Galanin-Like Peptides Exert Potent Vasoactive Functions in Vivo Sabine M

ORIGINAL ARTICLE

Galanin-Like Peptides Exert Potent Vasoactive Functions In Vivo Sabine M. Schmidhuber1, Radmila Santic1, Christina W. Tam3, Johann W. Bauer2, Barbara Kofler1 and Susan D. Brain3

The cutaneous vasculature plays a key role in the pathophysiology of inflammatory skin diseases. The vascular activity is under the control of the peripheral nervous system that includes locally released neuropeptides. Recently, we detected receptors for the neuropeptide galanin in association with dermal blood vessels, suggesting a role of the galanin-peptide-family in the regulation of the cutaneous microvasculature. Therefore, we have investigated galanin and galanin-like peptide (GALP), a neuropeptide previously only considered to be involved in metabolism and reproduction in the central nervous system, for vaso-modulatory activity in the murine skin in vivo. Picomole amounts of intradermally injected galanin and GALP decreased cutaneous blood flow and inhibited inflammatory edema formation. Both the full-length GALP (1–60) and the putative smaller proteolytic fragment GALP (3–32) showed similar effects. These activities are most likely mediated by galanin receptors galanin receptor subtype 2 (GalR2) and/or galanin receptor subtype 3 (GalR3), because reverse transcription-PCR analysis of murine skin revealed messenger RNA (mRNA) expression of GalR2 and GalR3 but not of galanin receptor subtype 1. The lack of galanin receptor mRNAs in endothelial and smooth muscle cells indicates a neuronal localization of these receptors around the vessels. These results indicate functional activity of GALP in the periphery in vivo and suggest a potential role as an inflammatory modulator. Journal of Investigative Dermatology (2007) 127, 716–721. doi:10.1038/sj.jid.5700569; published online 5 October 2006

INTRODUCTION mitogenic properties (Sethi et al., 1992; Wynick et al., 1993), The skin has an essential protective function in responding to and analgesic effects in response to nerve injury (Wiesenfeld- challenges from the environment. Cutaneous nerves transmit Hallin et al., 1992). nociceptive information to the central nervous system and A second peptide of the galanin family, galanin-like release biologically active neuropeptides (McDonald et al., peptide (GALP) shares the amino acids at position 9–21 with 1996). The efferent activities of sensory nerves, mediated via the first 13 amino acids of galanin, which are required to neuropeptides such as substance P (SP) and calcitonin gene- activate galanin receptors. The sequence of GALP (1–60) related peptide (CGRP), may contribute to the physiological contains a potential proteolytic cleavage site between two and pathophysiological modulation of skin responses basic amino acids at position 33, which might lead to shorter (Hokfelt and Ljungdahl, 1971; Brain, 1997). C-terminally amidated peptides. In addition, the first two Galanin is a 29 (30 human) amino-acid peptide involved amino acids of human GALP could be potentially removed in a variety of peripheral and central physiological and by dipeptidyl dipeptidase IV (Lang et al., 2005). Like galanin, pathophysiological processes (Bartfai et al., 1993), including GALP has potent species-specific and time-dependent effects secretion of hormones (McDonald et al., 1985; Bauer et al., in the central nervous system and has been suggested to 1986), inhibition of cardiac vagal action (Ulman et al., 1994), constitute a link between metabolism and reproduction (Gottsch et al., 2004). To our knowledge, GALP has not been shown to possess biological activity in the periphery. 1Department of Pediatrics, Paracelsus Private Medical University Salzburg, The effects of galanin and GALP are mediated by G-protein- 2 Salzburg, Austria; Department of Dermatology, Paracelsus Private Medical coupled receptor subtypes, which are uniquely distributed University Salzburg, Salzburg, Austria and 3Cardiovascular Division, Kings College London, New Hunt’s House, Guy’s Campus, London, UK throughout the central nervous system and periphery (Burgevin Correspondence: Dr Barbara Kofler, Department of Pediatrics, Paracelsus et al., 1995; Fathi et al., 1998; Kolakowski et al., 1998). Private Medical University Salzburg, Muellner Hauptstr. 48, Salzburg A-5020, Galanin has high affinity for all three galanin receptor sub- Austria. E-mail: [email protected] types, whereas GALP (1–60) displays high affinity only for Abbreviations: CGRP, calcitonin gene-related peptide; GALP, galanin-like galanin receptor subtype 2 (GalR2) and galanin receptor peptide; GalR2, galanin receptor subtype 2; GalR3, galanin receptor subtype subtype 3 (GalR3) (Ohtaki et al., 1999; Berger et al., 2004). 3; i.d., intradermally; mRNA, messenger RNA; RT, reverse transcription; SP, substance P Recently, it was shown that GALP (3–32) is a more potent Received 20 December 2005; revised 25 July 2006; accepted 30 July 2006; agonist on the GalR2 than the full-length peptide in neuro- published online 5 October 2006 blastoma cells expressing the GalR2 (Lang et al., 2005).

