ELSEVIER Journal of the Autonomic Nervous System 49 (1994) 207-216
Immunohistochemical studies on the intestinal nerve of Remak in the male chicken
M a s a h i r o S u z u k i a, Y a s u s h i g e O h m o r i a, T o h r u W a t a n a b e a,., I k u k o N a g a t s u b
Department of Veterinary Anatomy, School of Agriculture, Nagoya University, Nagoya 464-01, Japan, and b Department of Anatomy, School of Medicine, Fujita Health University, Toyoake, Japan
(Received 3 November 1993; revision received and accepted 14 February 1994)
Abstract
A peroxidase anti-peroxidase method was used to investigate and compare the distribution of neuropeptide and catecholamine synthesizing enzyme immunoreactive (IR) ganglion cells and nerve fibres in the intestinal nerve of Remak (INR) of male chickens. In the INR there were three kinds of ganglion cells: tyrosine hydroxylase (TH)-, aromatic L-amino acid decarboxylase (AADC)- and phenylethanolamine-N-methyltransferase (PNMT)-IR cells; AADC- and PNMT-IR but TH-immunonegative cells; and ganglion cells being immunoreactive for methionine enkephalin (mENK)- and somatostatin (SOM). The first one was distributed throughout the INR. The second was restricted in the ileojejunal region, and the last was localized in the rectal region. Substance P- and vasoactive intestinal polypeptide-IR nerve fibres were distributed in common but variable in number around three kinds of ganglion cells. Then TH-IR cells were characterized by the distribution of many calcitonin gene related peptide- and a few cholecystokinin-IR fibres, mENK and SOM-IR cells, and TH-immunonegative cells were distinguished by the distribution of SOM- and galanin-IR fibres. In addition, TH-immunonegative cells were characterized by the distribution of mENK- and neuropeptide Y-IR nerve fibres which were very few in number. Fig. 21 summarizes the connections described in the present study.
The intestinal nerve of Remak (INR) is a unique avian autonomic ganglionated nerve trunk which runs in the mesentery dorsal to the intes- tine from the cloaca to the duodenum and is primarily involved with the innervation of the intestine. The nerve trunk in the rectal region is
* Corresponding author.
thick and becomes thinner; the size of ganglion and the number of ganglion cells become smaller towards the duodenum [19]. The formaldehyde- induced fluorescence procedure showed that the rectal ganglionic clumps of the INR contain fluo- rescent neurons, have no fluorescent varicose fi- bres ending around them, and non-specifically fluorescent neurons with brightly fluorescent bas- ket works of varicose fibres surrounding them [1,2]. The pharmacology of the INR showed that the chicken rectum receives a dense non-adren-
208 M. Suzuki et al. /Journal o[" the Autonomk: Neruous System 49 (1994) 207-216
ergic and non-cholinergic excitatory innervation from the INR [17]. Ganglionic transmission oc- curs in ganglia of the INR but not in ganglia of the enteric plexuses, so the postganglionic neu- rons which are considered to be non-cholinergic are located in ganglia of the INR [9]. Ultrastruc- tural studies showed that the 1NR contains much more unmyelinated axons than myelinated ones which are not seen in the side branches [6], and that small granular vesicles" characterizing cate- cholamine (CA)-containing nerves could be demonstrated in the neurons of ileal ganglia, but not rectal ganglia of the INR [21].
In the immunohistochemical studies of the de- veloping chick hindgut, Epstein and colleagues [3,4] reported that the fact that methionine enkephalin (mENK)-immunoreactive (IR) neu- rons persist in ganglia of the INR and send processes into the rectum suggests that a major portion of mENK-IR fibres of the rectum is ex- trinsic, and that somatostatin (SOM)-IR cells ap- pear in the terminal part of the INR at embry- onic day 4 and send fibres into the wall of the hindgut, but no vasoactive intestinal polypeptide (VIP)-IR cells are found in ganglia of the INR.
Little information is available concerning the occurrence of neuropeptides as neurotransmitter in the INR, except for two reports already men- tioned. The present study conducts a series of immunohistochemical analysis and demonstrates neuropeptides in the INR of chickens. In addi- tion, special attention is paid to the distribution and frequency of peptidergic nerve fibres and the relationships between these fibres and peptide- containing neurons in the 1NR.
