Cell Tissue Res DOI 10.1007/s00441-013-1690-y

REGULAR ARTICLE

Site-specific gene expression and localization of growth factor ligand receptors RET, GFRα1 and GFRα2 in human adult colon M. Barrenschee & M. Böttner & I. Hellwig & J. Harde & J. H. Egberts & T. Becker & T. Wedel

Received: 6 March 2013 / Accepted: 27 June 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Two of the glial-cell-line-derived neurotrophic factor (GDNF) family ligands (GFLs), namely GDNF and neurturin (NRTN), are essential neurotropic factors for enteric nerve cells. Signal transduction is mediated by a receptor complex composed of GDNF family receptor alpha 1 (GFRα1) for GDNF or GFRα2 for NRTN, together with the tyrosine kinase receptor RET (rearranged during transfection). As both factors and their receptors are crucial for enteric neuron survival, we assess the site-specific gene expression of these GFLs and their corresponding receptors in human adult colon. Full-thickness colonic specimens were obtained after partial colectomy for non-obstructing colorectal carcinoma. Samples were processed for immunohistochemistry and co-localization studies. Site-specific gene expression was determined by real-time quantitative polymerase chain reaction in enteric ganglia and in circular and longitudinal muscle harvested by microdissection. Protein expression of the receptors was mainly localized in the myenteric and submucosal plexus. Dual-label immunohistochemistry with PGP 9.5 as a pan-neuronal marker detected immunoreactivity of the receptors in neuronal somata and ganglionic neuropil. RET immunoreactivity co-localized M. Barrenschee and M. Böttner contributed equally to the study This work was supported by research grants from the German Research Society (Deutsche Forschungsgemeinschaft DFG WE 2366/4-2) and the Faculty of Medicine, University of Kiel (F347022). The authors disclose no competing interests. The funding sources had no role in the study design, the management of data, or the writing of the paper. M. Barrenschee (*) : M. Böttner : I. Hellwig : J. Harde : T. Wedel Institute of Anatomy, Christian-Albrechts-University of Kiel, Otto-Hahn-Platz 8, D-24118 Kiel, Germany e-mail: [email protected] J. H. Egberts : T. Becker Department of General and Thoracic Surgery, University Hospital Schleswig-Holstein Campus Kiel, Arnold-Heller-Straße 3, D-24105 Kiel, Germany

with neuronal GFRα1 and GFRα2 signals. The dominant source of receptor mRNA expression was in myenteric ganglia, whereas both GFLs showed higher expression in smooth muscle layers. The distribution and expression pattern of GDNF and NRTN and their corresponding receptors in the human adult enteric nervous system indicate a role of both GFLs not only in development but also in the maintenance of neurons in adulthood. The data also provide a basis for the assessment of disturbed signaling components of the GDNF and NRTN system in enteric neuropathies underlying disorders of gastrointestinal motility. Keywords RET . GFRα . GDNF . Enteric nervous system . Colon . Human Abbreviations ARTN Artemin ENS Enteric nervous system GDNF Glial-cell-line-derived neurotropic factor GFRα GDNF family receptor alpha GI Gastrointestinal HSCR Hirschsprung´s disease LCM Laser capture microdissection NRTN Neurturin PSPN Persephin RET Rearranged during transfection

Introduction Glial-cell-line-derived neurotropic factor (GDNF), originally isolated from cell culture supernatants of a glioblastoma cell line (Lin et al. 1993), is known to act as a potent neurotropic factor for a variety of neuronal cell populations including dopaminergic, noradrenergic and cholinergic neurons, spinal cord motor neurons and hippocampal neurons (Unsicker 1996). GDNF-induced signal transduction is

Cell Tissue Res

mediated by the tyrosine kinase receptor called rearranged during transfection (RET) and the GDNF family receptor alpha 1 (GFRα1), a glycosyl-phosphatidylinositol-anchored protein. GDNF is an eponym for a neurotrophic factor family, named GDNF family ligands (GFLs), which is composed of four members: GDNF, neurturin (NRTN), artemin (ARTN) and persephin (PSPN). As demonstrated for GDNF, each ligand binds to a glycosyl-phosphatidylinositol-anchored protein acting as a co-receptor, which binds the ligand with high affinity, subsequently activating RET. NRTN binds with high affinity to the GDNF family receptor alpha 2 (GFRα2), which in turn mediates the autophosphorylation and activation of the RET receptor (Airaksinen and Saarma 2002). The enteric nervous system (ENS) represents a neuronal network composed of ganglia and interconnecting nerve fibers innervating the entire gastrointestinal (GI) tract. This network is organized into two major nerve plexus, i.e., the myenteric plexus and the submucosal plexus (Wedel et al. 1999). The finding that the ENS is able to work autonomously from the brain and spinal cord has led to the term “brain of the gut” or “second brain” (Gershon 1999). The major functions exerted by the ENS include the regulation of intestinal peristalsis, mucosal blood flow and ion and water transport (Costa et al. 2000; Gershon 2010). Consequently, compromised ENS function might cause functional GI diseases characterized by enteric neuropathies, as described for Hirschsprung´s disease, slow transit constipation, or chronic intestinal pseudoobstruction (Knowles et al. 2009, 2010). The indispensable importance of the GFLs GDNF and NRTN and their receptors for the ENS became evident in mice ablated for GDNF, NRTN, or their corresponding receptors. Thus, Gdnf (−/−), Gfra1 (−/−), or Ret (−/−) mice exhibit complete aganglionosis distal from the esophagus, demonstrating the key role of these factors for the development of the ENS (Gianino et al. 2003; Young et al. 2000). In vitro studies have delineated the underlying GDNF-mediated mechanisms, such as the regulation of the migration, survival and differentiation of neural-crest-derived cells in the embryonic GI tract (Burzynski et al. 2009; De Giorgio and Camilleri 2004). In addition, studies on postnatal myenteric neurons have revealed the functions of the GDNF system on the regulation of neuronal survival and differentiation at later time points of ENS development (Rodrigues et al. 2011; Schafer and Mestres 1999). Mice lacking Gfra2 do not display aganglionosis but show retarded growth and deficits in cholinergic nerve fibers in the enteric and parasympathetic nervous systems (Rossi et al. 1999). Similarly, Nrtn-deficient mice exhibit reduced myenteric plexus innervation and reduced GI motility (Heuckeroth et al. 1999). Thus, GFLs and their corresponding receptors exert crucial functions in ENS development. Congruent with the data obtained from in vitro studies, the importance of GFLs and their receptors in the human gut has become obvious in studies of Hirschsprung´s disease (HSCR,

