[1] 26w Distribution and Co-Localization of Calbindin D28g With VIP and Neuropeptide Y but not Somatostatin, Galanin and Substance P in the Enteric Nervous System of the Rat

[1] 57w Western blotting of colon samples resulted in a single immunoreactive band which co-migrated with purified bovine CaBP (mol.wt. 28 kD) and the single bands observed in rat brain samples (Fig. 1). Staining with pre-immune serum from the rabbit used to raise antibodies to CaBP produced no staining of either the pure CaBP or any of the samples.

[2] 48w Pre-incubation of the CaBP antiserum with 1 nmole/ml purified bovine cerebellar CaBP completely abolished the immunostaining in all regions. The absorpiton of the antiserum with VIP, somatostatin, substance P, and NPY did not affect the intensity of the immunostaining. The preabsorption of the remaining antibodies/antisera confirmed their specificity.

[3] 81w An intense staining of individual cell bodies was demonstrated in both the myenteric and submucous plexi. In the antral and corpus regions the immunoreactive neurons were confined to the myenteric plexus (Fig. 2a). Immunoreactive nerve fibers were demonstrable in the myenteric and deep muscular plexi although absent from the circular and longitudinal muscle. The pyloric sphincter muscle was also de-335 void of any CaBP-immunoreactive nerve fibers. Occasional immunoreactive epithelial cells were observed in both the corpus and antral mucosa (Fig. 2b).

[4] 106w In the duodenal region the number of CaBP positive neurons increased with 80% of the neurons in the submucous and 28% of the neurons in the myenteric plexus demonstrating CaBP immunoreactivity (see Table 2). Immunoreactive fibers were observed in the mucosa, interganglionic nerve trunks and both plexi (Fig. 2c). No immunoreactive fibers were observed in the circular or longitudinal muscle or around the Brunner's glands and no immunoreactive epithelial cells were seen in either these glands or in the mucosa. Positive fibers were often observed leading from a myenteric neuron into the interganglionic nerve trunk but they could not be traced to the point of termination.

[5] 69w The staining seen in the jejunal and ileal regions gave a similar pattern of immunoreactive cell bodies (see Table 2) in both plexi. The immunoreactive nerve fibers were present once again in the mucosa and the plexi but not in the muscle layers. The only significant change was an increase from 28% to 46% in the overall number of myenteric neurons in the ileal submucous plexus that contained CaBP.

[6] 48w The greatest amount of immunoreactivity was observed in the colon, especially in the submucous plexus and mucosa. The increased staining was represented by a marked thickening of the mucosal nerve fibers and an overall increase in the number of nerves seen in the plexus and mucosa (Fig. 2d).

[7] 32w The axonal projections from submucosal neurons were particularly easy to follow and most were seen to penetrate the muscularis mucosae and continue along the side of the crypts terminating on epithelial cells.

[8] 90w In the submucous plexus of the small intestine, approximately 50% of the CaBP-immunoreactive neurons demonstrated both VIP-and NPY-immunoreactivity (Table 2, Fig. 3a,b). The CaBP-immunoreactive nerve fibers in the lamina propria underlying the mucosal epithelium were also seen to be identical with VIP and NPY fibers. However, the CaBPcontaining nerve fibers in the two enteric plexi did not demonstrate VIP-and NPY-immunoreactivity. Without exception all the VIP-immunoreactive neurons contained CaBP and NPY. However a number of CaBP neurons did not contain VIP or NPY, the relative percentages are given in Table 2.

[9] 18w The submucosal neurons containing somatostatin or galanin-immunoreactivity formed two distinct populations and were not demonstrated to be CaBP-immunoreactive.

[10] 25w The CaBP-containing neurons in myenteric plexus did not display VIP-immunoreactivity and vice versa, although fibers containing VIP were seen to terminate on these cell bodies.

[11] 24w Although there was no co-localization of somatostatin within the CaBP-immunoreactive neurons in either plexus, in the stomach, the CaBP-immunoreactive epithelial cells were also somatostatin-immunoreactive.

[12] 39w Of the remaining peptides studied, nerve fibers containing galanin were seen to terminate on CaBP-containing neurons in the myenteric plexus. There was no evidence for an interaction between the substance P-containing neurons and nerve fibers and those containing CaBP.