2. Materials and methods
2.1. Animals and tissue preparations
Eight adult male chickens weighing 1.5-3.3 kg were used. Chickens were anesthetized with sodium pentobarbital (20 mg/kg, i.v.), and per- fused through the left common carotid artery with physiological saline solution, followed by Stefanini's fixative (2% paraformaldehyde and 15% saturated solution of picric acid in 0.1 M
phosphate buffer, pH 7.3) at a pressure of 100- 120 mm Hg. The INR from the levels of the jejunum to the cloaca was removed from the bird, given the shape of a helicoid on a plane surface, and then immersed in the same fresh fixative at 4°C. The tissue was dehydrated and embedded in paraffin wax (Tissue prep) in vacuo. Conse- quently, serial or mirror sections (4 p.m thick) were made of the longitudinal plane of the INR and placed on gelatin-coated glass slides. Sec- tions were deparaffinized with toluene and rehy- drated in the graded series of ethanol or N,N'-di- methylformamide.
2.2. Antisera
The antisera used included rabbit anti-syn- thetic bombesin (lncstar, Stillwater, MN), 1 : 1000; rabbit anti-synthetic porcine cholecystokinin (CCK)(CRB, Cambridge, UK) 1:4000; rabbit anti-synthetic rat calcitonin gene related peptide (CGRP)(Amersham, Buckinghamshire, UK) 1:1200; rabbit anti-porcine galanin (GAL)(CRB, Cheshire, UK) 1:1600; rabbit anti-synthetic leucine enkephalin (IENK)(UCB, Belgium) 1:2000; rabbit anti-synthetic methionine enkeph- alin (mENK)(CRB, Cambridge, UK) 1:4000; and (Incstar, Stillwater, MN; 1:1600-3200), rabbit anti-synthetic porcine neuropeptide Y (NPY) (Amersham, Buckinghamshire, UK) 1:1600; rab- bit anti-synthetic bovine neurotensin (NT)(Milab, Sweden) 1:4500; rabbit anti-synthetic human peptide histidine isoleucine (PHI) (Milab, Swe- den) 1:500; rabbit anti-synthetic human cyclic SOM (donated by Dr. S. Ito, Niigata, Japan) 1:2000-8000; rabbit anti-synthetic mammalian substance P (SP)(CRB, Cambridge, UK) 1 : 1000; rabbit anti-synthetic VIP (Incstar, Stillwater, MN) 1 : 2000; rabbit anti-rat adrenal tyrosine hydroxyl- ase (TH) I :2000; rabbit anti-bovine adrenal aro- matic L-amino acid decarboxylase (AADC) 1:2000; rabbit anti-bovine adrenal dopamine beta-hydroxylase (DBH) 1:2000; rabbit anti- bovine adrenal phenylethanolamine-N-methyl- transferase (PNMT) 1:2000 and rabbit anti-5-hy- droxytryptamine (5-HT) 1:8000.
The criteria which established the purity of each antigen were previously described for TH
M. Suzuki et al. /Journal of the Autonomic Nervous System 49 (1994) 207-216 209
[11], A A D C [8], DBH [13], PNMT [12], and 5-HT [14]. Specificity of antisera of immunohistochem- istry was demonstrated by the lack of immunore- activity after the primary antiserum had been preadsorbed overnight at 4°C with one of the antigens ( 1 0 - 4 0 / x g / m l ) . As two kinds of mENK antisera used, brought about much the same re- suit, they are described as a single antiserum.
2.3. Single immunostaining
The sections were treated by the peroxidase anti-peroxidase (PAP) method [15] or by the avidin-biotinylated peroxidase complex (ABC) method. The ABC components were purchased from Vector Laboratories (Burlingame, CA). De- paraffinized and rehydrated sections were placed for 30 min in 75% methanol containing 0.3% hydrogen peroxide or 0.3% hydrogen peroxide solution and then steeped for 30 min in 0.01 M phosphate buffer, pH 7.4 containing 0.8% NaC1 (PBS) and 0.25% Triton X-100 (PBST) for the PAP method or in PBS for the ABC method. Triton X-100 was omitted from procedures of the ABC method. Back-ground blocking was per- formed for 30 min with 5% normal goat serum (Cappel, PA) in PBS, followed by rinsing in PBST.