congenital megacolon), which causes severe constipation and intestinal obstruction (Wallace and Anderson 2011). The analysis of genetic defects has frequently demonstrated RET mutations in familiar forms of HSCR (Heanue and Pachnis 2007) characterized by an absence of enteric neurons in the submucosal and myenteric plexus (aganglionosis) of terminal gut regions (Amiel et al. 2008; Panza et al. 2012). Moreover, mutations in GFL genes (i.e., GDNF, NRTN, ARTN, or PSPN) have also been observed in HSCR patients and are thus thought to be involved in the pathogenesis of HSCR (Doray et al. 1998; Ruiz-Ferrer et al. 2011). Although the role of the GFLs GDNF and NRTN for the development of the ENS is well established, data concerning the localization of GDNF and NRTN and their corresponding receptors in the adult human colon are rare. Thus, our aim has been to systematically analyze the site-specific gene expression and localization of RET, GFRα1, GFRα2 and the corresponding GFLs GDNF and NRTN in the adult human colon.

Materials and methods Tissue source Segments of distal colon were obtained from patients (n=5, mean age: 68 y, 1 female, 4 males) who underwent partial colectomy for non-obstructing colorectal carcinoma. Anorectal evacuation and colonic motility disorders had previously been excluded. Full-thickness specimens were harvested at a safe distance (> 5 cm) from the tumor, thereby avoiding interference with the routine diagnostic workup by the Department of Pathology and ensuring retrieval of healthy control tissue. Specimens were immediately transferred from the operating theatre to the laboratory for tissue processing. The study of human tissue received approval from the Local Ethics Committee of the Faculty of Medicine, Christian-AlbrechtsUniversity of Kiel, Germany (B299/07). Laser capture microdissection and dissection of smooth muscle tissue As described previously (Bottner et al. 2010), full-thickness biopsies of the colonic wall were immediately frozen in isopentane and stored at −70 °C until use. Cryosections (14 μm) were placed on membrane-coated slides (polyethylene naphtalate, 1 μm; Zeiss MicroImaging, Göttingen, Germany) and regions of interest were visualized by ultra-rapid (ca. 30 s) staining with cresyl violet according to the manufacturer’s instructions (P.A.L.M. RNA Handling Protocols; Zeiss MicroImaging). Myenteric ganglia were identified by inverse light microscopy (Axiovert; Zeiss, Jena, Germany), excised by laser capture microdissection (LCM) and collected by

Cell Tissue Res

laser pressure catapulting (P.A.L.M. Microlaser Technologies, Bernried, Germany) in the cap of 0.5-ml reaction tubes. From each sample, 2 mm2 ganglionic tissue was collected, immediately dissolved in 200 μl RNA lysis buffer (PEQLAB, Erlangen, Germany) and stored at −70 °C. Longitudinal and circular muscle tissue was collected under stereomicroscopic control from cryosections (20 μm) excluding myenteric ganglia by carefully dissecting the tissue with a scalpel. For each sample, the tissue of six cryosections were collected, immediately dissolved in 200 μl RNA lysis buffer (PEQLAB) and stored at −70 °C. RNA extraction and reverse transcription Extraction of total RNA from myenteric ganglia and from circular and longitudinal muscle was performed by using the peqGOLD MicroSpin Total RNA Kit (PEQLAB) according to the manufacturer’s instructions. RNA was eluted in a volume of 15 μl H2O. Genomic DNA was digested for 15 min at room temperature by using 1.5 U DNAse I (Sigma-Aldrich, Munich, Germany). Reverse transcription was carried out in a total volume of 30 μl containing 375 ng random hexamer primer (GE Healthcare, Freiburg, Germany), 0.5 mM dNTPs (Promega, Mannheim, Germany), 0.01 M dithiothreitole, 1 × reaction buffer and 150 U Superscript II Reverse Transcriptase (Invitrogen, Karlsruhe, Germany). The annealing, elongation and denaturation steps were carried out at 25 °C for 10 min, at 42 °C for 50 min and at 70 °C for 15 min, respectively. Real-time quantitative polymerase chain rection Real-time quantitative polymerase chain rections (qPCR) were performed in 96-well plates in duplicate reactions. Each reaction (20 μl) contained 2 μl total cDNA, 900 nM primers, 225 nM hybridization probe and 1 × qPCR Master Mix Plus (Eurogentec, Cologne, Germany). qPCR product accumulation was monitored by an ABI Prism 7700 Sequence Detection System (TaqMan, Applied Biosystems, Calif., USA) over 45 cycles. Each cycle contained a denaturation phase of 15 s at 95 °C and a hybridization/elongation phase of 1 min at 60 °C. Monitoring of NRTN gene expression was performed by using the TaqMan Gene Expression Assay for human NRTN according to the manufactor’s instructions (Applied Biosystems, Foster City, Calif., USA). Other primers and probes are listed below: GFRA1, forward primer 5′-tcgggcaatacacacctctgt−3′, reverse primer 5′-cttggaggagcagccattga−3′, probe 5′tgaaaaagaaggtctcggtgcttcc−3′; GFRA2, forward primer 5′-tccc agggagtaacaaggtgatc−3′, reverse primer 5′-tacaaggccagtttc agcatcag−3′, probe 5′-ccccagcagagccagaccgtcg−3′; RET, forward primer 5′-aaggagatggcaaagggatcac−3′, reverse primer 5′ttgatgtcttgggtctccacaa−3′, probe 5′-aggaacttctccacctgctctccc3′; GDNF, forward primer 5′-tgaaaccaaggaggaactgatttt−3′, reverse primer 5′-gtcactcaccagccttctatttctg−3′, probe 5′-tactgc agcggctcttgcgatgcag−3′; HPRT (hypoxanthine phosphoribosyl