[13] 104w The large intestine or colon differed in that although there was some co-localization of VIP with CaBP in both neurons TABLE 2 QUANTIFICATION OF CaBP-IMMUNOREACTIVE NEURONS Region No. CaBP No. + Total No. % NPY/ Area Neurons CaBP Total VIP Corpus SMP* 36 --MYP~" 1236 390 31 --Antrum SMP 108 --MYP 1620 468 28 --Duodenum SMP 3042 2466 81 1350 MYP 5778 1590 27 --Jejunum SMP 2766 2103 76 1291 MYP 3560 1152 32 --Ileum SMP 2160 1584 73 1008 MYP 3816 1776 46 --Colon SMP 1656 1116 67 288 MYP 4602 2347 51 --*SMP=submucous plexus. ?MYP=myenteric plexus. % CaBP 55 61 63 26

[14] 29w and mucosal fibers, there was no evidence for the presence of NPY. In addition the percentage of CaBP cell bodies also containing VIP dropped to 26% (see Table 2).

[1] 57w The results demonstrate, for the first time, that CaBP is present in the mammalian peripheral innervation, specifically the myenteric and submucous neurons of the enteric innervation in the rat. Although CaBP has not previously been localized within the mammalian enteric innervation, the protein has recently been reported to occur in the innervation of the chicken cecum [14].

[2] 130w There has been some discussion as to the molecular weight forms of Calbindin D28k and the presence of two, electrophoretically distinct proteins, of molecular weights 27 and 29 kD has been reported [16]. We have also observed a double band of immunoreactivity but only when precautions were not taken to ensure that the separation of CaBP by PAGE was done either in the presence of excess calcium or, as presented here, in the presence of excess EDTA, suggesting that the two forms were artifacts of the preparation method. In the present study, the Western blot technique demonstrated a single immunoreactive band of molecular weight 28 kD, and this co-migrated with purified bovine cerebellar CaBP and extracts from various regions of the rat brain known to contain the 28 kD form.

[3] 61w The reports of two molecular weight forms of CaBP may also be the result of the use of antibodies raised to chick gut CaBP which detect two different calcium binding proteins (see [16] for example), whereas our antibody, which was raised to monkey cerebellar CaBP, detects only a single immunoreactive protein in two dimensional gel blots (M. R. Celio, personal communication).

[4] 58w The immunocytochemical results demonstrated that immunoreactive epithelial cells were detected only in the stomach. None of the intestinal absorptive cells were immunoreactive confirming the absence of this 28 kD protein in cells actively absorbing calcium. In the stomach, the epithelial cells detected by the CaBP antiserum were found to correspond to a proportion of the somatostatin-containing endocrine cells.

[5] 91w A number of CaBP-immunoreactive neurons in the myenteric plexus generated processes that were observed in the interganglionic nerve trunks. This fact coupled with the absence of positive nerve fibers in the muscle layers suggests these may be interneurons connecting the two enteric ganglia [6]. Another possibility is that the CaBP-positive neurons may give rise to afferent fibers leading to the sympathetic ganglia or via the vagus and spinal cord to the CNS, and hence may be sensory in function. The presence of such afferent fibers was not determined during this study.

[6] 98w The function of the CaBP-containing cell bodies in the submucous plexus is probably different due to the extensive co-localization with both VIP and NPY. A striking finding was that the pyloric sphincter and circular muscle, which were heavily innervated by NPY/VIPergic fibers, were totally devoid of CaBP-immunoreactivity. In addition, the VIP-immunoreactive cell bodies seen in the myenteric plexus did not contain CaBP. These data indicate that there are at least two groups of VIP neurons, one (those in the myenteric plexus) concerned with the control of smooth muscle and the second involved in secretomotor control of epithelial transport.

[7] 75w In the submucous plexus of the small intestine over 50% of the CaBP-containing cell bodies and the majority of mucosal nerve fibers were demonstrated to contain both VIP and NPY. In this case the CaBP neurons are likely to be associated with a secretomotor function because both VIP and NPY stimulate water and electrolyte movement across the intestinal epithelium. The remaining CaBP-immunoreactive cell bodies in this plexus may function as either interneurons or sensory neurons.

[8] 50w In the large intestine, a minimal co-localization with VIP in either neurons or mucosal nerves was demonstrable although the actual number of CaBP-immunoreactive fibers in the mucosa was greatly increased. This would suggest that the CaBP neurons in this area were functioning as sensory or interneurons rather than secretomotor neurons.

[9] 29w The existence of a complex interfacing system within the myenteric ganglion was demonstrated by the presence of VIP-and galanin-immunoreactive nerve fibers terminating on the CaBP-positive neurons in this plexus.

[10] 52w It is somewhat speculative at this time to comment on the intracellular role of CaBP in the enteric neurons. We have suggested that it may function as an intraneuronal calcium buffering system [1,2], and as such, may interact with, for example, calcium dependent processes involved in the release of neurotransmitters, including peptides.

