Abstract The effect of age on the proportion of nicoti-namide adenine dinucleotide phosphate diaphorase(NADPHd)-positive neurons was investigated in the my-enteric plexus of five different gastric areas of 1-day-,1-week-, 2-week-, 1-month- and 2-month-old rats. Pro-tein gene product 9.5 immunocytochemistry was used asa marker for the total enteric neuron population in orderto establish the percentage of gastric nitrergic neurons inrelation to age. The percentage of NADPHd-positiveneurons in the proximal parts of the rat stomach(34–38%) is significantly higher than in the antral part(29%). This difference persists in all the age groups in-vestigated. No significant relative increase with age ofNADPHd-positive neurons could be observed in any ofthe areas studied. These findings imply that the increasednitrergic response in the rat proximal stomach as seen inpharmacological studies cannot be explained by an in-creased relative number of nitrergic neurons.

Introduction

The involvement of nitric oxide (NO), together with va-soactive intestinal polypeptide (VIP), pituitary adenylatecyclase-activating peptide and adenosine triphosphate, inthe non-adrenergic, non-cholinergic relaxation within thegastrointestinal tract has been well documented and ac-cepted (Bartho et al. 1998; Bult et al. 1990; Fernandez etal. 1998; Grider et al. 1994; He and Goyal 1993; Li andRand 1990; Shuttleworth et al. 1991; Smits and Lefebvre1996). Convincing morphological, pharmacological andbiochemical evidence is at hand to state that both VIP

and NO are inhibitory (co-)transmitters in the rat stom-ach (Boeckxstaens et al. 1992; Curró et al. 1996;D’Amato et al. 1992; Forster and Southam 1993;Kamata et al. 1993; Lefebvre 1993). For the rat gastricfundus, it has been shown that NO is mainly involved inshort-lasting relaxations and in the initial phase of thesustained relaxation, whereas VIP mainly plays a role inthe subsequent phase of the sustained relaxation(Boeckxstaens et al. 1992; D’Amato et al. 1992; Li andRand 1990). Recent functional investigations of the in-nervation of the rat gastric fundus in 2-, 4- and 8-week-old rats (Smits and Lefebvre 1996) showed that the con-tractile responses to electrical stimulation of the cholin-ergic neurons and to exogenous acetylcholine decreasedto a similar extent from 2 to 8 weeks, pointing to a post-synaptic change in sensitivity to acetylcholine. Thismight be related to a decrease in the number of musca-rinic receptors or a decreased efficiency of the recep-tor–effector coupling. In contrast to the changes of thecholinergic responses, the nitrergic relaxations inducedby electrical stimulation increased from 2 to 8 weeks. Asthe responses to exogenous NO did not change with age,an increase in muscular sensitivity to NO was excluded.The possibility of a growing impact of the nitrergic in-nervation during postnatal development in the weaningperiod was considered (Smits and Lefebvre 1996). In therat colon, the higher density of nitrergic neurons in theproximal compared to the distal part was indeed correlat-ed with more pronounced nitrergic relaxations uponelectrical stimulation (Takahashi and Owyang 1998).The aim of this study was, therefore, to investigatewhether the postnatal functional increase in nitrergic in-nervation was, indeed, paralleled by an elevated propor-tion of the nitrergic neuronal population. A morpho-metric analysis was performed on whole mounts of ratstomach, applying nicotinamide adenine dinucleotidephosphate diaphorase (NADPHd) histochemistry andprotein gene product 9.5 (PGP9.5) immunocytochemis-try. In the above-cited study (Smits and Lefebvre 1996),the term “fundus” is applied to the white aglandular (orproventricular) region of the rat stomach, as is common-

J.-P. Timmermans (✉) · D. AdriaensenLaboratory of Cell Biology and Histology,University of Antwerp (RUCA), Groenenborgerlaan 171,B-2020 Antwerp, Belgiume-mail: jptimmer@ruca.ua.ac.beTel.: +32-3-2180300, Fax: +32-3-2180301

