology, Stockholm University, Sweden

Key words: nerve injury, extravasation, sensory ganglia, sympathetic ganglia, pain.

f

Department of Laboratory Medicine, Linköpings University Hospital, Linköping, Sweden g

AstraZeneca R&D, Södertälje, Sweden

Galanin is in most species a 29 (30 in human) amino acid peptide with an amidated C-terminal and was first isolated from the porcine intestine (Tatemoto et al., 1983). Galanin was subsequently isolated/cloned from bovine (see Rökaeus and Brownstein, 1986), rat (Vrontakis et al., 1987; Kaplan et al., 1988), human (Bersani et al., 1991; Evans and Shine, 1991; Schmidt et al., 1991; McKnight et al., 1992) and several other species showing a high degree of homology, whereby the N-terminal 14 amino acids are conserved in all species so far analyzed, with the exception of tuna fish (see Langel and Bartfai, 1998). Galanin derives from a large precursor molecule with 123 amino acids which also harbors a second peptide termed galanin message-associated peptide (GMAP) (Rökaeus and Brownstein, 1986). Initial studies showed that galanin is widely distributed both in the central and peripheral nervous system (Rökaeus et al., 1984; Ch’ng et al., 1985; Ekblad et al., 1985a; Melander et al., 1985, 1986; Skofitsch and Jacobowitz, 1985a,b, 1986; Furness et al., 1987). Several review articles (Bartfai et al., 1993; Merchenthaler et al., 1993; Crawley, 1995; Leibowitz, 1995; Kalra and Kalra, 1996; Zigmond et al., 1996; Fuxe et al., 1998; Kerr et al., 2000b; Xu et al., 2000; Gundlach, 2002; Vrontakis, 2002; Ubink et al., 2003) and books (Hökfelt et al., 1991, 1998) have reported on manifold functional aspects of galanin. In the sympathetic nervous system, i.e. in the rat superior cervical ganglion (SCG), galanin-like immunoreactivity (LI) is normally seen in only a few neurons (Strömberg et al., 1987). Following transection of the two efferent carotid nerves galanin is strongly upregulated in the neuronal cell bodies of rat and mouse SCG (Rao et al., 1993; Klimaschewski et al., 1994, 1996; Mohney et al., 1994;

h

The Harold Dorris Neurological Research Center, The Scripps Institute, La Jolla, CA, USA i

Biology Division, California Institute of Technology, Pasadena, CA, USA

Abstract—In most parts of the peripheral nervous system galanin is expressed at very low levels. To further understand the functional role of galanin, a mouse overexpressing galanin under the platelet-derived growth factor-B was generated, and high levels of galanin expression were observed in several peripheral tissues and spinal cord. Thus, a large proportion of neurons in autonomic and sensory ganglia were galanin-positive, as were most spinal motor neurons. Strong galanin-like immunoreactivity was also seen in nerve terminals in the corresponding target tissues, including skin, blood vessels, sweat and salivary glands, motor end-plates and the gray matter of the spinal cord. In transgenic superior cervical ganglia around half of all neuron profiles expressed galanin mRNA but axotomy did not cause a further increase, even if mRNA levels were increased in individual neurons. In transgenic dorsal root ganglia galanin mRNA was detected in around two thirds of all neuron profiles, including large ones, and after axotomy the percentage of galanin neuron profiles *Corresponding author. Tel: ⫹46-8-5248-7070; fax: ⫹46-8-331692. E-mail address: [email protected] (T. Hökfelt). Abbreviations: CGRP, calcitonin gene-related peptide; DRG, dorsal root ganglion; EDL, extensor digitorum longi; FITC, fluorescein isothiocyanate; GalOE, galanin overexpressing; GMAP, galanin messageassociated peptide; HPLC, high performance liquid chromatography; ir, immunoreactive; LI, like immunoreactivity; mRNA⫹, mRNA-positive; NOS, nitric oxide synthase; NP, neuron profile; NPY, neuropeptide Y; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; PDGF-B, platelet-derived growth factor B; RIA, radioimmunoassay; rOD, relative optical density; SCG, superior cervical ganglion; SSC, standard saline citrate; TSA, tyramide signal amplification; VAChT, vesicular acetylcholine transporter; VIP, vasoactive intestinal polypeptide; WT, wild type. 0306-4522/05$30.00⫹0.00 © 2005 Published by Elsevier Ltd on behalf of IBRO. doi:10.1016/j.neuroscience.2005.01.062

59

60

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Schreiber et al., 1994; Zhang et al., 1994; Shadiack et al., 1995). In sensory ganglia, such as dorsal root ganglia (DRGs), galanin is normally expressed only at low levels in a few small neurons, both in the rat (Ch’ng et al., 1985; Skofitsch and Jacobowitz, 1985a) and mouse (Corness et al., 1996; Shi et al., 2001). After axotomy there is a marked upregulation of galanin, mainly in small- and medium-sized neurons in both species (Hökfelt et al., 1987; Villar et al., 1989; Corness et al., 1996; Shi et al., 2001). Previous studies have shown that in adrenal chromaffin cells and adrenal nerve terminals of various species, galanin and/or GMAP are expressed at low to moderate levels (Bauer et al., 1986; Lundberg et al., 1986; PeltoHuikko, 1989; Rökaeus et al., 1990; Zentel et al., 1991). Since galanin under normal circumstances is expressed at very low levels in several systems, we and others (Cai et al., 1999; Steiner et al., 2001; Bacon et al., 2002; Holmes et al., 2003; see Crawley et al., 2002) have generated mice overexpressing galanin (GalOE mice), in our case with the platelet-derived growth factor-B (PDGF-B) promoter (Sasahara et al., 1991) linked to a large fragment of the galanin/GMAP gene including the second intron. It is hoped that such mice can provide some insight into the functional role of galanin released from endogenous stores, complementing multiple studies based on administration of exogenous galanin, and perhaps providing clues for new therapeutic strategies. The aim of the present work was therefore to define possible changes in behavioral phenotype of GalOE mice. In view of the broad distribution of PDGF-B (Sasahara et al., 1991) a similarly wide galanin expression can be expected. Therefore, to interpret possible phenotypic changes, it is necessary to perform a careful mapping of galanin in the GalOE mice. Here we focus on several peripheral tissues of our GalOE mouse, including sympathetic and sensory ganglia, adrenal gland, skin and muscle as well as spinal cord. We compared young adult GalOE mice with young wild type (WT) C57BL/6 mice, and in some cases also aged GalOE and WT mice. These mice were studied after transection of the carotid nerves or the sciatic nerve, and functional studies on extravasation, pain (formalin test), sweat secretion and muscle function were also carried out. Some of these results have previously been reported in abstract form (Holmberg et al., 2000). We have in a parallel paper also described the galanin distribution patterns in the brain of this GalOE mouse (Kuteeva et al., 2004).

EXPERIMENTAL PROCEDURES Generation of transgenic mice The 1.3 kb PDGF-B promoter (Collins et al., 1985; Sasahara et al., 1991) was excised from the psisCAT6a plasmid with XbaI and HindIII (New England Biolabs Inc., Beverly, MA, USA) and cloned into the XbaI/ HindIII restricted LITMUS 29 vector (Evans et al., 1995). The galanin/GMAP gene construct, including the second endogenous intron of the galanin gene and containing 5= SalI and 3= HindIII restriction sites, was ligated into the LITMUS/PDGF-B promoter plasmid. The galanin/GMAP fragment was created by

Fig. 1. (a– c) (a) Schematic drawing of the galanin/GMAP construct consisting of the PDGF-B promoter, the 5= untranslated region, a leader sequence (LS), a second intron from the galanin gene followed by the signaling sequence (black), the galanin/GMAP cDNA, and finally a 3= untranslated region. (b) Southern blot of GalOE DNA. A galanin plasmid was used as a control. (c) PCR products from genotyping. Lane 1 shows the 1 kb stair, lanes 2 and 3 the 445 bp WT band and the 248 bp GalOE band, respectively. In lane 3 a weak 445 bp band is seen indicating endogenous galanin.

polymerase chain reaction (PCR), from both mouse cDNA and genomic DNA. In a second round, the PCR products from the first round were used as overlapping templates, and the 5=- and 3=terminal primers contained restriction sites for SalI and HindIII, respectively. The PDGF-B promoter/galanin gene construct (see Fig. 1a) was excised from the XL-1 blue plasmid backbone using XbaI and HindIII. The excised PDGF-B promoter/galanin fragment was injected at a concentration of 2–3 ng/␮l into pronuclei from fertilized mouse oocytes (Hogan et al., 1986) from a cross between F1(C57Bl/6⫻CBA) mice. The experimental procedures were approved by the local ethics committee (Stockholms Norra Djurförsöksetiska Nämnd). The experiments conformed to the guidelines of the Society for Neuroscience on the ethical use of animals. The number of animals was minimized, and precautions were taken to minimize animal suffering.

DNA preparation and genotyping Mouse tail biopsies were lysed in a buffer [pH 8.0; 50 mM Tris, 100 mM EDTA, 100 mM NaCl, 1% (w/v) SDS] with 100 ␮g/ml Proteinase K (Boehringer Mannheim, Mannheim, Germany) overnight at 55 °C. DNA was extracted in 2-propanol and centrifuged at 13,000 r.p.m. for 10 min at 4 °C, washed in 70% ethanol and dissolved in Tris–EDTA buffer (10 mM, 1 mM; pH 8.0). DNA was amplified using PCR with specific primers for galanin 5=-TGCCTCCCTAGAGTCGACGAGGGATCCTCGTGCGCT-3= and 5=-AGGCATCCCAAGTCCCAGAGTGGCTGA-3= and TaqDNA polymerase (2.5 U; Sigma, St. Louis, MO, USA) at 95 °C for 45 s, at 54 °C for 1 min and at 72 °C for 1 min for 30 cycles, and finally at 72 °C for 10 min. The PCR product was separated on a 1.4% (w/v) agarose gel containing 5 ␮l ethidium bromide (10 mg/ml)/100 ml gel, and visualized with UV-light showing a 445 base pair WT band or a 248 base pair GalOE band.

Southern blot Genomic DNA (10 ␮g) from GalOE and WT mice was cleaved using a BamHI restriction enzyme (100 U/␮l) (NEB) in a 10⫻ BamHI buffer containing 1% bovine serum albumin overnight at 37 °C. As a control, 0.1 ng of plasmid was cleaved with the same enzyme, but for 1 h at 37 °C. As the size of the plasmid construct (ca 3⫻103 bp) represents approximately 1/106 of the size of the mouse genome (ca 3⫻109 bp), 0.1 ng of plasmid represents 10⫻ more copies of the gene than what is contained in 10 ␮g genomic DNA. DNA was separated on a 0.7% agarose gel containing ethidium bromide, then blotted over to a Hybond-N membrane

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77 (Amersham Pharmacia Biotech, Amersham, UK). A galanin cDNA probe was labeled with ␣-32P-dCTP (10.0 mCi/ml) (NEN, Boston, MA, USA) using a Rediprime II kit (Amersham Pharmacia Biotech). The membrane was prehybridized for 1 h at 65 °C in a hybridization solution containing 1⫻ Denhardt’s solution [0.02% bovine serum albumin (Sigma), 0.02% Ficoll (Pharmacia, Uppsala, Sweden), 0.02% polyvinylpyrrolidone), 0.02 M NaPO4 (pH 7.0), 1% N-lauroylsarcosine], 10% dextran sulfate (Pharmacia), and 50 mg/l denatured salmon testis DNA (Sigma). The probe was denatured, added to the hybridization solution and hybridized overnight at 65 °C. The membrane was then briefly washed in 2⫻ standard saline citrate (SSC; 1⫻ SSC⫽0.15 M NaCl; 0.15 M NaCitrate) at room temperature followed by 2⫻ SSC/1% sodiumdodecylate sulfate 2⫻30 min at 65 °C and then in 0.1% SSC for 5 min at room temperature. Results were examined using a phosphoimager (Fujifilm BAS-2500, Fuji, Japan) as well as after exposure to Super RX X-ray film (Fujifilm).