[1] 191w Samples of rat brain (cerebellum, hippocampus and cortex) or colon were rapidly dissected from a decapitated rat and homogenised in 10X vol./wt. 62.6 mM Tris-HCl (pH 7.6) containing 0.1 M dithiothreitol (DTT) and 1 mM EDTA. The colon samples were maintained in the homogenizing buffer at 75°C for 20 min prior to processing in order to inactivate any protease activity. The supernatant collected after centrifu-~ation at 38,000xg for 45 min was lyophylised and then redissolved in half the original volume of distilled water containing 10% glycerol, 2% SDS, 0.1 M DTT and 0.05% Bromophenol blue. Various amounts of these samples together with a CaBP standard purified from bovine cerebellum and suitable molecular weight markers were then run on a 7.5-17.5 linear gradient polyacrylamide gel using the discontinuous system of Laemmli [13] with the addition of 1 mM EDTA to all of the gel and running buffers. Following electrophoresis the gels were either stained with commassie blue or transferred, following the method of Towbin et al. [21], to nitrocellulose paper (Schleicher and Schuell, pore size 0.45 /zm) using a BioRad Trans-Blot ® apparatus with high field intensity (1 amp for 2 hours).

[2] 70w Staining of the electro-blots with antiserum to CaBP was performed as follows. Blots were washed briefly in Tris buffered saline (TBS-Tween, 20 mM Tris base in 500 mM saline, pH 7.4, containing 0.5% Tween 20). They were then blocked overnight at 4°C in this same buffer with the addition of 3% bovine serum and 12% goat serum (TBS-serum), followed by sequential 1 hour incubations at room temperature in TBSserum containing:

[3] 14w (1) rabbit anti-CaB P 1: 2000, (2) goat anti-rabbit IgG (Cappel) at 1:6000, and

[4] 55w (3) rabbit peroxidase anti-peroxidase (Cappel) at 1:3000. Triplicate washing steps between each antibody and before staining were completed with TBS-Tween. The CaBP immunoreactivity was visualised with 4-chloro-1 naphthol (60 mg dissolved in 20 ml ice cold methanol diluted to 100 ml with TBS containing 60/zl 3% H2Oz). Staining was stopped by repeated washing in TBS.

[5] 70w Tissue preparation. Samples of the antrum, corpus, duodenum, jejunum, ileum and colon of Wistar rats (N=5) were collected. The material was immersion fixed in Bouin's solution for 2 hr, dehydrated through graded alcohols and embedded in paraffin wax at 60°C. Five or three micron sections of the material embedded in wax were used for the immunostaining after removal of the wax by immersion in xylene and clearing in petroleum ether.

[6] 42w Immunostaining. The details of the antisera/antibodies used in this study are given in Table 1. In all cases the sections were incubated with the primary antiserum/antibody overnight at 4°C. In the single staining experiments the bound antibodies were localized using the avidin/biotin

[7] 68w TABLE 1 ANTISERA/ANTIBODIES Antigen Source Dilution Species Type Reference CaBP* KB 1:250 Rabbit Serum 1 VIP RPG 1:10 Mouse tcm 18 SS RPG 10/.~g/ml Mouse plgG 4 Galanin TMcD 1:2500 Rabbit Serum NPY JP 1:1000 Rabbit Serum SP JP 1:1000 Rabbit Serum *Monkey Cerebellar CaBP. tcm=tissue culture medium; plgG=ammonium sulfate purified IgG; KB = Dr. Baimbridge; RPG=(MRC) Regulatory Peptide Group; TMcD=Dr. McDonald, London, Ontario; JP=Professor Polak, London, England.

[8] 141w method. The biotinylated donkey anti-rabbit IgG (1:3000, Jackson Laboratories, PA) was applied for 1 hr at room temperature (RT) and the FITC conjugated avidin complex (1:4000, Vector Laboratories, CA) for 1 hr at RT. Finally the slides were washed in phosphate buffered saline (PBS), pH 7.4, and the coverslips applied in PBS/Glycerine (1:9). In order to determine co-localization of CaBP with peptides in the wax embedded material, serial 3 /x sections were collected. In experiments with the monoclonal antibodies to somatostatin and VIP, the initial staining sequence as was above. Then after washing in PBS, the same sections were incubated overnight with the relevant monoclonal antibody. The next day the sections were washed and incubated in rhodamine conjugated goat anti-mouse IgG (1:3000, Jackson Laboratories, PA) for 1 hr at RT. The slides were washed in PBS and coverslips applied as above.

[9] 51w In experiments involving the use of two antisera raised in rabbits, the serial 3/x sections were stained in pairs. The first section being stained with the CaBP antiserum and the second with the peptide antiserum. In both cases the bound primary antibodies were localized using the biotin/avidin FITC method as above.

[10] 33w In order to assess the percentage of enteric neurons containing CaBP and of the peptide co-localized with CaBP, neuronal counts were completed on 12 sections per area per peptide and matching CaBP staining.