R. LefebvreHeymans Institute of Pharmacology, University of Ghent,De Pintelaan 185, B-9000 Ghent, Belgium

Histochem Cell Biol (1999) 111:429–434 © Springer-Verlag 1999

O R I G I N A L PA P E R

Jean-Pierre Timmermans · Dirk AdriaensenRomain Lefebvre

Postnatal development of nitrergic neurons in the myenteric plexusof rat stomach

Accepted: 31 March 1999

ly used in functional studies. In the actual study, three re-gions were examined in this part, but two regions in theglandular part were investigated as well. In addition to 2,4 and 8 weeks, rats were also studied at 1 day and1 week postnatally.

Materials and methods

Wistar rats of five different postnatal age groups [1 day (n=7),1 week (n=6), 2 weeks (n=5), 1 month (n=5) and 2 months (n=6)]were used. The animals were killed by decapitation after short ex-posure to chloroform vapour. The stomachs were dissected out,rinsed with physiological salt solution and subsequently ligated atthe levels of the lower oesophageal and pyloric sphincter, andfilled with and immersed in 4% paraformaldehyde for 2 h at roomtemperature. After fixation, the stomachs were rinsed in phos-phate-buffered saline (PBS; 0.1 M), refilled with PBS and furtherprocessed in toto for NADPHd-staining. The PBS-filled stomachswere incubated in a solution containing nitroblue tetrazolium(0.25 mg/ml), β-NADPH (1 mg/ml) and Triton X-100 (0.5%) inTRIS-HCl buffer (0.1 M, pH 7.6) for 15 min at 37°C (Scherer-Singler et al. 1983).

Following NADPHd staining, the stomachs were cut openalong the lesser and greater curvatures, and whole mounts contain-ing the longitudinal muscle layer with the myenteric plexus ad-hered were prepared from five different regions (Fig. 1) and fur-ther processed for immunocytochemistry. The tissues were pre-treated with normal goat serum containing 1% Triton X-100 for2 h at room temperature. As a general marker for the enteric neu-rons, a mouse monoclonal antibody raised against PGP9.5 (17 h,room temperature, dilution 1:10,000, Clone 13C4; Biogenesis,UK) was used. A biotinylated sheep anti-mouse IgG (6 h, roomtemperature, dilution 1:100, RPN.1001; Amersham, UK) was usedas the secondary antiserum which was then coupled to a horse rad-ish peroxidase-conjugated streptavidin complex (17 h, room tem-perature, dilution 1:100, RPN.1233; Amersham). In order to visu-alize the immunoreaction, a 5% stock solution of aminoethylcarb-azol (AEC; 40 mg AEC dissolved in 10 ml dimethylformamide;SigmaChemie, Belgium) dissolved in 0.1 M sodium acetate(pH 5.6), to which 10 µl/ml of a 3% H2O2 solution was added, wasapplied for 10 min at room temperature. The preparations were ex-amined and photographed with a Leitz Orthoplan microscope.

For statistical analysis, ten distinct areas of the stomach, fiveventral and five dorsal, were selected for each age group (Fig. 1).The number of NADPHd-stained and PGP9.5-immunoreactive(IR) neurons was counted in randomly selected myenteric ganglia.In general, eight to ten whole mount preparations per area and perage group obtained from at least four different animals wereincluded in this study. On average, 3982 neurons (range2000–5280) per area were counted for each age group. As strongNADPHd staining might mask the PGP9.5 immunoreaction andassuming that nearly all neurons display immunoreactivity forPGP9.5, the summed number of the PGP9.5+/NADPHd+ and thePGP9.5+/NADPHd– neuronal populations was considered as thetotal number (100%) of counted neurons for each area and agegroup. Subsequently, the mean percentage of NADPHd-stainedneurons per area in the myenteric plexus of the five different agegroups was calculated. The data obtained from each area and eachage group were subjected to a mixed ANOVA test in order to eval-uate the effects of age and area on the nitrergic neuronal popula-tion. The proportion of nitrergic neurons compared to the totalnumber of enteric neurons was regarded as a ‘dependent variable’.Age, area and the interaction between age and area were enteredas ‘factors’, and individual and side (ventral/dorsal) as ‘randomvariables’. Degrees of freedom were adapted for statistical depen-dence by Satterthwaite formulas (SAS V6.07). A probability ofP<0.05 was set as the level of significance in all analyses.

