Smooth Muscle-Specific Expression of Neurotrophin-3 in the Embryonic and Neonatal Gastrointestinal Tract of the Mouse

256 words) Neurotrophin-3 (NT-3), which is essential for the survival of a large proportion of vagal sensory neurons, is expressed in the developing gastrointestinal (GI) tract where it could contribute to this survival, to other aspects of vagal afferent development, and to the development of spinal afferents, postganglionic sympathetic neurons and intrinsic neurons. Identifying the functions of this peripheral NT-3 requires a detailed understanding of the localization and timing of its expression in the developing GI tract. Therefore, X-gal staining of embryos and neonates expressing the lacZ reporter gene from the NT-3 locus was used to characterize the spatiotemporal pattern of NT-3 expression during development. NT-3 expression in the stomach, and intestines was largely restricted to smooth muscle of the outer GI wall and associated blood vessels. However, expression also occurred in the stomach lamina propria and esophageal epithelium. NT-3 expression occurred in GI wall mesenchyme by embryonic day (E)12.5 and became restricted to smooth muscle and lamina propria by E15.5 as these tissues differentiated, whereas expression in blood vessels and esophageal epithelium was first observed at E15.5. Immunohistochemical detection of β-galactosidase and cell type markers suggested myenteric elements, including neurons, glial cells, neural and glial precursors, and interstitial cells of Cajal did not express NT-3. Thus, NT-3 expression in the GI tract was largely restricted to smooth muscle at ages when vagal axons grow into the GI tract and vagal mechanoreceptors form in smooth muscle, consistent with roles in these processes, in vagal sensory neuron survival, and in development of intrinsic and other extrinsic GI wall innervation. Edward A. Fox and Jennifer McAdams 3 INTRODUCTION [Need to incorporate discussion of extrinsic and intrinsic myenteric elements that express trk and p75 receptors during development as targets for secreted NT-3 from smooth muscle.] Neurotrophin-3 (NT-3) is one of four members of the mammalian neurotrophin family of secreted proteins. The high affinity receptor for NT-3 is the receptor tyrosine kinase, trkC [Kaplan, 1991 #1183; Klein, 1991 #1184; Cordon-Cardo, 1991 #1182; Berkemeier, 1991 #1157; Glass, 1991 #1181; Soppet, 1991 #1173; Squinto, 1991 #1172; Lamballe, 1991 #1180], although it can also activate trkA or trkB [Berkemeier, 1991 #1157; Cordon-Cardo, 1991 #1182; Davies, 1996 #197; Farinas, 1998 #388; Glass, 1991 #1181; Huang, 1999 #149; Kaplan, 1991 #1183; Klein, 1991 #1184; Lamballe, 1991 #1180; Soppet, 1991 #1173; Squinto, 1991 #1172]. Additionally, p75 can be independently activated by each neurotrophin, or may partner with any one of the neurotrophin trk receptors to mediate specific responses [Chen, 2009 #2014]. NT-3 roles in peripheral sensory and sympathetic nervous systems and tissues Similar to other neurotrophins, NT-3 associated with the peripheral nervous system is essential for the survival of a large proportion of sensory and sympathetic neurons. In the initial characterizations, the survival effects of each neurotrophin were found to be associated with neurons mediating a different somatosensory modality and thus were described as “modality specific” [Farinas, 1999 #647; Snider, 1994 #788]. For example, Edward A. Fox and Jennifer McAdams 4 NT-3 supports large-diameter, parvalbumin-immunopositive dorsal root ganglion (DRG) neurons that convey proprioceptive signals [Ernfors, 1994b #599; Farinas, 1994 #227; Tessarollo, 1994 #220], whereas nerve growth factor is essential for survival of smalldiameter DRG neurons that transmit pain signals [Crowley, 1994 #1205; Smeyne, 1994 #1207; Stucky, 1999 #1043]. NT-3 acts sequentially to support sensory and sympathetic neuron survival, initially during gangliogenesis, subsequently during axon growth toward the target tissue, and then after target innervation [Farinas, 1996 #187; Francis, 1999 #1796; Kuruvilla, 2004 #1794; Patapoutian, 1999 #153; White, 1996 #1751; Zhou, 1996 #637]. For some pathways neurotrophin or nerve growth factor requirements