Surgical procedures and control groups Both WT and GalOE mice were anesthetized with pentobarbital prior to all surgery. The following groups were prepared. (I⫹II) Unilateral cut of the efferent carotid nerves of the SCG of young (4 – 8 months) (group I) and old (⬎15 months) (group II) mice (12 GalOE plus 12 WT mice); (III) unilateral transection of the sciatic nerve of young GalOE and WT mice (n⫽15⫹15); (IV) unilateral decentralization of the SCG by cutting the sympathetic trunk in old (⬎15 month old) GalOE and WT mice (n⫽4⫹8); (V) control GalOE and WT mice (n⫽12⫹12); (VI) fresh tissues from GalOE and WT mice for radioimmunoassay (RIA) (n⫽6⫹5); (VII) extravasation studies on GalOE and WT mice using mustard oil (n⫽7⫹7); (VII) formalin test on GalOE and WT mice (n⫽5⫹5); and (VIII) studies on muscle function of GalOE and WT mice (n⫽5⫹6). Surgically manipulated animals were re-anesthetized as above after a survival time of 7 days, and processed as described below.

In situ hybridization Oligonucleotides complementary to rat galanin nucleotides 152– 199, or alternatively a mixture of galanin nucleotides 241–273, 307–337, 324 –371, 361–393, 427– 459 and 487–519 (Vrontakis et al., 1987) was labeled with 35S-␥-dATP (NEN) at the 3=-end using terminal deoxynucleotidyltransferase (Amersham) and purified using QIAquick Nucleotide Removal Kit (Qiagen GmbH, Hilden, Germany). Animals from groups I, II, III and V were processed for in situ hybridization. The SCGs, DRGs, adrenal glands and spinal cord were rapidly dissected out, frozen on dry ice, cut in a cryostat (Microm, Heidelberg, Germany) at 14 ␮m thickness and thawmounted on Probe On slides (Fisher Scientific, Pittsburgh, PA, USA). The probe was diluted in a hybridization solution containing 50% deionized formamide (Ambion Inc., Austin, TX, USA), 4⫻ SSC, 1⫻ Denhardt’s solution, 50 mg/l denatured salmon testis DNA (Sigma) and 200 mM dithiotreitol (Sigma). Sections were air-dried overnight and then incubated with the oligonucleotide probe (0.5 ng) for 16 –18 h at 42 °C. After hybridization, sections were rinsed in 1⫻ SSC, 4⫻30 min at 55 °C followed by 1 h at room temperature, rapidly dehydrated, and then air-dried. Slides were dipped in liquid photo emulsion NTB2 (Kodak, Rochester, NY, USA). After exposure the sections were developed in Kodak D19 and fixed in Kodak 3000. Some of the sections were counterstained with Toluidine Blue. For control, slides were incubated with a hybridization cocktail containing an excess (100⫻) of cold probe. The slides were examined in a Nikon Microphot-FX microscope equipped with a dark field condenser and epi-polarization illumination. Photographs were taken with T-max black-and-white film (ASA 100; Kodak).

61

Immunohistochemistry Animals from groups I–V were processed for immunohistochemistry, that is perfused via the ascending aorta with 20 ml Tyrode’s Ca2⫹-free solution (37 °C) followed by 20 ml fixative containing 4% (w/v) paraformaldehyde and 14% (v/v) saturated picric acid in 0.16 M phosphate buffer (Pease, 1962; Zamboni and De Martino, 1967) and then with 50 ml of the same, but ice cold fixative. The SCGs, DRGs, skin of the foot paw, adrenal glands, skeletal muscle (soleus muscle) and spinal cord were dissected out and postfixed for 90 min at 4 °C. The tissues were then transferred to phosphate-buffered saline (PBS) containing 10% (w/v) sucrose, 0.01% azide and 0.02% bacitracin (Sigma), and stored overnight at 4 °C. The sections were cut on a cryostat (Microm) at 14 ␮m and thaw-mounted on gelatin– chrome–alum-coated slides. For indirect immunohistochemistry the sections were incubated overnight at 4 °C, in a humid chamber, with rabbit anti-galanin (1: 400 – 800) (Theodorsson and Rugarn, 2000), rabbit anti-vasoactive intestinal polypeptide (VIP) (1:400) (Prof. J. Fahrenkrug, unpublished observations), goat anti-calcitonin gene-related peptide (CGRP) (1: 800) (Arnel Products Co. Inc., New York, NY, USA), goat anti-nitric oxide synthase (NOS) (1:2000) (Herbison et al., 1996) and goat anti-vesicular acetylcholine transporter (VAChT) (1:1000 – 4000) (Arvidsson et al., 1997) or fluorescein isothiocyanate (FITC)-conjugated ␣-bungarotoxin (1:1500) (Sigma). Sections were rinsed in PBS, incubated with secondary antibodies conjugated to FITC (1:80) or lissamine rhodamine sulfonylcumarine (1:40), alternatively Rhodamine Red-X (1:100) or Cy3 (1:200) (all from Jackson ImmunoResearch Inc., West Grove, PA, USA) for 30 min at 37 °C, rinsed in PBS and mounted in PBS:glycerol (1:10) containing 0.1% paraphenylenediamine (Johnson and Nogueria Araujo, 1981; Platt and Michael, 1983). For double-immunohistochemistry the sections stained for galanin were re-incubated as described above with antiserum to CGRP, NOS, VAChT or with FITC-conjugated ␣bungarotoxin. In some experiments tyramide signal amplification method (TSA) (Adams, 1992), as described in detail in Brumovsky et al. (2002) and Kuteeva et al. (2004). Briefly, a commercial TSA-Plus Kit (NENLife Scientific Products, Boston, MA, USA) was used and the antibodies were diluted around 5–10⫻ more (galanin antiserum, 1:4000 – 8000; NOS, 1:10,000; VAChT, 1:10,000; VIP, 1:4000 – 8000). Sections were analyzed and photographed using a Nikon Microphot-FX microscope equipped with epi-fluorescence (Nikon Corporation, Tokyo, Japan) and appropriate filter combinations. Kodak T-max black-and-white film (400ASA; Kodak, Rochester, NY, USA) was used. Alternatively, sections were examined using a Biorad Radiance Plus (Biorad, Hempstead, UK) confocal scanning microscope installed on a Nikon E600 fluorescence microscope and equipped with argon and HeNe lasers. In some cases the films were scanned using a Nikon LS-2000 film scanner (Nikon Corporation). Scanned and digital images were imported into Adobe Photoshop 6.0 (Adobe System Incorporated, San Jose, CA, USA) and optimized for brightness and contrast.

Histochemical controls For in situ hybridization no signal was detected after incubation with a 100⫻ excess of unlabeled probe. For immunohistochemistry, preadsorption of peptide antisera with the respective peptide (10⫺5 or 10⫺6 M) abolished the staining patterns (not done for NOS and VAChT).

Grain density measurements and statistical analysis Grain density, measured as area occupied by grains in a segmented image, was monitored in the SCGs, DRGs and adrenal gland from WT and GalOE animals using a Nikon microscope connected to a Macintosh IIx (Apple Computer, Cupertino, CA, USA) equipped with NIH 1.6 program (courtesy W. Rasband,

62

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

NIMH, Bethesda, MD, USA), a Quick Capture frame grabber board (Data Translation, Marlboro, MA, USA) and a Cage-MTI 72 CCD camera (DAGE-MTI, Michigan City, IN, USA). Neuron profiles (NPs) with a clear nucleus with a three times higher grain density than the background were counted in control and axotomized SCGs and DRGs. Results are expressed as relative optical density (rOD). For each group five animals and four to five sections/ganglion were counted, except in aged animals (only four animals/group). Statistical analyses were performed using MannWhitney U test or Student’s t-test for independent groups.

Stereology To measure the total number of DRG neurons, stereological counting was performed (Gundersen et al., 1988a,b; Tandrup, 1993). Perfused WT and GalOE DRGs from control (n⫽4 per group) and axotomized (n⫽5 per group) mice were studied. Using the BioRad Radiance Plus confocal scanning microscope (see above), 14 ␮m thick sections, stained with 0.001% propidium iodide, were analyzed. The area and section thickness were measured by scanning at 10⫻ magnification. Using BioRad software the total volume for each ganglion was determined. Using a random starting point every sixth field was sampled. Using an x-y axis step size of 132 ␮m, and with a 60⫻ oil objective a 7 ␮m z-scan was performed, with a step size of 1 ␮m. All cells with clear nucleoli were counted, using an unbiased counting frame of 10,000 ␮m2. The total number of neurons was calculated as the product of the DRG volume and the numerical density. For further details, see Shi et al. (2001).

Sudomotor functional test Sudomotor function was evaluated by the silicone imprinting technique (Kennedy and Sakuta, 1984). Briefly, mice were placed (plantar surface up) over a 42 °C hot plate to stimulate sweating. Base and hardener (Silasoft® Normal, Detax GmbH & Co., Ettlingen, Germany) were mixed and spread onto the plantar surface of the hindpaw. After hardening they were evaluated for the number of sweat droplet impressions and the sweat output per gland at the main four pads of the paw under a dissecting microscope with transmitted light. The sweat output was calculated measuring the diameter of 20 sweat glands from the bottom flat surface of the paw pads using a grid.

Plasma extravasation assay Fifty milligrams/kilogram Evans Blue (Sigma) were administrated into the jugular vein, and 5 min later a 5% (v/v) mustard oil solution (Fluka, Sigma-Aldrich, Sweden) was injected intraplantarly into one hind paw, whereas mineral oil (Sigma), the vehicle, was injected into the contralateral hind paw (10 ␮l/paw). Thirty minutes later, the animals were killed by cervical dislocation, and punch biopsies were sampled from the hind paw plantar skin and weighed. Evans Blue was extracted from the tissue samples by incubation in 1 ml formamide for 24 h at 56 °C, and extravasated plasma protein was measured spectrophotometrically (Ultrospec 3000; Pharmacia Biotech, Uppsala, Sweden) at a wavelength of 620 nm. Result of plasma extravasation was expressed as absorbance/gram of tissue.

RIA

Formalin test

Tissues were cut into small pieces on ice, and 1 M acetic acid (Merck, Darmstadt, Germany) (10 ml/g tissue) was added and then boiled for 10 min. Samples were homogenized using a steel rod and a Vortex mixer (Scientific Industries Inc., New York, NY, USA) and centrifuged for 10 min at 8 °C and 1500⫻g. Supernatants were collected, and a second extraction using 10 ml distilled water/gram tissue was performed. The combined supernatants were lyophilized using a Speed Vac (Savant Instrument Inc., NY, USA) and stored at ⫺70 °C until analysis. Lyophilized samples were reconstituted in 1 ml phosphate buffer (0.05 M, pH 7.4) before assay. Galanin concentration was measured using a RIA for rat galanin (Theodorsson and Rugarn, 2000). High performance liquid chromatography (HPLC)-purified chloramine-T-125I-galanin was used as radioligand and synthetic rat galanin as calibrator (Neosystem, Strasbourg, France). Gammamaster 1277 and Multicalc 2000 software (LKB Wallac, Turku, Finland) were used for measuring radioactivity and evaluating results, respectively. Protein concentrations were measured using a modification of the Folin Phenol reagent method of Lowry et al. (1951) optimized for minimal (20 ␮l) sample volumes in wells of microtiter plates and analyzed at 690 nm on a UV Max Microplate Reader (Molecular Devices, Sunnyvale, CA, USA).

