Scholarly article on topic 'Short N-terminal galanin fragments are occurring naturally  in vivo'

Short N-terminal galanin fragments are occurring naturally in vivo Academic research paper on "Chemical sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Neuropeptides
OECD Field of science
Keywords
{Galanin / Receptor / Neuropeptide / "Affinity chromatography" / "High-performance liquid chromatography (HPLC)" / "Mass spectrometry (MS)" / "Post-translational modification (PTM)" / Autoradiography}

Abstract of research paper on Chemical sciences, author of scientific article — Robert Ihnatko, Elvar Theodorsson

Abstract The galanin family currently consists of four peptides, namely galanin, galanin-message associated peptide, galanin-like peptide and alarin. Unlike galanin that signals through three different G protein-coupled receptors; GAL1, GAL2, and GAL3, binding at its N-terminal end, the cognate receptors for other members of the galanin family are currently unknown. Research using short N-terminal galanin fragments generated either by enzymatic cleavage or solid-phase synthesis has revealed differences in their receptor binding properties exerting numerous biological effects distinct from galanin(1-29) itself. Our studies on tissue extracts derived from rat small intestine and bovine gut using chromatographic techniques and sensitive galanin(1-16)-specific radioimmunoassay revealed the presence of immunoreactive compounds reacting with antiserum against galanin(1-16) distributed in distinct elution volumes. These results suggested a possible presence of short N-terminal galanin fragments also in vivo. Moreover, employing immunoaffinity chromatography and reverse-phase high performance liquid chromatography (HPLC) followed by mass spectrometry allowed specific enrichment of these immunoreactive compounds from rat tissues and identification of their molecular structure. Indeed, our study revealed presence of several distinct short N-terminal galanin sequences in rat tissue. To prove their receptor binding, four of the identified sequences were synthetized, namely, galanin(1-13), galanin(1-16), galanin(1-20), galanin(6-20), and tested on coronal rat brain sections competing with 125I-labeled galanin(1-29). Our autoradiographs confirmed that galanin(1-13), galanin(1-16), and galanin(1-20) comprehensively displaced 125I-galanin(1-29) but galanin(6-20) did not. Here we show, for the first time, that short N-terminal galanin fragments occur naturally in rat tissues and that similar or identical galanin sequences can be present also in tissues of other species. Biological significance This study is first to provide an evidence of the presence of short N-terminal galanin fragments in vivo in a biological system and provides further foundations for the previous studies using synthetized short N-terminal galanin fragments.

Academic research paper on topic "Short N-terminal galanin fragments are occurring naturally in vivo"

Neuropeptides xxx (xxxx) xxx-xxx

Contents lists available at ScienceDirect

Neuropeptides

journal homepage: www.elsevier.com/locate/npep

Short N-terminal galanin fragments are occurring naturally in vivo

Robert Ihnatko*, Elvar Theodorsson

Department of Clinical Chemistry, Department of Clinical and Experimental Medicine, Linköping University, 58285 Linköping, Sweden

ARTICLE INFO

Keywords:

Galanin

Receptor

Neuropeptide

Affinity chromatography

High-performance liquid chromatography

(HPLC)

Mass spectrometry (MS) Post-translational modification (PTM) Autoradiography

ABSTRACT

The galanin family currently consists of four peptides, namely galanin, galanin-message associated peptide, galanin-like peptide and alarin. Unlike galanin that signals through three different G protein-coupled receptors; GALi, GAL2, and GAL3, binding at its N-terminal end, the cognate receptors for other members of the galanin family are currently unknown. Research using short N-terminal galanin fragments generated either by enzymatic cleavage or solid-phase synthesis has revealed differences in their receptor binding properties exerting numerous biological effects distinct from galanin(1-29) itself. Our studies on tissue extracts derived from rat small intestine and bovine gut using chromatographic techniques and sensitive galanin(1-16)-specific radioimmunoassay revealed the presence of immunoreactive compounds reacting with antiserum against galanin(1-16) distributed in distinct elution volumes. These results suggested a possible presence of short N-terminal galanin fragments also in vivo. Moreover, employing immunoaffinity chromatography and reverse-phase high performance liquid chromatography (HPLC) followed by mass spectrometry allowed specific enrichment of these immunoreactive compounds from rat tissues and identification of their molecular structure. Indeed, our study revealed presence of several distinct short N-terminal galanin sequences in rat tissue. To prove their receptor binding, four of the identified sequences were synthetized, namely, galanin(1-13), galanin(1-16), galanin(1-20), galanin(6-20), and tested on coronal rat brain sections competing with 125I-labeled galanin(1-29). Our autoradiographs confirmed that galanin(1-13), galanin(1-16), and galanin(1-20) comprehensively displaced 125I-galanin(1-29) but galanin (6-20) did not. Here we show, for the first time, that short N-terminal galanin fragments occur naturally in rat tissues and that similar or identical galanin sequences can be present also in tissues of other species. Biological significance: This study is first to provide an evidence of the presence of short N-terminal galanin fragments in vivo in a biological system and provides further foundations for the previous studies using synthetized short N-terminal galanin fragments.

1. Introduction

Galanin is a neuroendocrine peptide initially isolated from porcine intestine (Tatemoto et al., 1983). It has a widespread distribution throughout the central and peripheral nervous as well as endocrine systems exerting numerous physiological effects, including the control of food intake, learning, memory, nociception, anxiety, stress, depression, inflammation, and also exerting neurotrophic or neuroprotective action (Barreda-Gomez et al., 2014; Borbely et al., 2013; Kinney et al., 2003; Kormos and Gaszner, 2013; Kyrkouli et al., 1990; Lang and Kofler, 2011). The galanin family of peptides is encoded by two separate genes, galanin/galanin message associated peptide (GMAP) prepropeptide (GAL) and galanin-like peptide (GALP) (Evans et al., 1993; Ohtaki et al., 1999). The mature hormone is processed from a preproprecursor by cleavage of signal peptide and consequent proteo-lytic processing providing galanin(1-29) (30 amino acids in humans)

and GMAP (Evans et al., 1993). Excepting humans, all other species have galanin C-terminally amidated (Evans and Shine, 1991; Lang et al., 2015; Tatemoto et al., 1983). The N-terminal region of galanin (residues 1-14) is highly conserved between species and is involved in receptor binding mediating its biological effects (Land et al., 1991). The function of the C-terminal sequence of galanin that shows higher degree of inter-species variability is not fully elucidated. It has been proposed to serve as protector against proteolytic attacks and may play a role for the recognition of galanin(1-29) by galanin receptors (Bedecs et al., 1995a; Rossowski et al., 1990). Thus variability of C-terminal sequence of galanin can contribute to receptor binding differences observed between species.

So far three G protein-coupled galanin receptors have been identified; GAL1, GAL2, and GAL3. They trigger multiple signaling pathways, including stimulation of phospholipase C, particularly by GAL2, or inhibition of cAMP/PKA by GAL1 and GAL3 receptors respectively

☆ This work was supported by the County Council of Östergötland. * Corresponding author. E-mail address: robert.ihnatko@liu.se (R. Ihnatko).

http://dx.doi.org/10.10167j.npep.2017.03.005

Received 9 January 2017; Received in revised form 8 March 2017; Accepted 31 March 2017 0143-4179/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Ihnatko, R., Neuropeptides (2017), http://dx.doi.Org/10.1016/j.npep.2017.03.005

0 6 10 15 20 25 30 35 40 45 50 55 60 Fraction number

Fig. 1. Detection of immunoreactive compounds in peptide extract isolated from rat small intestine using rabbit antiserum K2 against galanin(1-16). A. The representative chromatogram from five independent experiments from HPLC separation of peptide extract from rat small intestine on C18 HPLC column. Gradient: 0-50% ACN with 0.1% TFA in 60 min (blue line). The arrows show the peaks with eluted immunoreactive material reacting with antiserum K2 in RIA. B. The representative results from five independent RIA analyses based on antiserum K2 of fractions from C18 HPLC separation of peptide extract from rat small intestine.

(Barreda-Gomez et al., 2014; Barreda-Gomez et al., 2005a; Lang et al., 2015; Rezaei et al., 2000; Webling et al., 2012). The distribution of galanin receptors is also tissue-specific. The presence of GAL1 has been shown in the hypothalamus, ventral hippocampal formation (HiFo), thalamus, amygdala, brainstem and spinal cord (Barreda-Gomez et al., 2005b; Burgevin et al., 1995; Gustafson et al., 1996; Habert-Ortoli et al., 1994; Lu et al., 2005). GAL2 is highly expressed in several brain regions, particularly in the hypothalamus, hippocampus, the anterior pituitary and lower levels also in the amygdala and regions of cortex (Fathi et al., 1997; Lu et al., 2005). GAL3 shows the highest expression in the preoptic/hypothalamic area and the pituitary (Lu et al., 2005; Mennicken et al., 2002).

Besides galanin, GALP, a 60 amino acids long neuropeptide binds to the galanin receptors. The structure of GALP is related to galanin since it contains the conserved sequence galanin(1-13) at position 9-21

within its amino acid sequence (Lang et al., 2005; Ohtaki et al., 1999).