may switch upon target innervation [Buchman, 1993 #235; Ernfors, 2001 #1750]. Several actions of NT-3 in addition to its effects on neuron survival have been characterized in the sensory and sympathetic nervous systems, including roles in neuronal differentiation [Ernfors, 2001 #1750], axon growth [Genc, 2004 #1755; Tucker, 2001 #1746], and nerve terminal formation, including size and structure [Lentz, 1999 #155; Ulupinar, 2000 #652], degree of contact with their accessory cells, and survival of their accessory cells [Albers, 1996 #192], receptor maintenance [Airaksinen, 1996 #204] and neurotransmission [Arvanov, 2000 #1822; Oestreicher, 2000 #748]. Further, NT-3 produced in peripheral tissues may have effects on non-neural tissues as has been observed in hair follicles where it stimulates growth during development and inhibits it in adulthood [Botchkarev, 2004 #1752]. Finally, exogenous NT-3 can affect sensory nerve activity [Mizisin, 1999 #142] as well as GI transit and myoelectric activity [Chai, 2003 #1754; Coulie, 2000 #1753; Parkman, 2003 #1757]. Edward A. Fox and Jennifer McAdams 5 NT-3 Roles in vagal sensory innervation of the GI tract NT-3, acting in part through activation of trkC, is essential for the survival of a large proportion of vagal sensory neurons: NT-3 and trkC homozygous mutants have 34-47 and 14% loss of neurons from the nodose-petrosal ganglion complex, respectively [Ernfors, 1994b #599; Farinas, 1994 #227; Liebl, 1997 #170; Tessarollo, 1997 #169]. Also, NT-3 supports survival of a different population of vagal sensory neurons than those supported by the other neurotrophins [ElShamy, 1997 #174]. However, in contrast to modality-specific regulation of somatosensory neuron survival by neurotrophins, their regulation of vagal sensory neuron survival may follow an “organ-specific” principle with each neurotrophin regulating survival of all the sensory receptor types that innervate an organ system [Brady, 1999 #384]. Consistent with this principle, different neurotrophins support survival of sensory innervation of different GI organs. Vagal sensory innervation of the esophagus appears to be dependent on NT-3 as the low threshold slowly adapting intraganglionic laminar ending (IGLE)-type mechanoreceptors that predominate in the muscle wall are reduced by 65 and 40% in NT-3 and trkC heterozygous mutants, respectively [Raab, 2003 #1678]. In contrast, the vagal sensory innervation of the small intestine appears to be dependent on NT-4 as homozygous NT4 mutants have almost complete loss of IGLEs from the small intestine, but normal IGLE density and structure in the stomach [Fox, 2001b #791]. The source of NT-3 regulating development of vagal sensory innervation of the GI tract Edward A. Fox and Jennifer McAdams 6 The effects of NT-3 on survival of vagal sensory neurons reviewed above and other potential effects of NT-3 on vagal afferent development could be mediated by its expression in the region of the developing nodose ganglion [Ernfors, 1992 #627], or by NT-3 present in embryonic and postnatal GI tract tissues innervated by vagal sensory neurons, similar to what has been observed in other sensory systems as described above. Indeed, NT-3 expression in the embryonic GI tract has been observed in the rat stomach [Scarisbrick, 1993 #1747], the mesenchyme of the avian gut [Chalazonitis, 1996 #205; Le Douarin, 1999 #1748], at the mesenchyme-epithelium boundary in the upper esophagus of the mouse [Patapoutian, 1999 #153], and in the postnatal mouse colon [Lommatzsch, 2005 #1756]. However, for the mouse, only the upper esophagus and colon have been examined, and in all of these species and organs that have been investigated the details of the locations and timing of NT-3 expression are lacking. It is highly probable that this NT-3 expression within GI tract tissues is involved in mediating NT-3 effects on development of vagal sensory innervation of the GI tract. Consistent with this hypothesis, trkC as well as trkB, which can also be activated by NT3 (see above) are expressed by nodose ganglion neurons from E13-E18 [Huber, 2000 #1290; Huang, 1999 #149; Ernfors, 1992 #627], and NT-3 undergoes retrograde transport in vagal sensory axons, albeit