Results

The myenteric plexus was well developed in all the agegroups investigated and both NADPHd-stained andPGP9.5-IR enteric neurons were present (Fig. 2). Sincethe statistical analysis showed no significant differencesin the percentages of NADPHd-stained neurons betweenthe ventral and dorsal side, these data were pooled (Fig. 3).The total number of enteric neurons counted in this studyamounted to almost 200,000, 34.7% of which showedNADPHd positivity (Fig. 3). The mean percentage ofNADPHd-stained neurons per area and age group rangedbetween 27.9% and 39.5% (Table 1).

No significant change in the proportion of the nitre-rgic neurons was observed within a distinct region as afunction of time, except for area 1, where a slight butsignificant (P=0.002) decrease could be noted as age in-creased (Table 1), notwithstanding a partial recuperationfrom 1 to 2 weeks. The mean percentage of nitrergicneurons for all age groups in the antral region (area 5:mean percentage=29.1%) was significantly (P=0.001)lower than in the proventricular (area 1: mean percent-age=36.3%; area 2: mean percentage=37.9%; area 3:mean percentage=35.5%) and fundic regions (area 4:mean percentage=34.1%; Table 1).

Discussion

Previous studies (Aimi et al. 1993; Barbiers et al. 1993;Belai et al. 1992; Dawson et al. 1991; Timmermans et al.1994; Young et al. 1992) have shown that NAPDHd his-tochemistry can be used to identify the nitrergic popula-tions within the enteric nervous system. PGP9.5-IR wasused as a marker for enteric neurons. There is still debateregarding the question whether a general marker for allenteric neurons does exist (Karaosmanoglu et al. 1996).The observation that PGP9.5 immunoreactivity is pres-

430

Fig. 1A–E Schematic drawings of the different regions of the ratstomach and of the corresponding areas (A1–A5, B1–B5) includedin the quantitative analysis. A Antrum, D duodenum, F fundic(glandular) region, O oesophagus, P proventricular (aglandular)region, T line of transition between glandular and non-glandularregion

ent in only about 80% of all the myenteric neurons ob-served in the gastrointestinal tract of the rat (Eaker andSallustio 1994) is not a limiting factor for the present ex-perimental approach since only ratios are taken into con-sideration. Furthermore, the problems encountered in thestudy of Karaosmanoglu and coworkers (1996) in which

431

Fig. 2 Low (a) and high (b) magnification micrographs of themyenteric plexus in a whole mount preparation of the proventricu-lar region of rat stomach at postnatal day 1 showing nicotinamideadenine dinucleotide phosphate diaphorase (NADPHd; purple-blue)- and protein gene product 9.5 (PGP9.5; red-brown)-labeledneurons. Also note an extensive network of NADPHd-positivenerve fibres in the muscle layer

a staining of the surrounding neuropil obscured the out-lines of the individual neurons, were largely avoided byapplying a higher dilution rate of the primary antiserum.