transferase, a house-keeping gene), forward primer 5′-tgaa cgtcttgctcgagatgtg-3′, reverse primer 5′-ccagcaggtcagcaaag aattt-3′, probe 5′-tgggaggccatcacattgtagcc-3′. The data were normalized to expression levels of the housekeeping gene HPRT that did not differ significantly between the tissue sources analyzed (data not shown). Data were expressed as relative mRNA expression and presented as mean ± SEM. Tissue processing and conventional immunohistochemistry After surgical removal, all specimens were transferred into PBS (phosphate-buffered saline, pH 7.2) at 37 °C to allow adaption and further dissection. Full-thickness rectangular tissue blocks (30 mm × 10 mm) were pinned out flat on a cork plate by fine needles with neither artificial stretching nor shortening, thereby preserving the original size. The longer border of the tissue block was oriented perpendicular to the gut axis and corresponded to the cutting surface for histologic sections, so that myocytes of the circular muscle layer were cut along their longitudinal axis. After fixation (4 % paraformaldehyde in PBS) for 24 h and dehydration, tissue blocks were transferred into paraffin wax and cut into sections (6 μm) for immunohistochemistry. Immunoreactive signals were visualized by using the avidin-biotin-complex system (VECTASTAIN Elite ABC Kit; Vector Laboratories, Burlingame, USA). Briefly, sections were incubated with 3 % hydrogen peroxide to block endogenous peroxidase activity, rinsed in TBS-buffer (TRIS-buffered saline: 10 mM TRIS, 50 mM NaCl, pH 7.4) and pretreated with citrate buffer (pH 6.0, 95 °C water bath, 20 min). Thereafter, samples were incubated overnight with a monoclonal mouse-anti-RET antibody (Ret01 clone 3 F8; 1:200; Imgenex, San Diego, USA), a polyclonal rabbit-anti-GFRα1 antibody (1:1000; antibodiesonline.com, Aachen, Germany), or a polyclonal rabbit-antiGFRα2 antibody (1:200; Abcam, Cambridge, UK) diluted in antibody diluent (Invitrogen). Sections were further incubated for 45 min with biotinylated goat anti-mouse IgG (1:400; DAKO, Hamburg, Germany) for sections previously incubated with mouse-anti-RET or with biotinylated goat anti-rabbit IgG (1:400; DAKO) for sections previously incubated with rabbit-anti-GFRα1 or rabbit-anti-GFRα2. After being washed three times with TBS, sections were incubated for 45 min with ABC conjugated with horseradish peroxidase. We used 3, 3’diaminobenzidine (DAKO) as a substrate chromogen. Sections were counterstained with Meyer’s hematoxylin. Omission of the primary or secondary antibody served as negative controls. Moreover, specificity of antibodies against GFRα1 and GFRα2 was verified with blocking peptide experiments. Blocking peptides for GFRα1 (antibodies-online.com) and GFRα2 (Abcam) were incubated in antibody diluent together with either GFRα1 or GFRα2 antibody for 30 min at 37 °C followed by the procedure described above. Blocking peptides almost completely abolished immunoreactive signals leaving

Cell Tissue Res

only a faint background (data not shown). According to the manufacturer, RET monoclonal antibody was made against a recombinant protein and hence a blocking peptide was not available. However, the specific staining pattern of RET clone01 has been described previously in the literature (Cerilli et al. 2002; Powers et al. 2003). Dual-label immunohistochemistry Paraffin-embedded tissue sections were pre-treated with citrate buffer (pH 6.0, 95 °C water bath) for 25 min followed by overnight incubation with either mouse-anti-RET (1:500; Imgenex), rabbit-anti-GFRα1 (1:500; antibodies-online.com), or rabbit-anti-GFRα2 antibody (1:200; Abcam) diluted in antibody diluent (Invitrogen) as primary antibodies. After being washed with TBS, sections were incubated with either a goat anti-rabbit AlexaFluor488 antibody or a goat-anti-mouse AlexaFluor488 antibody, diluted in antibody diluent (1:250, Invitrogen) as secondary antibodies for 2 h at room temperature. To visualize enteric neurons, sections were co-incubated with an antibody against mouse PGP 9.5 (1:1000; UCL, Wellow, UK) or rabbit PGP 9.5 (1:1000; Acris, Herford, Germany) depending on the primary antibody used. Secondary antibody incubation was carried out with a goat-anti-mouse AlexaFluor546 antibody or a goat anti-rabbit AlexaFluor546 antibody, respectively (1:250; Invitrogen) for 2 h at room temperature. For dual-labeling of receptors, mouse-anti-RET antibody, rabbit-anti-GFRα1 and rabbit-anti-GFRα2 antibody were used as primary antibodies. Goat anti-rabbit AlexaFluor488 and antimouse AlexaFluor546 were used as secondary antibodies. Finally, all sections were stained with 4,6-diamidino-2phenylindole (DAPI; Roche, Mannheim, Germany) to visualize cellular nuclei. Fluorescence signals were detected with a fluorescence microscope (Axiovert 200 M; Zeiss, Göttingen, Germany) coupled to a digital camera (Axiocam; Zeiss). Data were analyzed with Axiovision software (Zeiss). Overlays of immunoreactive signals were processed by using a co-localization tool (Zeiss).