this has only been demonstrated in adult rats [Helke, 1998 #167]. These findings are significant because neurotrophin effects on neuronal survival are thought to be dependent on retrograde transport of the internalized neurotrophin-trk receptor complex to the cell body (e.g., [Grimes, 1997 #1202; Grimes, 1996 #1203]; but see [MacInnis, 2002 #1204], although some NT-3 actions may not require this process [Kuruvilla, 2004 #1794]. Similarly, NT-3 expression Edward A. Fox and Jennifer McAdams 7 in the GI tract wall has been suggested to mediate survival and differentiation of a subset of enteric neuron innervation of the intestine (Chalazonitis, 2001). Consistent with these NT-3 roles, developing myenteric neurons express trk receptors [Lamballe, 1994 #645; Sternini, 1996 #807], and adult enteric neurons exhibited retrograde transport of radiolabeled NT-3 injected into their target tissues, the mucosa and the muscle wall, suggesting they may normally obtain NT-3 in this manner. Moreover, myenteric neuron density was increased or decreased with overexpression or knockout of NT-3, respectively, and submucosal plexus neurons also exhibited reduced density in NT-3 deficient mice. Investigation of the roles NT-3 produced by peripheral tissues plays in development of vagal sensory innervation of the GI tract, or for that matter, its roles in development of other extrinsic and intrinsic innervation of the GI tract, requires a much more detailed understanding of the spatiotemporal pattern of NT-3 expression in the developing GI tract than is currently available. Knowledge of which tissues express NT-3 and receive vagal sensory innervation would identify the vagal sensory pathways likely to be regulated by NT-3, and determination of when this expression occurred would distinguish the stage(s) of development at which NT-3 might regulate them (e.g., axon arrival vs. receptor differentiation). Therefore, our aims were to identify which tissues/cell types within the GI tract express NT-3, and to characterize the time course of this expression at ages when vagal axons grow into the GI tract and their nerve terminals begin to differentiate. These findings will provide a basis for formulating and testing hypotheses about the roles of peripheral NT-3 expression in GI vagal afferent development, including the organ-specific principle of neurotrophin action. Edward A. Fox and Jennifer McAdams 8 MATERIALS AND METHODS Subjects The NT-3 mouse [Farinas, 1994 #227] obtained on an ICR out bred background was backcrossed up to 11 generations with C57BL/6 mice obtained from Harlan Industries (Indianapolis, IN) and bred in our lab. Mice were maintained at 23oC on a 14:10 hr light:dark schedule, with ad libitum access to tap water and Laboratory Rodent Diet 5001 (PMI Nutrition International. St. Louis, MO). All procedures were conducted in accordance with Principles of Laboratory Animal Care (NIH publication No. 86-23, revised 1985) and American Association for Accreditation of Laboratory Animal Care guidelines and were approved by the Purdue University Animal Care and Use Committee. Rationale for using NT-3 mice. Mice that had the LacZ reporter gene “knocked into” the NT-3 locus (NT-3 mice; [Farinas, 1994 #227], and thus expressed βgalactosidase under control of the NT-3 promoter were utilized to characterize NT-3 expression in the developing GI tract. The rationale for using NT-3 mice in such studies has been discussed extensively (e.g., [Farinas, 1996 #187; Vigers, 2000 #1685; Vigers, 2003 #1682]. Of particular relevance, NT-3 mice have been employed in numerous studies of detailed NT-3 expression patterns in many different organs and tissues [Bennett, 1999 #1825; Farinas, 1996 #187; Farinas, 2001 #1826; Hess, 2007 #1828; Tojo, 1995 #1824; Vigers, 2000 #1685; Vigers, 2003 #1682; Wilkinson, 1996 #188; Yee, 2003 #1359]. Moreover, these mice have yielded higher resolution cellular Edward A. Fox and Jennifer McAdams 9 expression patterns than had been achieved with in situ hybridization (e.g., somatosensory system; [Farinas, 1994 #227; Jones, 1994 #414]. Also, the patterns of NT-3 expression obtained in these mice accurately reproduced the patterns derived using methods that