The sections were incubated with the primary antiserum for 40 h at 4°C. After rinsing in PBST for 15 min, the sections were incubated in 2% goat anti-rabbit IgG serum (Miles Scientific, Is- rael) in PBS or biotinylated anti-rabbit IgG for 30-60 min at room temperature. Following a secondary rinse in PBST, the sections were incu- bated in 1-2.5% rabbit PAP complex (ICN im- munobiological, IL) in PBS or ABC complex for
30-60 min at room temperature and, after a final rinse, visualized with 0.02% 3,3'-diaminobenzi- dine tetrahydrochloride (DAB, Wako Chemical, Japan) and 0.005% hydrogen peroxide in 0.05 M Tris buffer, pH 7.4. Development time was 10-30 min. The sections were counterstained with methyl green.
2.4. Double immunostaining
Tissue procedure for the double immunostain- ing was the same as that described for the single staining. After immunostaining by the first spe- cific antiserum, the sections were incubated with another specific antiserum and reacted with the PAP method. In order to obtain a different color, DAB added 0.014% CoC12 was employed as a chromogen for grayish-blue reaction products [7]. The sections were washed in PBS, mounted in Entellan without the counterstaining, and exam- ined under a standard light microscope.
2.5. Morphometric methods
To determine the mean area of nerve fibres being immunoreactive for eight kinds of anti- serum around three kinds of ganglion cell (TH( + ), T H ( - ) and mENK( + )) in the INR, 83 single staining and 26 double staining sections were selected from the rectal and ileojejunal parts of the INR. The representative three regions selected in each section were photographed at ×200 magnification as original images under a light microscope. With a monochrome TV cam- era (Ikegami IF-103H, Tokyo, Japan), an image of an original film was written on image memory
Table 1 Area of neuropeptide-IR nerve fibres per 10,000 ~zm 2 around three kinds of ganglion cells
VIP 491.1 + 40.9 ~,b SP 584.3 + 52.3 a CGRP 448.6 + 39.9 a,b SOM 156.9 +_ 19.8 c SP 425.2 + 58.8 a.b VIP 504.0 + 93.1 a.b SP 117.1 _+ 16.4 c,d VIP 108.2 _+ 34.4 c,J CGRP 324.4 ± 27.1 b GAL 55.0 + 14.3 d VIP 91.7 + 26.8 c,~ SP 102.1 + 36.4 c,d CCK 40.2± 6.2 ~ SOM 35.2+ 7.3 d CCK 32.3 + 2.8 d mENK 37.6+ 4.5 d
GAL 35.0 + 4.7 d NPY 15.8_+ 3.0 d
Values (~m 2) represent the mean + S.E. Means with different superscript letters are significantly different from each other (P < 0.01).
210 M. Suzuki et al. /Journal o f the Autonomic NerL~ous System 49 (1994) 207-216
ii i ~ ~ ~ ' ~ ' ~ - ' ~ , V ii!!~
M. Suzuki et aL / Journal of the Autonomic Nervous System 49 (1994) 207-216 211
of the monochrome image analyzer (Nireco LU Z EX III, Tokyo, Japan). The image analyzer has 1024 × 1024 pixels and recognizes images of each pixel by grading brightness from 0 (darkest) to 255 (brightest). The image of the original black and white light micrograph was projected as an assembly of a total of 1,048,576 pixels possessing respective original grading brightness. The area (/~m 2) of immunostained nerve fibres per 10,000 /zm 2 in each region was measured using the im- age analyzer and mean _+ S.E. was calculated. All results were compared with each other (paired Student's t-test) (Table 1). The mean areas were divided into seven grades which were defined as follows: 1;60/xm 2 or less, 2;60-120 /xm 2, 3;120- 200 /xm 2, 4;200-300 /xm 2, 5;300-420 /xm 2, 6;420-560 /zm 2, and 7;560 /zm 2 and over per 10,000/~m 2 each. Considering a ganglion cell as a circle a grade of the mean area was determined as 20 degree in the central angle and was shown with a figure of the synaptic vesicle. The data obtained are summarized in schematic illustra- tions (Fig. 21). The difference of more than two vesicles is significant (P < 0.01).
