Absence of protective role of afferent nerves in early intestinal mucosal alterations induced by abdominal irradiation in rats.

Purpose : To assess the early effects of primary afferent nerve suppression by systemic treatment with the neurotoxin capsaicin in an acute model of abdominal irradiation in rats (10 Gy, gamma). Materials and methods : Changes in myeloperoxidase (MPO) activity, calcitonin gene-related peptide (CGRP) tissue content, number of mast cells and apoptotic cells were determined in jejunum and ileum in four groups of rat male Wistar (vehicle sham-irradiated, vehicle irradiated, capsaicin sham-irradiated and capsaicin irradiated) at 1 and 3 days post-irradiation. Results : In vehicle irradiated rats, CGRP was significantly increased from the first day after irradiation in jejunal mucosa; MPO activity increased in both segments at day 3 but not at day 1 after irradiation; the number of detectable mucosal mast cells dropped to nearly zero on days 1 and 3, while the apoptotic cells in the intestinal mucosa were significantly increased at day 1. Similar results were obtained for mast cells and apoptosis in capsaicin irradiated rats as compared to capsaicin sham-irradiated rats, while MPO activity was significantly increased and CGRP concentration in jejunal mucosa significantly decreased from the first day in these rats in comparison with capsaicin sham-irradiated rats. Conclusions : Intestinal sensory innervation seems not to have a major protective role against a radiation-induced intestinal inflammatory reaction.

Some studies also reported the eå ects of both total Results: In vehicle irradiated rats, CGRP was signi cantly increased from the rst day after irradiation in jejunal mucosa; body or abdominal irradiation on mast cell degranula-MPO activity increased in both segments at day 3 but not at tion (Harari et al. 1994, Sedgwick andFerguson 1994), day 1 after irradiation; the number of detectable mucosal mast and which suggested a role for mast cell mediators in cells dropped to nearly zero on days 1 and 3, while the apoptotic the in ammatory response. cells in the intestinal mucosa were signi cantly increased at day It has been reported that abdominal irradiation 1. Similar results were obtained for mast cells and apoptosis in capsaicin irradiated rats as compared to capsaicin sham-induces early neurally mediated intestinal responses. in dogs (Erickson et al. 1994). In the rat colon, Conclusions : Intestinal sensory innervation seems not to have a major protective role against a radiation-induced intestinal neurally evoked electrolyte transport has been found in ammatory reaction.
to be dramatically reduced on the rst day after exposure to ionizing radiation (François et al. 1998). These alterations of intestinal neural functions are 1. Introduction supported by increased expression of transmitters of Application of local doses of ionizing radiation for neural aå erents, such as substance P (SP), after whole radiotherapy of abdominal and pelvic malignancies body irradiation (Esposito et al. 1996) and vasoactive leads to acute gastrointestinal symptoms seen within intestinal peptide (VIP) and substance P after irradihours to days, as well as to chronic radiation enteropaation of upper abdomen in rats (Hockerfelt et al. thy within months to years (Yeoh et al. 1993). Acute 2000. Furthermore, an increase in CGRP-immunoradiation enteritis is characterized by diarrhea, nausea reactive nerve bres has been also reported after and vomiting while chronic enteropathy is dominated whole body irradiation in rats (Zhang et al. 1998). by haemorrhage and ulceration. The origin of these Other authors have demonstrated modi cations in symptoms is not clear but they are classically attributed neural functions after diå erent modes of irradiation to the breakdown of epithelial integrity associated with in human studies, suggesting a signi cant role for a decrease in water absorption combined with excessnerves in in ammatory response of radiation-induced ive secretion and loss of sodium and chloride, leading enteropathy. Radiotherapy of the abdominal sphere has been shown to modify the innervation of the radiation on the neural control of intestinal function were injected subcutaneously on 4 consecutive days with capsaicin (Sigma Chemical, St Quentin-even though this was suggested over 40 years ago (Conard 1956).