716 Journal of Investigative Dermatology (2007), Volume 127 & 2006 The Society for Investigative Dermatology SM Schmidhuber et al. GALP in Cutaneous Microvasculature

The upregulation of galanin and galanin-binding sites in a 250 rat skin upon inflammation indicates a role of the galanin Galanin peptides system in the plasticity of the cutaneous microvasculature l/g) 200 + Vehicle (Ji et al., 1995). Accordingly, galanin inhibits plasma * + Salbutamol 100 nmol extravasation induced by antidromic C-fiber stimulation in 150 + Endothelin-110 pmol the rat hind paw (Xu et al., 1991) and histamine-induced + Galanin 1 pmol 100 ** plasma extravasation in the pigeon skin (Jancso et al., 2000). + GALP (1−60) 10 pmol ** ** − Furthermore, galanin-overexpressing mice showed a signifi- 50 ** ** + GALP (1 32) 10 pmol cant decrease in plasma extravasation upon activation of + GALP (3 − 32) 10 pmol Plasma extravasation ( Plasma extravasation neurogenic inflammation following mustard oil treatment 0 + GALP (19 −37) 10 pmol Tyrode (Holmberg et al., 2005). Consistent with these vascular SP + CGRP effects, we were able to demonstrate that galanin-binding b sites are present around blood vessels (Kofler et al., 2004). 300

Although there is evidence for a role of galanin in l/g) + Galanin 3 pmol modulating inflammatory edema in the periphery, the 200 possible functional significance of GALP in the periphery has not been investigated. Moreover, the mechanism of ** edema formation and the galanin receptor subtype involved, 100 are so far unclear. The purpose of this study was to analyze the effects of galanin and GALP on cutaneous microvascu- 0 lature in mice. We have examined the ability of galanin and ( Plasma extravasation Tyrode GALP to modulate inflammatory edema formation and blood Histamine + CGRP flow in the murine skin in vivo. We have also examined the Figure 1. Effects on inflammatory edema of the galanin-peptide-family. skin for the presence of galanin receptors, that both galanin (a) Edema formation was induced by i.d. injection of SP (300 pmol) and and GALP can mediate their effects through. CGRP (10 pmol), and the effect of coinjected salbutamol, endothelin-1 (n ¼ 4), and galanin peptides (n ¼ 6) is shown. Results are expressed as plasma 7 125 RESULTS extravasation (ml/g), mean SEM measured by the I-BSA method. Responses that are significantly different from the corresponding SP þ CGRP- Effects of galanin and GALP peptides on inflammatory edema treated sites are shown (*Po0.05, **Po0.01; Dunnett’s). (b) Edema formation formation was induced by histamine (3 nmol) and CGRP (10 pmol), and the effect of Plasma extravasation was induced in skin by the coinjection coinjected galanin (3 pmol) is shown. Results are expressed as plasma of the potent mediator of increased microvascular perme- extravasation (ml/g), mean7SEM (n ¼ 4) measured by the 125I-BSA method. ability SP and the vasodilator CGRP, which act in a Responses that are significantly different from the corresponding synergistic manner to induce an acute inflammatory histamine þ CGRP-treated sites (**Po0.01; Dunnett’s). edema in skin (Brain and Williams, 1985; Cao et al., 1999) (Figure 1a). The b-adrenergic anti-permeability agent salbutamol (Teixeira et al., 1995) was used (Figure 1a), and the potent vasoconstrictor endothelin (Brain et al., 1989) as 100 ** positive controls for the assay. It can be seen that galanin, at a ** ** 1 pmol dose, acted in a similar manner to endothelin to 75 abolish this edema. The effect of the full-length mature GALP ** ** (1–60), the truncated peptides (1–32) and (3–32), as well as the 50 peptide fragment GALP (19–37), which lacks the galanin receptor-binding domain, were also examined at a dose of 25 10 pmol/site for their ability to modulate edema formation