3. Results
3.1. Rectal part of the INR
Ganglion cells in the rectal part of the INR exhibited immunoreactivities for two specific an-
tisera against neuropeptides (mENK and SOM) and three specific antisera against catecholamine synthesizing enzymes (TH, AADC and PNMT) out of 17 kinds of antiserum tested. About one- half of the ganglion cells in the rectal part showed immunoreactivity for T H (Fig. la). The TH-IR ganglion cells also were clearly immunoreactive with AADC (Fig. lb), but a good few of them showed immunoreactivity for PNMT (Fig. lc). The remaining half of ganglion cells in the rectal part of the INR were mENK-IR cells. Many mENK-IR ganglion cells (Fig. 2a) also showed immunoreactivity for SOM according to the ex- amination in the serial sections (Fig. 2b). Hence ganglion cells showing immunoreactivity for both mENK and SOM are expressed as mENK-IR ganglion cells. They seem to be smaller cells than TH-IR ones and become more numerous in the rectal part of the INR towards the cloaca. As shown in Fig. 1, TH-, AADC- and PNMT-IR nerve fibres are never found in the rectal part of the INR.
On the other hand, nerve fibres in the rectal part of the INR immunoreacted with six antisera for neuropeptides (VIP, SP, CGRP, SOM, CCK and GAL) out of 17 antisera tested. We cannot clearly distinguish whether the structures showing immunoreactivity for neuropeptides are varicosi- ties on nerve fibres or nerve terminals, but here- after we shall use the term 'nerve fibres' to de- scribe them.
Fig. l a - c . Three serial sections from the rectal part of the INR in chickens, immunostaining with anti-TH (a), A A D C (b) and PNMT (c) sera. TH- IR ganglion cells (a: arrows) are also A A D C - I R ones (b: arrows), but part of them are PNMT-IR cells (a,b,c: arrowhead) but others not (a,b,c: arrows). Bars = 20 /~m
Fig. 2a -b . Two serial sections from the rectal part of the INR, immunosta ined with an t i -mENK (a) and SOM (b) sera. A good many mENK-IR ganglion cells also show immunoreact ion for SOM. A few SOM-IR nerve fibres are slightly dotted around and among ganglion cells irrespective of immunoreactivity for SOM. Bars = 20 ~ m
Figs. 3-6. Double immunostaining sections from the rectal part of the INR. The PAP method with DAB for two kinds of neuropeptides, SP (3,4) or VIP (5,6) followed by the PAP method with DAB added the cobalt for TH or mENK, respectively. The immunoreact ive cells are gray. 3. SP-IR nerve fibres distributed densely around TH-IR cells, but not densely around TH-IN cells. 4. SP-IR nerve fibres sur rounded compactly around mE NK-IR cells. 5. TH-IR cells are compactly distributed with VIP-IR fibres. 6. mENK-IR cells are densely surrounded with VIP-IR fibres. Bars = 20 /~m
Figs. 7-9. Double immunostaining sections from the rectal part of the INR. The PAP method with DAB for CGRP (7), CCK (8) and G A L (9) followed by the PAP method with DAB added the cobalt for TH (7,8) or m E N K (9). The immunoreact ive ganglion cells show gray. (7) TH- IR cells are densely distributed with CGRP- IR fibres. (8) A few CCK-IR nerve fibres are slightly dotted around or among TH-IR cells. (9) GAL- IR fibres are slightly dotted among and around mENK-IR cells. Bars = 20 /~m
212 M. Suzuki et al. /Journal of the Autonomic Nerl,ous System 49 (1994) 207-210
M. Suzuki et al. / Journal of the Autonomic Nert~ous System 49 (1994) 207-216 213
In the doub le immunos t a in ing m e t h o d by the use of an t i se ra aga ins t n e u r o p e p t i d e s and TH, S P - I R nerve f ibres were dense ly d i s t r ibu ted a r o u n d and among gangl ion cells r ega rd le s s of immunoreac t iv i ty for T H or m E N K (Figs. 3 and 4). T h e a r ea of S P - I R nerve f ibres a r o u n d m E N K - I R cells was the la rges t in the whole of the I N R (Tab le 1). V I P - I R nerve f ibres showed a dense d i s t r ibu t ion among and a r o u n d gangl ion cells i r respec t ive of immunoreac t iv i ty for T H or m E N K (Figs. 5 and 6). Many C G R P - I R nerve f ibres d i s t r ibu ted dense ly a r o u n d T H - I R cells (Fig. 7). C G R P - I R nerve f ibres were obse rved also wi thin nerve bundles . A few C C K - I R nerve f ibres were occas iona l ly found a r o u n d and among T H - I R gangl ion cells (Fig. 8). T h e a r ea of these nerve f ibres was s ignif icant ly smal le r t han those of the VIP- , SP- and C G R P - I R nerve f ibres men- t i oned above (Tab le 1). G A L - I R nerve f ibres were found as dots a m o n g or a r o u n d m E N K - I R cells (Fig. 9), bu t a small n u m b e r of these cells were a p p a r e n t l y devoid of con tac t with G A L - I R fibres. In the sect ions i m m u n o s t a i n e d with S O M ant i se rum, S O M - I R nerve f ibres were found as dots among and a r o u n d gangl ion cells r ega rd le s s of immunoreac t iv i ty for SOM, bu t they showed a t endency to con tac t wi th S O M - I R cells (Fig. 2b). The a r ea of S O M - I R nerve f ibres was the small- est in the rec ta l pa r t (Tab le 1). m E N K - I R nerve f ibres were ha rd ly found at all in the rec ta l pa r t of the INR.