Interestingly, several studies have reported different binding properties and distinct functional effects induced by short N-terminal fragments of galanin and GALP that were generated only artificially either by enzymatic cleavage or a solid-phase synthesis until now (Crawley et al., 1990; Floren et al., 2000; Kinney et al., 1998; Land et al., 1991; Lang et al., 2005; Parker et al., 1995; Runesson et al., 2009; Smith et al., 1998; Todd et al., 2000; Wang et al., 1997; Wirz et al., 2005; Wynick et al., 1993). For example, full length galanin(1-29) has shown different distribution of binding sites in the rat brain compared to galanin(1-15) (Hedlund et al., 1992; Melander et al., 1988). Galanin (1-15) has more pronounced electrophysiological effects on pyramidal neurons in the dorsal HiFo than galanin(1-16) and full length galanin(1-29) (Xu et al., 1999). However, the last two showed stronger hyperpo-larizing effect on neurons of locus coeruleus (Ma et al., 2001).

Different effects of galanin(1-29) and short N-terminal galanin fragments were also observed in interaction with other signaling systems in the brain. Galanin(2-11) was shown to have distinct effect on serotonin signaling than galanin(1-29) (Mazarati et al., 2005). Galanin(1-15) influences mood regulation, inducing strong depression-like and anxiogenic-like effects in rats via possible GAL1-GAL2 heteroreceptor complex (Borroto-Escuela et al., 2014; Millon et al., 2014). An interactions between neuropeptide Y, its cognate Y2 receptor subtype and the galanin(1-15) via galanin receptors has been observed in the nucleus of the solitary tract showing a facilitating effect of galanin(1-15) in the central cardiovascular control (Diaz-Cabiale et al., 2010; Diaz-Cabiale et al., 2005). Moreover, an enhancing effect of galanin(1-15) on antidepressant activity of the 5-HT1A receptor agonist 8-OH-DPAT has also been observed (Millon et al., 2016).

In fact, galanin(1-15) has been shown to exert a multitude of biological effects in several biological systems (Borroto-Escuela et al., 2014; Diaz-Cabiale et al., 2002; Diaz-Cabiale et al., 2010; Diaz-Cabiale et al., 2005; Diaz et al., 1996; Hedlund et al., 1994; Millon et al., 2016; Millon et al., 2014; Narvaez et al., 2000; Narvaez et al., 1994). Taken together, the results from the abovementioned studies suggest a presence of additional tissue-specific galanin receptors with preference for short N-terminal sequences of galanin similar to a finding of a hyperpolarizing, galanin(1-15)-selective receptor on CA3 pyramidal neurons in HiFo (Xu et al., 1999).

So far, however, there is no evidence of the presence of such short N-terminal galanin fragments in vivo. In search of such fragments in a biological system we used chromatographic techniques and sensitive and specific radioimmunoassay (RIA) based on antiserum K2 raised against galanin(1-16) in our laboratory (Hilke et al., 2006) for analysis of tissue extracts from rat small intestine and bovine gut. The immunoreactive material was characterized by means of reverse-phase high-performance liquid chromatography (HPLC) and enriched by immunoaffinity chromatography to high degree of purity. The identity and structure of purified immunoreactive compounds was determined using nano-flow liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). Finally, the sequences of several identified compounds with potential physiological relevance were synthetized and their ability of receptor displacement of 125I-labeled galanin(1-29) was tested using coronal sections of the rat brain.

2. Results

2.1. Reverse-phase HPLC of tissue extracts

The extracts derived from rat small intestine and bovine gut were separated by reverse-phase HPLC and the collected fractions (1 mL/ min) were analyzed in RIA based on antiserum K2 (Hilke et al., 2006). Reverse-phase HPLC of the extracts from rat small intestine showed that immunoreactive compounds reacting with antiserum K2 were eluted into two distinct elution volumes with different retention time (Fig. 1 A, B). Moreover, several fractions from reverse-phase HPLC of tissue extracts from bovine gut displayed also significant signal in RIA (Fig. 2). In this case the immunoreactive material was distributed in the volumes eluted within 19-31 min indicating that the compounds reacting with antiserum K2 are present also in other species. However, for further analysis and sequencing we used only rat tissue.

2.2. Affinity purification of immunoreactive compounds

A single one-dimensional separation of tissue extracts by reversephase HPLC did not provide quantity and purity of material containing immunoreactive compounds reacting with antiserum K2 sufficient for mass spectrometric analysis. Therefore, tissue extracts from rat small intestine were subjected to affinity chromatography using purified immunoglobulins reacting with galanin(1-16) from antiserum K2. The affinity purified immunoreactive material was further separated by

Fig. 2. Detection of immunoreactive compounds reacting with antiserum K2 in tissue extract from bovine gut. The extract was separated on C18 HPLC column using a gradient from 0% to 80% ACN with 0.1% TFA in 60 min and collected fractions (1 mL/min) were analyzed in RIA based on antiserum K2. The figure shows representative results from two independent experiments.

reverse-phase HPLC and collected fractions (1 mL/min) were analyzed in RIA based on antiserum K2. Reverse-phase HPLC and RIA showed that immunoreactive compounds reacting with antiserum K2 were distributed in three distinct fractions with different retention time, eluting at 24 min, 30 min, and 32 min respectively (Fig. 3). The abundances of the immunoreactive compounds in the analyzed fractions, however, cannot be inferred from the presented RIA results since they may reflect binding affinity of the particular molecular form to antiserum K2. The results from RIA show the highest concentration of the immunoreactive material in the fraction 32. Our results from mass spectrometric analysis has revealed that this fraction contained the highest abundance of galanin(1-16), which is the sequence the antiserum K2 was originally raised against. It is therefore not surprising that this fraction displayed the highest signal.

2.3. LC-MS/MS and database search

The immunoreactive compounds distributed in the fraction 24, 30 and 32 from reverse-phase HPLC were subjected to nano-flow liquid chromatography coupled with Orbitrap Velos Pro mass spectrometer. Searching of acquired MS/MS data against non-redundant database of complete rat proteome using PEAKS DB search algorithm revealed 27 distinct peptide species that covered galanin amino acid sequence in various extent. The analysis showed the presence of twelve short galanin sequences in the fraction 24, twenty-six in the fraction 30 and fourteen in the fraction 32 respectively. The fraction 24 contained exclusively peptides with truncated N-terminal end of galanin with the longest peptide galanin(3-20) and the shortest galanin(13-20), whereas fraction 30 and 32 contained peptides with truncated either N- or C-terminal end at various extent. The fractions 30 and 32 contained also full length C-terminally amidated galanin(1-29) (Table 1). The presence of small amount of C-terminally amidated galanin(21-29) in all analyzed fractions indicates that this fragment was generated by trypsin

Fig. 3. Distribution of immunoreactive compounds from extracts of rat small intestine after affinity purification using column with purified IgGs reacting against galanin(1-16) from antiserum K2. The fractions with immunoreactive material were identified in RIA based on K2 antiserum. The figure shows representative RIA results from four independent analyses. The gradient used in reverse-phase HPLC: 10-80% ACN with 0.1% TFA in 80 min.

cleavage prior to mass spectrometry of galanin(1-29) in the fraction 30 and 32 or galanin(3-29) in the fraction 24 respectively at the position of Arg52 within the amino acid sequence of galanin prepropeptide (Table 1). Although this peptide passed the exclusion criteria, its appearance is obviously generated by added trypsin before the MS-based sequencing and therefore this peptide is not listed in the table of galanin peptides with potential physiological functions (Table 2). The short C-terminal fragment ELPLEVEEGR which was found in small amount in the fraction 30 and 32 (Table 1) was similarly generated by cleavage at Arg64 by added trypsin prior to mass spectrometry. This fragment, however, did not pass the exclusion criteria. An example, the MS/MS spectrum derived by collision-induced dissociation (CID) of the peptide GWTLNSAGYLLGPHAID is shown in Fig. 4.

2.4. Bioinformatics

The identified amino acid sequences of short N-terminal galanin fragments were subjected to bioinformatics analysis in order to find peptides with potential physiological relevance and exclude those peptides which could be generated by accidental cleavage during the process of extraction from tissue or during the analytical procedures. For this purpose, the amino acid sequence of galanin prepropeptide (Fig. 5 A) was submitted to in silico cleavage experiment using PeptideCutter bioinformatics tool (http://web.expasy.org/peptide_ cutter/, Swiss Institute of Bioinformatics) to identify cleavage sites within the sequence that can be targeted by known proteases. The list of proteases with at least one cleavage site within the analyzed galanin sequence along with positions and number of cleavages are summarized in Table 3.

The identified galanin sequences were examined for the uniqueness, excluding those which amino acid sequence started at a position known to be a target for the tested proteases in the in silico cleavage

experiment. The exception has been allowed only for Arg32 and Arg52, as these two sites could be a target for trypsin used in the procedure of sample preparation for mass spectrometry. The amino acid sequence between the position Arg32 and Arg52 corresponds to galanin (1-20) but the analysis showed the presence either shorter or even longer galanin sequences that exceeded the position at Arg52 indicating that treatment of the samples with trypsin did not exert a detrimental effect on the analyzed galanin peptides. The presence of the two abovementioned C-terminal galanin fragments generated prior to mass spectrometry by added trypsin was only in minor extent. The comprehensive list of cleavage sites within the amino acid sequence of the identified short N-terminal galanin fragments for the tested proteases can be found in the Supplemental Table 1 B.