Fallavier, France; 40 mmol/l in 80% NaCl 0.9%, 10% ethanol and 10% tween 80). A total dose of A role for the neurotransmitters of sensory aå erents, such as tachykinins or calcitonin gene-related peptide 125 mg/kg was administered under anaesthesia (ketamine 60 mg/kg, ip; xylazine 5 mg/kg, ip) in doses of (CGRP), in the control of intestinal in ammation is now accepted (Sharkey 1992, Maggi 1997. Despite 10, 25, 40 and 50 mg/kg on days 1-4, respectively. Control animals received equal volumes of vehicle. some con icting reports, it is well established that capsaicin-sensitive aå erents play a protective role The eå ectiveness of capsaicin treatment was assessed according to Gamse (Gamse 1982) by means of the against experimental intestinal in ammation in animals (Reinshagen et al. 1996, McCaå erty et al. eye-wiping test, which consists of impaired chemosensitivity of corneal aå erents to one drop of 1% NH 4 OH 1997, Mazelin et al. 1998a, Holzer 1998. In addition, a proin ammatory role of sympathetic nerves has instilled into the eye. Animals treated with capsaicin that showed any wiping movement were excluded been shown (McCaå erty et al. 1997). CGRP is released from sensory nerve bres in the gut wall during the from the study. Capsaicin-treated and vehicle-treated animals were irradiated 10 days after treatment. time course of experimental colitis in animals (Eysselein et al. 1992, Keates et al. 1998) and in in ammatory bowel disease in humans (Eysselein et al. 2.3. Irradiation protocol 1992). However, in contrast to the eå ects observed with tachykinin receptor antagonists (Mazelin et al. Rats were exposed to abdominal gamma ( 60 Co source, 3900 Ci) irradiation with a single dose of 1998b), CGRP receptor antagonists have been found to aggravate (Reinshagen et al. 1998) and CGRP to 10 Gy (0.96 Gy/min) under anaesthesia induced by pentobarbital (60 mg/kg; i.p.). The irradiation source decrease (Mazelin et al. 1999) the severity of experimental colitis, suggesting that CGRP exerts mucosal was placed in the vertical position and collimated to delimit an irradiation eld (6 cm) and animals were protection during intestinal in ammation. Since early intestinal alterations induced by abdominal irradiation irradiated in a zone comprised between xyphoid cartilage and iliac crests. Sham-irradiated rats were involve both in ammatory and neuronal pathways, the question arises as to whether sensory aå erents play anaesthetised in the same conditions but not exposed to the radioactive source. a role in these alterations.
The aim of this study was to determine whether Four groups of rats (six rats in each group) were examined and sacri ced 1 and 3 days post-capsaicin-sensitive denervation modi es some parameters which are characteristic of the early mucosal irradiation: vehicle sham-irradiated, vehicle irradiated, capsaicin sham-irradiated and capsaicin alterations induced by abdominal irradiation in jejunum and ileum in rats. In addition tissue concen-irradiated groups. trations of CGRP, a mediator characteristic of capsaicin-sensitive aå erents (Sternini et al. 1987), 2.4. Histological study were measured in tissue extracts of intestinal wall.
Two pieces (5 mm) of both jejunum and ileum were taken from each rat at 10 cm from the ligament 2. Materials and methods of Treitz and 5 cm from the ileo-cecal junction, 2.1. Animals respectively. Tissues were xed in 10% neutralized formalin and Carnoy's solution, respectively. The Experiments were performed in male Wistar rats samples were cleared in a vacuum in ltration pro-( Janvier, Le Genest Saint Isle, France) initially cessor, embedded in paraae n blocks and 5 mm sections weighing 200-250 g. Rats were housed under conwere made with a rotating microtome. Formalin trolled lighting conditions, with standard diet (Usine xative was used for routine histological analysis with d'Alimentation Rationnelle, Epinay-sur-Orge, hematoxylin and eosin staining, whereas Carnoy's France) and water provided ad libitum. All protocols xative was used for identi cation of intestinal mast were conducted according to the French regulations cells with Alcian blue-Safranin O staining (Roberts for animal experimentation (Ministry of Agriculture, et al. 1990). The number of mucosal mast cells was Act 87-848, 19 October 1987).