The formalin test was performed based on the experiments by Dubuisson et al. (1977). Briefly, the animals were habituated to the observation chamber for 30 min prior to testing. The chamber contained six individual acrylic enclosures (125 mm⫻100 mm, height 150 mm) which were ventilated and maintained at a temperature of 23–25 °C during the experiment. Formalin (20 ␮l of 2.7% formalin, corresponding to 1% paraformaldehyde, in 0.9% saline) was injected s.c. into the dorsum of the left hind paw. The experiments were video-recorded and analyzed off-line. Pain behavior was quantified as time spent licking the formalin-injected paw. Formalin injection normally induces a biphasic response, and the first phase (0 –5 min after formalin) and the second phase (15–30 min after formalin) were scored separately. In addition, the behavior during the interlude (5–15 min) was recorded.

HPLC Reverse-phase HPLC was performed using LKB Ultrapac TSK ODS-120T (5 ␮m, 4.6⫻250 mm) eluted with a 40 min linear gradient of acetonitrile (20 – 40%) in water containing 0.1% trifluoroacetic acid. Two Pharmacia P300 pumps were controlled by a Pharmacia GP250 gradient programmer. Samples were passed through Millipore GS filters (0.22 mm) before chromatography to remove particulate matter. Samples of 200 ␮l were injected. Fractions of 0.5 ml were collected at an elution rate of 1.0 ml/min. Each fraction was lyophilized and re-dissolved in 100 ␮l of distilled water before analysis. The fractions were assayed for immunoreactivity in the tubes used for their collection.

Muscle physiology Isolated fast-twitch extensor digitorum longi (EDL) and slow-twitch soleus muscles were studied as described earlier (Johansson et al., 2003). Isolated muscles were mounted in a chamber with oxygenated Tyrode’s solution, and stimulated by brief current pulses. The force production was measured under isometric conditions. The number of muscles was five to six throughout. Data are presented as mean⫾S.E.M.

RESULTS In the present study GalOE mice under the PDGF-B promoter were generated resulting in three founders (B3, N8 and O4), and a line of transgenic animals was established from each founder. In the present study all experiments were performed on the B3 founder line, with the exception of the analysis of old animals, where also N8 and O4 founder lines were used. Immunohistochemical analysis showed a somewhat stronger staining for galanin in the B3 line which was therefore chosen for the

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

63

Table 1. Levels of galanin-LI in peripheral tissues of young WT and GalOE mice expressed as pmol GAL/mg total protein Tissue SCG Adrenal gland DRG Hind paw Spinal cord (thoracic) Spinal cord (lumbar)

WT (n⫽3) 0.07⫾0 2.07⫾0.65 0.07⫾0.09 0.11⫾0.07 0.47⫾0.11 0.75⫾0.42

GaLOE (n⫽4) 1.27⫾0.43 2.34⫾0.68 0.69⫾0.36 0.22⫾0.16 1.81⫾0.96 2.07⫾1.42

Fold-increase 14.1 1.1 9.5 2.0 3.9 2.8

Data expressed as mean values (⫾SD).

further experiments. Southern blot showed that in the B3-founder approximately five copies of the construct have been inserted, as the hybridization intensity from 0.1 ng of the plasmid control was approximately twice as strong as that from 10 ␮g genomic DNA (see Experimental Procedures for calculation) (Fig. 1b). For routine breeding the animals were genotyped using PCR, with specific primers generating a 445 bp band for the WT and a 248 bp band for the GalOE mouse (Fig. 1c). The RIA experiments showed significantly higher galanin levels in tissues of the GalOE as compared with WT mice; e.g. in SCGs and DRGs the galanin levels were 14- and eight-fold increased, respectively (Table 1). HPLC chromatography showed galanin-immunoreactive (ir) components eluting in the same position as synthetic rat galanin in all tissues examined (Fig. 2a–f). With regard to histochemistry both in situ hybridization and immunohistochemistry were performed, but only the former results were quantified. Galanin expression in young and old SCGs In young WT mice, following transection of the two carotid nerves, the percentage of mRNA-positive (mRNA⫹) NPs was increased from ⬃4% to ⬃40%, with a more than doubling of the rOD mice (Fig. 3a, d; Table 2). However, in the GalOE mice ⬃50% of the NPs were galanin mRNA⫹ both with and without axotomy (Fig. cf. 3j with 3g; Table 2), but there was a significant increase (⬃60%) in rOD (Table 2). Both small and large neurons were positive. The percentages of NPs were not significantly different between WT (39.0⫾6.4) and GalOE (49⫾8.2) mice after axotomy, nor were the rODs (99.5⫾25.7 versus 125.9⫾15.3) (Table 2). Immunohistochemical analysis of the SCGs in WT and GalOE mice showed similar expression patterns as seen with in situ hybridization. In young WT mice only low numbers of neurons were seen, and a few, slender galanin-ir processes, possibly axons (Fig. 4a), in contrast to many VAChT-positive fibers (Fig. 4b, c). There were more galanin-ir neurons and patches of processes in the young GalOE mice (Fig. 4d). The latter processes had the morphological features of nerve terminals. Indeed doublestaining with VAChT (Fig. 4e) in the GalOE mice revealed a high degree of coexistence (Fig. 4f), strongly suggesting that they are preganglionic fibers. In aged animals there was a distinct increase in galanin mRNA levels in WT SCGs, and ⬃60% of the NPs were galanin mRNA⫹, and in GalOE mice there was a further increase to ⬃85% (Table 2). Also the rOD was significantly

higher in GalOE than in WT mice (56.5⫾12.1 versus 26.6⫾12.1) (Table 2). In aged WT (Figs. 3b, e; 4g) and GalOE (Figs. 3h, k; 4j) many galanin ir cell bodies were seen. However, densely packed processed forming clusters were only observed in GalOE (Figs. 3h; 4j, m) but not WT (Figs. 3b, e; 4g) mice. These processes, as in young GalOE mice, mostly colocalized VAChT (Fig. 4k, l, n, o). In aged WT mice there were, just as in young WT mice (Fig. 4a– c), many more VAChT than galanin-positive fibers (Fig. 4g–i). Following transection of the sympathetic trunk (decentralization) most of these fibers disappeared, and the numerous galanin-IR cell bodies could now be more clearly seen (c.f. Fig. 3k with 3h). Galanin expression in stellate and local parasympathetic ganglia In stellate ganglia some neurons in WT and many in GalOE (Fig. 5a) mice expressed galanin (quantification and lesion were not performed). Double-staining in the GalOE mouse showed that a small proportion of these galanin neurons co-expressed VIP (Fig. 5b, c), a marker for sympathetic cholinergic neurons (Lundberg et al., 1979). There were also some VIP-positive, galanin-negative neurons (Fig. 5c). In local parasympathetic ganglia in the salivary gland of GalOE mice many neurons expressed galanin, and several of these were ir for VIP (Fig. 5d–f) or NOS (Fig. 5g–i) (no triple-staining was performed). Galanin expression in young and old DRGs after peripheral nerve injury Only a low percentage of NPs was galanin mRNA⫹ in young WT DRGs, but axotomy caused a marked increase in both percentage (from ⬃4% to ⬃52%) of NPs and rOD (from ⬃40 to ⬃60) in (Table 2). In the GalOE mouse around two thirds of the NPs showed significant labeling, but here axotomy induced a small, but significant decrease in percentage NPs (from ⬃70% to ⬃50%) but not in rOD (Table 2). The immunohistochemical analysis revealed similar findings, that is only single galanin-positive cells in WT mice (Fig. 3c), a strong increase after axotomy (Fig. 3f), and a very high proportion in GalOE mice both without (Fig. 3i) and with axotomy (Fig. 3l). Both small and large neurons were labeled, both in controls and after axotomy (Fig. 3i, j). Stereological total neuron count showed no difference between the WT (9934⫾585) and GalOE (9858⫾587) mice, and after axotomy there was a statistically significant

64

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 2. (a–f) HPLC chromatography shows galanin-ir component eluting in the position of synthetic rat galanin in SCG (a, b), DRG (c), spinal cord (d) and hind paw (e, f) of WT (a, e) and GalOE (b– d, f) mouse.

reduction of total number of neurons in both WT (7873⫾225) and GalOE (8394⫾229) mice. When comparing the reduction in number of neurons in WT versus GalOE mice, there was a 20.7⫾2.3% reduction in the WT and a 14.8⫾2.3% reduction in the GalOE. This difference is statistically significant (P⬍0.004), suggesting that overexpression of galanin can have a modest rescue effect. Double-immunohistochemistry showed colocalization of galanin- (Fig. 6a) and CGRP- (Fig. 6b) LIs in a major subpopulation of DRG neurons (Fig. 6c). However, in addition galanin-LI was also seen in large, CGRP-negative DRG cell bodies (Fig. 6c). Aged GalOE mice showed, when compared with young GalOE mice, a reduction of NPs expressing galanin mRNA in DRGs (from ⬃70 to ⬃50%), but no statistically significant change was seen in young versus old WT DRGs (3.8 vs 1.9%) (Table 2). The rOD was

lower in aged WT than in aged GalOE DRGs (⬃18 vs. ⬃60), but there was no difference in aged versus young GalOE mice (⬃60 vs. ⬃64%) (Table 2). Galanin expression in spinal cord In WT spinal cord galanin-LI was mainly seen in a dense fiber network in the superficial layers of the dorsal horns and, in lower numbers, in the lateral sympathetic columns and in the remaining parts of the gray matter, including ventral horns (Fig. 7a). In the GalOE mouse galanin-LI (Fig. 7b) and galanin mRNA (Fig. 7b, inset) were seen in ␣-motor neurons at all spinal levels, whereas no galanin mRNA or peptide (Fig. 7a) expression could be detected in these neurons in WT mice. There appeared in general to be more numerous and more strongly fluorescent fibers in the entire gray matter of GalOE (Fig. 7b) than WT (Fig. 7a) mice. Galanin-ir

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

65

Fig. 3. (a-l) Galanin mRNA (a, d, g, j) and galanin-LI (b, e, h, k) in SCGs and galanin-LI in DRGs (c, f, i, l). (a, d, g, j) In young adult mice, a few galanin mRNA⫹ neurons (arrowhead) are seen in control WT SCG (a), whereas GalOE SCG has high numbers (g). Following transection of the two carotid nerves there is a dramatic increase in numbers of galanin mRNA⫹ neurons (arrowheads) in WT SCG (d), whereas axotomized GalOE SCG (j) is similar to unlesioned GalOE SCG (g). (b, e, h, k) In aged mice, galanin-LI is seen in a moderate number of neurons in the WT SCG, together with sparsely distributed galanin-ir processes (arrowheads) (b, e). Aged GalOE SCGs have a dense galanin-IR fiber network and many galanin-ir cell bodies, the latter being difficult to discern (h). However, following decentralization of the GalOE SCG most of the galanin-positive fibers disappear, revealing many strongly galanin-IR neurons (arrowheads) (k). (c, f, i, l) In young adult mice a few weakly (small arrowhead) and more strongly (big arrowhead) galanin-ir neurons (arrowhead) are seen in WT DRG (c), with a strong upregulation after axotomy (f). In GalOE DRG many neurons, both small (small arrowheads) and large (big arrowheads) ones are galanin-positive (i). Axotomized GalOE DRG shows a somewhat less intense galanin-LI in some neurons (cf. l with i). Scale bars⫽50 ␮m in a (a⫽c–l) and b.

sympathetic preganglionic neurons could be seen only at high thoracic level in aged (Fig. 7d), but not young GalOE mice, and were identified by double-labeling with NOS (Fig. 6d–f). Only a small proportion of the NOS-positive neurons contained galanin-LI. Galaninpositive cells were also seen in the white matter of GalOE (Fig. 8b), but not WT (Fig. 8a) mice. In skeletal muscles strongly galanin-ir fibers were seen associated with ␣-bungarotoxin-labeled motor end-plates of GalOE mice (Fig. 6g), and following transection of the sciatic nerve these galanin-IR fibers disappeared (data not shown).