Four of the identified short N-terminal galanin sequences were selected for peptide synthesis, namely galanin(1-13), galanin(1-16), galanin(1-20) and galanin(6-20) and tested for binding at galanin receptors on coronal rat brain sections based on a competition with 125I-labeled galanin(1-29).

The search of post-translationally modified (PTM) sites contained in MS/MS spectra using PEAKS PTM algorithm combined with ASCORE's probabilistic approach and scoring algorithm (Beausoleil et al., 2006) built in Scaffold PTM software revealed also two asparagine (Asn) residues at the position 37 and 50 within the amino acid sequence of several identified short N-terminal galanin peptides deamidated. The most abundant galanin fragments, however, did not possess any deamidated Asn residues. The highest number of modified peptides was found in the fraction 30 and in less extent in the fraction 24. On the other hand, the presence of deamidated peptides containing Asn37 and Asn50 residues in the fraction 32 was rare (Table 4). The physiological relevance of deamidation of N-terminal galanin homologs may merit further investigation.

2.5. Receptor displacement test and autoradiography

Receptor binding of HPLC purified 125I-labeled rat galanin(1-29) was tested in two different concentrations on freshly prepared coronal rat brain sections prior to receptor displacement test using the synthetized galanin fragments. The concentration 1.5nmol/L was chosen for further binding experiments and autoradiography.

According the rat brain atlas of Paxinos and Watson (Paxinos and Watson, 1998) the highest density of binding sites for galanin(1-29) was found in the magnocellular preoptic nucleus, piriform cortex, island of Calleja, cortex-amygdala transition zone, supraschiastic nucleus, fornix, zona incerta, stria terminalis and medularis, ventro-lateral, central medial, dorsomedial, paraventricular, and anteroven-tricular thalamic nucleus, and anterior and lateral hypothalamic area. The autoradiographs of the rat brain sections used in the competitive receptor displacement test has shown displacement of 125I-labeled galanin 1-29 with three of four tested short N-terminal galanin fragments, particularly with galanin(1-13), galanin(1-16) and galanin (1-20) respectively (Fig. 6 C-E). No displacement was observed with galanin(6-20) (Fig. 6 F).

3. Discussion

The present work provides evidence for the presence of a handful of naturally-occurring short N-terminal galanin fragments in the rat small intestine which bind to the antiserum K2 raised against galanin(1-16) in our laboratory (Hilke et al., 2006). Probing of tissue material separated in reverse-phase HPLC in RIA showed that immunoreactive compounds that react with antiserum K2 were distributed in two elution volumes (Fig. 1 A, B), suggesting distinct physicochemical properties of the eluted immunoreactive compounds. Moreover, probing of tissue extracts from bovine gut separated in reverse-phase HPLC in RIA showed that immunoreactive compounds that react with antiserum K2 are present also in other species (Fig. 2). However, for further enrichment

Table 1

List of peptide sequences found in the selected fractions from reverse-phase HPLC with eluted immunoreactive material along with the mass and spectral counts. The peptide score — 10lgP > 20 was considered for confident peptide identification. Abbreviations: Uniq = unique peptide, #Spec = MS/MS spectral counts for the particular peptide from quadruplicates, (a) = amidation, Y = yes.

Peptide Uniq — 10lgP Mass Length m/z #Spec Start End

Fraction #24

SAGYLLGPHAIDNHR Y 87.99 1619.817 15 810.915 13 38 52

AGYLLGPHAIDNHR Y 82.86 1532.785 14 767.310 3 39 52

TLNSAGYLLGPHAIDNHR Y 79.48 1947.991 18 975.001 3 35 52

NSAGYLLGPHAIDNHR Y 78.56 1733.860 16 867.939 5 37 52

GYLLGPHAIDNHR Y 75.30 1461.748 13 731.882 5 40 52

LNSAGYLLGPHAIDNHR Y 74.71 1846.944 17 924.477 4 36 52

YLLGPHAIDNHR Y 65.51 1404.726 12 703.368 4 41 52

LLGPHAIDNHR Y 53.89 1241.663 11 621.834 3 42 52

LGPHAIDNHR Y 50.58 1128.579 10 565.297 3 43 52

GPHAIDNHR Y 42.55 1015.495 9 508.752 5 44 52

SFSDKHGLT(a) Y 33.10 989.493 9 495.753 2 53 61

PHAIDNHR Y 28.86 958.473 8 480.243 1 45 52

Fraction #30

GWTLNSAGYLLGPHAIDNHRSF Y 106.08 2425.193 22 1213.604 6 33 54

GWTLNSAGYLLGPHAID Y 105.92 1783.889 17 892.952 179 33 49

GWTLNSAGYLLGPHAIDNH Y 104.87 2034.991 19 1018.505 8 33 51

GWTLNSAGYLLGPHAIDNHR Y 97.67 2191.092 20 1096.551 15 33 52

GWTLNSAGYLLGPH Y 89.97 1484.741 14 743.380 9 33 46

GWTLNSAGYLLGPHAI Y 88.37 1668.862 16 835.437 3 33 48

GWTLNSAGYLLGPHA Y 85.13 1555.778 15 778.888 2 33 47

GWTLNSAGYLLGP Y 80.11 1347.682 13 674.848 6 33 45

SAGYLLGPHAIDNHR Y 76.98 1619.817 15 810.917 9 38 52

GWTLNSAGYLLG Y 72.05 1250.629 12 626.321 2 33 44

GWTLNSAGYLLGPHAIDN Y 71.24 1897.932 18 949.982 10 33 50

TLNSAGYLLGPHAIDNHR Y 69.48 1947.991 18 975.001 5 35 52

LLGPHAIDNHR Y 64.67 1241.663 11 621.837 7 42 52

GWTLNSAGYLL Y 60.77 1193.608 11 597.811 3 33 43

LGPHAIDNHR Y 52.93 1128.579 10 565.296 5 43 52

ELPLEVEEGR Y 52.80 1169.593 10 585.804 2 65 74

GYLLGPHAIDNHR Y 52.62 1461.748 13 731.880 3 40 52

AGYLLGPHAIDNHR Y 50.05 1532.785 14 767.400 2 39 52

NSAGYLLGPHAIDNHR Y 48.05 1733.860 16 867.937 3 37 52

GWTLNSAGY Y 47.91 967.440 9 484.727 2 33 41

GPHAIDNHR Y 45.01 1015.495 9 508.756 5 44 52

GWTLNSAGYLLGPHAIDNHRSFSD Y 41.79 2627.252 24 876.760 2 33 56

YLLGPHAIDNHR Y 33.91 1404.726 12 703.371 2 41 52

GWTLNSAGYLLGPHAIDNHRSFSDKHGLT(a) Y 32.19 3162.575 29 1055.199 2 33 61

PHAIDNHR Y 29.86 958.473 8 480.244 1 45 52

SFSDKHGLT(a) Y 29.15 989.493 9 495.754 1 53 61

Fraction #32

GWTLNSAGYLLGPHAID Y 91.89 1783.889 17 892.952 2 33 49

GWTLNSAGYLLGPHAI Y 86.15 1668.862 16 835.438 30 33 48

GWTLNSAGYLLGPH Y 79.06 1484.741 14 743.378 3 43 46

SAGYLLGPHAIDNHR Y 75.83 1619.817 15 810.915 3 38 52

LLGPHAIDNHR Y 60.77 1241.663 11 621.839 6 42 52

GWTLNSAGYLLGPHAIDNHR Y 59.38 2191.092 20 1096.555 4 33 52

LGPHAIDNHR Y 54.11 1128.580 10 565.293 2 43 52

GWTLNSAGYLLGPHAIDNH Y 44.58 2034.991 19 679.336 2 33 51

GPHAIDNHR Y 39.87 1015.495 9 508.753 2 44 52

ELPLEVEEGR Y 39.03 1169.593 10 585.804 1 65 74

GYLLGPHAIDNHR Y 27.25 1461.748 13 731.882 1 40 52

GWTLNSAGYLLGPHAIDNHRSFSDKHGLT(a) Y 26.87 3162.575 29 1055.199 1 33 61

TLNSAGYLLGPHAIDNHR Y 24.98 1947.991 18 975.005 1 35 52

SFSDKHGLT(a) Y 24.83 989.493 9 495.754 1 53 61

and structural analysis of the immunoreactive compounds we used rat tissue.

As the concentration of neuropeptides in rat tissues is very low we developed multistep purification procedure using specific enrichment of galanin peptides on affinity column with purified immunoglobulins reacting with galanin(1-16) from antiserum K2. Combining the im-munoaffinity purification with reverse-phase HPLC chromatography we could concentrate short N-terminal galanin peptides to purity sufficient for structural analysis by mass spectrometry. The multistep purification procedure mentioned above allowed not only the enrichment of the immunoreactive material but also revealed that short N-terminal galanin peptides are distributed in three distinct elution volumes

indicating even higher level of diversity in their physicochemical properties (Fig. 3). Using ice-cold treatment to chill the tissue immediately after dissection followed by rapid boiling to inactivate proteases and having protease inhibitors present during all relevant experimental procedures we protected the biological material against enzymatic degradation processes in order to get maximally relevant data.