determined under a light microscope ( Ö 40 objective) within 10 randomly selected elds of the mucosa 2.2. Capsaicin treatment (Wallace et al. 1992). The number of apoptotic cells in the crypts of small intestine was determined on Sensory denervation was performed according to the protocol described by Holzer (Holzer 1991). Rats hematoxylin and eosin stained sections. The number Intestinal sensory nerves and irradiation of apoptotic cells were counted over complete sec-The RIA for CGRP was performed using rabbit antibody RAS 6006 raised against rat CGRP. The tions (between 100-200 crypts) and data are expressed as the number of apoptotic cells per 50 sensitivity of this assay is 3 pg/tube. In brief, aliquots of 100 ml of rehydrated antiserum were incubated for whole crypts per animal in order to standardize the procedure (Merritt et al. 1996, Potten and Grant 16-24 h at 4ß C with standards or the unknown samples in a nal volume of 200 ml; 15 000 cpm/ 1998). Histological analyses were carried out by a pathologist (Dr J. Pomiès, Histotox, La Rochelle, 100 ml of 125 I-CGRP (speci c activity: 1717 Ci/ mmole) was added and incubated once again for France) who was unaware of the treatment groups.
16-24 h at 4ß C. To perform the second antibody separation, 100 ml of dilute goat anti-rabbit IgG 2.5. MPO activity measurement serum (GARGG 500) and 100 ml of dilute normal rabbit serum (NRS 500) were added. Precipitates Intestinal segments (8-10 mm in length) adjacent to those used for microscopic evaluation were rinsed were allowed to form for 2 h at room temperature, then 500 ml of RIA buå er was added. RIA tubes and stored at -80ß C for no more than one week. MPO activity was determined using the method of were centrifuged at 1700 g for 20 min at 4ß C. Then the supernatants were discarded by aspiration and Bradley et al. (1982). Brie y, samples were homogenized with a T25 Ultraturrax in phosphate buå er the 125 I radioactivity in the pellet was determined in a gamma counter (Cobra II auto-gamma, Packard, (50 mm, pH 6.0), snap-frozen and thawed three times, and centrifuged for 15 min at 8000 g at 4ß C. The Rungis, France). pellets were sonicated with 500 ml of hexadecyl trimethylammonium bromide (HTAB 0.5%, w/v, in 50 mm 2.7. Statistical analysis phosphate buå er). After centrifugation for 15 min at 8000 g at 4ß C, 100 ml of supernatant was mixed with Results are expressed as means Ô SEM. Multiple 3 ml of buå er containing phosphate buå er 50 mm, comparisons were performed with non-parametric 0.167 mg/ml of o-dianisidine dihydrochloride and analysis of variance and followed by Dunn's multiple 0.0005% hydrogen peroxide. Absorbance at 450 nm comparison post-test. Statistical signi cance was was determined with a spectrophotometer (DU-640, accepted if p < 0.05. Beckman, Gagny, France) for 2 min. One unit of MPO was de ned as the quantity catalysing the decomposition of 1 mmol of hydrogen peroxide to 3. Results water per minute at 25ß C. MPO from human leuko-3.1. Eå ects of irradiation cytes was used as standard. Protein content was measured by absorbance at 750 nm with a Bio-Rad Following 10 Gy abdominal irradiation, food inprotein assay kit (Bio-Rad, Ivry sur Seine, France). take was markedly decreased by 54.8Ô 6.2% and MPO activity was expressed in units per gram of 88.2Ô 4.0% on days 1 and 3 respectively (n 5 4). This protein.