Galanin expression in the adrenal gland Galanin mRNA was detected at moderately high levels in the medulla of young WT and GalOE mice, without any significant difference in grain density (rOD 10.3⫾4.3 versus 11.5⫾5.3) (chromaffin cells and ganglion neurons were not measured separately). The same was true for aged WT (rOD 5.8⫾2.1) and GalOE (rOD 12.2⫾8.3) mice, and for young versus aged mice of both types (data not shown). Galanin-LI was detectable in most of the chromaffin cells in the medulla with no apparent difference in intensity between young WT and transgenes. Galanin-ir

66

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Table 2. Percentage of GAL mRNA-positive neuron profiles and grain density (rOD) in SCGs and DRGs of WT and GALOE mice Tissue

SCG young WT OE SCG old WT OE DRG young WT OE DRG old WT OE

Neuron profiles (%)

rOD

Control

Axotomy

Control

Axotomy

3.7⫾3.3a 52.7⫾9.1e

39.0⫾6.4b 49.0⫾8.2f

40.7⫾18.0c 74.9⫾14.0g

99.5⫾25.7d 125.9⫾15.3h

59.1⫾9.3 84.0⫾2.9

n.d. n.d.

26.6⫾12.1 56.5⫾12.1

n.d. n.d.

3.8⫾1.5i 69.9⫾6.4m

51.6⫾3.6j 52.6⫾7.0n

40.1⫾9.8k 63.7⫾13.0o

1.9⫾0.4 49.4⫾3.4

n.d. n.d.

18.4⫾7.6 59.3⫾17.8

60.7⫾12.0l 58.9⫾10.9p n.d. n.d.

All values are reported as mean⫾S.D. n.d. ⫽ not determined; n.s. ⫽ not significant; WT SCG young vs. old, P⫽ 0.05. b WT SCG control vs. axo, P⫽0.02. c WT SCG young vs. old, n.s. d WT SCG control vs. axo, P⫽0.02. e OE SCG young vs. old, P⫽0.021. f OE SCG control vs. axo, n.s. g OE SCG young vs. old, n.s. h OE SCG control vs. axo, P⫽0.003. i WT DRG young vs. old, P⫽0.042. j WT DRG control vs. axo, P⫽ 0.001. k WT DRG young vs. old, P⫽0.001. l WT DRG control vs. axo, P⫽0.039. m OE DRG young vs. old, P⫽0.006. n OE DRG control vs. axo, P⫽ 0.003. o OE DRG young vs. old, n.s. p OE DRG control vs. axo, n.s. a

fibers were occasionally seen in cortex of GalOE, but not WT mice (data not shown). No obvious difference in the number of galanin fibers in the medulla was observed among the four groups.

and 93 and 1.5 and 3.2, respectively (Table 3). However, no difference was seen between the averages of the WT and GalOE mice, neither 10 nor 20 min after the onset of stimulation (Table 3).

Galanin in sweat and salivary glands, and sweat secretion

Galanin in sensory fibers and plasma extravasation and pain behavior

In WT mice only single galanin-positive fibers could be observed in sweat glands and around related blood vessels (Fig. 9a), whereas in the GalOE mouse numerous strongly fluorescent nerve endings were seen in these structures (Figs. 6h, k, m; 9b). This was true both for young and old GalOE mice, although there was a tendency for more galanin-positive fibers in old glands. In sweat glands most galanin-ir fibers associated with the acini were VAChT-positive (Fig. 6h–j), strongly suggesting that these fibers are cholinergic sympathetic fibers. The density of VAChT-positive fibers was similar in young and old mice of both WT and GalOE mice. Along small blood vessels inside the acini, galanin coexisted with CGRP (Fig. 6k–m). Galanin was observed in a plexus of nerves around the acini of salivary glands of GalOE mice (Fig. 5j, m) and most of these fibers were also VIP-positive (Fig. 5k, l and n, o). Evaluation of sudomotor function after physiological stimulation showed values for the total number of sweat glands and a sweat output per gland ranging between 68

In the hind paw of the GalOE mouse, many galanin-positive fibers were seen closely associated with, and penetrating into the epithelium (Figs. 6m; 9b), around blood vessel walls (Figs. 6k, m; 9c, e) and in small nerves (Fig. 9c). CGRP-ir fibers had a similar localization (Figs. 6l; 9d), and in most of these cases CGRP and galanin coexisted (Figs. 6m; 9e, f). This distribution was similar in young and old GalOE mice. The functional role of galanin in these sensory fibers was analyzed in plasma extravasation experiments. The results from these experiments showed a significant reduction (⬃35%) of plasma extravasation in GalOE mice as compared with WT mice, whereas no significant difference was observed on the control side (Fig. 10). In the formalin test a typical biphasic response was seen in the WT mice (Fig. 11). A stronger pain behavior was seen in the GalOE mice throughout the test, with significant differences between the two groups during the interlude (P⬍0.05) and second phase (P⬍0.02) (Fig. 11).

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

67

Fig. 4. (a-o) Immunofluorescence micrographs of SCGs of young WT (a– c), young GalOE (d–f), old WT (g–i) and old GalOE (j– o) mice after double-staining for galanin (a, d, g, j, m) and VAChT (b, e, h, k, n); c, f, i, l, o show, respectively, the merged pictures. (a– c) In the young WT SCG low number of galanin-IR cell bodies are seen and only few processes (a). There are clusters of VAChT-positive nerve terminals (b) which surround galanin-positive and -negative cell bodies (c). (d–f) In young GalOE mouse there are many galanin-positive cell bodies and also nerve terminals (small arrowheads) (a) together with many VAChT-positive fibers (e), and in some of these galanin and VAChT coexist (small arrowheads). Arrows point to VAChT-only-positive fibers (f). (g–i) In the old WT mouse there are only few fibers (arrow in g) but many galanin-positive cell bodies (g). The many VAChT-positive terminals (h) surround galanin-positive cell bodies (i). The single galanin-positive fiber (arrows) lacks VAChT and may represent an axon. (j– o) In the old GalOE mouse galanin-positive nerve terminals form dense clusters (arrowheads in j, m), and they often overlap and coexist with the VAChT-positive terminals (arrowheads in k, l, n, o). There are, however, still both VAChT-only-positive fibers (arrows in j–l) as well as, occasionally, galanin-only-positive processes (arrows in m– o). Scale bars⫽50 ␮m (bar in j is valid for a–i, and k, l) and 10 ␮m (m– o).

Galanin in the motor end-plate, and muscle function In the GalOE mouse most motor end-plates in the soleus muscle, characterized by ␣-bungarotoxin binding, were galanin-ir (Fig. 6g). This was not seen in the WT mouse (data not shown). In EDL and soleus muscles there was no difference between the groups regarding muscle length or weight or cross-sectional area. The extensor digiti longi muscle. There was no difference (P⬎0.05) in (i) speed of contraction and relaxation in isometric twitches or tetani, (ii) force frequency-relationship, or (iii) force production during fatiguing stimulation

(300 ms, 70 Hz tetani given every 2 s). However, after fatiguing stimulation, force recovered faster in WT muscle and was significantly stronger at 30 min (P⬍0.05) (Fig. 12a). The soleus muscle. WT muscles had significantly shorter (P⬍0.01) contraction times than GalOE muscles both in twitches (57.4⫾1.3 vs. 67.0⫾1.9) and tetani, but the relaxation speed was not different. There was no difference in the force frequency-relationship, or force during fatiguing stimulation (600 ms, 50 Hz tetani given every 2 s) between the groups. After fatiguing stimulation, force re-

68

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 5. (a-o) Immunofluorescence micrographs of stellate ganglion (a– c), of local parasympathetic ganglia in salivary gland (d–i) and of salivary gland (j– o) of GalOE mice after double-staining for galanin (a, d, g, j, m) and VIP (b, e, k, n) or nNOS (h); c, f, i, l and o show, respectively, the merged pictures. (a– c) Many neurons in the stellate ganglion express galanin (a) and only a few are VIP-ir. Two out of five VIP-ir cells are galanin-positive (yellow) and three are VIP-alone-ir (red) (c). (d–f) Several galanin positive neurons are seen (d), and most are VIP-positive (yellow) (e, f). Arrow points to small nerve bundle. (g–i) Two galanin/nNOS positive cells are seen (yellow). (j– o) A moderately dense galanin-positive fiber network is seen (j) with a VIP-ir network with a similar distribution (k), and many of these fibers are identical (l), as more clearly seen in the corresponding high power magnifications as indicated by box in l (m– o). Scale bars⫽50 ␮m (a– c, d–f, j–l) and 10 ␮m (m– o).

covered faster in WT muscles and was significantly stronger at 30 min (P⬍0.05) (Fig. 12b). Controls None of the transcript signals described were seen after incubation with a hybridization cocktail with an added excess of cold probe. None of the ir structures described could be observed after incubation with an excess of the immunogenic peptide (Fig. 8c). Fairly strong, unspecific fluorescence was observed, to varying degrees, in the epidermis, especially the keratinized layer, and in sweat gland acini (Fig. 9a, b).

DISCUSSION To further understand the role of the 29-amino acid peptide galanin (Tatemoto et al., 1983) as a messenger molecule

in the nervous and endocrine systems, a mouse overexpressing galanin under the PDGF-B promoter was generated. We had previously attempted to overexpress galanin under two other promoters (two forms of neuron-specific enolase), without success. Therefore we turned to the widely expressed PDGF-B promoter (Sasahara et al., 1991), which had been shown by Games et al. (1995) to produce robust overexpression of mutated human ␤-amyloid precursor protein in mouse brain. We now show a wide expression also of galanin in the periphery and spinal cord, and in a parallel paper in the brain (Kuteeva et al., 2004). Interestingly, we also observed galanin-positive cells in the spinal white matter, presumably representing glial cells. The wide distribution of overexpressed galanin requires a careful analysis of the expression patterns to allow interpretation of phenotypic changes. Despite this, such inter-

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

69

Fig. 6. (a-m) Immunofluorescence micrographs showing galanin-LI in GalOE mice (green) (a, d, g, h, k) doubled-stained for CGRP (red) in DRG (b, c), for NOS (red) in preganglionic neurons in the thoracic spinal cord (e, f), for bungarotoxin (red) in motor end-plates (g), for VAChT (red) in sweat gland (i, j) and for CGRP (red) in sweat glands and skin (l, m). (a– c) Most DRG neurons are galanin-positive (a), and virtually all of them are CGRP-ir (b, c), but there are large galanin-positive neurons (arrowheads) apparently lacking CGRP-LI. (d–f) Some preganglionic neurons contain both galaninand NOS-LI (arrows). (g) Virtually all motor end-plates, characterized by bungarotoxin, are galanin ir. (h–m) In the hind paw numerous galanin-ir fibers are seen in sweat glands and most are VAChT-ir (h–j), whereas outside the glands most galanin fibers are CGRP-positive (arrowheads) (k–m). Arrows point to galanin terminals apparently lacking CGRP in close contact to the acini. Scale bars⫽50 ␮m in a (a⫽b⫽c), d (d⫽e⫽f), h (h⫽i⫽j), k (k⫽l⫽m) and 10 ␮m in g.

pretations should be considered preliminary, since, for example, galanin overexpressed in the sympathetic neurons may influence the results in our ‘sensory models’ (extravasation and formalin tests, see below). General considerations The present study shows high galanin expression in neuronal cell bodies in autonomic ganglia, DRGs and motoneurons in the transgenic mouse, and, in agreement, high levels of galanin in nerve terminals in peripheral tissues innervated by these ganglia. These fibers represent

either autonomic, sensory or motor nerve terminals, as evidenced by double-labeling with antibodies to, respectively, the cholinergic marker VAChT, the sensory marker CGRP, ␣-bungarotoxin, a marker for motor endplates, as well as NOS as a marker for parasympathetic autonomic neurons and sympathetic preganglionic neurons, and VIP for cholinergic sympathetic and parasympathetic neurons. HPLC chromatography shows identity between overexpressed and synthetic galanin in all tissues studied. The galanin expression patterns recorded in untreated GalOE animals represent at least four cases, (i)