3.2. Ileojejunal part o f I N R
G a n g l i o n cells of the i leoje junal pa r t of the I N R exhib i ted pr inc ipa l ly immunoreac t iv i t i e s for th ree specif ic an t i se ra agains t c a t echo l amine syn- thes iz ing enzymes (TH, A A D C and P N M T ) out of 17 kinds of an t i se ra tes ted (Fig. 10). Some of t hem did not show any immunoreac t iv i ty for T H (Fig. 10a). Consequent ly , gangl ion cells of the i leoje junal pa r t of the I N R were d iv ided into two groups by the immunoreac t iv i ty for TH. H e n c e they are desc r ibed as T H - I R and T H - i m - munone ga t i ve ( IN) cells. T H - I R gangl ion cells were much la rger in n u m b e r than T H - I N cells, which seem to be smal le r in size than T H - I R cells (Fig. 10a,b,c).
O n the o t h e r hand , nerve f ibres in the i leojeju- nal pa r t i m m u n o r e a c t e d with e ight an t i se ra for n e u r o p e p t i d e s ( C G R P , SOM, VIP, SP, CCK, G A L , m E N K and NPY) out of 17 kinds of anti- sera tes ted . Both S P - I R (Figs. 11 and 12) and V I P - I R (Figs. 13 and 14) nerve f ibres were ob- served among and a r o u n d a lmost all the gangl ion cells, tha t is T H - I R (Figs. 11 and 13) and T H - I N cells (Figs. 12 and 14) in the i leo je junal par t . However , a reas of these nerve f ibres were signifi- cant ly smal le r than those in the rec ta l pa r t (Table 1). C G R P - I R nerve f ibres showed a dense distr i- bu t ion and res t r i c ted a r o u n d T H - I R cells (Fig. 15). The a rea of C G R P - I R nerve f ibres was the larges t in the i leoje junal pa r t (Table 1). A few
Fig. 10a-c. Three serial sections from the ileal part of the INR, immunostained with anti-TH (a), PNMT (b) and AADC (c) sera. Ganglion cells being immunoreactive for both AADC (c) and PNMT (b) are divided into two groups by the immunoreactivity for TH (a). Bars = 20/xm
Figs. 11-14. Double immunostaining sections from the jejunal (11,13,14) or ileal (12) part of the INR. The PAP method with DAB for two kinds of peptides, SP (11,12) and VIP (13,14) followed by the PAP method with DAB added the cobalt for TH. TH-IR ganglion cells are gray. 11. SP-IR fibres disseminated fairly among TH-IR cells. 12. SP-IR fibres disseminated fairly among TH-IN cells. 13. TH-IR cells are densely distributed with VIP-IR fibres. 14. VIP-IR fibres distributed fairly among TH-IN cells. Bars = 20 p.m
Figs. 15-20. Double immunostaining sections from the ileal part of the INR. The PAP method with DAB for CGRP (15), CCK (16), SOM (17), mENK (18), GAL (19) and NPY (20), followed by the PAP method with DAB added the cobalt for TH. TH-IR ganglion cells show gray. (15) TH-IR cells are densely distributed with CGRP-IR fibres. (16) A few CCK-IR fibres are found as dots among TH-IR cells. (17) SOM-IR fibres distributed fairly around TH-IN cells. (18) mENK-IR fibres are slightly dotted among TH-IN cells. (19) GAL-IR fibres are seen as dots among TH-IN cells. (20) NPY-IR fibres are found as a few dots among TH-IN cells. Bars = 20 p.m
214 M. Suzuki et al. /Journal o[the Autonomic Nerl'ous A~vstem 49 (1994) 207-21~
CCK-IR nerve fibres were distinguished as dots mainly among T H - I R cells (Fig. 16). The remain- ing four others (SOM, GAL, m E N K and NPY) were localized to TH- IN cells. SOM-IR nerve fibres fairly distributed around TH- IN ganglion cells (Fig. 17). The area of SOM-IR nerve fibres was significantly larger than that around mENK- IR cells in the rectal part (Table 1). m E N K - I R nerve fibres were distinguished as dots among TH-IN cells (Fig. 18). G A L - I R nerve fibres were found as dots among TH-IN cells (Fig. 19). NPY- IR nerve fibres were detected as dots among TH-IN cells (Fig. 20). The area of the latter was by far the smallest of the whole of the INR (Table 1). In the whole of the INR, ganglion ceils and nerve fibres did not show any immunoreactiv- ity with bombesin, 1ENK, NT, PHI, DBH and 5-HT antisera in the present study (negative re- suits not shown).