Based on earlier pharmacological studies showing adecreased sensitivity to acetylcholine and an increasednitrergic response with age in the rat gastric fundus(Smits and Lefebvre 1996), it was hypothetised that thisenhanced nitrergic response might be reflected in an in-creased percentage of neurons expressing the NO-syn-thesising enzyme. Such an increase has been reported forthe rat colon up to 3 weeks (Matini et al. 1997) or even 6months postnatally (Belai et al. 1995). However, in the

present and to our knowledge most extensive quantita-tive study so far, no significant increase with age in therelative proportion of NADPHd-positive neurons couldbe seen in the investigated regions in the stomachs of1-day-, 1-week-, 2-week-, 1-month- and 2-month-oldrats. Although the pharmacological study (Smits andLefebvre 1996) comprised data from neurons belongingonly to areas 1–3 in the present study, our data allow usto conclude that the increase in the nitrergic functionalresponse observed from 2 to 8 weeks postnatally cannotbe explained by an elevated relative proportion of nitre-rgic neurons. Also in rat ileum no significant increase in

432

Table 1 Proportion of nicotinamide adenine dinucleotide phosphate diaphorase (NADPHd)-positive neurons for each area and each agegroup. Data are expressed as mean percentage±SD.(n Number of animals)

1 day 1 week 2 weeks 1 month 2 months

Area 1 % NADPHd 38.55±2.59 35.08±2.7 37.05±3.15 36.18±1.55 34.45±1.51n 6 5 4 5 6

Area 2 % NADPHd 39.06±2.28 35.51±2.29 39.48±2.32 38.13±1.87 37.07±1.89n 6 5 4 5 6

Area 3 % NADPHd 35.82±1.49 33.74±2.09 34.65±3.23 37.28±0.92 36.05±1.11n 5 5 4 4 4

Area 4 % NADPHd 33.9±1.94 34±1.79 33.66±1.75 34.67±2.34 34.05±1.42n 5 5 4 4 4

Area 5 % NADPHd 29.01±1.46 27.87±2.22 30.08±1.68 29.56±0.87 28.72±1.61n 4 5 3 5 5

Fig. 3 Histogram showing the counted numbers of NADPHd-positive and PGP9.5-immunoreactive/NADPHd-negative neurons per areaand per age group. Each column represents the sum of all counted neurons per area and age

mean percentage of NADPHd-stained myenteric neuronsper ganglion was noted postnatally (Belai et al. 1995).Since no increase in the percentage of nitrergic neuronscould be observed, an increase in density of the nitrergicpopulation during the postnatal period under study canbe excluded. Therefore, other possibilities which have tobe considered are: (1) a higher NO release per neuron,(2) an increasing branching of nitrergic axons, not paral-leled by an increase in nitrergic neuronal somata, result-ing in a higher density of the nitrergic innervation of theouter muscle layer and (3) an increase in muscular NOsynthase resulting in muscular NO production has beensuggested. In the canine lower oesophageal sphincter, amyogenic NO synthase was described that was NADPHd-reactive but was not recognized by antibodies againstneural, inducible and endothelial NO synthase (Salapateket al. 1998). However, no muscular NADPHd reactivitywas seen in the actual study. Recently, the expression ofendothelial NO synthase was reported in rabbit gastricand human intestinal smooth muscle cells (Teng et al.1998).

The proportion of NADPHd-positive myenteric neu-rons in the rat stomach is comparable to that in the ratproximal colon (31%) and slightly higher than that in therat ileum (23%; Belai et al. 1995). Similar numbers havebeen reported for the mouse (26%; Sang and Young1996) and guinea-pig (21%; Costa et al. 1996) small in-testine. These numbers appear, however, to be only halfof the reported proportion of myenteric nitrergic neuronsin the adult human stomach, where they account forabout 50–60% of the total myenteric population(Manneschi et al. 1998).

In addition, the results of the present study demon-strate that the distribution of NADPHd-positive entericneurons differs between the antral and other regions ofthe stomach. In all age groups, the percentage of nitre-rgic neurons in the antrum of 1-day- to 2-month-old ratsis significantly lower than in the proventricular and fun-dic regions. The higher proportion of relaxant nitrergicneurons in the proximal part of the stomach might be re-lated to the profound relaxation that has to occur in thisregion upon food intake (Kumar 1992). In conclusion,the morphological data do not support the hypothesisthat an increased number of nitrergic myenteric neuronsis responsible for the increased nitrergic response in therat proximal stomach from 2 to 8 weeks postnatally.