% Inhibition of SP+CGRP 0 (Figure 1a). Although GALP (1–60) and GALP (3–32) demon- 0.1 0.3 1 3 10 strated an inhibitory effect, GALP (1–32) exhibited a lesser but Galanin (pmol) albeit significant inhibitory response and GALP (19–37) was Figure 2. The dose-related effect of galanin on edema formation induced by inactive (Figure 1a). i.d. injected SP (300 pmol) with CGRP (10 pmol). The response to increasing Further studies were then carried out to learn the dose- doses of galanin (0.1–10pmol) coinjected with SP þ CGRP is shown as dependent activities. Results for galanin are shown in % inhibition (mean7SEM.) of edema formation measured by the 125I-BSA Figure 2, where galanin (0.1–10pmol) acted in a dose- method. (n ¼ 10). Responses that are significantly different from the dependent manner to inhibit edema formation, although corresponding SP þ CGRP-treated sites are shown (**Po0.01; Dunnett’s). higher doses had less effect, giving a bell-shaped dose– response curve. The ability of galanin to inhibit non- neurogenic edema formation was investigated through the The dose-related effect of the two GALP peptides that study of histamine coinjected with CGRP-induced responses. showed potent anti-edema activity, GALP (1–60) (1–100 pmol) Figure 1b shows that galanin also inhibited histamine- and and GALP (3–32) (1–100 pmol) is shown in Figure 3. Again CGRP-induced edema formation. bell-shaped dose–response curves were observed.

www.jidonline.org 717 SM Schmidhuber et al. GALP in Cutaneous Microvasculature

ab a 50 100 100 ** 40 75 75 ** ** ** 30 ** ** ** * 50 50 ** ** 20 ** to Tyrode ** ** *

* Decrease in % 25 25 10 clearance compared clearance 0 0 0 % Inhibition of SP+CGRP % Inhibition of SP+CGRP 0.1 0.3 1 3 10 100 1 3 10 30 100 1 3 10 30 100 Galanin (pmol) GALP (1− 60) (pmol) GALP (3 − 32) (pmol) b 50 c 40 ** 30 ** * 20

to Tyrode * * * Decrease in % 10 clearance compared clearance 0 1 10 30 100 250 500 Figure 3. Effect of GALP on inflammatory edema. Edema formation was GALP (3 − 32) (pmol) induced in response to i.d. injected SP (300 pmol) with CGRP (10 pmol). The Figure 4. The dose-related effect of galanin and GALP on blood flow. The response in increasing doses of (a) GALP (1–60) (n ¼ 6) and (b) GALP (3–32) dose–response effect of (a) galanin (n ¼ 8) and (b) GALP (3–32) (n ¼ 10) is (n ¼ 8) coinjected with SP þ CGRP is shown as % inhibition (mean7SEM.) of shown as decrease in % clearance (mean7SEM) compared with vehicle edema formation measured by the 125I-BSA method. Responses that are (Tyrode-injected) skin. Results that are significantly different from significantly different from the corresponding SP þ CGRP-treated sites are Tyrode-injected sites are shown (*Po0.05; **Po0.01). shown (*Po0.05, **Po0.01; Dunnett’s). (c) The effect of GALP (3–32) (10 pmol) on plasma extravasation induced by SP þ CGRP as observed by Evans Blue. Note that the intensity of blueing to SP þ CGRP is reduced in the presence of GALP (3–32). GALP (3–32) alone had no effect when compared to 50 vehicle. ** 40 Salbutamol (100 nmol) ** ** ET-1 (10 pmol) 30 ** The anti-edema effect of GALP (3–32) was confirmed using Galanin (3 pmol) GALP (1− 60) (100 pmol) a second technique (Evans Blue) to assess plasma extravasa- 20 tion (Figure 3c). GALP (1− 32) (100 pmol) GALP (3 − 32) (100 pmol) compared to Tyrode compared to 10