Results Site-specific gene expression of RET, GFRA1 and GFRA2 and their corresponding GFLs GDNF and NRTN Site-specific mRNA expression of RET, GFRA1, GFRA2 and the GFLs GDNF and NRTN was detected by real-time qPCR of RNA isolated from myenteric ganglia and from circular muscle and longitudinal muscle (Fig. 1). Analysis of mRNA expression profiles revealed higher levels of RET (Fig. 1a), GFRA1 (Fig. 1b) and GFRA2 (Fig. 1c) in myenteric ganglia compared with the muscle layers, whereas GDNF and NRTN exhibited more pronounced expression in both muscle layers compared with myenteric ganglia (Fig. 1d, e). The

mRNA expression level of GFRA1 was up to two-fold higher in myenteric ganglia compared with the expression level in the circular and longitudinal muscle (Fig. 1b). Similar results were obtained for GFRA2 mRNA expression (Fig. 1c). RET expression levels in myenteric ganglia were also significantly higher, with about 18-fold increased expression compared with that of the circular muscle and nearly 21-fold higher expression compared with that of the longitudinal muscle (Fig. 1a).

Localization of RET, GFRα1 and GFRα2 Localization of RET, GFRα1 and GFRα2 was determined by immunohistochemistry applied to colonic full-thickness sections. Consistent with the findings of gene expression studies, RET immunoreactivity was detected in ganglia of the myenteric (Fig. 2a) and submucosal (Fig. 2b) plexus. No apparent RET-immunoreactive signals were found in the muscle layers. RET immunoreactivity showed strongest signals in most of the neuronal somata, whereas the neuropil surrounding the ganglion cells was stained faintly. GFRα1 immunoreactivity was detected in both the myenteric (Fig. 2c) and the submucosal (Fig. 2d) plexus. Faint granular staining was observed throughout the ganglia. In contrast to RETimmunoreactive signals, weak GFRα1-immunoreactive signals were also observed in the circular and longitudinal muscle layers. As demonstrated for GFRα1 and RET immunoreactivity, GFRα2 immunoreactivity was also found in both the myenteric (Fig. 2e) and the submucosal (Fig. 2f) plexus. Immunoreactive signals were detected in neuronal somata and to a smaller extent throughout the ganglionic neuropil. GFRα2 immunoreactivity also displayed faint signals in the circular and longitudinal muscle layers. To confirm RET, GFRα1 and GFRα2 expression in enteric nerve cells, we performed dual-label immunocytochemistry with PGP 9.5 as a pan-neuronal marker for visualization of neuronal somata and processes. RET-immunoreactive signals were detected in the myenteric plexus (Fig. 3a-c) and in the submucosal plexus (Fig. 3d-f). Not all PGP-9.5-positive neurons exhibited RET immunoreactivity but if they were RETpositive, the immunoreactive signals revealed robust staining of neuronal somata. The surrounding neuropil displayed weaker immunoreactive signals. GFRα1-immunoreactive signals were also found in both the myenteric (Fig. 4a-c) and submucosal (Fig. 4d-f) plexus and as for RET, some PGP-9.5-positive neurons lacked GFRα1 immunoreactivity. When neurons showed GFRα1-positive immunoreacivity, GFRα1 displayed strong immunoreactive signals in neuronal somata and weaker immunoreactivity in the neuropil. GFRα2-immunoreactive signals were observed in the myenteric (Fig. 5a-c) and submucosal (Fig. 5d-f) plexus. GFRα2 displayed strong granular immunoreactivity in neuronal somata and weaker immunoreactivity in the neuropil.

Cell Tissue Res

Fig. 1 Site-specific mRNA expression of RET (rearranged during transfection), GFRA1 (GDNF family receptor alpha 1), GFRA2, GDNF (glial-cell-line-derived neurotropic factor) and NRTN (neurturin) in myenteric ganglia and muscle layers of the adult human colon. Analysis of site-specific mRNA expression profiles in microdissected myenteric ganglia (MP), circular muscle (CM) and longitudinal muscle (LM) reveals myenteric ganglia as the main source of all three receptors,

whereas GDNF and NRTN are mainly expressed in the colonic musculature. Levels of mRNA were determined by real-time quantitative polymerase chain reaction and expression of target genes was normalized to mRNA expression of the house-keeping gene HPRT (hypoxanthine phosphoribosyl transferase). Data are shown as means ± SEM, n=5-7 per experimental group

Co-localization of RET with GFRα1 and GFRα2

Dual-label immunohistochemistry with PGP 9.5 as a panneuronal marker has detected strong immunoreactive signals for RET, GFRα1 and GFRα2 in neuronal somata and weaker staining in the ganglionic neuropil. Co-localization studies of RET with either GFRα1 or GFRα2 has revealed the main location of these receptors within neuronal somata.

To investigate the distribution pattern of RET, GFRα1 and GFRα2 within enteric ganglia, co-localization studies of RET with GFRα1 and GFRα2 were performed for the myenteric plexus. RET was co-localized with GFRα1 (Fig. 6a-c) and with GFRα2 (Fig. 6d-f). For all GFL receptors, the strongest immunoreactive signals were detected in neuronal somata. However, whereas RET immunoreactivity decorated homogeneously the entire neuronal somata, both GFRα1- and GFRα2-immunoreactive signals were restricted to confined pericaryal regions and displayed a punctuate staining pattern.