Acknowledgements The authors would like to thank J. VanDaele, L. Svensson and D. De Rijck for technical assistance andDrs. F. Adriaensen and E. Goetghebeur for help with the statisticalanalysis. Financial support was provided by a grant within theframework of IUAP P4/16.

References

Aimi Y, Kimura H, Kinoshita T, Minami Y, Fujimura M, VincentSR (1993) Histochemical localization of nitric oxide synthasein rat enteric nervous system. Neuroscience 53:553–560

Barbiers M, Timmermans J-P, Scheuermann DW, Adriaensen D,Mayer B, De Groodt-Lasseel MHA (1993) Distribution and

morphological features of nitrergic neurons in the large intes-tine. Histochemistry 100:27–34

Bartho L, Lenard L, Szigeti R (1998) Nitric oxide and ATP co-me-diate the NANC relaxant response in the guinea-pig taeniacaeci. Naunyn Schmiedebergs Arch Pharmacol 358:496–499

Belai A, Schmidt HHHW, Hoyle CHV, Hassall CJS, Saffrey MJ,Moss J, Fostermann U, Murad F, Burnstock G (1992) Colocal-ization of nitric oxide synthase and NADPH-diaphorase in themyenteric plexus of the rat gut. Neurosci Letts 143:60–64

Belai A, Cooper S, Burnstock G (1995) Effect of age on NADPH-diaphorase-containing myenteric neurones of rat ileum andproximal colon. Cell Tissue Res 279:379–383

Boeckxstaens GE, Pelckmans PA, De Man JG, Bult H, HermanAG, Van Maercke YM (1992) Evidence for a differential re-lease of nitric oxide and vasoactive intestinal polypeptide bynonadrenergic noncholinergic nerves in the rat gastric fundus.Arch Int Pharmacodyn 318:107–115

Bult H, Boeckxstaens GE, Pelckmans PA, Jordaens FH, VanMaercke YM, Herman AG (1990) Nitric oxide as an inhibitorynon-adrenergic non-cholinergic neurotransmitter. Nature 345:346–347

Costa M, Brookes SJH, Steele PA, Gibbins I, Burcher E, KandiahCJ (1996) Neurochemical classification of myenteric neuronsin the guinea-pig ileum. Neuroscience 75:949–967

Curró D, Volpe AR, Preziosi P (1996) Nitric oxide synthase activi-ty and non-adrenergic non-cholinergic relaxation in the ratgastric fundus. Br J Pharmacol 117:717–723

D’Amato M, Curró D, Montuschi P (1992) Evidence for dualcomponents in the non-adrenergic non-cholinergic relaxationin the rat gastric fundus: role of endogenous nitric oxide andvasoactive intestinal polypeptide. J Auton Nerv Syst 37:175–186

Dawson TM, Bredt DS, Fotuhi M, Hwang PM, Synder SH (1991)Nitric oxide synthase and neuronal NADPH diaphorase areidentical in brain and peripheral tissues. Proc Natl Acad SciUSA 88:7797–7801

Eaker EY, Sallustio JE (1994) The distribution of a novel interme-diate filament protein defines subpopulations of myentericneurons in rat intestine. Gastroenterology 107:666–674

Fernandez E, Guo X, Vergara P, Jimenez M (1998) Evidence sup-porting a role for ATP as non-adrenergic non-cholinergic in-hibitory transmitter in the porcine ileum. Life Sci 62:1303–1315

Forster ER, Southam E (1993) The intrinsic and vagal extrinsic in-nervation of the rat stomach contains nitric oxide synthase.Neuroreport 4:275–278

Grider JR, Katsoulis S, Schmidt WE, Jin J-G (1994) Regulation ofthe descending relaxation phase of intestinal peristalsis by PA-CAP. J Auton Nerv Syst 50:151–159