reveal endogenous expression (Farinas et al, 1996). Tissue preparation Matings were set up between heterozygous NT-3 mice, or between heterozygous NT-3 mice and wild-type mice. Noon of the day a copulatory plug was observed was designated E0.5 and the day of birth as postnatal day (P)0. Embryos were harvested on E12.5, E13.5, E15.5 and E17.5, and GI tracts were dissected from P4 mice. Results were based on staining patterns in 3 or more independent litters at each age studied, and multiple embryos or neonates from each litter were examined. Homozygous and heterozygous mutants and wild types were genotyped by PCR using DNA extracted from a tail sample for mice, or the yolk sac for embryos (PCR primer sequences: wild-type allele forward: ACT ACG GCA ACA GAG ACG CTA C, reverse: ACA GGC TCT CAC TGT CAC ACA C; mutant allele forward: GTG CCA GCG GGG CTG CTA AAG CGC, reverse: CTG CAT TCT AGT TGT GGT TTG TCC AAA CTC ATC). Embryos. Pregnant mice were euthanized by cervical dislocation, and then the uterus was removed and placed in chilled phosphate buffered saline (PBS) on ice. Each embryo was dissected free of their deciduum and yolk sac, and then the skin and muscle enclosing the abdomen were peeled back to expose the abdominal organs. Embryos were then fixed for 30 min on ice with1% paraformaldehyde, 0.02% Edward A. Fox and Jennifer McAdams 10 gluteraldehyde, 0.5 mM EGTA, and 2 mM MgCl2 in 0.1M sodium phosphate buffer, pH 7.4. Postnatal mice. Neonatal mice were deeply anesthetized with a lethal dose of methohexital sodium (Brevital Sodium, King Pharmaceuticals, Inc., Bristol, TN; 100 mg/kg i.p.). When mice were unresponsive to nociceptive stimuli, the abdomen and thorax were exposed and animals were perfused transcardially at a flow rate of 3 ml/minute with 0.9% saline for 5 min at RT followed by the same fixative used for embryos for 30 min at 4oC. Histochemical staining of β-galactosidase Immediately after fixation the embryos or abdominal organs were stained with X-gal as previously described [Fox, 2000 #333]. After staining was completed, tissue was postfixed 48 hr in 4% paraformaldehyde at 4oC, washed with PBS and then transferred to 10% buffered formalin at 4oC for a minimum of 5 days. Then the tissues were embedded in paraffin, sectioned at 8 μm thickness, air dried on gelatin-coated slides, alternate ribbons of sections were counterstained with 0.1% neutral red, and all sections were dehydrated in a series of graded alcohols (70%, 95%, 2 x 100%; 2 min each), cleared in xylene (3 x 2 min) and coverslipped with Cytoseal (Richard Alan Scientific, Kalamazoo, Michigan). Additionally, IHC detection of β-galactosidase in the E17.5 gut using the methods described below was employed to validate the results obtained by Xgal staining. Antibody characterization Edward A. Fox and Jennifer McAdams 11 Information on primary antibodies used in mouse embryonic tissues, including immunogens, is summarized in Table 1. Additional information on controls for specificity of each of these antibodies is provided below. The β-galactosidase antibody employed was produced in guinea pigs and its specificity characterized by Yee et al. [Yee, 2003 #1359]. No immunoreactivity was detected after applying the antibody to mouse lingual tissue that did not contain βgalactosidase. Moreover, application of this antibody to tissues from several LacZ transgenic strains with tissues expressing β-galactosidase resulted in strong immunoreactivity. Additionally, when the β-galactosidase immune sera were preadsorbed with β-galactosidase, no immunoreactivity was detectable after its application to β-galactosidase expressing mouse lingual tissue. In the present study, this antibody also showed complete correspondence with X-gal staining of βgalactosidase expression in developing gastric and intestinal smooth muscle, esophageal epithelium and gastric lamina propria. Goat polyclonal anti-c-Kit (lot numbers H2907, C2108) was raised against a peptide from the carboxy terminus of mouse c-Kit. The molecular weight of precursor c-Kit is 120 kDa and the mature form is 145 kDa. In Western blots of lysates (some immunoprecipitated) from several cell lines (manufacturer technical information; [Jahn, 