Effect of galanin and GALP peptides on skin blood flow Decrease in % clearance There are two obvious mechanisms via galanin peptides 0

could be acting to modulate inflammatory edema formation. Figure 5. Effect of galanin and GALP peptides on blood flow in cutaneous Either by inhibition of microvascular permeability (as dorsal microvasculature. The responses of salbutamol and ET-1 as negative salbutamol) or as an indirect mechanism through reduction and positive controls, respectively, are shown alongside responses of the of skin blood flow (as endothelin). To investigate possible most potent concentration (according to Figure 4) of galanin and GALP effects on blood flow, a 99 m technetium clearance peptides (GALP (1–60), GALP (1–32), and GALP (3–32)). Responses are 7 technique was utilized. Injection of galanin, GALP (1–60), shown as decrease in % clearance (mean SEM) compared with vehicle and GALP (3–32) all induced dose-dependent reductions in (Tyrode-injected) skin (n ¼ 8). Results that are significantly different from clearance at Tyrode-injected sites are shown (**Po0.01). inhibiting cutaneous blood flow (Figure 4; data for GALP 1–60 not shown). Figure 5 shows the expected lack of effect of salbutamol, and the potent constrictor effect of endothelin- 1, and again GALP (1–32) was not effective (Figure 5). separately used for galanin receptor-specific RT-PCR analysis. Neither primary human endothelial cells (human dermal Expression of galanin receptor subtypes microvascular endothelial cell) nor smooth muscle cells Reverse transcription (RT)-PCR analysis of dorsal murine skin (human umbilical artery smooth muscle cell) showed an messenger RNA (mRNA) revealed no expression of galanin expression of galanin receptor subtype 1, GalR2, or GalR3 receptor subtype 1 (Figure 6). GalR2 was consistently mRNA (data not shown). This finding was also supported expressed at substantial levels (Figure 6). GalR3 was detected by a receptor membrane-binding study. Both cell types at low levels of mRNA expression (Figure 6). To determine were below the detection limit, which was defined which cell types are expressing galanin receptor mRNA in the as p0.006 pmol bound galanin/mg membran protein. As microcvasulature, endothelial and smooth muscle cells were a control for the assay, SH-SY5Y/GalR2 were used with a