70

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 7. (a– c) Galanin-LI (a, b) and mRNA (c) in spinal cord of young adult mice. Galanin-LI is seen in the gray matter, especially in the superficial layers of the dorsal horn (DH) (big arrowheads) in both WT (a) and GalOE (b) mouse, with a clearly higher intensity and density in the latter (cf. b with a). Note ir fibers in the dorsal columns (DC) (small arrowheads) and along the medial aspects of the ventral horn (VH) (short arrow). In the GalOE mouse galanin-LI (b) and mRNA (c) are seen in motor neurons (long arrows). Scale bar⫽50 ␮m (a⫽b⫽c).

overexpressed galanin in neurons with detectable galanin in the WT mouse; (ii) overexpressed galanin in neurons that have the capacity to synthesize galanin, but where galanin can only be detected after certain stimuli, e.g. nerve injury or colchicine/vinblastine treatment; (iii) overexpressed galanin in ‘non-galanin’ neurons, that is likely de novo (ectopic) expression. Galanin may also be expressed in neurons not overexpressing galanin (iv). Examples from the present study may, respectively, be (i) small DRG neurons; (ii) SCG and small DRG neurons, and motoneurons, both after nerve injury; and (iii) large, CGRP-negative DRG neurons, sympathetic cholinergic and parasympathetic neurons. No example for category (iv) can be found with certainty in this study, but in the brain several such systems seem to exist, e.g. in the hypothalamus (Kuteeva et al., 2004). In group (ii) WT neurons there may be an ongoing, low synthesis, often not detectable with our histochemical techniques, and this cannot be excluded, even for group III; that is, it is difficult to prove de novo expression. Injury adds a further degree of complexity, as shown in the quantitative evaluation (Table 2). For example, in young WT and GalOE DRGs the percentage positive NPs (⬃52 vs ⬃53%) and rODs (⬃61 vs. ⬃59) is similar after axotomy. There is even a decrease when comparing control GalOE with axotomized GalOE DRG neurons (from ⬃70 to ⬃53%). Also, there is no difference between rODs in control and axotomized OE DRGs (⬃64 vs. ⬃59). It may be speculated that galanin overexpressed under the PDGF-B promoter in DRGs is markedly down-regulated by nerve injury, that is the galanin in the ⬃53% GalOE DRG neurons could exclusively reflect upregulation of ‘WT’ galanin, and the ⬃70% in the control OE DRGs could then

theoretically be down to undetectable levels after axotomy. Interestingly, colchicine treatment, which like axotomy upregulates galanin synthesis, causes a decrease in cholecystokinin expression in rat cortex (Cortes et al., 1990). Mitosis inhibitors, such as colchicine and vinblastine, block axonal transport and induce a ‘chemical axotomy’ (Kashiba et al., 1992). Also age by itself plays a role in some systems. Thus, old WT and GalOE SCGs have more galanin than young ones (⬃59 vs. 4% and ⬃84 vs. 53%, respectively) which is not seen in DRGs. Moreover, preganglionic neurons at certain spinal levels appear to have higher galanin levels in old GalOE mice. In the case of the SCG of GalOE mice, however, there was an increase in rOD after axotomy (from ⬃75 to ⬃125), even if the percentage of galanin-positive NPs did not differ, suggesting that upregulation can occur under the PDGF-␤ promoter, thus possibly contrasting the situation in DRGs (see above). In terms of functionality, presence of any of the three types of galanin receptors, Gal-R1, -R2 and -R3 (Iismaa and Shine, 1999; Branchek et al., 2000), is critical. Since in most of these peripheral systems and spinal cord galanin is also present in the WT mouse, albeit in much lower amounts, and since in some of these systems it can be upregulated, e.g. after injury, it seems reasonable to assume that receptors often are present and can be activated by released, overexpressed and even ectopic galanin. In other systems, such as the sympathetic cholinergic neurons innervating sweat glands, receptors may not be present, and ectopically released galanin may remain without any effects.

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 8. (a– c) Immunofluorescence micrographs of ventral spinal cord of wild-type (a), GalOE (b, c) mice after incubation with galanin antiserum (a, b) and galanin antiserum preadsorbed with galanin (10⫺5M) (c). (a) In the WT ventral horn numerous galanin-IR nerve terminals are seen, but no positive structures in the white matter. (b, c) In the GalOE mouse there are, in addition, positive cell bodies in the ventral horn (thick arrows) and also numerous small fluorescent cell bodies in the white matter (thin arrows in b). None of these structures can be seen after incubation with control serum (c). Scale bars⫽100 ␮m (a⫽b⫽c).

Sympathetic ganglia The present results show a large increase in percentage and intensity of galanin expressing SCG neurons in young GalOE as compared with young WT mice, but there is no further increase in percentage of galanin-positive NPs in the GalOE SCG after nerve transection distal to SCG. In

71

the young WT mouse this procedure causes a marked upregulation, as it does in rat (Klimaschewski et al., 1994; Mohney et al., 1994; Schreiber et al., 1994; Zhang et al., 1994; Klimaschewski et al., 1995; Shadiack et al., 1995; Shadiack and Zigmond, 1998), and Shadiack et al. (1995) have shown that the axotomized neurons are the ones that upregulate galanin synthesis. It should be noted that the axotomy of the SCG was (in fact, has to be) made very close to the ganglion. It is therefore likely that a proportion of the SCG neurons dies. An interesting finding is the dense, patchy galaninpositive network of processes seen in young, and especially in aged GalOE SCGs. These fibers are presumably cholinergic and of preganglionic origin, since virtually all of them disappeared after decentralization and since they colocalized galanin and VAChT. There were also some galanin-positive, VAChT-negative fibers, probably representing axonal processes of the galanin-containing sympathetic neurons, although dendritic processes cannot be excluded. The possibility that the patches of galaninpositive terminals represent axonal sprouting was considered, since DRG neurons cultured from GAL knock-out mice have a slower and less branched neurite outgrowth than WT neurons (Holmes et al., 2000; see also Zigmond, 2001). However, the dense galanin-positive network may, instead, reflect ‘filling up’ of the preganglionic fiber network with galanin peptide, since the VAChT-positive terminals approximately have the same density in young and old GalOE and WT mice, and since a few galanin-positive cell bodies were observed in the sympathetic lateral cell column, however only in aged GalOE mice. This seems to be restricted to preganglionic neurons at certain spinal levels, since, for example, we could not detect any increase in preganglionic galanin fibers in the adrenal medulla when comparing WT and GalOE mice. Also, the increase in galanin levels in the SCG cell bodies of GalOE mouse is interesting in relation to the increases in galanin in the preganglionic fibers. It has been shown by Mohney and Zigmond (1998) that VIP can enhance the expression of VIP in SCG neurons. These authors raise the possibility that VIP present in preganglionic neurons can alter VIP expression in the postsynaptic SCG neurons in response to increased sympathetic activity. It remains to be studied to what extent the high galanin levels in neuronal cell bodies and terminals in SCGs of GalOE mice have functional consequences. Sensory ganglia and spinal cord Galanin has been extensively studied in DRGs, and is normally seen in a few small neurons in adult rat (Ch’ng et al., 1985; Skofitsch and Jacobowitz, 1985b) and mice (Corness et al., 1996; Shi et al., 2001). After nerve injury galanin is upregulated in CGRP-positive DRG neurons in both species (Villar et al., 1989; Zhang et al., 1993b; Corness et al., 1996; Shi et al., 2001). However, in the GalOE mouse galanin is present in the vast majority of neurons, including CGRP-negative ones, in agreement with a study of neonatal rat DRGs, where PDGF-B mRNA was detected in the majority of DRG neurons (Eccleston et

72

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 9. (a–f) Galanin- (a– c, e), and CGRP- (d, f) LIs in the hind paw of WT (a) and GalOE (b–f) mice, whereby e and f show the same section after double-staining (galanin⫹CGRP). (a) In WT mouse galanin-LI is only seen in a few fibers (arrow) in the epithelium (ep) and among sweat glands (sg). bv, blood vessel; c, cornified epithelium. Note strong signal in sg acini. This fluorescence is unspecific, that is does not disappear after incubation with preadsorbed galanin antiserum. (b) In the GalOE mouse, intensely stained galanin-IR fibers are seen both in epidermis close to and inside the epithelium (arrows), in and around sg and around bv. (c) Many galain-ir fibers (arrows) are seen around a bv and in a small nerve (n). (d) Also CGRP-ir fibers are abundant in the skin and seen inside (arrows) and below (arrowheads) the epithelium. (e, f) Many nerve terminals (arrows) around a bv colocalize galanin and CGRP. Scale bars⫽100 ␮m in a and b (a⫽b⫽d), 50 ␮m in c and 25 ␮m in e (e⫽f).

However, in the dorsal horn the galanin fibers in the GalOE mouse are still fairly sparse (Fig. 7b) considering the very high percentage of galanin neuronal cell bodies in the DRGs, including most large neuronal cell bodies (Fig. 6a– c). In contrast, the many large DRG neurons upregulating neuropeptide Y (NPY) synthesis after axotomy (Wakisaka et al., 1991), give rise to a much denser NPY fiber network in the deeper layers of the dorsal horn (Wakisaka et al., 1991; Zhang et al., 1993a) than seen for galanin in our GalOE mice. Galanin is expressed in motor neurons and in motor end-plates in the GalOE mouse, but not in WT mice or rats. However, after axotomy rat motoneurons are galanin-positive (Moore, 1989; Johnson et al., 1992; Zhang et al., 1993c).

al., 1993). This suggests some ectopic expression of galanin and involvement in sensory neurons with different modalities. The stereological analysis indicates that sciatic nerve transection causes a loss of DRG neurons, as shown previously (Shi et al., 2001), and that this loss is modestly, but still significantly, smaller in the GalOE mice. This is in agreement with a developmental/trophic role of galanin (Holmes et al., 2000; Wynick et al., 2001; Zigmond, 2001; Wynick and Bacon, 2002), probably related to GalR2 receptors (Mahoney et al., 2003), and that there is a 15% reduction in the number of DRG neurons in galanin knockout mice (Holmes et al., 2000). Galanin-LI in the spinal cord was present in the gray matter, as reported earlier in rat (Rökaeus et al., 1984; Ch’ng et al., 1985; Skofitsch and Jacobowitz, 1985a; Melander et al., 1986) and mouse (Shi et al., 2001), but with a clearly higher fiber density and staining intensity in GalOE as compared with WT mice, in agreement with the immunohistochemical and RIA analyses of the DRGs.

Sensory neurons and extravasation Extravasation due to increased vascular permeability, here induced by mustard oil and used as an assay of sensory

Table 3. Sudomotor function in WT and GalOE mice (SG, sweat gland) Time (min)

No of SGs (n ⫽ 20) SG diameter Mean and SD are shown

WT

GalOE

10

20

10

20

75⫾5 2.2⫾0.1

81⫾2 2.1⫾0.1

83⫾6 2.0⫾0.1

83⫾6 2.0⫾0.1

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Fig. 10. Effect of mustard oil on plasma extravasation using Evans Blue dye. There is a significant decrease in plasma extravasation in GalOE mice as compared with WT mice after mustard oil administration (filled bars). No differences between the two groups are seen on the contralateral side (open bars). The difference between groups is shown as mean⫾S.E.M. and is significant (P⬍0.05) (Student’s t-test for independent groups).

function, is an important component in cutaneous inflammation (Jancsó et al., 1967). Substance P released from sensory neurons is a major mediator of this response (Lembeck and Holzer, 1979; Lembeck et al., 1982; Lundberg et al., 1983), and can be inhibited by intradermal injection of galanin (Xu et al., 1991). In agreement, the present results show that extravasation is significantly reduced in the GalOE as compared with WT mouse. We propose that mustard oil induces release of both galanin and substance P from sensory nerve terminals of GalOE mice, and that the high galanin levels in the CGRP/substance P-positive, sensory fibers of the GalOE mouse, via a presynaptic action, inhibits release of substance P, thereby reducing extravasation. However, we cannot exclude involvement of galanin from other sources, e.g. the sympathetic nerves. Evidence for a presynaptic action of galanin has also been presented in studies both on the peripheral (Ekblad et al., 1985b; Giuliani et al., 1989) and central (Tsuda et al., 1989) nervous system, but the receptor(s) involved has not been defined. Recent evidence by Hua et al. (2004) suggests involvement of a presynaptic

Fig. 11. Pain response in the formalin test. Filled bars show WT, and open bars GalOE mice. Pain behavior was quantified as licking of the formalin-injected paw for 30 min. Phase I⫽0 –5 min, Interlude⫽5–15 min and Phase II⫽15–30 min after formalin. No difference between the two groups is seen in phase I, but GalOE mice show an increased licking both during interlude and phase II, as compared with WT. The difference between groups is shown as mean⫾S.E.M. and is significant (Mann-Whitney U test) during the interlude (P⬍0.05) and during Phase II (P⬍0.02).