Further analysis of the purified immunoreactive compounds by mass spectrometry revealed their primary amino acid structure (Table 1). Interestingly, the fraction 24 contained exclusively truncated galanin species lacking several amino acids from N-terminus within the position 35-45 of the amino acid sequence of galanin prepropeptide.

Table 2

List of galanin sequences with potential physiological relevance identified in the immunoreactive material from rat small intestine that passed the exclusion criteria (see the section Bioinformatics in Material and methods for more details). The abundance of the particular sequences is expressed as the sum of spectral counts in the fraction 24, 30, and 32 analyzed in quadruplicates. The peptide sequences of galanin(1-13), galanin(1-16), galanin(1-20) and galanin(6-20) were synthetized and tested in receptor displacement test on the rat brain sections competing with 125I-labeled rat galanin(1-29).

Name Sequence Abundance

Galan n 1-9) GWTLNSAGY 2

Galan n 1-11) GWTLNSAGYLL 3

Galan n 1-12) GWTLNSAGYLLG 2

Galan n 1-13) GWTLNSAGYLLGP 6

Galan n 1-14) GWTLNSAGYLLGPH 12

Galan n 1-15) GWTLNSAGYLLGPHA 2

Galan n 1-16) GWTLNSAGYLLGPHAI 33

Galan n 1-17) GWTLNSAGYLLGPHAID 181

Galan n 1-18) GWTLNSAGYLLGPHAIDN 10

Galan n 1-19) GWTLNSAGYLLGPHAIDNH 10

Galan n 1-20) GWTLNSAGYLLGPHAIDNHR 19

Galan n 1-22) GWTLNSAGYLLGPHAIDNHRSF 6

Galan n 1-24) GWTLNSAGYLLGPHAIDNHRSFSD 2

Galan n 1-29) GWTLNSAGYLLGPHAIDNHRSFSDKHGLT 3

Galan n 13-20) PHAIDNHR 2

Galan n 8-20) GYLLGPHAIDNHR 9

Galan n 7-20) AGYLLGPHAIDNHR 5

Galan n 6-20) SAGYLLGPHAIDNHR 25

However, for a potential physiological relevance, the vast majority of N-terminally truncated galanin species eluting in this fraction were not further considered. These short galanin sequences did not pass the exclusion criteria in our bioinformatics analysis since the most of them started at a cleavage site for the proteases we have tested in our in silico cleavage experiment (Table 3, see the section Experimental procedures and Supplemental material for more details). It is likely that most of the short N-terminal galanin sequences present in the fraction 24 were products of galanin turnover.

In contrast, the vast majority of identified short N-terminal galanin fragments in the fraction 30 and 32 possessed intact N-terminal end suggesting physiological activity. The presence of several short N-terminal galanin peptides in the analyzed fractions, some of them in

abundance clearly above the others, confirmed the ability of our affinity column for a specific enrichment of immunoreactive galanin compounds from tissue extracts.

The samples with enriched immunoreactive galanin compounds contained protein impurities that were nonspecifically bound to the column despite of vigorous washing. The impurities, if undigested, would have deteriorating effect on performance of the nano-LC column coupled to the mass spectrometer. The use of trypsin was therefore essential for a successful mass spectrometric analysis and sequencing. The digestion with trypsin did not exert detrimental effect on analyzed peptides of primary interest because the cleavage sites for trypsin consists of two Arg residues at the position 32 and 52 are determining the peptide corresponding to galanin(1-20). We could either identify shorter or longer amino acid sequences that exceed the trypsin cleavage site at Arg52, such as galanin(1-22) or galanin(1-24) and galanin(1-29) (Table 1). The presence of small amount of C-terminally amidated galanin(21-29) in all analyzed fractions indicates that this fragment was generated by trypsin cleavage prior to mass spectrometry of galanin(1-29) in the fraction 30 and 32 respectively or galanin(3-29) in the fraction 24 at the position of Arg52 within the amino acid sequence of galanin prepropeptide (Table 1). Although this C-terminal fragment passed the exclusion criteria in our bioinformatics analysis, its appearance is clearly due to tryptic processing required for mass spectrometry and therefore this peptide is not further considered and listed within the galanin peptides with potential physiological functions (Table 2). The last mentioned is applicable also for the short C-terminal fragment ELPLEVEEGR generated by cleavage at Arg64 by added trypsin required for mass spectrometry based sequencing. This short fragment, however, did not pass the exclusion criteria and is therefore not listed in Table 2.

The most abundant short N-terminal galanin peptides in this study were galanin(1-17), galanin(1-16), galanin(1-20), galanin(1-14), and galanin(1-19). Furthermore, galanin(1-18), galanin(1-22), galanin(1-13), galanin(1-11), galanin(1-15), galanin(1-12), and galanin(1-9) were also identified, but in lower abundances. The identified short N-terminal galanin sequences may exert numerous physiological effects. Some of them, prepared by peptide synthesis, have already been tested in several studies using in vitro and in vivo biological systems. The most extensively studied have been galanin(1-15) and galanin(1-16) but

Fig. 4. Tandem mass spectra derived by collision-induced dissociation (CID) of the peptide GWTLNSAGYLLGPHAID, m/z = 892.952, - 10lgP = 105.95 by PEAKS DB. The b-ions (in blue) are CID fragment ions from N-terminus and y-ions (in red) are CID fragment ions from C-terminal end of the identified peptide.

10 20 30 40 50

MARGSVILLA WLLLVATLSA TLGLGMPTKE KRGWTLNSAG YLLGPHAIDN

60 70 80 90 100

HRSFSDKHGL TGKRELPLEV EEGRLGSVAV PLPESNIVRT IMEFLSFLHL

110 120 KEAGALDSLP GIPLATSSED LEQS

B. Fraction 24

si ■

KRGWTLNSAG YLLGPHAIDN HRSFSDKHGL TGKRELPLEV

C. Fraction 30

KRGWTLNSAG YLLGPHAIDN HRSFSDKHGL TGKRELPLEV EEGRLGSVAV

D. Fraction 32

KRGWTLNSAG YLLGPHAIDN HRSFSDKHGL TGKRELPLEV EEGRLGSVAV

i-i-» -

Fig. 5. A. The amino acid sequence of galanin prepropeptide (P106831 GALA_RAT) OS = Rattus norvegicus GN = Gal PE = 1 SV = 1. The sequence of galanin(1-29) is underlined; The graphical view of the sequence coverage of matched mass spectra by PEAKS DB in the fraction 24 (Fig. 5 B), in the fraction 30 (Fig. 5 C), and the fraction 32 (Fig. 5 D) respectively with — 10lgP > 20 set for confident peptide identification. The list of short N-terminal galanin peptides in these fractions is shown in Table 1. (a) = amidation.

Table 3

List of proteases with at least one cleavage site within the analyzed sequence of galanin prepropeptide shown in Fig. 5 A.

Enzyme No. of cleavages Positions of cleavage sites

Arg-C proteinase 6 3 32 52 64 74 89

Asp-N endopeptidase 4 48 55 106 119

Asp-N endopeptidase + N-terminal Glu 14 29 48 55 64 68 70 71 83 92 101 106 118 119 121

Chymotrypsin-high specificity (C-term to [FYW], not before P) 6 11 34 41 54 94 97

Chymotrypsin-low specificity (C-term to [FYWML], not before 32 1 8 9 11 12 13 14 18 22 24 34 36 41 42 43 46 51 54 58 60 68 75 92 94 95 97 98 99 100 106

P) 114 121

Glutamyl endopeptidase 10 30 65 69 71 72 84 93 102 119 122

LysC 5 29 31 57 63 101

Pepsin (pH 1.3) 33 7 8 9 11 12 13 14 17 18 21 22 23 24 35 36 41 42 67 75 93 94 95 96 97 98 99 100 105 106 109

113 120 121

Pepsin (pH > 2) 35 7 8 9 10 11 12 13 14 17 18 21 22 23 24 35 36 40 41 42 67 75 93 94 95 96 97 98 99 100 105

106 109 113 120 121

Trypsin 11 3 29 31 32 52 57 63 64 74 89 101

undergo spontaneous deamidation where the side chain amide group of glutamine or asparagine is transformed into an acidic carboxylate group (Geiger and Clarke, 1987). It has been suggested that deamidation can act as molecular clock for the timing of biological events (Robinson and Robinson, 2001). In case of the identified galanin peptides, deamidation is likely to set off catabolism of those peptides or some other corrective processes or to form a peptides with a new function. The deamidated peptides were mostly present in the fraction 30 and, in less extent, in the fraction 24 (Table 4). However, the comparison of total spectral counts of the identified galanin fragments with the number of spectra with deamidated Asn residues revealed that deamidated pep-tides were present only to a minor extent. As an example, there were only forty-one spectra (nearly 16%) with deamidated Asn37 from the total of 257 matched spectra containing this site found in the fraction 30 (see Table 1 and Table 4). Whether deamidation of Asn37 and Asn50 residues occurred in vitro during the experimental procedure or may have some physiological relevance in vivo remains unclear.

Taken together, this study provides the first evidence that short N-terminal galanin sequences occur naturally in vivo in biological systems and provides further foundations for the previous studies using synthetized short N-terminal galanin fragments.