718 Journal of Investigative Dermatology (2007), Volume 127 SM Schmidhuber et al. GALP in Cutaneous Microvasculature

12 3456 that galanin-overexpressing mice show reduced edema GalR1 formation following chemical stimulation of neurogenic inflammation in the mouse paw, although they did not GalR2 measure blood flow changes. To further clarify the vasocon- strictive potency of galanin and GALP, skin blood flow was GalR3 measured using a 99 m technetium clearance assay, which GAPDH was initially developed in order to assess the vasoconstrictor potential of the neuropeptide Y (Chu et al., 2003). In this Figure 6. RT-PCR analysis of murine skin. RT-PCR analysis of murine skin assay, we show that galanin, GALP (1–60), and GALP (3–32) mRNA for galanin receptor subtype 1, GalR2, and GalR3 expression. Lane 1: decreased skin blood flow and inflammatory edema, whereas 100 bp ladder; lane 2: negative control; lane 3–5: dorsal murine skin of three salbutamol, a b2-adrenergic receptor agonist which decreases different animals; and lane 6: murine brain cDNA. blood vessel leakiness (Kwan et al., 2001), was only active in the edema model. This suggests that the primary activity of galanin, GALP (1–60), and GALP (3–32) is that of cutaneous binding capacity of 3.270.3 pmol bound galanin/mg mem- vasoconstriction. However, we cannot rule out the possibility brane protein (data not shown). of a concomitant permeability decrease. Pharmacologically, the galanin and the GALP effects DISCUSSION showed bell-shaped dose–response curves in the plasma This study demonstrates the potent ability of galanin and extravasation and vasoconstriction assays. It is possible that GALP to inhibit inflammatory edema, which is most likely whereas at lower doses galanin is decreasing in blood flow, at secondary to vasoconstricton and inhibition of blood flow in higher doses other mechanisms mediating direct effects in the cutaneous microvasculature. Thus, we show that either promoting vascular permeability become active. the full-length peptide GALP (1–60) or GALP (3–32) is able to We have recently shown the presence of galanin-binding exert vasoactive effects in the cutaneous microcirculation. sites in human skin. They were detected in a high density There is evidence that galanin inhibits plasma extravasa- around small vessels, indicating that galanin receptors tion in experimental models of neurogenic inflammation located around these vessels mediate the vascular activity induced either chemically or by C-fiber stimulation (Xu et al., of galanin (Kofler et al., 2004). The type of galanin receptor 1991; Green et al., 1992; Holmberg et al., 2005), and that could not be specified by the galanin receptor autoradio- SP-induced plasma extravasation was inhibited by galanin in graphy used. In this study, we demonstrate that dermal the rat hind paw (Xu et al., 1991). Galanin-induced inhibition endothelial and smooth muscle cells do not express galanin of plasma extravasation could be mediated by presynaptic or receptors. These observations suggest that galanin receptors postsynaptic effects. Previously, galanin has been suggested around blood vessels in the human dermal microvasculature to modulate neurogenic edema formation via a presynaptic are most likely located on neurons surrounding the vessels, mechanism by inhibiting release of SP and CGRP (Holmberg indicating that the effect on the dermal microvasculature is et al., 2005). The experiments presented in our work utilizing indirect owing to mediators secondarily released by other an experimental design with coinjection of SP and CGRP cells (e.g. nerve cells). However, we cannot exclude that with galanin or GALP indicate that the effect of galanin species- (human/mouse) and tissue-specific (skin/umbilical on plasma extravasation is primarily a postsynaptic effect in cord) differences of galanin receptor location exist. our system. We show that the potency order of galanin peptide Possible downstream postsynaptic consequences could be fragments on edema formation was as follows: galanin an interaction with the signaling of SP and CGRP, or a direct oGALP (3–32)oGALP (1–60)5GALP (1–32), and this is in vasoconstrictor activity of galanin. A first indication of an accordance with the affinities for the human GalR2 receptor involvement of galanin in the regulation of skin blood flow reported previously (Lang et al., 2005). Moreover, the came from the observation that mustard oil-mediated substantial expression of GalR2 in the murine skin further vasodilatation in pigeon skin was potentiated by a galanin implies a more likely GalR2 than GalR3-mediated mechan- antagonist (Santha et al., 1999). The vasoconstrictor potential ism. In other model systems, GalR2 is upregulated following of galanin was confirmed in the hamster cheek pouch, inflammation whereas galanin receptor subtype 1 is down- measured by the change of arteriolar diameter (Dagar et al., regulated, suggesting that under inflammatory conditions, 2003). This group found a similar potent vasoconstrictor there may be preferential activity of the GalR2 receptor effect of human galanin when administered alone as (Xu et al., 1996). presented in this study, but also importantly that micelle We demonstrate here, that active members of the galanin formation can amplify its vasoactive effects in vivo. In our family of peptides possess activities that are in the region of studies, a direct vasoactive effect was supported by a series of 10- to 100-fold more potent than those of the established experiments with histamine. Histamine, a potent mediator of potent constrictor agent neuropeptide Y (Chu et al., 2003). increased microvascular permeability, was co-injected with Thus, our findings suggest a far-reaching influence in the vasodilator CGRP to induce edema formation. This edema pathophysiology of inflammatory skin disorders and should formation was inhibited by galanin coinjection. In further prompt experiments evaluating the therapeutic potential of support of our findings, Holmberg et al. (2005) have reported galanin and GALP in the skin.

www.jidonline.org 719 SM Schmidhuber et al. GALP in Cutaneous Microvasculature

MATERIALS AND METHODS solution) were injected i.d. Animals were left for 30 minutes to Mice allow plasma extravasation to occur. Animals were then killed by Normal female CD-1 mice (22–27g, 8–12 weeks) were obtained cervical dislocation. Dorsal skin was removed and photographed to from Charles River, UK. All mice were maintained on normal diet, identify sites of plasma extravasation. with free access to food and water, in a climatically controlled environment. Animals were anesthetized with urethane (25% w v1; Measurement of skin blood flow 2.5 g kg1 intraperitoneally) and the dorsal skin was shaved. Blood flow changes were measured using a 99 m technetium Injection sites were chosen according to a randomized site pattern clearance technique (Chu et al., 2003). Briefly, test agents (Galanin, on the dorsal skin of the anesthetized mouse. Experiments involving GALP (1–60), GALP (1–32), GALP (3–32), endothelin-1, and mice were conducted under the Animals (Scientific Procedures) Act, salbutamol) were made up in Tyrode’s solution and an equal 1986. amount of 99 m technetium (aprox. 200 kBq) was added to all samples, and kept on ice until use. Test agents (50 ml/site) were Cell lines injected i.d., with an identical amount placed into a vial for Primary human dermal microvascular endothelial cells were isolated measurement of the total radioactivity. The mouse was killed after a from normal adult foreskin and cultured as described previously by clearance period (15 minutes) and the skin was removed and the Nguyen et al. (2002). Primary human umbilical artery smooth injected sites punched out for measurement of the remaining muscle cells were purchased from Promocell (Heidelberg, Germany) radioactivity. Initially, the amount of 99 m technetium cleared from and cultured in smooth muscle basal medium 2 (Promocell, each injection-site was calculated by comparing counts in skin with Heidelberg, Germany) at 371C in a humified atmosphere of 5% counts in the respective paired aliquot of total radioactivity. From