[11] 38w The sections were screened using a Zeiss Axiophot microscope equipped with epi-fluorescence. The FITC staining was visualized using a filter of band width 465-490 nm (green) and the rhodamine with a filter of band width 510-560 nm (red).

[12] 170w Control incubations involved pre-incubation of the individual antisera/antibodies with the relevant parent antigen and related antigens at a concentration of 1-I0 nmoles/ml of the stock dilution, e.g., the CaBP antiserum was preincubated with 1 nmole of CaBP overnight at 4°C and this absorbed antiserum was then used as the first layer for the immunostaining. The CaBP antiserum was also preincubated with 10 nmoles/ml of vasoactive intestinal polypeptide (VIP), somatostatin, substance P, and neuropeptide Y (NPY). In the case of neuropeptides, the VIP monoclonal antibody was preabsorbed with 1 nmole of VIP or 10 nmole of one of the following: gastric inhibitory polypeptide, secretin and pan- creatic glucagon. Other control incubations involved the use of normal rabbit serum (NRS) or an inappropriate monoclonal antibody (anti-ferredoxin, AF) in place of the normal first layer. During the double staining experiments, on the 3 /x serial sections, initial staining was completed and the second primary antibody/antiserum replaced by AF, NRS or PBS and the sections incubated with the rhodamine conjugated anti-mouse or anti-rabbit IgG.

[1] 215w neuropeptide Y but not somatostatin, galanin and substance P in the enteric nervous system of the rat. PEPTIDES 9(2)333-338, 1988.--Calbindin D2~k, previously demonstrated in the mammalian central nervous system, has been localized to discrete neurons in the enteric nervous system of the rat. Calbindin D2.k is present in cell bodies in both the myenteric and submucous plexi and in interganglionic nerve fibers in all regions of the gastrointestinal tract. Immunoreactive nerve fibers were also detected in the mucosal region, although none were observed in the pyloric sphincter, circular or longitudinal muscle layers. The highest concentration of immunoreactivity was present in the submucosal plexus and mucosa of the colon. Western blot analysis of the protein detected by the antiserum confirmed that it comigrated with purified calbindin D2sk and the single immunoreactive band seen in extracts from rat brain. The colocalization of calbindin D~sk with components of the peptidergic innervation was also investigated. Of the peptides studied the neurons containing both vasoactive intestinal polypeptide and neuropeptide Y in the submucous plexus were seen to exhibit calbindin D~sk immunoreactivity. The neurons containing somatostatin, galanin and substance P did not demonstrate co-localization. In the stomach, calbindin D2sk was detected within a small number of epithelial cells which were found to correspond to a sub-population of the somatostatin-immunoreactive endocrine cells.

[2] 5w Neuropeptide Y VIP Rat Coexistence

[3] 78w A 28 kD Vitamin D dependent calcium binding protein was first isolated from chick intestinal mucosa by Wasserman and Taylor [22]. Subsequently, in mammals and chick, an analogous protein was demonstrated in a number of other tissues associated with a calcium transporting function, such as kidney [12,20] and bone [4]. A protein which was identical by immunological and physiochemical parameters has also been found in non-calcium transporting tissues such as brain [1,2,5,8,9,11,12], pancreatic B-cells [17] and pituitary [5].

[4] 26w A comparison of the primary amino acid sequences of the chick gut and bovine cerebellum 28 kD proteins confirmed that the two proteins are homologous [19,24].

[5] 68w In studies concerned with the identity of the calcium transport protein in mammalian intestinal epithelial cells, a smaller, and genetically unrelated, 9 kD Vitamin D depend-ent protein has been identified [7,10], which replaces the larger form seen in the avian intestinal epithelium. In order to distinguish between the two different forms of calcium binding proteins the nomenclature of Calbindin Dzsk (CaBP) and Calbindin Dgk has been adopted [23].

[6] 61w Within the brain CaBP is found in a number of distinct neuronal populations but no clear association between the presence of CaBP and any single neuronal function or neurotransmitter content has been established [2,8,9,11]. In many cases CaBP co-exists in neurons in which a peptide neurotransmitter/neuromodulator has been demonstrated, such as the opiate peptide containing neurons of the dentate gyrus [15].

[7] 71w In view of the localization of CaBP to neurons within the central nervous system that also contain peptidergic trans- mitters, the present study investigated the possible presence of the protein within the mammalian peripheral innervation. The enteric innervation of the rat was investigated by both immunocytochemistry and Western blot analysis of mucosal extracts for CaBP immunoreactivity. The distribution of this immunoreactive CaBP was compared with that of immunocytochemically identified peptidergic neurons.