2002 #1984] and cultured fetal cells [Nobuhisa, 2003 #1983] stained both the precursor and mature bands. Moreover, immunostaining using this antibody in tissue sections was prevented by pre-incubation of the antibody with c-Kit blocking peptide [SchransStassen, 1999 #1985]. Additionally, this antibody has been used to label cells in the GI tract wall of the mouse, rat and guinea pig with distributions and morphologies Edward A. Fox and Jennifer McAdams 12 consistent with those of the several ICC classes [Fox, 2000 #689; Ho, 2003 #1986; Pham, 2002 #1982]. The mouse IgG2b neuronal protein HuC/HuD (HuC/D) monoclonal (clone 16A11) antibody (lot number 42804A) recognizes an epitope within the carboxy terminal domain of HuD, resulting in labeling of neuronal cell bodies and nuclei [Marusich, 1994 #1964]. This antibody recognizes neuronal proteins of the Elav family, including HuC and HuD, which are RNA-binding proteins, and Hel-N1. The presence of Hu proteins acts as a marker for both newly committed postmitotic neurons and mature neurons [Barami, 1995 #1961; Marusich, 1994 #1964; Young, 2005 #1962]. The staining pattern of cellular morphology and distribution observed in the enteric nervous system in the present study was the same as previously described [Lin, 2002 #1966; Murphy, 2007 #1965; Young, 2005 #1962]. Importantly, the staining pattern of biotin-conjugated HuC/D monoclonal antibody was identical to that of the unconjugated antibody [Hoff, 2008 #1963]. The polyclonal rabbit antibody directed against p75 (lot number 606031680), the low affinity neurotrophin receptor, was raised against the extracellular fragment specified by exon 3 sequences of the murine p75 gene (manufacturer’s technical information). Specificity of this antibody was originally characterized by Huber and Chao [Huber, 1995 #1976] and in Western blots it labeled a single band of about 75 kDa in brain [Huber, 1995 #1976], embryonic testis [Campagnolo, 2001 #1974], and importantly, GI myenteric plexus [Lin, 2005 #1977]. In the present study the morphology and distribution of immunoreactivity produced using anti-p75 was consistent with glial and Edward A. Fox and Jennifer McAdams 13 neural precursors, and with staining patterns produced by the same antibody in similar preparations [Lin, 2005 #1977; Vannucchi, 2000 #1971; Yan, 2008 #1970]. A polyclonal rabbit antibody directed against the established glial marker S100 (lot number 15317) was employed, which stains the cytoplasm, labeling glial somata and processes. Specificity was demonstrated by two-dimensional immunoelectrophoresis, which identified one distinct double peak with human and cow brain extracts corresponding to S100a and –b and found no reaction with human plasma or cow serum (manufacturer’s technical information). Further, it was found using an indirect ELISA that anti-S100 did not react with human plasma or cow serum. IHC detection of sites of NT-3 expression (β-galactosidase protein) IHC co-localization of an antibody that detects LacZ expression (β-galactosidase) with an antibody selective for either neurons (anti-HuC/D), glia (anti-s100), undifferentiated neural crest precursors (anti-p75), or interstitial cells of Cajal (ICCs; anti-c-Kit) in heterozygous NT-3lacZ embryos was utilized to determine whether NT-3 expression occurred in these cell types. Tissue sections from wild-types were stained in parallel to examine the specificity of the β-galactosidase antibody. Embryos harvested at E17.5 as described above were immersion fixed in 4% paraformaldehyde at 4C for 24 hr, rinsed in PBS, incubated in 15% sucrose PBS for 1 hr at 4C, followed by 30% sucrose overnight (ON) at 4C, then a 1:1 mixture of 30% sucrose and OCT compound (Tissue-Tek, Miles Inc) ON at 4C, and frozen in 100% OCT using liquid nitrogen. Sections were cut at 10 μm thickness and air dried on gelatin-coated slides. Additionally, a small number of wild-type embryos were harvested at E14.5 for Edward A. Fox and Jennifer McAdams 14 evaluation of NT-3-LIR at an earlier stage of expression (see Results) and were treated the same as described above for E17.5 embryos. IHC procedures. All steps were done at RT unless indicated. Tissue sections were washed in