Discussion The present study demonstrates the expression and localization of RET, GFRα1, GFRα2 and their corresponding GFLs GDNF and NRTN in the adult human colon. Site-specific mRNA expression of RET, GFRA1 and GFRA2 has been detected in myenteric ganglia and in the circular and longitudinal muscle, with the highest expression levels occurring in the enteric ganglia, whereas the main source of GDNF and NRTN has been identified in the colonic muscle layers. Confirmatory to the gene expression studies, the localization of RET, GFRα1 and GFRα2 has been determined by immunohistochemistry in the myenteric and the submucosal plexus.

mRNA expression and localization of RET, GFRA1, GFRA2 and their corresponding GFLs, GDNF and NRTN in the adult human colon The site-specific expression of mRNA for RET, GFRA1 and GFRA2 isolated from myenteric ganglia and from circular and longitudinal muscle of the adult human colon has revealed the enteric ganglia as the dominant source for the GFL receptors, whereas both GFLs are mainly expressed in the muscle layers. This observation is in agreement with previous studies that have found RET and GFRA1 to be expressed in ganglia of the mature ENS in animals and of the ENS in infant colonic tissue by using in situ hybridization (Golden et al. 1999; Wartiovaara et al. 1998). Further, GFRA2 has previously been found to be expressed in enteric neurons of the adult mouse GI tract (Golden et al. 1999). In addition, our observation of the major mRNA expression of GDNF in the smooth muscle layers of the human adult colon is confirmatory with regard to investigations of the role of smooth muscle cells in rat postnatal ENS performed by

Cell Tissue Res Fig. 2 Localization of RET, GFRα1 and GFRα2 in enteric ganglia of the adult human colon. RET-immunopositive signals were identified in myenteric (a) and submucosal (b) ganglia; robust staining was observed in neuronal somata and weaker signals in the surrounding neuropil. GFRα1 immunoreactivity was detected in myenteric (c) and submucosal (d) ganglia with granular staining being confined to neuronal somata and a faint staining of the neuropil. GFRα1 immunoreactivitiy was also observed to a minor extent within muscle layers (c). GFRα2 immunoreactivity was detected in myenteric (e) and submucosal (f) ganglia. Immunoreactive signals were mainly observed in neuronal somata but also decorated the ganglionic neuropil. GFRα2 also displayed faint signals in the circular and longitudinal muscle layers. Hematoxylin counterstaining. Bars 20 μm

Rodrigues et al. (2011) who identified intestinal smooth muscle as the dominant source of GDNF in vitro. Moreover, NRTN has recently been found to be expressed in the muscularis externa of the GI tract of rats at 6 and 24 months of age (Korsak et al. 2012). The distinct spatial complementary expression pattern of GFLs and their receptors in the human adult colon suggests that the “neurotrophic factor concept” originally postulated for the central nervous system might also apply to the ENS. This concept in its basic form states that innervated tissues produce a signal specifically directed toward the innervating neurons, which in turn express receptors for these targetderived neurotrophic factors for selective limitation of neuronal death (Korsching 1993). Translated into the present context, this would mean that the intestinal musculature (innervated tissue) produces GDNF/NRTN to act on their corresponding receptors (GFRα1/RET or GFRα2/RET), which are expressed by enteric ganglia (nerves) to ensure maintenance and survival during adulthood.

We have shown that RET mRNA expression levels are considerably higher in comparison with levels of GFRA1 or GFRA2 expression, an observation that has been confirmed by co-localization studies of RET, GFRα1 and GFRα2. In the myenteric ganglia, all RET-positive neurons are colocalized either with GFRα1 or GFRα2 but whereas RET immunoreactivity decorates homogeneously the entire neuronal somata, GFRα1 and GFRα2 display only punctuate immunoreactive signals. Both the dominant mRNA expression levels and abundant distribution of RET compared with GFRα1 and GFRα2 can be explained by both the GDNFGFRα1 and NRTN-GFRα2 pathways using RET for signal transduction, as do other GFLs (e.g., ARTN) expressed in the ENS (Maruccio et al. 2008). Moreover, RET is also known to be able to signal independently of GFLs (Tsui-Pierchala et al. 2002) and vice versa (Trupp et al. 1999). In addition to the mRNA expression of RET, GFRA1 and GFRA2 in myenteric ganglia, we have found detectable amounts of GFRA1 and GFRA2 mRNA and moderate

Cell Tissue Res

Fig. 3 Co-localization of RET with the pan-neuronal marker PGP 9.5 in enteric ganglia of the adult human colon. RET (green) was localized in both myenteric (a–c) and submucosal (d–f) ganglia. RET-immunoreactive signals co-localized with PGP-9.5-immunoreactive (red)

signals (merged in c, f, yellow) giving evidence that RET was expressed by enteric nerve cells (blue DAPI [4,6-diamidino-2-phenylindole] staining of nuclei). Bars 20 μm

immunoreactive signals in the muscular layers of the human adult colon. The immunoreactive signals can be explained by target-derived soluble GFRα1. Ledda et al. (2002) have

demonstrated that soluble forms of GFRα1 (sGFRα1) are produced by targets of c-Ret-expressing sensory and sympathetic neurons, which are able to stimulate neurite outgrowth

Fig. 4 Co-localization of GFRα1 with the pan-neuronal marker PGP 9.5 in enteric ganglia of the adult human colon. GFRα1 (green) was localized in both myenteric (a–c) and submucosal (d–f) ganglia. GFRα1-immunoreactive signals co-localized with PGP-9.5-immunoreactive (red) signals