He XD, Goyal RK (1993) Nitric oxide involvement in the peptideVIP-associated inhibitory junction potential in the guinea-pigileum. J Physiol 461:485–499

Kamata K, Kohzuki M, Misawa M, Kasuya Y (1993) Involvementof nitric oxide pathway in non-adrenergic non-cholinergic(NANC) relaxation in the rat stomach: differential innervationof NANC nerves in the longitudinal and circular muscle of thefundus. Gen Pharmacol 24:1403–1410

Karaosmanoglu T, Aygun B, Wade P, Gershon MD (1996) Region-al differences in the number of neurons in the myenteric plex-us of the guinea pig small intestine and colon: an evaluation ofmarkers used to count neurons. Anat Rec 244:470–480

Kumar D (1992) Gastric motor physiology and pathophysiology.In: Gustavsson S, Kumar D, Graham DY (eds) The stomach.Churchill Livingstone, Edinburgh, pp 129–156

Lefebvre RA (1993). Non-adrenergic non-cholinergic neurotrans-mission in the proximal stomach. Gen Pharmacol 24:257–266

Li CG, Rand MJ (1990) Nitric oxide and vasoactive intestinalpolypeptide mediate non-adrenergic, non-cholinergic inhibito-ry transmission to smooth muscle of the rat gastric fundus. EurJ Pharmacol 191:303–309

433

Manneschi LI, Vannucchi MG, Bechi P, Faussone-Pellegrini MS(1998) Neuron density and distribution of NADPH-diaphorasepositive neurons in the human stomach. Neurosci Lett 250:169–172

Matini P, Mayer B, Faussone-Pellegrini MS (1997) Neurochemi-cal differentiation of rat enteric neurons during pre- and post-natal life. Cell Tissue Res 288:11–23

Salapatek AMF, Wang Y-F, Mao Y-F, Lam A, Daniel EE (1998)Myogenic nitric oxide synthase activity in canine lower esoph-ageal sphincter: morphological and functional evidence. Br JPharmacol 123:1055–1064

Sang Q, Young HM (1996) Chemical coding of neurons in the my-enteric plexus and external muscle of the small and large intes-tine of the mouse. Cell Tissue Res 284:39–53

Scherer-Singler U, Vincent SR, Kimura H, McGeer EG (1983)Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J Neurosci Methods 9:229–234

Shuttleworth CWR, Murphy R, Furness JB (1991) Evidence thatnitric oxide participates in non-adrenergic inhibitory transmis-sion to intestinal muscle in the guinea-pig. Neurosci Lett130:77–80

Smits GJM, Lefebvre RA (1996) Development of cholinergic andinhibitory non-adrenergic non-cholinergic responses in the ratgastric fundus. Br J Pharmacol 118:1987–1994

Takahashi T, Owyang C (1998) Regional differences in thenitrergic innervation between the proximal and distal colon inrats. Gastroenterology 115:1504–1512

Teng B-Q, Murthy KS, Kuemmerle JF, Grider JR, Sase K, MichelT, Makhlouf GM (1998) Expression of endothelial nitric oxidesynthase in human and rabbit gastrointestinal smooth musclecells. Am J Physiol 275:G342-G351

Timmermans J-P, Barbiers M, Scheuermann DW, Bogers JJ,Adriaensen D, Fekete E, Mayer B, Van Marck EA, De Groodt-Lasseel MHA (1994) Nitric oxide synthase immunoreactivityin the enteric nervous system of the developing human diges-tive tract. Cell Tissue Res 275:235–245

Young HM, Furness JB, Shuttleworth CWR, Bredt DS, Snyder SH(1992) Co-localization of nitric oxide synthase immunoreac-tivity and NADPH-diaphorase staining in neurons of the guin-ea pig small intestine. Histochemistry 97:375–378

434