4. Experimental procedures 4.1. Animals

Thirty-four male Sprague Dawley rats (Charles River, Leiden, Netherlands), weighing 200-250 g were housed in standard cages and kept at constant temperature (21 ± 1 °C) under a controlled 12 h light/dark cycle with free access to standard rat chow and water. For each preparation of tissue extract, four rats were anesthetized with 4% isoflurane and sacrificed. The gut from each animal was instantly dissected and kept on ice until the extraction procedure (< 30 min). Another set of animals were used for receptor binding experiments. The brains were instantly removed from the scull, frozen on dry ice and

Table 4

Position and spectral counts of the most frequent post-translationally modified sites within the position 33-61 of the amino acid sequence of rat galanin (Uniprot_P10683) found using PEAKS PTM and Scaffold PTM algorithm in the fraction 24, 30, and 32 respectively. (The complete list of PTMs found using these bioinformatics tools within galanin sequence can be found in Supplemental Table 2).

Site Modification Best ASCORE Localization probability Spectral counts Fraction 24 Fraction 30 Fraction 32

W34 Oxidation 73.31 100% 0 105 9

W34 Dioxidation 140.00 100% 0 83 12

W34 Trp ^ kynurenine 1000.00 100% 0 24 10

N37 Deamidation 1000.00 100% 11 41 2

N50 Deamidation 1000.00 100% 13 34 3

T61 Amidation 1000.00 100% 2 5 3

other galanin fragments have also been of interest (Bedecs et al., 1995b; Borroto-Escuela et al., 2014; Diaz-Cabiale et al., 2000a; Diaz-Cabiale et al., 1998; Diaz et al., 1996; Fisone et al., 1989; Fuxe et al., 1998; Girotti et al., 1993; Gu et al., 1995; Hedlund et al., 1994; Kinney et al., 1998; Land et al., 1991; Lang et al., 2005; Marcos et al., 2001; Millon et al., 2016; Narvaez et al., 2000; Narvaez et al., 1994; Runesson et al., 2009).

The potential physiological relevance of the N-terminal galanin fragments and their relative abundances in vivo cannot be inferred simply from the relative abundances found in the current study since they may simply reflect the binding affinity of the particular short N-terminal galanin fragments to specific galanin(1-16)-reactive immuno-globulins bound to the immunoaffinity column. As an example, we found galanin(1-15) in quite low abundance but it has been shown to exert a variety of biological functions (Borroto-Escuela et al., 2014; Diaz-Cabiale et al., 2000b; Diaz et al., 1996; Hedlund et al., 1994; Millon et al., 2016; Millon et al., 2014; Narvaez et al., 1994; Xu et al., 1999).

We demonstrated binding of several of the identified naturally-occurring short N-terminal galanin fragments at galanin receptors in the rat brain. We tested four peptides synthetized according the identified sequences from tissue extracts in the receptor displacement assay competing with 125I-labeled galanin(1-29). The autoradiographs of the coronal brain sections showed displacement of 125I-labeled gala-nin(1-29) with galanin(1-13), galanin(1-16) and galanin(1-20) respectively (Fig. 6 C-E). No displacement was observed with galanin(6-20) (Fig. 6 F). This peptide, however, is lacking N-terminal GWTLN sequence, thus our results are in agreement with previously published studies confirming the importance of these amino acids at the N-terminal end of galanin for the receptor binding (Berthold et al., 1997; Land et al., 1991).

Using mass spectrometry and bioinformatics analysis we also found two asparagine residues, particularly Asn37 and Asn50 within the amino acid sequence of several identified peptides deamidated. It is well known that asparagine residues in proteins and peptides can

-0.20 mm -0.80 mm -1.40 mm -2.00 mm -2.60 mm

Fig. 6. Autoradiographs of coronal sections of the rat brain showing galanin(1-29) binding sites visualized using 125I-labeled galanin(1-29) (A), receptor displacement of 125I-labeled galanin(1-29) with unlabeled galanin(1-29) used as negative control (B) and the tested peptides identified in the immunoreactive material, particularly galanin(1-13) (C), galanin(1-16) (D), and galanin(1-20) (E). Galanin(6-20) fragment was not able of displacement of galanin(1-29) from its receptors (F). The coordinates determine the position from Bregma.

coronal sections were cut using a cryostat and used in binding experiments (see below).

All animal procedures were approved by the Animal Care and Use Committee at Linkoping University and conformed to the standards in the Guide for the Care and Use of Laboratory Animals (Swedish Research Council).

4.2. Preparation of tissue extracts

Small intestine freshly obtained from four rats was washed in ice-cold phosphate buffer containing 0.04 mol/L Na2HPO4, 0.0096 mol/L NaH2PO4 and 0.000513 mol/L EDTA, pH7.4, supplemented with protease inhibitors (Roche Diagnostics Scandinavia AB, Stockholm, Sweden). The tissue was then transferred in 20 mL of boiling acetic acid (1 mmol/L), cut on fragments of about 5 mm long and boiled for 5 min.

After boiling, the tissue was homogenized using a polytron (CAT X520D, Scientific Industries, New York, NY, USA). The suspension was centrifuged at 2000 x g for 10 min at 4 °C and the supernatant was applied on a C-18 Sep-Pak cartridge (Merck AB, Stockholm, Sweden) preconditioned with 10 mL 100% acetonitrile (ACN, Sigma-Aldrich Sweden AB, Stockholm, Sweden) followed by 10 mL 0.1% trifluoroa-cetic acid (TFA, Merck AB, Stockholm, Sweden). Peptide fraction was eluted from the cartridges with 2 mL of 50% ACN supplemented with 0.1% TFA. The eluate was used in analytical reverse-phase HPLC. Identical procedure was used for extraction of peptides from bovine gut excepting of the size of cut tissue of about 15-20 mm long.

4.3. Reverse-phase HPLC of tissue extracts

The eluate from Sep-Pac was diluted with 4 volumes of 0.1% TFA

Fig. 6. (continued)

(buffer A). For the analysis, 5 mL of diluted eluate were transferred into a clean tube and the volume was reduced to 500 |iL using a SpeedVac (SC210A, Savant). The sample was then loaded on C-18 HPLC column (Genesis C18, 4 |im particle size, Jones Chromatography Ltd., Hengoed, Mid Glamorgan, UK). The tissue extracts from rat small intestine were separated by gradient elution from 0% to 50% ACN supplemented with 0.1% TFA in 60 min or gradient from 0 to 80% ACN with 0.1% TFA in 60 min in case of extracts from bovine gut. The separated material was collected in 60 fractions (1 mL/min), lyophilized and the immunor-eactivity of the collected fractions was tested in galanin(1-16)-specific RIA assay.

4.4. 125I labeling of rat galanin(1-16) and galanin(1-29)

Peptides used as a tracer in RIA assays as well as a ligand in receptor displacement test and autoradiography were labeled using chloramineT method. Ten micrograms of galanin(1-16) or galanin(1-29) (both

from Bachem, Bubendorf, Switzerland) were dissolved in 10 |iL of 0.25 mol/L phosphate buffer, pH 7.4 and mixed with 10 |iL (1 mCi) 125I (PerkinElmer, Groningen, Netherlands). The reaction mixture was supplemented with 10 |iL of 1.5 mg/mL chloramine T (Merck, AB, Stockholm, Sweden) in 0.25 mol/L phosphate buffer, pH 7.4, vortexed and incubated 15-20 s. The reaction was stopped by adding 10 |iL of 1.5 mg/mL sodium bisulphite (Merck AB, Stockholm, Sweden) in 0.25 mol/L phosphate buffer, pH 7.4. After mixing, 200 |iL of 0.02% BSA solution in phosphate buffer (0.25 mol/L, pH 7.4) were added and the labeled peptides were purified by reverse-phase HPLC on Nucleosil C18, 5 |im particle size, 4.6 x 300 mm column (Merck AB, Stockholm, Sweden). The fractions were eluted with linear gradient from 20% to 50% ACN during 40 min and collected in 40 fractions (1 mL/min). The fraction with specific activity by means of purified radioligand was determined using a gamma counter (Wizard 2, 2470, PerkinElmer) and used in experiments.

4.5. RIA assays

The lyophilized samples from reverse-phase HPLC were reconstituted in 0.1 mL of 0.05 mol/L phosphate buffer, pH 7.4 and the concentration of galanin(1-16)-like immunoreactivity was measured using antiserum K2 against galanin(1-16) raised in our laboratory, as described previously (38). 125I labeled galanin(1-16) was used as radioligand and unlabeled galanin(1-16) as calibrator with serial twofold dilutions starting from 8000 pmol/L to 7.8 pmol/L. The final dilution of antiserum was 1:12,500. After 48 h of incubation at 4 °C, 100 |iL of anti-rabbit IgG Sac-Cel (Immunodiagnostic Systems Limited, Boldon, UK) was added, vortexed and incubated for 30 min at room temperature. Immunoreactivity was measured using a gamma counter (Wizard 2, 2470, PerkinElmer).