CO2. Stable transfected SH-SY5Y neuroblastoma cells with the this, the clearance at test agent-injected sites was then calculated by human galanin receptor GalR2 (SH-SY5Y/GalR2) were cultivated as comparing with the Tyrode-value (which was normalized to 100 for recently described (Berger et al., 2004). As the galanin receptor each experiment), and expressed as % change in clearance expression is under control of a tetracycline-regulated expression compared to Tyrode, with positive numbers indicating a decreased system (T-Rex System, Invitrogen Corporation, CA), receptor blood flow. expression was induced overnight with 1 mg/ml tetracycline. Receptor-binding assay Agents Membrane preparation and radioligand-binding assay were per- SP, CGRP, endothelin-1, and salbutamol were from Sigma (Poole, formed as described previously (Berger et al., 2004). Radio-labeled UK), as well as all others agents unless specified. Galanin (rat) and galanin binding to membrane preparation (15 mg) was carried out in GALP (1–60) (human) were purchased from Bachem (Bubendorf, duplicates in a total volume of 120 ml of binding buffer containing 125 Switzerland). GALP (1–32)-amide (human) GALP (3–32)-amide 50 pM [ I]galanin (2.000 Ci/mmol; Amersham Pharmacia Biotech (human), GALP (19–37) (human) were custom synthesized by (Little Chalfont, UK)) and non-specific binding was determined in the NeoMPS Inc. (Strasbourg, France). All peptides were dissolved in presence of 1 mM human galanin. distilled water. The stock solutions (10 nM) were stored at 201C and further diluted in Tyrode’s solution (137 mM NaCl, 2.7 mM KCl, RT-PCR analysis

0.5 mM MgCl2, 0.4 mM NaH2PO4, 11.9 mM NaHCO3, and 5.6 mM Murine skin tissues where homogenized within 10 minutes after glucose) just prior use. withdrawal of a freshly killed animal in Tri Reagent (Molecular Research Center Inc., Cincinnati, OH), using an Ultra-Turrax T25 Measurement of inflammatory edema formation (IKA Werke GmbH & Co., KG, Germany). Total RNA from tissues Plasma extravasation was used as an index of inflammatory edema and cell lines was isolated according to the instructions of formation and measured as described previously (Cao et al., 1999). the manufacturer. RNA (1 mg) was reverse transcribed using 200 U Briefly, test agents (Galanin, GALP (1–60), GALP (1–32), GALP SUPERSCRIPT IITM reverse transcriptase (Life Technologies Inc., (3–32), GALP (19–37), endothelin-1, and salbutamol) were diluted in Gaithersburg, MD). PCR amplification using 100 ng of cDNA was Tyrode’s solution and stored on ice. 125I-BSA (45 kBq in 100 mlof performed using Thermo Start polymerase (ABgene, Surrey, UK) in saline) was administered intravenously into the tail vein, and the presence of 10 pmol of each primer. Primers and PCR conditions 5 minutes later test agents (50 ml/site) were injected intradermally are listed as Supplementary Material in Table S1. The PCR products (i.d.). Plasma extravasation was allowed for 30 minutes and then a were analyzed by electrophoresis on a 2% agarose gel stained with blood sample (0.5 ml) in a heparine-coated syringe was obtained via ethidium bromide. cardiac puncture and centrifuged at 10,000 g for 4 minutes to obtain plasma. The mice were then killed, the dorsal skin was removed, and Statistical analysis the injected sites punched out (8 mm). The amount of plasma Results for functional studies are shown mainly as mean7SEM. extravasated (mlg1 tissue) was calculated by comparing the amount Statistical analyses were performed on original data by one-way of radioactivity in each skin site with that in 100 ml plasma from the analysis of variance, followed by Dunnett’s or Bonferroni multiple same animal. comparison test. Po0.05 was considered as significant. N represents In a separate set of experiments, edema formation was observed the number of animals. by the extravascular accumulation of intravenously injected Evans 1 Blue (0.1 ml of 2.5% w v in saline) that binds to endogenous CONFLICT OF INTEREST plasma albumin. After 5 minutes, test agents (50 ml in Tyrode The authors state no conflict of interest.