(merged in c, f, yellow) giving evidence that GFRα1 was expressed by enteric nerve cells. GFRα1 immunoreactivity was stronger in neuronal somata and weaker in the surrounding neuropil (blue DAPI staining of nuclei). Bars 20 μm

Cell Tissue Res

Fig. 5 Co-localization of GFRα2 with the pan-neuronal marker PGP 9.5 in enteric ganglia of the adult human colon. GFRα2 (green) was localized in both myenteric (a–c) and submucosal (d–f) ganglia. GFRα1-immunoreactive signals co-localized with PGP-9.5-immunoreactive (red) signals

(merged in c, f, yellow) giving evidence that GFRα1 was expressed by enteric nerve cells. GFRα1 immunoreactivity was stronger in neuronal somata and weaker in the surrounding neuropil (blue DAPI staining of nuclei). Bars 20 μm

in the presence of GDNF. Thus, enteric smooth muscle cells that are known to express GDNF (Nosrat et al. 1997; Rodrigues et al. 2011) might harbor sGFRα1, thereby

contributing to enteric neurite outgrowth. However, the presence of GFRA1 or GFRA2 mRNA within muscle layers cannot be explained by the soluble form of the receptor.

Fig. 6 Co-localization of RET with GFRα1 and GFRα2 in myenteric ganglia of the adult human colon. RET (red) co-localizes with GFRα1 (green, a–c) and GFRα2 (green, d–f). Co-localization is depicted in merged images for RET with GFRα1 (yellow, c) and for RET with

GFRα2 (yellow, f). Whereas RET immunoreactivity is homogeneously distributed within neuronal somata, both GFRα1- and GFRα2-immunoreactive signals are limited to confined pericaryal regions displaying a punctuate staining pattern (blue DAPI staining of nuclei). Bars 20 μm

Cell Tissue Res

Therefore, muscle cells themselves probably express minor amounts of GLF receptors, as shown previously for GFRA1 in human striated muscle (Yamamoto et al. 1999). Relevance of RET, GFRα1 and GFRα2 for peripheral/ enteric neuropathies RET, GFRα1 and GFRα2 have been demonstrated in several nerve tissues, e.g., dorsal root ganglia (Golden et al. 1999; Josephson et al. 2001), Schwann cells (Hase et al. 2005; Hoke et al. 2002), anterior horn of the spinal cord (Mitsuma et al. 1999) and ventral midbrain (Golden et al. 1999) and the abnormal expression pattern of these receptors affecting GFL/GFRα/RET signaling is associated with distinct neuropathies in humans, e.g., activating point mutations in RET cause multiple endocrine neoplasia type 2 (Mologni 2011; Takahashi 2001) or familial medullary thyroid carcinoma (Wells and Santoro 2009). In addition, germ-line and inactivating mutations of RET or GDNF lead to complete absence of enteric ganglia (aganglionosis) causative for HSCR (Arighi et al. 2005; Gershon 2010). Mutations in other GFL genes, e.g., NRTN, ARTN, or PSPN, have also been reported in HSCR patients and are thus thought to be involved in the pathogenesis of HSCR (Doray et al. 1998; Ruiz-Ferrer et al. 2011). Moreover, as has recently been suggested, diminished GFRA1 or RET expression directly affects the survival of enteric neurons (Uesaka et al. 2007, 2008). Despite the crucial role of the GDNF/NRTN system for the development and maintenance of the ENS, limited data are available on the localization and distribution of the GFL receptors RET, GFRα1 and GFRα2 in the GI tract. Most studies have focused on animal models (Chalazonitis et al. 1998; Golden et al. 1999; Heanue and Pachnis 2007; Heuckeroth et al. 1999; Rossi et al. 1999) or isolated cell cultures (Rodrigues et al. 2011; Rossi et al. 1999; Taraviras et al. 1999). Studies of the GDNF/NRTN system in the human GI tract have been confined either to GI tissue collected from infants (Karim et al. 2006; Martucciello et al. 1995) or to pathologic conditions, e.g., HSCR (Arighi et al. 2005; Doray et al. 1998; Rusinek et al. 2011; Takahashi 2001). Concluding remarks In this study, we have demonstrated, for the first time, the sitespecific gene expression and distributional pattern of RET, GFRα1 and GFRa2 with their corresponding GFLs GDNF and NRTN in the human adult colon. As GDNF/GFRα1-RET and NRTN/GFRα2-RET pathways are crucial for the development, survival and maintenance of enteric neurons, we assume that distinct alterations of these pathways not only lead to HSCR (complete aganglionosis) but might also be associated with other enteric neuropathies characterized by a partial loss of enteric neurons (hypoganglionosis), e.g., slow-

transit constipation or diverticular disease (De Giorgio and Camilleri 2004; Wedel et al. 2010). The data of this study provide a basis to investigate further whether altered GDNF/GFRα1-RET or NRTN/GFRα2-RET pathways are linked to a spectrum of enteric neuropathies in which neuronal degeneration or shortfall has taken place. Acknowledgements The authors thank Karin Stengel, Inka Geurink, Bettina Facompré and Clemens Franke (Institute of Anatomy, ChristianAlbrechts-University of Kiel) for their excellent technical assistance.