4.6. Binding of galanin(1-16) to NHS-activated Sepharose

Rat galanin(1-16) (8.4 |Jg, Bachem, Bubendorf, Switzerland) was dissolved in 2 mL of coupling buffer containing 0.05 mol/L sodium phosphate, 0.15 mol/L NaCl, pH7.2 and added to 4 mL of NHS-Sepharose 4 Fast Flow (GE Healthcare Life Sciences, Uppsala, Sweden) activated with 1 mmol/L HCl and washed with coupling buffer instantly prior incubation with peptide. The peptide with activated medium was incubated for 18 h at 4 °C with gentle endover-end agitation. After incubation, the medium with bound peptide was separated by centrifugation (1000 x g, 1 min, 4 °C) and incubated with 100 mmol/L Tris-HCl, pH 7.5 for 2 h at 4 °C to block free NHS groups. After incubation, the medium with immobilized peptide was washed with 50 mmol/L acetate buffer, pH 4.5 and 100 mmol/L Tris-HCl, pH 7.5. The washing step was repeated 3-times to remove any unbound peptide. The medium with immobilized peptide was then used for purification of galanin(1-16)-specific IgG clones from antiserum K2.

4.7. Purification of specific immunoglobulins (IgG) reacting against galanin (1-16) from rabbit antiserum K2

The galanin(1-16)-specific IgGs were purified from rabbit antiserum K2 by immunoaffinity chromatography using a column (Econo-Pac, Bio-Rad Laboratories, Hercules, CA, USA) with synthetic rat galanin(1-16) (Bachem, Bubendorf, Switzerland) covalently immobilized on sepharose support via N-hydroxysuccinimide spacer, as described above. The antisera were diluted 1:1 with 5 mmol/L Tris-HCl, pH 7.5 (loading buffer) and loaded on the column with immobilized galanin(1-16) preconditioned with 10 bed volumes of loading buffer at flow rate 0.5 mL/min. After loading, the column was washed with 3 bed volumes of 50 mmol/L Tris-HCl, pH 7.5 containing 500 mmol/L NaCl and 0.1% Nonidet P-40 followed by wash with 3 bed volumes of loading buffer and the galanin specific immunoglobulins were eluted with 100 mmol/ L glycine, pH 2.7 (Supplemental Fig. 1). The eluate was instantly neutralized to pH 7.4 with 1.5 mol/L Tris-HCl. The specific IgGs were used for purification of immunoreactive compounds from tissue extracts of rat small intestine.

4.8. Affinity purification of immunoreactive compounds

The immunoreactive galanin compounds from tissue extracts were purified using a column with immobilized specific immunoglobulins reacting against galanin(1-16). To obtain sufficient quantity and purity of immunoreactive compounds, tissue extracts from 34 male Sprague Dawley rats, weighing 200-300 g were used in total. For each purification, 3-4 animals were sacrificed and extracts from small intestine were prepared using the method described above. The extracts were diluted 1:1 with 5 mmol/L Tris-HCl, pH 7.5 (loading buffer) and loaded on the column with immobilized galanin-specific immunoglobulins preconditioned with 10 bed volumes of loading buffer. After loading, the column was washed with 3 bed volumes of buffer containing 50 mmol/L Tris-

HCl, pH7.5 and 0.1% Nonidet P-40 to remove unspecific binding followed by wash with 3 bed volumes of loading buffer. Finally, the specifically bound compounds were eluted with 100 mmol/L glycine, pH 2.7. After elution, the column was equilibrated with at least 10 bed volumes of loading buffer and was ready for next purification. The eluates from each purification procedure were pooled, lyophilized and stored at — 80 °C until analysis.

4.9. Reverse-phase HPLC of affinity-purified immunoreactive compounds

The lyophilized material containing immunoreactive compounds from affinity chromatography was dissolved in 100 |iL of buffer containing 10% ACN and 0.1% TFA and loaded on C18 HPLC column (LiChrospher WP 300, RP-18, 5 |jm particle size, Merck AB, Stockholm, Sweden). The compounds were separated by gradient elution from 10% to 80% ACN in 80 min and collected in 1 ml/min fractions. The volume of 100 |iL from each fraction was transferred into a clean tube, lyophilized and used in RIA assay to detect the fractions containing the immunoreactive compounds. The rest of the volume of each fraction was kept at — 80 °C. Only the fractions showing immunoreactivity in RIA were further analyzed by mass spectrometry.

4.10. Sample preparation for mass spectrometry

The volume of 100 |iL of each HPLC fractions containing immunoreactive compounds was lyophilized and dissolved in 50 |iL of 40 mmol/ L NH4HCO3 (Sigma-Aldrich AB, Stockholm, Sweden) in 10% ACN for LC-MS (Sigma-Aldrich AB, Stockholm, Sweden). The samples were then digested with trypsin (0.1 |Jg, Trypsin Gold, Promega, Madison, WI, USA) and incubated for 16 h at 37 °C. The peptides were purified and concentrated using ZipTip C18 (Merck Millipore, Darmstadt, Germany) and eluted in 5 |iL 0.1% formic acid in 50% ACN.

4.11. LC-MS/MS

The purified peptides were loaded on an Easy-nanoLC II system coupled online to an LTQ Orbitrap Velos Pro hybrid mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA) and separated by reversephase chromatography on a 20 x 0.1 mm C18 pre-column followed by a 100 x 0.075 mm C18 analytical column, particle size 5 |jm (NanoSeparations, Nieuwkoop, Netherlands) at a flow rate 300 nL/ min using a gradient starting from 2% ACN with 0.1% formic acid (FA) to 100% ACN + 0.1% FA in 60 min. The Orbitrap mass spectrometer was operated in a positive mode with normalized collision energy of 30% in collision-induced dissociation (CID). Dynamic exclusion of sequenced peptides was set to 60 s with enabled predictive automatic gain control and charge state-filtering disqualifying singly charged peptides. The samples prepared from each fraction containing immu-noreactive material were analyzed in quadruplicates.

4.12. Database search

Raw files from the mass spectrometer were searched using PEAKS version 8.0 (Bioinformatics Solutions Inc., Waterloo, ON, Canada) against a Uniprot Rat 10116 database from 20140626 with the search parameters as follows: digestion enzyme trypsin with maximum number of missed cleavages 2; parent ion mass error tolerance 20.0 ppm, fragment ion mass error tolerance 0.5 and oxidation of methionine and amidation as variable modification.

4.13. Bioinformatics analysis

In order to discriminate those short N-terminal galanin sequences that might be generated by proteolytic cleavage by intestinal peptidases and may not have a physiological relevance, an in silico cleavage of the amino acid sequence of galanin prepropeptide (Fig. 5A) was performed

using PeptideCutter (http://web.expasy.org/peptide_cutter/, Swiss Institute of Bioinformatics) to find potential cleavage sites for known proteases. The list of tested proteases with at least one cleavage site within the amino acid sequence of galanin prepropeptide is shown in Table 3. The identified galanin sequences that started at such cleavage site were not further considered. The only exception were Arg32 and Arg52 as these two arginine residues are cleavage sites for trypsin required for mass spectrometry-based sequencing. The resulting list contained short N-terminal galanin sequences with potential physiological relevance (Table 2) and four of short N-terminal galanin sequences from the list, namely galanin(1-13), galanin(1-16), galanin(1-20), and galanin(6-20) were selected for peptide synthesis and receptor binding tests. The complete list of enzymes tested in PeptideCutter can be found in Supplemental Table 1 A.

4.14. PTM analysis

The MS/MS spectral data generated by mass spectrometer were searched using PTM algorithm built in PEAKS 8 software (Bioinformatics Solutions Inc., Waterloo, ON, Canada) and the results exported in MZID files with complementary MGF files containing the analyzed spectral data were loaded to Scaffold PTM (Proteome Software Inc., Portland, Oregon, USA). The PTM sites contained in MS/MS spectra were annotated using ASCORE's probabilistic approach and scoring algorithm (39) with filtering of data set at confidence level > 95% for the analyzed PTMs.

4.15. Solid-phase synthesis

Four of the identified short N-terminal galanin sequences, particularly of galanin(1-13), galanin(1-16), galanin(1-20) and galanin(6-20) were selected and synthetized by solid-phase synthesis (CASLO, Lyngby, Denmark). The peptides were purified by reverse-phase HPLC to > 95% purity. The amino acid composition and sequence of each peptide was confirmed by mass spectrometry.