73

Fig. 12. Force recovery after fatiguing stimulation is impaired in muscles of GalOE mice. Mean data (⫾S.E.M.) of the relative force produced during recovery of fast-twitch EDL muscles (A) and slow-twitch soleus muscles (B). Filled circles, WT mice; open circles, GalOE mice. In each muscle the relative force is expressed as percent of the force at the start of fatiguing stimulation. Time 0 min represents the end of fatiguing stimulation.

GalR1 receptor. In fact, DRG neurons express both Gal-R1 and -R2 receptor mRNA (Xu et al., 1996; Shi et al., 1997; O’Donnell et al., 1999; Kerekes et al., 2003). The possibility of a postsynaptic action of galanin causing vasoconstriction should, however, also be considered (Courtice et al., 1994). Sensory neurons and pain The formalin test used here monitors nociception as a consequence of tissue and nerve injury (Dubuisson and Dennis, 1977). The pain response to formalin injection normally is biphasic, as also observed here. However, there is a significant increase in paw licking in the GalOE as compared with WT mice in the second phase, indicating that overexpression of galanin in sensory fibers enhances nociceptive pain behavior. This is in agreement with the reduced pain threshold seen after intrathecal administration of ‘low’ concentrations of galanin in normal rats (Kerr et al., 2000b; Reeve et al., 2000; Liu et al., 2001), and that galanin knock-out mice show an increased pain threshold (Kerr et al., 2000a), and also with earlier studies showing excitatory effects of low doses of galanin given intrathecally onto the spinal cord (Wiesenfeld-Hallin et al., 1988, 1989; Kuraishi et al., 1991). This pronociceptive effect of galanin may in rat be mediated via GAL-R2 receptors, whereas the GAL-R1 receptor exerts an anti-nociceptive effect after nerve injury (Pooga et al., 1998; Liu et al., 2001; Wiesenfeld-Hallin and Xu, 2001; Liu and Hökfelt, 2002; Hua et al., 2004). The present GalOE mouse has been analyzed in other types of pain tests (hot plate and tail flick), and shown to exhibit a small but significantly elevated heat nociceptive threshold (Hygge-Blakeman et al., 2001), suggesting that in these particular tests galanin has a modest inhibitory role in pain modulation at the spinal cord level. Also in another GalOE mouse, based on a DBH promoter construct (Steiner et al., 2001), galanin appears to exert an inhibitory role at the spinal cord level (Hygge-Blakeman et al., 2004), and to reduce spinal cord sensitization to C-fiber stimulation (Grass et al., 2003). An inhibitory role is also reported in studies on two transgenic mice, which overexpress galanin in specific populations of DRG neurons,

74

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

either in inducible or constitutive manner (Bacon et al., 2002; Holmes et al., 2003). Thus, depending on pain model galanin can exert both anti- and pronociceptive effects, as discussed by Liu and Hökfelt (2002). Somatic motor neurons and muscle function In view of presence of galanin in the motor end-plate of GalOE mice, physiological tests were carried out on the EDL and soleus muscles, showing a slower force recovery as compared with WT mice (Westerblad et al., 2000). The longer contraction times seen in GalOE slow-twitch soleus muscle could be a consequence of alterations in either sacroplasmic reticulum Ca2⫹ handling or cross-bridge kinetics. These findings suggest that a galanin receptor demonstrated in skeletal muscle (Fathi et al., 1997; Wang et al., 1997) is functional and that skeletal muscle fibers in GalOE mice may cope less well with excessive muscular exercise. Cholinergic sympathetic neuron and sweat secretion Even if galanin in sweat glands was observed in virtually all sympathetic cholinergic fibers, identified by the cholinergic marker VAChT (Bejanin et al., 1994; Roghani et al., 1994; Schäfer et al., 1994), no certain difference in sweat secretion was seen between WT and GalOE mice. As discussed above, galanin receptors may not be present on these cholinergic, sympathetic fibers (presynaptic) or in the sweat gland cells (postsynaptic). Finally, also local VIP- and NOS-positive parasympathetic ganglion neuron in salivary glands express galanin in GalOE mice, and there is a corresponding galanin innervation of the secretory acini. Possible functional effects of this galanin overexpression have not been monitored. Concluding remarks Our study demonstrates galanin overexpression in several neuronal systems in the periphery and spinal cord. In most of these systems galanin is presumably synthesized also in the WT mouse, albeit at much lower levels. However, in large CGRP-negative DRG neurons and perhaps in sympathetic and parasympathetic cholinergic neurons presence of galanin may represent de novo (ectopic) expression. Age-dependent increases in galanin expression were seen in sympathetic, but not sensory, WT and GalOE neurons, and in preganglionic GalOE neurons. In WT mice axotomy caused a marked upregulation of galanin expression in SCGs and DRGs. Moreover, even if the percentage of NPs were similar in control axotomized SCGs, there was an increase in rOD suggesting increased galanin synthesis after nerve injury. The axotomy-induced DRG neuron loss in WT mice is modestly attenuated in GalOE mice, indicating a small rescuing effect of galanin, in agreement with the studies of Wynick and collaborators on galanin knockout mice (Wynick and Bacon, 2002). The GalOE mice also seemed to cope less well with excessive muscular exercise. In two functional tests (extravasation and pain) differences were found between WT and GalOE mice, suggesting that some of the overexpressed galanin stores can be

released and influence physiological processes, presumably via an action on receptors. As discussed, these results are in agreement with studies on WT mice in the literature. The lack of difference between GalOE and WT mice in sweat secretion may reflect absence of galanin receptors in sweat glands. Thus, to better understand these events, it will be necessary to define presence, localization, levels and type of galanin receptors in the GalOE mouse. Acknowledgments—This study was supported by Marianne and Marcus Wallenbergs’ Stiftelse, the Swedish MRC (04X-2887; K2001-33X-07464-16A), the Swedish Cancer Society, Stiftelsen Sigurd och Elsa Goljes Minne, Knut and Alice Wallenbergs Stiftelse, and a Bristol-Myers Squibb Unrestricted Neuroscience Grant. We would like to express our sincere thanks for the generous gift of the PBGF-B promoter plasmid by Prof. T Collins, Harvard Medical School, Boston, USA. For the generous donation of antisera we thank Prof. P. Emson, Babraham Institute, Babraham, Cambridge, UK (NOS), Prof. B. Meister, Karolinska Institutet, Stockholm, Sweden (VAChT) and Prof. J. Fahrenkrug, Bispebjerg’s Hospital, Copenhagen, Denmark (VIP).

REFERENCES Adams JC (1992) Biotin amplification of biotin and horseradish peroxidase signals in histochemical stains. J Histochem Cytochem 40:1457–1463. Arvidsson U, Riedl M, Elde R, Meister B (1997) Vesicular acetylcholine transporter (VAChT) protein: a novel and unique marker for cholinergic neurons in the central and peripheral nervous systems. J Comp Neurol 378:454 – 467. Bacon A, Holmes FE, Small CJ, Ghatei M, Mahoney S, Bloom S, Wynick D (2002) Transgenic over-expression of galanin in injured primary sensory neurons. NeuroReport 13:2129 –2132. Bartfai T, Hökfelt T, Langel U (1993) Galanin: a neuroendocrine peptide. Crit Rev Neurobiol 7:229 –274. Bauer FE, Hacker GW, Terenghi G, Adrian TE, Polak JM, Bloom SR (1986) Localization and molecular forms of galanin in human adrenals: elevated levels in pheochromocytomas. J Clin Endocrinol Metab 63:1372 Bejanin S, Cervini R, Mallet J, Berrard S (1994) A unique gene organization for two cholinergic markers, choline acetyltransferase and a putative vesicular transporter of acetylcholine. J Biol Chem 269:21944 –21947. Bersani M, Johnsen AH, Höjrup P, Dunning BE, Andreasen JJ, Holst JJ (1991) Human galanin: primary structure and identification of two molecular forms. FEBS Lett 283:189 –194. Branchek TA, Smith KE, Gerald C, Walker MW (2000) Galanin receptor subtypes. Trends Pharmacol Sci 21:109 –117. Brumovsky PR, Shi TJ, Matsuda H, Kopp J, Villar MJ, Hokfelt T (2002) NPY Y1 receptors are present in axonal processes of DRG neurons. Exp Neurol 174:1–10. Cai A, Hayes JD, Patel N, Hyde JF (1999) Targeted overexpression of galanin in lactotrophs of transgenic mice induces hyperprolactinemia and pituitary hyperplasia. Endocrinology 140:4955– 4964. Ch’ng JL, Christofides ND, Anand P, Gibson SJ, Allen YS, Su HC, Tatemoto K, Morrison JF, Polak JM, Bloom SR (1985) Distribution of galanin immunoreactivity in the central nervous system and the responses of galanin-containing neuronal pathways to injury. Neuroscience 16:343–354. Collins T, Ginsburg D, Boss JM, Orkin SH, Pober JS (1985) Cultured human endothelial cells express platelet-derived growth factor B chain: cDNA cloning and structural analysis. Nature 316:748 –750. Corness J, Shi TJ, Xu ZQ, Brulet P, Hökfelt T (1996) Influence of leukemia inhibitory factor on galanin/GMAP and neuropeptide Y