References Airaksinen MS, Saarma M (2002) The GDNF family: signalling, biological functions and therapeutic value. Nat Rev Neurosci 3:383–394 Amiel J, Sproat-Emison E, Garcia-Barcelo M, Lantieri F, Burzynski G, Borrego S, Pelet A, Arnold S, Miao X, Griseri P, Brooks AS, Antinolo G, de Pontual L, Clement-Ziza M, Munnich A, Kashuk C, West K, Wong KK, Lyonnet S, Chakravarti A, Tam PK, Ceccherini I, Hofstra RM, Fernandez R (2008) Hirschsprung disease, associated syndromes and genetics: a review. J Med Genet 45:1–14 Arighi E, Borrello MG, Sariola H (2005) RET tyrosine kinase signaling in development and cancer. Cytokine Growth Factor Rev 16:441–467 Bottner M, Bar F, Von Koschitzky H, Tafazzoli K, Roblick UJ, Bruch HP, Wedel T (2010) Laser microdissection as a new tool to investigate site-specific gene expression in enteric ganglia of the human intestine. Neurogastroenterol Motil 22:e152 Burzynski G, Shepherd IT, Enomoto H (2009) Genetic model system studies of the development of the enteric nervous system, gut motility and Hirschsprung’s disease. Neurogastroenterol Motil 21:113–127 Cerilli LA, Mills SE, Rumpel CA, Dudley TH, Moskaluk CA (2002) Interpretation of RET immunostaining in follicular lesions of the thyroid. Am J Clin Pathol 118:186–193 Chalazonitis A, Rothman TP, Chen J, Vinson EN, MacLennan AJ, Gershon MD (1998) Promotion of the development of enteric neurons and glia by neuropoietic cytokines: interactions with neurotrophin-3. Dev Biol 198:343–365 Costa M, Brookes SJ, Hennig GW (2000) Anatomy and physiology of the enteric nervous system. Gut 47 (Suppl 4):iv15-iv26 De Giorgio R, Camilleri M (2004) Human enteric neuropathies: morphology and molecular pathology. Neurogastroenterol Motil 16:515–531 Doray B, Salomon R, Amiel J, Pelet A, Touraine R, Billaud M, Attie T, Bachy B, Munnich A, Lyonnet S (1998) Mutation of the RET ligand, neurturin, supports multigenic inheritance in Hirschsprung disease. Hum Mol Genet 7:1449–1452 Gershon MD (1999) The enteric nervous system: a second brain. Hosp Pract (Minneap) 34:31–32 Gershon MD (2010) Developmental determinants of the independence and complexity of the enteric nervous system. Trends Neurosci 33:446–456 Gianino S, Grider JR, Cresswell J, Enomoto H, Heuckeroth RO (2003) GDNF availability determines enteric neuron number by controlling precursor proliferation. Development 130:2187–2198 Golden JP, DeMaro JA, Osborne PA, Milbrandt J, Johnson EM Jr (1999) Expression of neurturin, GDNF, and GDNF family-receptor mRNA in the developing and mature mouse. Exp Neurol 158:504–528 Hase A, Saito F, Yamada H, Arai K, Shimizu T, Matsumura K (2005) Characterization of glial cell line-derived neurotrophic factor family receptor alpha-1 in peripheral nerve Schwann cells. J Neurochem 95:537–543

Cell Tissue Res Heanue TA, Pachnis V (2007) Enteric nervous system development and Hirschsprung’s disease: advances in genetic and stem cell studies. Nat Rev Neurosci 8:466–479 Heuckeroth RO, Enomoto H, Grider JR, Golden JP, Hanke JA, Jackman A, Molliver DC, Bardgett ME, Snider WD, Johnson EM Jr, Milbrandt J (1999) Gene targeting reveals a critical role for neurturin in the development and maintenance of enteric, sensory, and parasympathetic neurons. Neuron 22:253–263 Hoke A, Gordon T, Zochodne DW, Sulaiman OA (2002) A decline in glial cell-line-derived neurotrophic factor expression is associated with impaired regeneration after long-term Schwann cell denervation. Exp Neurol 173:77–85 Josephson A, Widenfalk J, Trifunovski A, Widmer HR, Olson L, Spenger C (2001) GDNF and NGF family members and receptors in human fetal and adult spinal cord and dorsal root ganglia. J Comp Neurol 440:204–217 Karim S, Hession C, Marconi S, Gang DL, Otis CN (2006) The identification of ganglion cells in Hirschsprung disease by the immunohistochemical detection of Ret oncoprotein. Am J Clin Pathol 126:49–54 Knowles CH, De Giorgio R, Kapur RP, Bruder E, Farrugia G, Geboes K, Gershon MD, Hutson J, Lindberg G, Martin JE, Meier-Ruge WA, Milla PJ, Smith VV, Vandervinden JM, Veress B, Wedel T (2009) Gastrointestinal neuromuscular pathology: guidelines for histological techniques and reporting on behalf of the Gastro 2009 International Working Group. Acta Neuropathol 118:271–301 Knowles CH, De Giorgio R, Kapur RP, Bruder E, Farrugia G, Geboes K, Lindberg G, Martin JE, Meier-Ruge WA, Milla PJ, Smith VV, Vandervinden JM, Veress B, Wedel T (2010) The London Classification of gastrointestinal neuromuscular pathology: report on behalf of the Gastro 2009 International Working Group. Gut 59:882–887 Korsak K, Dolatshad NF, Silva AT, Saffrey MJ (2012) Ageing of enteric neurons: oxidative stress, neurotrophic factors and antioxidant enzymes. Chem Cent J 6:80 Korsching S (1993) The neurotrophic factor concept: a reexamination. J Neurosci 13:2739–2748 Ledda F, Paratcha G, Ibanez CF (2002) Target-derived GFRalpha1 as an attractive guidance signal for developing sensory and sympathetic axons via activation of Cdk5. Neuron 36:387–401 Lin LF, Doherty DH, Lile JD, Bektesh S, Collins F (1993) GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science 260:1130–1132 Martucciello G, Favre A, Takahashi M, Jasonni V (1995) Immunohistochemical localization of RET protein in Hirschsprung’s disease. J Pediatr Surg 30:433–436 Maruccio L, Lucini C, Russo F, Antonucci R, Castaldo L (2008) The development of avian enteric nervous system: distribution of artemin immunoreactivity. Acta Histochem 110:163–171 Mitsuma N, Yamamoto M, Li M, Ito Y, Mitsuma T, Mutoh T, Takahashi M, Sobue G (1999) Expression of GDNF receptor (RET and GDNFR-alpha) mRNAs in the spinal cord of patients with amyotrophic lateral sclerosis. Brain Res 820:77–85 Mologni L (2011) Development of RET kinase inhibitors for targeted cancer therapy. Curr Med Chem 18:162–175 Nosrat CA, Tomac A, Hoffer BJ, Olson L (1997) Cellular and developmental patterns of expression of Ret and glial cell line-derived neurotrophic factor receptor alpha mRNAs. Exp Brain Res 115:410–422 Panza E, Knowles CH, Graziano C, Thapar N, Burns AJ, Seri M, Stanghellini V, De Giorgio R (2012) Genetics of human enteric neuropathies. Prog Neurobiol 96:176–189 Powers JF, Brachold JM, Tischler AS (2003) Ret protein expression in adrenal medullary hyperplasia and pheochromocytoma. Endocr Pathol 14:351–361 Rodrigues DM, Li AY, Nair DG, Blennerhassett MG (2011) Glial cell line-derived neurotrophic factor is a key neurotrophin in the