attenuated food intake was associated with a reduction in body weight (7.7Ô 0.6% and 9.3Ô 0.8% decrease on days 1 and 3, respectively; n 5 6). 2.6. Radioimmunoassay for CGRP CGRP concentrations were determined in control and irradiated rats using jejunal and ileal segments 3.1.1. Mast cell and structural histology. In the two intestinal segments investigated, the number of histo-(5 cm in length) adjacent to those collected for MPO activity and histology determinations. Tissues were logically detectable mucosal mast cells decreased dramatically (Õ 95%) on the rst day after irradiation dissected on ice and separated into muscular and mucosal layers, weighed and boiled in 20 vols (w/v) and mast cells were nearly undetectable on the third day ( gure 1). Routine histology revealed neither of 0.1 N HCl for 10 min. They were then homogenized using a T25 Ultraturrax and centrifuged at destruction of the continuity of the epithelium nor any alterations characteristic of in ammation of both 1700 g for 10 min. The supernatants were stored at Õ 20ß C for later neutralization and radioimmuno-mucosa and muscular layer of the small intestine on day 1 and on day 3. In the small intestine there was assay (RIA). Neutralization of tissue extracts was performed by addition of NaOH (1 M) to a nal pH a three-fold increase in apoptotic cells in the crypts in vehicle irradiated group (18.3Ô 1.7 apoptotic of 7. CGRP immunoreactivity was directly determined on the neutralized tissue homogenate super-cells/50 crypts; n 5 12, p< 0.05) as compared with the vehicle sham-irradiated group (5.9Ô 1.1 apoptotic natant according to a RIA protocol (Peninsula Laboratories, St Helens, UK) using rabbit antiserum. cells/50 crypts) at 1 day after irradiation. However vehicle-irradiated group values. * vehicle sham-irradiated rats; + capsaicin shamirradiated rats; vehicle-irradiated rats; capsaicinirradiated rats. (means Ô SEM, n 5 6). *p < 0.05,

Eå ects of irradiation after sensory
signi cantly diå erent from corresponding control values.

denervation
Capsaicin pretreatment did not signi cantly modify the irradiation-induced decreases in food at 3 days the number of apoptotic cells was similar intake (6.8Ô 0.9% and 8.0Ô 1.1% on days 1 and 3 to vehicle sham-irradiated group (6.2Ô 0.6, n 5 6). respectively; n 5 4; means Ô SEM) or body weight (51.01 Ô 4.6% and 87.6Ô 2.5% on days 1 and 3, respectively, after irradiation; n 5 6, means Ô SEM). 3.1.2. MPO activity. On the rst day after irradiation, no signi cant change in MPO activity was observed at the two levels investigated. On the third day, 3.2.1. Mast cells and routine histology. The marked MPO activity dramatically increased at the two decrease in the number of mast cells observed in the levels, the greatest increase (32-fold) being observed small intestine on the rst and third days after in the jejunum (318.5Ô 55.0 versus 10.1Ô 2.3 U/g irradiation was not signi cantly modi ed by capsaicin protein; gure 2). pretreatment ( gure 1). As observed in vehicletreated rats, neither epithelial disruption nor marked in ammatory processes were detected in either the 3.1.3. CGRP content. On day 1, jejunal mucosa CGRP levels were increased ( p< 0.05 versus control; mucosa or external muscle layers on days 1 and 3 after irradiation in capsaicin-treated rats. Capsaicin table 1). At this time no changes were seen in ileal CGRP content (table 1). However, on the third day pretreatment alone did not cause apoptosis in the small intestine and levels were similar to those after irradiation, CGRP content signi cantly increased in small intestine ( jejunum and ileum) both described in the previous section (5.9Ô 0.9, n 5 12, in the capsaicin sham-irradiated group and 5.9Ô 1.1, in mucosa and muscle layers ( p< 0.05 versus control). Values are means Ô SEM, n 5 6 ( pmol/g tissue). *p < 0.05, caspaicin treatment compared with vehicle. †p< 0.05, irradiated animals compared with sham-irradiated animals.
n 5 12, in the vehicle sham-irradiated group). pared with vehicle group at the two sites investigated (table 1). Capsaicin treatment (22.9Ô 3.0 apoptotic cells/50 crypts; n 5 12) did not modify the increased number of apoptotic cells observed in vehicle irradiated rats (18.3Ô 1.7 apoptotic cells/50 crypts; n 5 12) from the 4. Discussion rst day after irradiation.