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77 expression in mouse primary sensory neurons after axotomy. Exp Brain Res 112:79 – 88. Cortes R, Ceccatelli S, Schalling M, Hökfelt T (1990) Differential effects of intracerebroventricular colchicine administration on the expression of mRNAs for neuropeptides and neurotransmitter enzymes, with special emphasis on galanin: an in situ hybridization study. Synapse 6:369 –391. Courtice GP, Hales JR, Potter EK (1994) Selective regional vasoconstriction underlying pressor effects of galanin in anaesthetized possums compared with cats. J Physiol 481(Pt 2):439 – 445. Crawley J, Mufson E, Hohmann J, Teklemichael D, Steiner R, Holmberg K, Xu Z, Blakeman K, Xu X, Wiesenfeld-Hallin Z, Bartfai T, Hökfelt T (2002) Galanin overexpressing transgenic mice. Neuropeptides 36:145 Crawley JN (1995) Biological actions of galanin. Regul Pept 59:1–16. Dubuisson D, Dennis SG (1977) The formalin test: a quantitative study of the analgesic effects of morphine, meperidine and brain stem stimulation in rats and cats. Pain 4:161–174. Eccleston PA, Funa K, Heldin CH (1993) Expression of platelet-derived growth factor (PDGF) and PDGF alpha- and beta-receptors in the peripheral nervous system: an analysis of sciatic nerve and dorsal root ganglia. Dev Biol 155:459 – 470. Ekblad E, Håkanson R, Rökaeus Å, Sundler F (1985a) Galanin fibers in the rat gut: distribution, origin and projections. Neuroscience 16:355–363. Ekblad E, Håkanson R, Sundler F, Wahlestedt C (1985b) Galanin: neuromodulatory and direct contractile effects on smooth muscle preparations. Br J Pharmacol 86:241–246. Evans HF, Shine J (1991) Human galanin: molecular cloning reveals a unique structure. Endocrinology 129:1682–1684. Evans PD, Cook SN, Riggs PD, Noren CJ (1995) LITMUS: multipurpose cloning vectors with a novel system for bidirectional in vitro transcription. Biotechniques 19:130 –135. Fathi Z, Cunningham AM, Iben LG, Battaglino PB, Ward SA, Nichol KA, Pine KA, Wang J, Goldstein ME, Iismaa TP, Zimanyi IA (1997) Cloning, pharmacological characterization and distribution of a novel galanin receptor. Mol Brain Res 51:49 –59. Furness JB, Costa M, Rökaeus A, McDonald TJ, Brooks B (1987) Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell Tissue Res 250:607– 615. Fuxe K, Jansson A, Diaz-Cabiale Z, Andersson A, Tinner B, Finnman UB, Misane I, Razani H, Wang FH, Agnati LF, Ögren SO (1998) Galanin modulates 5-hydroxytryptamine functions. Focus on galanin and galanin fragment/5-hydroxytryptamine1A receptor interactions in the brain. Ann N Y Acad Sci 863:274 –290. Games D, Adams D, Alessandrini R, Barbour R, Berthelette P, Blackwell C, Carr T, Clemens J, Donaldson T, Gillespie F, et al (1995) Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 373:523–527. Giuliani S, Amann R, Papini AM, Maggi CA, Meli A (1989) Modulatory action of galanin on responses due to antidromic activation of peripheral terminals of capsaicin-sensitive sensory nerves. Eur J Pharmacol 163:91–96. Grass S, Crawley JN, Xu XJ, Wiesenfeld-Hallin Z (2003) Reduced spinal cord sensitization to C-fibre stimulation in mice over-expressing galanin. Eur J Neurosci 17:1829 –1832. Gundersen HJG, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N, Möller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sörensen FB, Vesterby A, West MJ (1988a) The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS 96:857– 881. Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N, Möller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sörensen FB, Vesterby A, West MJ (1988b) Some new simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS 96:379 –394.

75

Gundlach AL (2002) Galanin/GALP and galanin receptors: role in central control of feeding, body weight/obesity and reproduction? Eur J Pharmacol 440:255–268. Herbison AE, Simonian SX, Norris PJ, Emson PC (1996) Relationship of neuronal nitric oxide synthase immunoreactivity to GnRH neurons in the ovariectomized and intact female rat. J Neuroendocrinol 8:73– 82. Hogan B, Costanini F, Lacy E (1986) Manipulating the mouse embryo: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Hökfelt T, Bartfai T, Crawley J (eds.) (1998) Galanin: basic research discoveries and therapeutic implications. Ann NY Acad Sci, Vol. 863. New York: NYAS. Hökfelt T, Bartfai T, Jacobowitz D, Ottoson D (eds.) (1991) Galanin: a new multifunctional peptide in the neuro-endocrine system. Wenner-Gren Center International Symposium Series, Vol. 58. London: MacMillan. Hökfelt T, Wiesenfeld-Hallin Z, Villar M, Melander T (1987) Increase of galanin-like immunoreactivity in rat dorsal root ganglion cells after peripheral axotomy. Neurosci Lett 83:217–220. Holmberg K, Kahl U, Lendahl U, Kokaia Z, Nanobashvili A, Lindvall O, Ekström P, Bartfai T, Hökfelt T (2000) Creation and characterization of a mouse over-expressing galanin. Abstr Soc Neurosci 26:27 Holmes FE, Bacon A, Pope RJ, Vanderplank PA, Kerr NC, Sukumaran M, Pachnis V, Wynick D (2003) Transgenic overexpression of galanin in the dorsal root ganglia modulates pain-related behavior. Proc Natl Acad Sci USA 100:6180 – 6185. Holmes FE, Mahoney S, King VR, Bacon A, Kerr NC, Pachnis V, Curtis R, Priestley JV, Wynick D (2000) Targeted disruption of the galanin gene reduces the number of sensory neurons and their regenerative capacity. Proc Natl Acad Sci USA 97:11563–11568. Hua XY, Hayes CS, Hofer A, Fitzsimmons B, Kille K, Langel Ü, Bartfai T, Yaksh TL (2004) Galanin acts at GalR1 receptors in spinal antinociception synergy with morphine and AP-5. J Pharmacol Exp Ther 308:574 –582. Hygge-Blakeman K, Brumovsky P, Hao JX, Xu XJ, Hökfelt T, Crawley JN, Wiesenfeld-Hallin Z (2004) Galanin over-expression decreases the development of neuropathic pain-like behaviors in mice after partial sciatic nerve injury. Brain Res 1025:152–158. Hygge-Blakeman K, Holmberg K, Hao JX, Xu XJ, Kahl U, Lendahl U, Bartfai T, Wiesenfeld-Hallin Z, Hökfelt T (2001) Mice over-expressing galanin have elevated heat nociceptive threshold. NeuroReport 12:423– 425. Iismaa TP, Shine J (1999) Galanin and galanin receptors. Results Probl Cell Differ 26:257–291. Jancsó N, Jancsó-Gábor A, Szolcsányi J (1967) Direct evidence for neurogenic inflammation and its prevention by denervation and by pretreatment with capsaicin. Br J Pharmacol 31:138 –151. Johansson C, Lunde PK, Gothe S, Lännergren J, Westerblad H (2003) Isometric force and endurance in skeletal muscle of mice devoid of all known thyroid hormone receptors. J Physiol 547:789 –796. Johnson DG, Nogueria Araujo GM (1981) A simple method of reducing the fading of immunofluorescence during microscopy. J Immunol Methods 43:349 –350. Johnson H, Hökfelt T, Ulfhake B (1992) Galanin- and CGRP-like immunoreactivity coexist in rat spinal motoneurons. NeuroReport 3:303–306. Kalra SP, Kalra PS (1996) Nutritional infertility: the role of the interconnected hypothalamic neuropeptide Y-galanin-opioid network. Front Neuroendocrinol 17:371– 401. Kaplan LM, Spindel ER, Isselbacher KJ, Chin WW (1988) Tissuespecific expression of the rat galanin gene. Proc Natl Acad Sci USA 85:1065–1069. Kashiba H, Senba E, Kawai Y, Ueda Y, Tohyama M (1992) Axonal blockade induces the expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons. Brain Res 577:19 –28.

76

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77

Kennedy WR, Sakuta M (1984) Collateral reinnervation of sweat glands. Ann Neurol 15:73–78. Kerekes N, Mennicken F, O’Donnell D, Hökfelt T, Hill RH (2003) Galanin increases membrane excitability and enhances Ca2⫹ currents in adult, acutely dissociated dorsal root ganglion neurons. Eur J Neurosci 18:2957–2966. Kerr BJ, Cafferty WB, Gupta YK, Bacon A, Wynick D, McMahon SB, Thompson SW (2000a) Galanin knockout mice reveal nociceptive deficits following peripheral nerve injury. Eur J Neurosci 12:793– 802. Kerr BJ, Wynick D, Thompson SW, McMahon SB (2000b) The biological role of galanin in normal and neuropathic states. Prog Brain Res 129:219 –230. Klimaschewski L, Grohmann I, Heym C (1996) Target-dependent plasticity of galanin and vasoactive intestinal peptide in the rat superior cervical ganglion after nerve lesion and re-innervation. Neuroscience 72:265–272. Klimaschewski L, Tran TD, Nobiling R, Heym C (1994) Plasticity of postganglionic sympathetic neurons in the rat superior cervical ganglion after axotomy. Microsc Res Tech 29:120 –130. Klimaschewski L, Unsicker K, Heym C (1995) Vasoactive intestinal peptide but not galanin promotes survival of neonatal rat sympathetic neurons and neurite outgrowth of PC12 cells. Neurosci Lett 195:133–136. Kuraishi Y, Kawamura M, Yamaguchi T, Houtani T, Kawabata S, Futaki S, Fujii N, Satoh M (1991) Intrathecal injections of galanin and its antiserum affect nociceptive response of rat to mechanical, but not thermal, stimuli. Pain 44:321–324. Kuteeva E, Calza L, Holmberg K, Theodorsson E, Ögren SO, Hökfelt T (2004) Distribution of galanin and galanin transcript in the brain of a galanin-overexpressing transgenic mouse. J Chem Neuroanat 28:185–216. Langel U, Bartfai T (1998) Chemistry and molecular biology of galanin receptor ligands. Ann N Y Acad Sci 863:86 –93. Leibowitz SF (1995) Brain peptides and obesity: pharmacologic treatment. Obes Res 3(Suppl 4):573S–589S. Lembeck F, Donnerer J, Bartho L (1982) Inhibition of neurogenic vasodilation and plasma extravasation by substance P antagonists, somatostatin and [D-Met2, Pro5]enkephalineamide Eur J Pharmacol 85:171–176. Lembeck F, Holzer P (1979) Substance P as neurogenic mediator of antidromic vasodilation and neurogenic plasma extravasation. Arch Pharmacol 310:175–183. Liu HX, Brumovsky P, Schmidt R, Brown W, Payza K, Hodzic L, Pou C, Godbout C, Hökfelt T (2001) Receptor subtype-specific pronociceptive and analgesic actions of galanin in the spinal cord: selective actions via GalR1 and GalR2 receptors. Proc Natl Acad Sci USA 98:9960 –9964. Liu HX, Hökfelt T (2002) The participation of galanin in pain processing at the spinal level. Trends Pharmacol Sci 23:468 – 474. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurements with the Folin phenol reagent. J Biol Chem 193:265–275. Lundberg JM, Hökfelt T, Hemsen A, Theodorsson-Norheim E, Pernow J, Hamberger B, Goldstein M (1986) Neuropeptide Y-like immunoreactivity in adrenaline cells of adrenal medulla and in tumors and plasma of pheochromocytoma patients. Regul Pept 13:169 –182. Lundberg JM, Hökfelt T, Schultzberg M, Uvnäs-Wallensten K, Köhler C, Said SI (1979) Occurrence of vasoactive intestinal polypeptide (VIP)-like immunoreactivity in certain cholinergic neurons of the cat: evidence from combined immunohistochemistry and acetylcholinesterase staining. Neuroscience 4:1539 –1559. Lundberg JM, Saria A, Brodin E, Rosell S, Folkers K (1983) A substance P antagonist inhibits vagally induced increase in vascular permeability and bronchial smooth muscle contraction in the guinea pig. Proc Natl Acad Sci USA 80:1120 –1124. Mahoney SA, Hosking R, Farrant S, Holmes FE, Jacoby AS, Shine J, Iismaa TP, Scott MK, Schmidt R, Wynick D (2003) The second