4.16. Receptor displacement test and autoradiography

Male Sprague Dawley rats weighing 200-250 g were housed in standard cages with 12 h light/dark cycle with water and food available at libitum. Animals were anesthetized with 4% isoflurane and sacrificed. The brain was instantly removed, placed in a pre-chilled brain slicer and two coronary sections were performed in order to dissect the brain section containing the entire hypothalamus and surrounding brain structures. The dissected brain section was further cut in 14 |jm thick slices using a cryostat (CM3050, Leica Biosystems AB, Stockholm, Sweden), thaw-mounted on glass slides (4 slices/slide) and kept at — 20 °C in the cryostat during the whole procedure. The slides with sections were placed in an open polypropylene box and dried in a vacuum desiccator overnight. Dried brain sections were pre-incubated in buffer containing 50 mmol/L Tris-HCl, 5 mmol/L MgCl2, 2 mmol/L EGTA, pH 7.4 (Sigma-Aldrich AB, Stockholm, Sweden) and protease inhibitor cocktail (Roche Diagnostics Scandinavia AB, Stockholm, Sweden) for 30 min in order to remove unwanted tissue proteases. The sections were consequently incubated in the same buffer containing 0.05% BSA (incubation buffer) and 1.5 nmol/L 125I-labeled rat galanin (1-29) (Bachem) to determine total binding. Another set of brain slices were incubated in incubation buffer with either 1.5 nmol/L 125I-labeled rat galanin(1-29) and 1 ^mol/L of unlabeled galanin(1-29) to determine nonspecific binding. Finally, another sets of brain sections were co-incubated with 1.5 nmol/L 125I-labeled galanin(1-29) and 1 ^mol/L of synthetized galanin(1-13), galanin(1-16), galanin(1-20) and galanin(6-20), incubated separately. After incubation, the sections were washed two-times in ice-cold incubation buffer without radioligand for 5 min and briefly rinsed by dipping in ice-cold deionized water in order to remove salts present in the incubation buffer. The sections were dried

with gentle flow of cold air and placed in a vacuum desiccator for 1 h. The dried sections were placed in a cassette and exposed to sensitive film (Amersham Hyperfilm, GE Healthcare Life Sciences) for 4 days.

Conflict of interest

The authors declare that they have no potential conflict of interest with the contents of this article.

Author contributions

RI designed the study and concept, conducted the experiments, analyzed and interpreted the results and wrote the paper. ET contributed to the concept and read the manuscript. Both authors approved the final version of the manuscript.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.npep.2017.03.005.

References

Barreda-Gomez, G., Giralt, M.T., Rodriguez-Puertas, R., 2005a. Effects of central galanin administration on muscarinic cholinergic and galanin receptor G protein coupling. Neuropeptides 39, 157-160. Barreda-Gomez, G., Giralt, M.T., Rodriguez-Puertas, R., 2005b. G protein-coupled galanin receptor distribution in the rat central nervous system. Neuropeptides 39, 153-156. Barreda-Gomez, G., Giralt, M.T., Pazos, A., Rodriguez-Puertas, R., 2014. Galanin

activated Gi/o-proteins in human and rat central nervous systems. Neuropeptides 48, 295-304.

Beausoleil, S.A., Villen, J., Gerber, S.A., Rush, J., Gygi, S.P., 2006. A probability-based approach for high-throughput protein phosphorylation analysis and site localization. Nat. Biotechnol. 24, 1285-1292. Bedecs, K., Berthold, M., Bartfai, T., 1995a. Galanin—10 years with a neuroendocrine

peptide. Int. J. Biochem. Cell Biol. 27, 337-349. Bedecs, K., Langel, U., Bartfai, T., 1995b. Metabolism of galanin and galanin (1-16) in isolated cerebrospinal fluid and spinal cord membranes from rat. Neuropeptides 29, 137-143.

Berthold, M., Kahl, U., Jureus, A., Kask, K., Nordvall, G., Langel, U., Bartfai, T., 1997. Mutagenesis and ligand modification studies on galanin binding to its GTP-binding-protein-coupled receptor GalR1. Eur. J. Biochem./FEBS 249, 601-606. Borbely, E., Scheich, B., Helyes, Z., 2013. Neuropeptides in learning and memory.

Neuropeptides 47, 439-450. Borroto-Escuela, D.O., Narvaez, M., Di Palma, M., Calvo, F., Rodriguez, D., Millon, C., Carlsson, J., Agnati, L.F., Garriga, P., Diaz-Cabiale, Z., Fuxe, K., 2014. Preferential activation by galanin 1-15 fragment of the GalR1 protomer of a GalR1 -GalR2 heteroreceptor complex. Biochem. Biophys. Res. Commun. 452, 347-353. Burgevin, M.C., Loquet, I., Quarteronet, D., Habert-Ortoli, E., 1995. Cloning, pharmacological characterization, and anatomical distribution of a rat cDNA encoding for a galanin receptor. J. Mol. Neurosci.: MN 6, 33-41. Crawley, J.N., Austin, M.C., Fiske, S.M., Martin, B., Consolo, S., Berthold, M., Langel, U., Fisone, G., Bartfai, T., 1990. Activity of centrally administered galanin fragments on stimulation of feeding behavior and on galanin receptor binding in the rat hypothalamus. J. Neurosci. 10, 3695-3700. Diaz, Z., Narvaez, J.A., Hedlund, P.B., Aguirre, J.A., Gonzalez-Baron, S., Fuxe, K., 1996. Centrally infused galanin-(1-15) but not galanin-(1-29) reduces the baroreceptor reflex sensitivity in the rat. Brain Res. 741, 32-37. Diaz-Cabiale, Z., Narvaez, J.A., Marcos, P., Cordon, M.P., Covenas, R., Fuxe, K., GonzalezBaron, S., 1998. Galanin and NH2-terminal galanin fragments in central cardiovascular regulation. Ann. N. Y. Acad. Sci. 863, 421-424. Diaz-Cabiale, Z., Narvaez, J.A., Finnman, U.B., Bellido, I., Ogren, S.O., Fuxe, K., 2000a. Galanin-(1-16) modulates 5-HT1A receptors in the ventral limbic cortex of the rat. Neuroreport 11, 515-519. Diaz-Cabiale, Z., Narvaez, J.A., Yanaihara, N., Gonzalez-Baron, S., Fuxe, K., 2000b. Galanin/alpha2-receptor interactions in central cardiovascular control. Neuropharmacology 39, 1377-1385. Diaz-Cabiale, Z., Cordon, M.P., Covenas, R., Rivera, A., Yanaihara, N., Fuxe, K., GonzalezBaron, S., Narvaez, J.A., 2002. Propranolol blocks the tachycardia induced by galanin (1-15) but not by galanin (1-29). Regul. Pept. 107, 29-36. Diaz-Cabiale, Z., Parrado, C., Vela, C., Razani, H., Covenas, R., Fuxe, K., Narvaez, J.A., 2005. Role of galanin and galanin(1-15) on central cardiovascular control. Neuropeptides 39, 185-190. Diaz-Cabiale, Z., Parrado, C., Narvaez, M., Millon, C., Puigcerver, A., Fuxe, K., Narvaez, J.A., 2010. The galanin N-terminal fragment (1-15) interacts with neuropeptide Y in central cardiovascular control: Involvement of the NPY Y2 receptor subtype. Regul. Pept. 163, 130-136.

Evans, H.F., Shine, J., 1991. Human galanin: molecular cloning reveals a unique structure. Endocrinology 129, 1682-1684.

Evans, H., Baumgartner, M., Shine, J., Herzog, H., 1993. Genomic organization and localization of the gene encoding human preprogalanin. Genomics 18, 473-477.

Fathi, Z., Cunningham, A.M., Iben, L.G., Battaglino, P.B., Ward, S.A., Nichol, K.A., Pine, K.A., Wang, J., Goldstein, M.E., Iismaa, T.P., Zimanyi, I.A., 1997. Cloning, pharmacological characterization and distribution of a novel galanin receptor. Brain Res. Mol. Brain Res. 51, 49-59.

Fisone, G., Berthold, M., Bedecs, K., Unden, A., Bartfai, T., Bertorelli, R., Consolo, S., Crawley, J., Martin, B., Nilsson, S., et al., 1989. N-terminal galanin-(1-16) fragment is an agonist at the hippocampal galanin receptor. Proc. Natl. Acad. Sci. U. S. A. 86, 9588-9591.

Floren, A., Land, T., Langel, U., 2000. Galanin receptor subtypes and ligand binding. Neuropeptides 34, 331-337.

Fuxe, K., Jansson, A., Diaz-Cabiale, Z., Andersson, A., Tinner, B., Finnman, U.B., Misane, I., Razani, H., Wang, F.H., Agnati, L.F., Ogren, S.O., 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.

Geiger, T., Clarke, S., 1987. Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides. Succinimide-linked reactions that contribute to protein degradation. J. Biol. Chem. 262, 785-794.

Girotti, P., Bertorelli, R., Fisone, G., Land, T., Langel, U., Consolo, S., Bartfai, T., 1993. N-terminal galanin fragments inhibit the hippocampal release of acetylcholine in vivo. Brain Res. 612, 258-262.

Gu, Z.F., Pradhan, T.K., Coy, D.H., Jensen, R.T., 1995. Interaction of galanin fragments with galanin receptors on isolated smooth muscle cells from guinea pig stomach: identification of a novel galanin receptor subtype. J. Pharmacol. Exp. Ther. 272, 371-378.

Gustafson, E.L., Smith, K.E., Durkin, M.M., Gerald, C., Branchek, T.A., 1996. Distribution of a rat galanin receptor mRNA in rat brain. Neuroreport 7, 953-957.

Habert-Ortoli, E., Amiranoff, B., Loquet, I., Laburthe, M., Mayaux, J.F., 1994. Molecular cloning of a functional human galanin receptor. Proc. Natl. Acad. Sci. U. S. A. 91, 9780-9783.

Hedlund, P.B., Yanaihara, N., Fuxe, K., 1992. Evidence for specific N-terminal galanin fragment binding sites in the rat brain. Eur. J. Pharmacol. 224, 203-205.