720 Journal of Investigative Dermatology (2007), Volume 127 SM Schmidhuber et al. GALP in Cutaneous Microvasculature

ACKNOWLEDGMENTS Jancso G, Santha P, Horvath V, Pierau F (2000) Inhibitory neurogenic This study was supported by a grant of the Austrian Science Foundation modulation of histamine-induced cutaneous plasma extravasation in the (P14906), the Salzburg ‘‘Auslandsstipendim fu¨r kurzfristige wissenschaftliche pigeon. Regul Peptide 95:75–80 Arbeiten im Ausland,’’ and the British Heart Foundation and the BBSRC. We Ji RR, Zhang X, Zhang Q, Dagerlind A, Nilsson S, Wiesenfeld-Hallin Z et al. thank the Department of Nuclear Medicine, Guy’s Hospital, London, UK for (1995) Central and peripheral expression of galanin in response to 99 m technetium. Primary human dermal endothelia cells (HDMEC) were inflammation. Neuroscience 68:563–76 kindly provided by Dr Sepp Norbert, Department of Dermatology, University of Innsbruck, Austria. Kofler B, Berger A, Santic R, Moritz K, Almer D, Tuechler C et al. (2004) Expression of neuropeptide galanin and galanin receptors in human skin. J Invest Dermatol 122:1050–3 SUPPLEMENTARY MATERIAL Kolakowski LF Jr, O’Neill GP, Howard AD, Broussard SR, Sullivan SR, Table S1. Primers and PCR conditions. Feighner SD et al. (1998) Molecular characterization and expression of cloned human galanin receptors GALR2 and GALR3. J Neurochem 71:2239–51 REFERENCES Kwan ML, Gomez AD, Baluk P, Hashizume H, McDonald DM (2001) Airway Bartfai T, Hokfelt T, Langel U (1993) Galanin –a neuroendocrine peptide. Crit vasculature after mycoplasma infection: chronic leakiness and selective Rev Neurobiol 7:229–74 hypersensitivity to substance P. Am J Physiol Lung Cell Mol Physiol Bauer FE, Ginsberg L, Venetikou M, MacKay DJ, Burrin JM, Bloom SR (1986) 280:286–97 Growth hormone release in man induced by galanin, a new hypot- Lang R, Berger A, Santic R, Geisberger R, Hermann A, Herzog H et al. (2005) halamic peptide. Lancet 2:192–5 Pharmacological and functional characterization of galanin-like peptide Berger A, Lang R, Moritz K, Santic R, Hermann A, Sperl W et al. (2004) fragments as potent galanin receptor agonists. Neuropeptides 39:179–84 Galanin receptor subtype GalR2 mediates apoptosis in SH-SY5Y McDonald DM, Bowden JJ, Baluk P, Bunnett NW (1996) Neurogenic neuroblastoma cells. Endocrinology 145:500–7 inflammation. A model for studying efferent actions of sensory nerves. Brain SD (1997) Sensory neuropeptides: their role in inflammation and wound Adv Exp Med Biol 410:453–62 healing. Immunopharmacology 37:133–52 McDonald TJ, Dupre J, Tatemoto K, Greenberg GR, Radziuk J, Mutt V (1985) Brain SD, Crossman DC, Buckley TL, Williams TJ (1989) Endothelin-1: Galanin inhibits insulin secretion and induces hyperglycemia in dogs. demonstration of potent effects on the microcirculation of humans and Diabetes 34:192–6 other species. J Cardiovasc Pharmacol 13(Suppl 5):147–9 Nguyen VA, Ebner S, Furhapter C, Romani N, Kolle D, Fritsch P et al. (2002) Brain SD, Williams TJ (1985) Inflammatory oedema induced by synergism Adhesion of dendritic cells derived from CD34+ progenitors to resting between calcitonin gene-related peptide (CGRP) and mediators of human dermal microvascular endothelial cells is down-regulated upon increased vascular permeability. Br J Pharmacol 86:855–60 