4. Discussion
The present study shows that ganglion cells of the INR exhibited immunoreactivity for antisera against three catecholamine synthesizing en- zymes. The formaldehyde-induced fluorescence procedure showed that the pararectal ganglia of the INR contain fluorescent and non-fluorescent neurons, and the latter is surrounded with brightly fluorescent basket works of varicose fibres [1,2]. However, no TH-, AADC- and PNMT-IR nerve fibres were observed in any part of the INR in this study. It is still unsettled what the fluorescent basket works correspond to in the INR. There is a report that [3H]5-HT labeled cells and neurites were found in ganglia of the INR and in neural connections with hindgut in chick embryo [5]. Although we cannot detect any immunoreactivity for 5-HT in any part of the INR in this study, the immunoreactivity for 5-HT should not be ex- cluded from the INR.
Ultrastructurally small granular vesicles, gen- erally accepted as evidence of adrenergic nerves, could be demonstrated in the neurons of juxta- ileal ganglia but not the juxta-rectal ganglia of the INR [21]. According to the present results, about one-half of ganglion cells in the rectal part of the
INR showed immunoreactivity for TH, AADC and PNMT. Therefore, further studies may estab- lish whether or not ganglion cells containing small granular vesicles and showing immunoreactivity for PNMT are located in the INR of chicken.
The present results demonstrate that neu- ropept ide-IR nerve fibres were common to T H - I R ganglion cells in the whole of the INR, whereas nerve fibres distributed around TH-IN cells in the ileojejunal part of the INR were clearly dif- ferent from those innervating to TH- IR cells ex- cept for SP- and VIP-IR fibres. Consequently, it is probable that AADC- and PNMT-IR but TH- IN cells are established as a type of ganglion cells in the INR. Recently Lunam [10] reports fibres expressing TH- and opioid-immunoreactivity in the varicosities in the juxta-rectal ganglia of the INR while this was not the case in this paper. Despite repeated attempts, we could not show the clear immunoreactivity for DBH in any part of the INR in the present study. This result should be due to the validity of antiserum for DBH used here.
This is the first report of the presence of TH-IN but AADC- and PNMT-IR ganglion cells in the ileojejunal part of the INR. The presence of two catecholamine biosynthetic enzymes in the ganglion cells in question indicates synthesis rather than the uptake of dopamine a n d / o r nor- adrenaline. However, because immunoreactivity for anti-TH serum has not been found in the ganglion cell in question in the ileojejunal part, the uptake of DOPA by the ganglion cells is a possibility. The significance of these ceils remains to be determined.