Hedlund, P.B., Finnman, U.B., Yanaihara, N., Fuxe, K., 1994. Galanin-(1-15), but not galanin-(1-29), modulates 5-HT1A receptors in the dorsal hippocampus of the rat brain: possible existence of galanin receptor subtypes. Brain Res. 634, 163-167.

Hilke, S., Hokfelt, T., Theodorsson, E., 2006. A short estrogen-responsive N-terminal galanin homologue found in rat brain and gut with antiserum raised against rat galanin(1-16). Neurochem. Res. 31, 177-188.

Kinney, G.A., Emmerson, P.J., Miller, R.J., 1998. Galanin receptor-mediated inhibition of glutamate release in the arcuate nucleus of the hypothalamus. J. Neurosci. 18, 3489-3500.

Kinney, J.W., Starosta, G., Crawley, J.N., 2003. Central galanin administration blocks consolidation of spatial learning. Neurobiol. Learn. Mem. 80, 42-54.

Kormos, V., Gaszner, B., 2013. Role of neuropeptides in anxiety, stress, and depression: from animals to humans. Neuropeptides 47, 401-419.

Kyrkouli, S.E., Stanley, B.G., Seirafi, R.D., Leibowitz, S.F., 1990. Stimulation of feeding by galanin: anatomical localization and behavioral specificity of this peptide's effects in the brain. Peptides 11, 995-1001.

Land, T., Langel, U., Low, M., Berthold, M., Unden, A., Bartfai, T., 1991. Linear and cyclic N-terminal galanin fragments and analogs as ligands at the hypothalamic galanin receptor. Int. J. Pept. Protein Res. 38, 267-272.

Lang, R., Kofler, B., 2011. The galanin peptide family in inflammation. Neuropeptides 45, 1-8.

Lang, R., Berger, A., Santic, R., Geisberger, R., Hermann, A., Herzog, H., Kofler, B., 2005. Pharmacological and functional characterization of galanin-like peptide fragments as potent galanin receptor agonists. Neuropeptides 39, 179-184.

Lang, R., Gundlach, A.L., Holmes, F.E., Hobson, S.A., Wynick, D., Hokfelt, T., Kofler, B., 2015. Physiology, signaling, and pharmacology of galanin peptides and receptors: three decades of emerging diversity. Pharmacol. Rev. 67, 118-175.

Lu, X., Mazarati, A., Sanna, P., Shinmei, S., Bartfai, T., 2005. Distribution and differential regulation of galanin receptor subtypes in rat brain: effects of seizure activity. Neuropeptides 39, 147-152.

Ma, X., Tong, Y.G., Schmidt, R., Brown, W., Payza, K., Hodzic, L., Pou, C., Godbout, C., Hokfelt, T., Xu, Z.Q., 2001. Effects of galanin receptor agonists on locus coeruleus neurons. Brain Res. 919, 169-174.

Marcos, P., Diaz-Cabiale, Z., Cordon, M.P., Covenas, R., Yanaihara, N., Fuxe, K., GonzalezBaron, S., Narvaez, J.A., 2001. Central galanin and N-terminal galanin fragment induce c-Fos immunoreactivity in the medulla oblongata of the anesthetized rat.

Peptides 22, 1501-1509.

Mazarati, A.M., Baldwin, R.A., Shinmei, S., Sankar, R., 2005. In vivo interaction between serotonin and galanin receptors types 1 and 2 in the dorsal raphe: implication for limbic seizures. J. Neurochem. 95, 1495-1503.

Melander, T., Kohler, C., Nilsson, S., Hokfelt, T., Brodin, E., Theodorsson, E., Bartfai, T., 1988. Autoradiographic quantitation and anatomical mapping of 125I-galanin binding sites in the rat central nervous system. J. Chem. Neuroanat. 1, 213-233.

Mennicken, F., Hoffert, C., Pelletier, M., Ahmad, S., O'Donnell, D., 2002. Restricted distribution of galanin receptor 3 (GalR3) mRNA in the adult rat central nervous system. J. Chem. Neuroanat. 24, 257-268.

Millon, C., Flores-Burgess, A., Narvaez, M., Borroto-Escuela, D.O., Santin, L., Parrado, C., Narvaez, J.A., Fuxe, K., Diaz-Cabiale, Z., 2014. A role for galanin N-terminal fragment (1-15) in anxiety- and depression-related behaviors in rats. Int. J. Neuropharmacol. 18.

Millon, C., Flores-Burgess, A., Narvaez, M., Borroto-Escuela, D.O., Santin, L., Gago, B., Narvaez, J.A., Fuxe, K., Diaz-Cabiale, Z., 2016. Galanin (1-15) enhances the antidepressant effects of the 5-HT1A receptor agonist 8-OH-DPAT: involvement of the raphe-hippocampal 5-HT neuron system. Brain Struct. Funct. 221, 4491-4504.

Narvaez, J.A., Diaz, Z., Aguirre, J.A., Gonzalez-Baron, S., Yanaihara, N., Fuxe, K., Hedlund, P.B., 1994. Intracisternally injected galanin-(1-15) modulates the cardiovascular responses of galanin-(1-29) and the 5-HT1A receptor agonist 8-OH-DPAT. Eur. J. Pharmacol. 257, 257-265.

Narvaez, J.A., Diaz-Cabiale, Z., Hedlund, P.B., Aguirre, J.A., Covenas, R., Gonzalez-Baron, S., Fuxe, K., 2000. The galanin receptor antagonist M40 blocks the central cardiovascular actions of the galanin N-terminal fragment (1-15). Eur. J. Pharmacol. 399, 1,97-1203.

Ohtaki, T., Kumano, S., Ishibashi, Y., Ogi, K., Matsui, H., Harada, M., Kitada, C.,

Kurokawa, T., Onda, H., Fujino, M., 1999. Isolation and cDNA cloning of a novel galanin-like peptide (GALP) from porcine hypothalamus. J. Biol. Chem. 274, 37041-37045.

Parker, E.M., Izzarelli, D.G., Nowak, H.P., Mahle, C.D., Iben, L.G., Wang, J., Goldstein, M.E., 1995. Cloning and characterization of the rat GALR1 galanin receptor from Rin14B insulinoma cells. Brain Res. Mol. Brain Res. 34, 179-189.

Paxinos, G., Watson, C., 1998. The Rat Brain in Stereotaxic Coordinates, fourth ed. Academic Press.

Rezaei, K., Saar, K., Soomets, U., Valkna, A., Nasman, J., Zorko, M., Akerman, K.,

Schroeder, T., Bartfai, T., Langel, U., 2000. Role of third intracellular loop of galanin receptor type 1 in signal transduction. Neuropeptides 34, 25-31.

Robinson, N.E., Robinson, A.B., 2001. Molecular clocks. Proc. Natl. Acad. Sci. U. S. A. 98, 944-949.

Rossowski, W.J., Rossowski, T.M., Zacharia, S., Ertan, A., Coy, D.H., 1990. Galanin binding sites in rat gastric and jejunal smooth muscle membrane preparations. Peptides 11, 333-338.

Runesson, J., Saar, I., Lundstrom, L., Jarv, J., Langel, U., 2009. A novel GalR2-specific peptide agonist. Neuropeptides 43, 187-192.

Smith, K.E., Walker, M.W., Artymyshyn, R., Bard, J., Borowsky, B., Tamm, J.A., Yao, W.J., Vaysse, P.J., Branchek, T.A., Gerald, C., Jones, K.A., 1998. Cloned human and rat galanin GALR3 receptors. Pharmacology and activation of G-protein inwardly rectifying K+ channels. J. Biol. Chem. 273, 23321-23326.

Tatemoto, K., Rokaeus, A., Jornvall, H., McDonald, T.J., Mutt, V., 1983. Galanin - a novel biologically active peptide from porcine intestine. FEBS Lett. 164, 124-128.

Todd, J.F., Edwards, C.M., Ghatei, M.A., Bloom, S.R., 2000. The differential effects of galanin-(1-30) and -(3-30) on anterior pituitary hormone secretion in vivo in humans. American journal of physiology. Endocrinol. Metab. 278, E1060-E1066.

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.

Webling, K.E., Runesson, J., Bartfai, T., Langel, U., 2012. Galanin receptors and ligands. Front. Endocrinol. (Lausanne) 3, 146.

Wirz, S.A., Davis, C.N., Lu, X., Zal, T., Bartfai, T., 2005. Homodimerization and

internalization of galanin type 1 receptor in living CHO cells. Neuropeptides 39, 535-546.

Wynick, D., Smith, D.M., Ghatei, M., Akinsanya, K., Bhogal, R., Purkiss, P., Byfield, P., Yanaihara, N., Bloom, S.R., 1993. Characterization of a high-affinity galanin receptor in the rat anterior pituitary: absence of biological effect and reduced membrane binding of the antagonist M15 differentiate it from the brain/gut receptor. Proc. Natl. Acad. Sci. U. S. A. 90, 4231-4235.

Xu, Z.Q., Ma, X., Soomets, U., Langel, U., Hokfelt, T., 1999. Electrophysiological evidence for a hyperpolarizing, galanin (1-15)-selective receptor on hippocampal CA3 pyramidal neurons. Proc. Natl. Acad. Sci. U. S. A. 96, 14583-14587.