The present study provides evidence that a single dose (10 Gy) of abdominal irradiation increases MPO activity and concentrations of CGRP and decreases 3.2.2. MPO activity. In capsaicin-treated rats on the mast cell numbers in the small intestine. The routine rst day after irradiation, in contrast to that observed histology did not detect any epithelial disruption 1 in vehicle animals, there was a signi cant increase days and 3 days after irradiation, but number of (4.5-and 3-fold) in MPO activity ( gure 2). In addiapoptotic cells increased in crypts of the small tion, in the jejunum, the increase in MPO activity intestine 1 day after irradiation. was diå erent from both capsaicin sham-irradiated The absence of epithelial damage, which agrees and vehicle irradiated rats. However, the increase in with observations of others (Empey et al. 1992, MPO activity observed on the third day after irradi- MacNaughton et al. 1997) does not exclude functional ation in vehicle rats was not signi cantly modi ed alterations of the mucosa. The appearance of apopby capsaicin pretreatment ( gure 2). totic cells is in agreement with other studies (Arai et al. 1996, Ruifrok et al. 1997 as is the rapid disappearance of mast cells (Cummins et al. 1989, 3.2.3. CGRP content. Capsaicin pretreatment signi cantly decreased levels of CGRP in the muscle layers Harari et al. 1994). However no in ltration of neutrophils was seen histologically at any time, which did of both jejunum and ileum as compared with control vehicle values ( p< 0.05), but not in the mucosal not correlate with increased MPO activity at 3 days. Buell and Harding (1989) found an increase in layer. One day after irradiation, pretreatment with capsaicin signi cantly ( p< 0.05) decreased CGRP neutrophil in ltration within 12 h after 10 Gy abdominal irradiation in rats. At 24 h, in agreement with content in jejunum mucosa compared with vehicle pretreatment; in ileum mucosa, CGRP concentra-the present study, there was no evidence of in ltration. In contrast, in other models of intestinal tions tended to be lower in the capsaicin rats than in vehicle animals, but this decrease did not reach in ammation, such as that provoked by administration of trinitrobenzenosulphonic acid (TNBS), acute statistical signi cance.
Three days after irradiation, there was no diå er-in ammatory responses have been observed with an important increase of neutrophil in ltration at the ence between the capsaicin group and the vehicle group in the mucosa of both jejunum and ileum; in site of TNBS administration (Miller et al. 1993). In the present study, the irradiation was external and contrast, CGRP contents in muscle layers were signicantly decreased ( p< 0.05) in capsaicin group com-was not to a speci cally localized site of the gut, like TNBS-induced administration, but over a large of rats tissue levels of substance P were decreased in the ileum (Esposito et al. 1996). abdominal eld. In ltration of neutrophils was In the small intestine CGRP is localized in both observed histologically only infrequently on 5-mm extrinsic and intrinsic neurons (Sternini et al. 1987) sections and so, it is diae cult to compare such measand measurement of CGRP levels in intestinal tissue urements to those of MPO activity which uses a whilst providing tissue concentration of the peptide much larger tissue sample. However, no change in does not relate directly to neuronal activity. small intestine MPO activity has been observed 24 h Nevertheless, the increase of tissue CGRP concentraand 48 h after whole-body irradiation (MacNaughton tions 1 day after irradiation may re ect either blocked et al. , MacNaughton et al. 1998) and, moreover, CGRP release, increased synthesis or decreased cataa decrease in MPO activity has been reported at the bolism. Consequently, we suggest that, according to colonic level 48 h after irradiation (MacNaughton the increase of CGRP levels in jejunum 1 day after et al. 1998). On the other hand, an increased jejunal irradiation in vehicle-treated animals compared to MPO activity has been shown 2 h post-irradiation the decrease seen in capsaicin-treated animals 1 day (MacNaughton and Prud'homme-Lalonde 1995).