maturation and partially depends on CD11a-CD18, CD11b-CD18 and Burgevin MC, Loquet I, Quarteronet D, Habert-Ortoli E (1995) Cloning, CD36. Eur J Immunol 32:3638–50 pharmacological characterization, and anatomical distribution of a rat Ohtaki T, Kumano S, Ishibashi Y, Ogi K, Matsui H, Harada M et al. (1999) cDNA encoding for a galanin receptor. J Mol Neurosci 6:33–41 Isolation and cDNA cloning of a novel galanin-like peptide (GALP) from Cao T, Gerard NP, Brain SD (1999) Use of NK(1) knockout mice to analyze porcine hypothalamus. J Biol Chem 274:37041–5 substance P-induced edema formation. Am J Physiol 277:476–81 Santha P, Pierau FK, Jancso G (1999) Galanin mediated inhibitory nervous Chu DQ, Cox HM, Costa SK, Herzog H, Brain SD (2003) The ability of modulation of cutaneous vascular reactions. Acta Physiol Hung neuropeptide Y to mediate responses in the murine cutaneous 86:279–85 microvasculature: an analysis of the contribution of Y1 and Y2 receptors. Sethi T, Langdon S, Smyth J, Rozengurt E (1992) Growth of small cell lung Br J Pharmacol 140:422–30 cancer cells: stimulation by multiple neuropeptides and inhibition by Dagar S, Onyuksel H, Akhter S, Krishnadas A, Rubinstein I (2003) Human broad spectrum antagonists in vitro and in vivo. Cancer Res 52:2737–42 galanin expresses amphipathic properties that modulate its vasoreactivity Teixeira MM, Williams TJ, Hellewell PG (1995) Anti-inflammatory effects of a in vivo. Peptides 24:1373–80 short-acting and a long-acting beta 2-adrenoceptor agonist in guinea pig Fathi Z, Battaglino PM, Iben LG, Li H, Baker E, Zhang D et al. (1998) skin. Eur J Pharmacol 272:185–93 Molecular characterization, pharmacological properties and chromo- Ulman LG, Potter EK, McCloskey DI (1994) Functional effects of a family of somal localization of the human GALR2 galanin receptor. Brain Res Mol galanin antagonists on the cardiovascular system in anaesthetised cats. Brain Res 58:156–69 Regul Peptide 51:17–23 Gottsch ML, Clifton DK, Steiner RA (2004) Galanin-like peptide as a link in Wiesenfeld-Hallin Z, Xu XJ, Langel U, Bedecs K, Hokfelt T, Bartfai T (1992) the integration of metabolism and reproduction. Trends Endocrinol Galanin-mediated control of pain: enhanced role after nerve injury. Proc Metab 15:215–21 Natl Acad Sci USA 89:3334–7 Green PG, Basbaum AI, Levine JD (1992) Sensory neuropeptide interactions Wynick D, Hammond PJ, Akinsanya KO, Bloom SR (1993) Galanin regulates in the production of plasma extravasation in the rat. Neuroscience basal and oestrogen-stimulated lactotroph function. Nature 364:529–32 50:745–9 Xu XJ, Hao JX, Wiesenfeld-Hallin Z, Hakanson R, Folkers K, Hokfelt T (1991) Hokfelt T, Ljungdahl A (1971) Light and electron microscopic autoradio- Spantide II, a novel tachykinin antagonist, and galanin inhibit plasma graphy on spinal cord slices after incubation with labeled glycine. Brain extravasation induced by antidromic C-fiber stimulation in rat hind paw. Res 32:189–94 Neuroscience 42:731–7 Holmberg K, Kuteeva E, Brumovsky P, Kahl U, Karlstrom H, Lucas GA et al. Xu ZQ, Shi TJ, Landry M, Hokfelt T (1996) Evidence for galanin receptors in (2005) Generation and phenotypic characterization of a galanin over- primary sensory neurones and effect of axotomy and inflammation. expressing mouse. Neuroscience 133:59–77 Neuroreport 8:237–42