PBS, incubated 1 hr in blocking solution (10% normal donkey serum, 0.5% triton X-100, 2% BSA, and 0.1% sodium azide), and then incubated ON in the first primary antibody at 4C. These, as well as all other primary and secondary antibodies employed in the present experiments were diluted with 1% BSA, 2% normal donkey serum, 0.3% Triton X-100, and 0.1% sodium azide. After washing with PBS, sections were incubated in the first secondary antibody for 2 hr and then washed again with PBS. Next, the sections were incubated for 1 hr in blocking solution and then in the second primary antibody ON at 4C. Sections were washed in PBS and then incubated in the second secondary antibody for 2 hr. After washing in PBS sections were mounted in glycerol, coverslipped and sealed with nail polish. The secondary antibodies used included donkey rhodamine red X (RRX) conjugated anti-goat or anti-guinea pig IgG (H+L) and donkey fluorescein iso-thiocyanate (FITC) – conjugated anti-rabbit, antigoat or anti-guinea pig IgG (H+L; Jackson ImmunoResearch Laboratories, Inc.). The mouse monoclonal HuC/D antibody employed was tagged with biotin, which avoided use of an anti-mouse secondary and the associated high background. To label this antibody streptavidin tagged with tetramethyl rhodamine iso-thiocyanate (TRITC) or FITC was applied in the same manner as described for secondary antibodies. To control for antibody interactions, the majority of staining runs for each pair of primary antibodies included several sections that were taken through the entire double-staining protocol, omitting the first primary antibody on some sections and the second primary Edward A. Fox and Jennifer McAdams 15 on other sections (incubation in primary antibody diluent alone was substituted). Further, to control for non-specific staining by secondary antibodies, staining produced by each primary antibody alone was compared with staining following omission of that primary antibody (incubation in primary antibody diluent alone was substituted) with all other protocol steps kept constant. [Emphasize the outcomes of these control expts in the Results section] Microscopy and Photomicrography X-gal-stained tissue was examined with standard bright-field or differential interference contrast illumination and rhodamine and fluorescein fluorescence was visualized using standard filter sets (Leica DM5000 microscope; fluorescence filter cubes L5 and Y3, respectively). Photomicrographs were acquired directly with a video camera (Spot RT Slider; Diagnostic Instruments, Inc., Sterling Heights, MI). The fluorescence-stained tissues were also scanned with an Olympus BX-DSU spinning disk confocal microscope operated using Slidebook software (v.4.1, Intelligent Imaging Innovations, USA). Image processing, three-dimensional reconstructions and twodimensional projections of z-series stacks of optical sections were also performed using the Slidebook software. Standard fluorescence microscopy was utilized to identify regions of tissue sections with potential for co-localization of two different primary antibodies based on the apparent overlap of their respective rhodamine or fluorescein labels. Regions to be imaged were randomly selected and then a series of confocal scanning optical sections through the z-axis of each region were collected. The series of sections collected at each site was examined to determine whether both labels were Edward A. Fox and Jennifer McAdams 16 present in the same tissue elements. Three criteria were employed to determine colocalization: (1) In composite images of the FITC and TRITC or Texas Red channels, elements exhibited a yellow pseudocolor as a result of the combined green pseudocolored FITC channel and red pseudocolored TRITC or Texas Red channel, (2) The separate greenand red-labeled elements were in focus as determined by scanning through the z-series, and (3) The shapes of the greenand red-labeled elements exhibited partial or complete matching of in focus tissue elements. Photoshop software (version 6.0 Adobe Systems, Mountain View, CA) was used to a) apply scale bars and text, b) adjust brightness and contrast, c) apply color correction, and d) organize the final layouts for printing.