after irradiation, the eå ect of sensory aå erents against Thus it seems that data concerning changes in MPO irradiation-induced in ammation may depend in part after irradiation are still controversial and the diå eron CGRP at this time. In support of this, a recent ences between studies seem to depend upon the study showed an increase in CGRP-immunoreactive intestinal segment, the mode of irradiation and the nerve bres in mesenteric arteries in rats, 24 h after time at which the observations were performed. a whole-body irradiation; the authors suggested a Destruction of sensory aå erents by capsaicin did protective function of CGRP against irradiation not modify the mast cell depletion observed after (Zhang et al. 1998). irradiation. According to the anatomical relationships In contrast, on the third day after irradiation we between sensory neurons and mast cells, and the did not observe any protective action of sensory well-known ability of substance P, a major tachykiaå erents on intestinal in ammation as assessed by nergic mediator of sensory aå erents, to degranulate MPO activity and it may be postulated that the mast cells (Maggi 1997), one can speculate that protective action of aå erents is overwhelmed by substance P, which is rapidly released from the several pro-in ammatory factors. Finally, our study intestinal wall after irradiation (Esposito et al. 1996) shows that ablation of sensory nerves by capsaicin could be responsible in part for the irradiationdoes not aå ect the marked enhancement of mucosal induced mast cell degranulation. Such an hypothesis apoptosis observed on the rst day after irradiation. is not supported by our results showing that mast This suggests that intestinal sensory innervation may cell depletion occurs independently of the sensory not be involved in the protection of the intestinal innervation. Similarly, Cummins et al. (1994) mucosa against radiation-induced apoptosis. Some observed that mast cell degranulation associated with data for a role of innervation in apoptosis have been weaning in rats was not modi ed after destruction of already reported for organs or tissues other than the sensory aå erents by capsaicin. digestive tract. For example, in newborn rats some In contrast to the marked mast cell depletion no muscle bres in developing striated muscle undergo change in MPO activity occurred on the rst day apoptosis due to lost innervation resulting from the post-irradiation. However, a signi cant increase of retraction of nerve terminals (Trachtenberg 1998). MPO activity was found in capsaicin-treated animals, Moreover, experimental denervation induces a sevsuggesting a role for sensory aå erents in the early eral-fold increase in the magnitude of bre apoptosis. stage of irradiation-induced in ammation. A protect-One important property of CGRP is vasodilatation; ive role of sensory neurons has been observed in thus a reduced release may eå ectively create tissue diå erent animal models where in ammation was anoxia and subsequent apoptosis of intestinal epitheinduced by compounds such as TNBS (Reinshagen lial cells as has been demonstrated in cases of ischet al. 1996ischet al. , Mazelin et al. 1998a) and sodium dextran aemia-reperfusion in mice (Farber et al. 1999). Despite sulphate (Domek et al. 1997) or ricin (Shea-Donohue the increase in CGRP levels after irradiation, the et al. 1997). The main mediators of sensory aå erents irradiation-induced apoptosis does not seem to be are CGRP and tachykinins. Substance P is released associated with CGRP, since capsaicin treatment, from extrinsic and intrinsic neurons during an experiwhich destroys CGRP immunoreactive sensory neumental in ammation in animals (Reinshagen et al. rons, did not modify irradiation-induced apoptosis. 1995) and the number of substance P nerve bres is In conclusion, sensory aå erents do not play a increased in ulcerative colitis in humans (Keranen major protective role in the early (1 day) mucosal alterations induced by abdominal irradiation, since et al. 1995). However, following total-body irradiation Intestinal sensory nerves and irradiation induced colitis. Annals New York Academy of Sciences, 657, only MPO activity was slightly, yet signi cantly, 319-327. increased at this time in capsaicin-treated rats while Farber , A., Connors, J. P., Friedlander, R. M., Wagner, other parameters, such as apoptosis and mast cell R. J., P owell, R. J. and Cronenwett, J. L., 1999, A numbers remained unchanged. speci c inhibitor of apoptosis decreases tissue injury after intestinal ischemia-reperfusion in mice. Journal of Vascular Surgery, 30, 752-760. Forsgren, S., H ockerfelt, U., Norrgard, O., H enriksson,