postnatal enteric nervous system. Neurogastroenterol Motil 23:e44–e56 Rossi J, Luukko K, Poteryaev D, Laurikainen A, Sun YF, Laakso T, Eerikainen S, Tuominen R, Lakso M, Rauvala H, Arumae U, Pasternack M, Saarma M, Airaksinen MS (1999) Retarded growth and deficits in the enteric and parasympathetic nervous system in mice lacking GFR alpha2, a functional neurturin receptor. Neuron 22:243–252 Ruiz-Ferrer M, Torroglosa A, Luzon-Toro B, Fernandez RM, Antinolo G, Mulligan LM, Borrego S (2011) Novel mutations at RET ligand genes preventing receptor activation are associated to Hirschsprung’s disease. J Mol Med (Berl) 89:471–480 Rusinek D, Szpak-Ulczok S, Jarzab B (2011) Gene expression profile of human thyroid cancer in relation to its mutational status. J Mol Endocrinol 47:R91–R103 Schafer KH, Mestres P (1999) The GDNF-induced neurite outgrowth and neuronal survival in dissociated myenteric plexus cultures of the rat small intestine decreases postnatally. Exp Brain Res 125:447–452 Takahashi M (2001) The GDNF/RET signaling pathway and human diseases. Cytokine Growth Factor Rev 12:361–373 Taraviras S, Marcos-Gutierrez CV, Durbec P, Jani H, Grigoriou M, Sukumaran M, Wang LC, Hynes M, Raisman G, Pachnis V (1999) Signalling by the RET receptor tyrosine kinase and its role in the development of the mammalian enteric nervous system. Development 126:2785–2797 Trupp M, Scott R, Whittemore SR, Ibanez CF (1999) Ret-dependent and -independent mechanisms of glial cell line-derived neurotrophic factor signaling in neuronal cells. J Biol Chem 274:20885–20894 Tsui-Pierchala BA, Milbrandt J, Johnson EM Jr (2002) NGF utilizes cRet via a novel GFL-independent, inter-RTK signaling mechanism to maintain the trophic status of mature sympathetic neurons. Neuron 33:261–273 Uesaka T, Jain S, Yonemura S, Uchiyama Y, Milbrandt J, Enomoto H (2007) Conditional ablation of GFRalpha1 in postmigratory enteric neurons triggers unconventional neuronal death in the colon and causes a Hirschsprung’s disease phenotype. Development 134:2171–2181 Uesaka T, Nagashimada M, Yonemura S, Enomoto H (2008) Diminished Ret expression compromises neuronal survival in the colon and causes intestinal aganglionosis in mice. J Clin Invest 118:1890– 1898 Unsicker K (1996) GDNF: a cytokine at the interface of TGF-betas and neurotrophins. Cell Tissue Res 286:175–178 Wallace AS, Anderson RB (2011) Genetic interactions and modifier genes in Hirschsprung’s disease. World J Gastroenterol 17:4937– 4944 Wartiovaara K, Salo M, Sainio K, Rintala R, Sariola H (1998) Distribution of glial cell line-derived neurotrophic factor mRNA in human colon suggests roles for muscularis mucosae in innervation. J Pediatr Surg 33:1501–1506 Wedel T, Roblick U, Gleiss J, Schiedeck T, Bruch HP, Kuhnel W, Krammer HJ (1999) Organization of the enteric nervous system in the human colon demonstrated by wholemount immunohistochemistry with special reference to the submucous plexus. Ann Anat 181:327–337 Wedel T, Busing V, Heinrichs G, Nohroudi K, Bruch HP, Roblick UJ, Bottner M (2010) Diverticular disease is associated with an enteric neuropathy as revealed by morphometric analysis. Neurogastroenterol Motil 22:e493–e404 Wells SA Jr, Santoro M (2009) Targeting the RET pathway in thyroid cancer. Clin Cancer Res 15:7119–7123 Yamamoto M, Mitsuma N, Inukai A, Ito Y, Li M, Mitsuma T, Sobue G (1999) Expression of GDNF and GDNFR-alpha mRNAs in muscles of patients with motor neuron diseases. Neurochem Res 24:785–790 Young HM, Hearn CJ, Newgreen DF (2000) Embryology and development of the enteric nervous system. Gut 47 (Suppl 4):iv12–iv26