Although no direct evidence is available from the present study for the hypothesis that mENK- and SOM-IR ganglion cells correspond to non- fluorescent neurons reported by Cantino et al. [2], the immunohistochemistry techniques used may not be sufficiently sensitive to disprove this hypothesis. There is a report that SOM- and mENK-IR cell bodies appeared in the rectal part of ganglia of the INR at embryonic day 4 and 6, respectively, but no VIP-IR cells were found in the INR [3,4]. In addition, it is also reported that SOM-IR cells are found in large numbers in ganglia of the INR at the time of hatching [4]. In
M. Suzuki et aL /Journal of the Autonomic Nervous System 49 (1994) 207-216 215
the present study no SOM-IR ganglion ceils were found in the ileojejunal part of the INR. This discrepancy between our data and those of Ep- stein and Poulsen [4] may be attributable to the difference in age of materials. Excepting SOM-IR cells in the ileojejunal part at hatching, these results are supported by the present experiment. Immunoreactivities for mENK and SOM were exhibited within the same ganglion cell and were restricted in the rectal part of the INR; mENK-IR nerve fibres innervated ganglion cells were re- stricted to the ileojejunal region, and SOM-IR nerve fibres were distributed throughout the whole of the INR. Bearing these results in mind it is possible that a small number of mENK-IR ganglion cells may not be SOM-IR cells. If this is true, there may be three kinds of ganglion cell: mENK-, SOM- and m E N K / S O M - I R cells, and mENK-IR cells may send the ascending fibres to the ileojejunal part of the INR, and SOM-IR cells may act as the interconnection between gan- glion cells within the INR.
It is very interesting to describe that the results that VIP-IR neurons were identified in the sub- mucous plexus throughout the gut [18], and that VIP-IR innervation of the smooth muscle was more predominant in large intestine, whereas
small intestine VIP-IR nerves were less numer- ous in the smooth muscle [16] support that there is significant difference in area of VIP-IR nerve fibres between in the rectal and in the ileojejunal parts of the INR described in this study.
Our findings indicate that the INR is unusu- ally rich in TH-IR ganglion cells, and that mENK-IR cells are contained only in the rectal part. Therefore, the INR is considered to be a chain of sympathetic ganglions rather than parasympathetic ones (as previously described [20]).
In conclusion, our observations indicate that three kinds of ganglion cell (TH( + ), T H ( - ) and m E N K ( + ) ) are found in the INR, and that they are characterized by the distribution pattern of nerve fibres immunoreacted with eight antisera (SP, VIP, CGRP, SOM, CCK, mENK, GAL and NPY). The relationships between ganglion cells and nerve fibres, including the statistical analysis, are summarized in Fig. 21.
Acknowledgements
We thank Dr. Roy Fuller Intestinal Microecol- ogy Consultant, Reading for linguistic correc-
CCK
RECTAL I LEOJEJ UNAL
CCK mENK
Fig. 21. Schematic representations of ganglion cells in the INR innervated by the varicose nerve fibres. TH-, AADC- and PNMT-IR cells and mENK- and SOM-IR cells are observed in the rectal part. Around the former, SP- (thin stripe) and VIP-IR (thick stripe) nerve fibres are more abundant than CGRP-IR (thin mesh) ones, and a few CCK-IR (close) fibres also are found. Around the latter, areas of SP- and VIP-IR fibres are by far larger than those of GAL- (open) and SOM-IR (thick mesh) fibres. TH-, AADC- and PNMT-IR cells and AADC- and PNMT-IR but not TH-IR cells are discerned in the ileojejunal part. Around the former, CGRP-IR fibres are the most abundant, VIP- and SP-IR fibres come second, next in order is CCK-IR ones. Around the latter, SOM-IR fibres are the most abundant, VIP- and SP-IR fibres show much the same area, and just a few GAL-, mENK- (course dotted line) and NPY-IR (fine broken line) fibres also are detected. For further details, see text.
216 M. Suzuki et al. / Journal of" tile Autonomic Nercous System 49 (1994) 207-216
tions. This study was supported in part by Grant- in-Aid for General Science Research (C) (No. 05660343) of the Ministry of Education, Science and Culture of Japan.
References
[1] Bennett, T. and Malmfors, T., The adrenergic nervous system of the domestic fowl (Gallus domesticus L.), Z. Zellforsch., 106 (1970) 22-50.
[2] Cantino, D., Barasa, A. and Guglielmone, R., Cate- cholamine-containing neurons in Remak's ganglion: a developmental and tissue culture study, J. Neurocytol,, 11 (1982) 763-778.
[3] Epstein, M.L. and Dahl, J.L., Development of enkephalin in the rectum and the ganglion of Remak of the chick, Peptides, 3 (1982) 77-82.
[4] Epstein, M.L. and Poulsen, K.T., Appearance of somato- statin and vasoactive intestinal peptide along the devel- oping chicken gut, J. Comp. Neurol., 311 (1991) 168-178.