galanin receptor GalR2 plays a key role in neurite outgrowth from adult sensory neurons. J Neurosci 23:416 – 421. McKnight GL, Karlsen AE, Kowalyk S, Mathewes SL, Sheppard PO, O’Hara PJ, Taborsky GJ Jr (1992) Sequence of human galanin and its inhibition of glucose-stimulated insulin secretion from RIN cells. Diabetes 41:82– 87. Melander T, Hökfelt T, Rökaeus A, Fahrenkrug J, Tatemoto K, Mutt V (1985) Distribution of galanin-like immunoreactivity in the gastrointestinal tract of several mammalian species. Cell Tissue Res 239:253–270. Melander T, Hökfelt T, Rökaeus Å (1986) Distribution of galanin like immunoreactivity in the rat central nervous system. J Comp Neurol 248:475–517. Merchenthaler I, Lopez FJ, Negro-Vilar A (1993) Anatomy and physiology of central galanin-containing pathways. Prog Neurobiol 40:711–769. Mohney RP, Siegel RE, Zigmond RE (1994) Galanin and vasoactive intestinal peptide messenger RNA expression increase following axotomy of the adult rat SCG. J Neurobiol 25:108 –118. Mohney RP, Zigmond RE (1998) Vasoactive intestinal peptide enhances its own expression in sympathetic neurons after injury. J Neurosci 18:5285–5293. Moore RY (1989) Cranial motor neurons contain either galanin- or calcitonin gene- related peptidelike immunoreactivity. J Comp Neurol 282:512–522. O’Donnell D, Ahmad S, Wahlestedt C, Walker P (1999) Expression of the novel galanin receptor subtype GALR2 in the adult rat CNS: distinct distribution from GALR1. J Comp Neurol 409:469 – 481. Pease PC (1962) Buffered formaldehyde as a killing agent and primary fixative for electron microscopy. Anat Rec 142:342. Pelto-Huikko M (1989) Immunocytochemical localization of neuropeptides in the adrenal medulla. J Electron Microsc Tech 12:364 –379. Platt JL, Michael AF (1983) Retardation of fading and enhancement of intensity of immunofluorescence by p-phenylenediamine. J Histochem Cytochem 31:840 – 842. Pooga M, Soomets U, Hallbrink M, Valkna A, Saar K, Rezaei K, Kahl U, Hao JX, Xu XJ, Wiesenfeld-Hallin Z, Hökfelt T, Bartfai T, Langel ¨ (1998) Cell penetrating PNA constructs regulate galanin receptor U levels and modify pain transmission in vivo. Nat Biotechnol 16:857– 861. Rao MS, Sun Y, Escary JL, Perreau J, Tresser S, Patterson PH, Zigmond RE, Brulet P, Landis SC (1993) Leukemia inhibitory factor mediates an injury response but not a target-directed developmental transmitter switch in sympathetic neurons. Neuron 11: 1175–1185. Reeve AJ, Walker K, Urban L, Fox A (2000) Excitatory effects of galanin in the spinal cord of intact, anaesthetized rats. Neurosci Lett 295:25–28. Roghani A, Feldman J, Kohan SA, Shirzadi A, Gundersen CB, Brecha N, Edwards RH (1994) Molecular cloning of a putative vesicular transporter for acetylcholine. Proc Natl Acad Sci USA 91:10620 –10624. Rökaeus A, Melander T, Hökfelt T, Lundberg JM, Tatemoto K, Carlquist M, Mutt V (1984) A galanin-like peptide in the central nervous system and intestine of the rat. Neurosci Lett 47:161–166. Rökaeus Å, Brownstein MJ (1986) Construction of a porcine adrenal medullary cDNA library and nucleotide sequence analysis of two clones encoding a galanin precursor. Proc Natl Acad Sci USA 83:6287– 6291. Rökaeus Å, Pruss RM, Eiden LE (1990) Galanin gene expression in chromaffin cells is controlled by calcium and protein kinase signaling pathways. Endocrinology 127:3096 –3102. Sasahara M, Fries JWU, Raines EW, Gown AM, Westrum LE, Frosch MP, Bonthrin DT, Ross R, Collins T (1991) PDGFB-chain in neurons of the central nervous system, posterior pituitary, and in a transgenic model. Cell 64:217–227. Schäfer MK, Weihe E, Varoqui H, Eiden LE, Erickson JD (1994) Distribution of the vesicular acetylcholine transporter (VAChT) in

K. Holmberg et al. / Neuroscience 133 (2005) 59 –77 the central and peripheral nervous systems of the rat. J Mol Neurosci 5:1–26. Schmidt VE, Kratzin H, Eckart K, Drevs D, Mundkowski G, Clemens A, Katsoulis S, Schäfer H, Gallwitz B, Creutzfeld W (1991) Isolation and primary structure of pituitary human galanin, a 30-residue nonamidated neuropeptide. Proc Natl Acad Sci USA 88:11435–11439. Schreiber RC, Hyatt-Sachs H, Bennett TA, Zigmond RE (1994) Galanin expression increases in adult rat sympathetic neurons after axotomy. Neuroscience 60:17–27. Shadiack AM, Vaccariello SA, Zigmond RE (1995) Galanin expression in sympathetic ganglia after partial axotomy is highly localized to those neurons that are axotomized. Neuroscience 65:1119 –1127. Shadiack AM, Zigmond RE (1998) Galanin induced in sympathetic neurons after axotomy is anterogradely transported toward regenerating nerve endings. Neuropeptides 32:257–264. Shi TJ, Tandrup T, Bergman E, Xu ZQ, Ulfhake B, Hökfelt T (2001) Effect of peripheral nerve injury on dorsal root ganglion neurons in the C57 BL/6J mouse: marked changes both in cell numbers and neuropeptide expression. Neuroscience 105:249 –263. Shi TJS, Zhang X, Holmberg K, Xu ZQ, Hökfelt T (1997) Expression and regulation of galanin-R2 receptors in rat primary sensory neurons: effect of axotomy and inflammation. Neurosci Lett 237:57– 60. Skofitsch G, Jacobowitz DM (1985a) Galanin-like immunoreactivity in capsaicin sensitive sensory neurons and ganglia. Brain Res Bull 15:191–195. Skofitsch G, Jacobowitz DM (1985b) Immunohistochemical mapping of galanin-like neurons in the rat central nervous system. Peptides 6:509 –546. Skofitsch G, Jacobowitz DM (1986) Quantitative distribution of galanin-like immunoreactivity in the rat central nervous system. Peptides 7:609 – 613. Steiner RA, Hohmann JG, Holmes A, Wrenn CC, Cadd G, Jureus A, Clifton DK, Luo M, Gutshall M, Ma SY, Mufson EJ, Crawley JN (2001) Galanin transgenic mice display cognitive and neurochemical deficits characteristic of Alzheimer’s disease. Proc Natl Acad Sci USA 98:4184 – 4189. Strömberg I, Björklund H, Melander T, Rökaeus A, Hökfelt T, Olson L (1987) Galanin-immunoreactive nerves in the rat iris: alterations induced by denervations. Cell Tissue Res 250:267–275. Tandrup T (1993) A method for unbiased and efficient estimation of number and mean volume of specified neuron subtypes in rat dorsal root ganglion. J Comp Neurol 329:269 –276. Tatemoto K, Rökaeus Å, Jörnvall H, McDonald TJ, Mutt V (1983) Galanin: a novel biologically active peptide from porcine intestine. FEBS Lett 164:124 –128. Theodorsson E, Rugarn O (2000) Radioimmunoassay for rat galanin: immunochemical and chromatographic characterization of immunoreactivity in tissue extracts. Scand J Clin Lab Invest 60:411– 418. Tsuda K, Yokoo H, Goldstein M (1989) Neuropeptide Y and galanin in norepinephrine release in hypothalamic slices. Hypertension 14:81– 86. Ubink R, Calza L, Hökfelt T (2003) ‘Neuro’-peptides in glia: focus on NPY and galanin. Trends Neurosci 26:604 – 609. Villar MJ, Cortés R, Theodorsson E, Wiesenfeld-Hallin Z, Schalling M, Fahrenkrug J, Emson PC, Hökfelt T (1989) Neuropeptide expression in rat dorsal root ganglion and spinal cord after peripheral nerve injury with special reference to galanin. Neuroscience 33:587– 604. Vrontakis ME (2002) Galanin: a biologically active peptide. Curr Drug Targets 1:297–317.

77

Vrontakis ME, Peden LM, Duckworth ML, Friesen HG (1987) Isolation and characterization of a complementary DNA (galanin) clone from estrogen-induced pituitary tumor messenger RNA. J Biol Chem 262:16755–16758. Wakisaka S, Kajander KC, Bennett GJ (1991) Increased neuropeptide Y (NPY)-like immunoreactivity in rat sensory neurons following peripheral axotomy. Neurosci Lett 124:200 –203. Wang S, Hashemi T, He C, Strader C, Bayne M (1997) Molecular cloning and pharmacological characterization of a new galanin receptor subtype. Mol Pharmacol 52:337–343. Westerblad H, Bruton JD, Allen DG, Lännergren J (2000) Functional significance of Ca2⫹ in long-lasting fatigue of skeletal muscle. Eur J Appl Physiol 83:166 –174. Wiesenfeld-Hallin Z, Villar MJ, Hökfelt T (1988) Intrathecal galanin at low doses increases spinal reflex excitability in rats more to thermal than mechanical stimuli. Exp Brain Res 71:663– 666. Wiesenfeld-Hallin Z, Xu XJ (2001) Neuropeptides in neuropathic and inflammatory pain with special emphasis on cholecystokinin and galanin. Eur J Pharmacol 429:49 –59. Wiesenfeld-Hallin Z, Xu XJ, Villar MJ, Hökfelt T (1989) The effect of intrathecal galanin on the flexor reflex in rat: increased depression after sciatic nerve section. Neurosci Lett 105:149 –154. Wynick D, Bacon A (2002) Targeted disruption of galanin: new insights from knock-out studies. Neuropeptides 36:132–144. Wynick D, Thompson SW, McMahon SB (2001) The role of galanin as a multi-functional neuropeptide in the nervous system. Curr Opin Pharmacol 1:73–77. Xu XJ, Hao JX, Wiesenfeld-Hallin Z, Håkanson R, Folkers K, Hökfelt T (1991) Spantide II, a novel tachykinin antagonist, and galanin inhibit plasma extravasation induced by antidromic C-fiber stimulation in rat hindpaw. Neuroscience 42:731–737. Xu XJ, Hökfelt T, Bartfai T, Wiesenfeld-Hallin Z (2000) Galanin and spinal nociceptive mechanisms: recent advances and therapeutic implications. Neuropeptides 34:137–147. Xu Z-Q, Shi T-J, Landry M, Hökfelt T (1996) Evidence for galanin receptors in primary sensory neurons and effect of axotomy and inflammation. NeuroReport 8:237–242. Zamboni I, De Martino C (1967) Buffered picric acid formaldehyde. A new rapid fixative for electron microscopy. J Cell Biol 35:148A. Zentel HJ, Nohr D, Albrecht R, Jeurissen SH, Vainio O, Weihe E (1991) Peptidergic innervation of the bursa fabricii: interrelation with T-lymphocyte subsets. Int J Neurosci 59:177–188. Zhang X, Dagerlind Å, Bao L, Ji RR, Lundberg M, Hökfelt T (1994) Increased expression of galanin in the rat superior cervical ganglion after pre- and postganglionic nerve lesions. Exp Neurol 127:9–22. Zhang X, Meister B, Elde R, Verge VM, Hökfelt T (1993a) Large calibre primary afferent neurons projecting to the gracile nucleus express neuropeptide Y after sciatic nerve lesions: an immunohistochemical and in situ hybridization study in rats. Eur J Neurosci 5:1510 –1519. Zhang X, Nicholas AP, Hökfelt T (1993b) Ultrastructural studies on peptides in the dorsal horn of the spinal cord: I. Co-existence of galanin with other peptides in primary afferents in normal rats. Neuroscience 57:365–384. Zhang X, Verge VM, Wiesenfeld-Hallin Z, Piehl F, Hökfelt T (1993c) Expression of neuropeptides and neuropeptide mRNAs in spinal cord after axotomy in the rat, with special reference to motoneurons and galanin. Exp Brain Res 93:450 – 461. Zigmond RE (2001) Can galanin also be considered as growth-associated protein 3.2? Trends Neurosci 24:494 – 496. Zigmond RE, Hyatt-Sachs H, Mohney RP, Schreiber RC, Shadiack AM, Sun Y, Vaccariello SA (1996) Changes in neuropeptide phenotype after axotomy of adult peripheral neurons and the role of leukemia inhibitory factor. Perspect Dev Neurobiol 4:75–90.

(Accepted 26 January 2005) (Available online 20 April 2005)