[5] Gershon, M.D., Epstein, M.L. and Hegstrand, L., Colo- nization of the chick gut by progenitors of enteric sero- tonergic neurons: Distribution, differentiation, and matu- ration within the gut, Dev. Biol., 77 (1980) 41-51.
[6] Hodgkiss, J.P., The unmyelinated fibre spectrum of the main trunk and side branches of the intestinal nerve in the chicken (Gallus gallus var. domesticus), J. Anat., 148 (1986) 99-110.
[7] Hsu, S. and Soban, E., Color modification of di- aminobenzidine (DAB) precipitation by metallic ions and its application for double immunobistochemistry, J. His- tochem. Cytochem., 30 (1982) 1079-1082.
[8] lchinose, H., Kojima, K., Togari, A., Kato, Y., Parvez, S.. Parvez, H. and Nagatsu, T., Simple purification of aro- matic L-amino acid decarboxylase from human pheochro- mocytoma using high-performance liquid chromatogra- phy, Anal. Biochem., 150 (1985) 408-414.
[9] Kanazawa, T., Ohashi, H. and Takewaki, T., Evidence that cell bodies of non-cholinergic, excitatory neurones which supply the smooth muscle of the chicken rectum arc located in the ganglia of Remak's nerve, Br. J. Pharmacol., 71 (1980) 519-524.
[10] Lunam, C.A., Calbindin, tyrosine hydroxylase and opioid-like immunoreactivity in the intestinal nerve of
Remak of the domestic fowl, J. Auton. Nerv. Syst., 44 (1993) 189-196.
[11] Mogi, M., Harada, M., Kojima, K., Kikuchi, K., Nagatsu, 1. and Nagatsu, T., Effects of repeated injection of t- methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) on striatal tyrosine hydroxylase activity in vitro and tyrosine hydroxylase content, Neurosci. Res , 80 (1987) 213-218.
[12] Nagatsu, I., Phenylethanolamine-N-methyltransferase, In: M.A. Hayat (Ed.), Electron Microscopy of Enzymes. Principles and Methods, Van Nostrand Reinhold, New York, 1977, pp. 52-71.
[13] Nagatsu, I., Immunohistochemistry of biogenic amines and immunoenzyme histochemistry of catecholamine- synthesizing enzymes: application for axoplasmic trans- port and neuronal localization, In: S. Parvez, T. Nagatsu, I. Nagatsu, H. Parvez (Eds.), Methods in Biogenic Amines Research, Elsevier, Amsterdam, 1983, pp. 873-909.
[14] Nagatsu, I., Sakai, M., Yoshida, M. and Nagatsu, T., Aromatic-L-amino acid decarboxylase-immunoreactive neurons in and around the cerebrospinal fluid-contacting neurons of the central canal do not contain dopamine or serotonin in the mouse and rat spinal cord, Brain Res., 475 (1988) 91-1(12.
[15] Sternberger, L.A.. Immunocytochemistry, 2nd Edn., John Wiley&Sons, New York, 1979, pp. 104-109.
[16] Sundler, F., Alumets, J., Fahrenkrug, J., H~kanson, R. and Schaffalitzky de Muckadell, O.B., Cellular localiza- tion and ontogeny of immunoreactive vasoactive intesti- nal polypeptide(VIP) in the chicken gut, Cell Tissue Res., 196 (1979) 193-201.
[17] Takewaki, T., Ohashi, H. and Okada, T., Non-cholinergic and non-adrenergic mechanisms in the contraction and relaxation of the chicken rectum, Jap. J. Pharmacol., 27 (1977) 105-115.
[18] Vaillant, C., Dimaline, R. and Dockray, G.J., The distri- bution and cellular origin of vasoactive intestinal polypeptide in the avian gastrointestinal tract and pan- creas, Cell Tissue Res., 211 (1980) 511-523.
[19] Watanabe, T., Comparative and topographical anatomy of the fowl. LXIV. Sympathetic nervous system of the fowl. Part 2. Nervus intestinalis, Jpn. J. Vet. Sci., 34 (1972) 3(13-313 (in Japanese with English summary).
[2(I] Yntema, C.L. and Hammond, W.S., Experiments on the origin and development of the sacral autonomic nerves in the chick embryo, J. Exp. Zool., 129 (1955) 375-413.
[21] Young, H.M., The ultrastructure of the intestinal nerve of Remak in the domestic fowl. Cell Tissue Res., 260 (1990) 601 616.
