A882 AGA ABSTRACTS

4878

STRESS REACTIVATION COLITIS AND THE ROLE OF THEAUTONOMIC NERVOUS SYSTEM (ANS).Paul Saunders, Markad Kamath, Kevan Jacobson, IDRP, McMaster Univ,Hamilton, ON, Canada; Dept of Medicine, McMaster Univ, Hamilton, ON,Canada.

The autonomic nervous system (ANS) plays an important role in modu­lating ones response to stress. We have previously demonstrated in thestress reactivation model of colitis that pretreatment with hexamethonium,a nicotinic ganglionic blocking agent, prevents stress-induced reactivationcolitis. In this study we assessed neurocardiac response to stress. Biopo­tential transmitters were surgically implanted subcutaneously, for laterrecording of ECG' s in conscious rats. Once rats had recovered (~l week),intrarectal (ir) dinitrobenzene sulfonic acid (DNB), 20 mg dissolved in 250J.Ll 40% ethanol was administered. Six weeks later rats were subjected to3hr restraint stress (RS). Twenty-min ECG recordings were taken at 5 min,2hr into the stress and 30 min post stress. Heart rate variability as measuredfrom the ECG's was subjected to power spectral analysis and low fre­quency (LF. predominantly sympathetic) and high frequency (HF, predom­inantly parasympathetic) spectrums determined. One hr post stress ratsreceived ir DNB, 20 mg dissolved in 250 J.Ll saline. Two days later the ratswere sacrificed and their colons examined. At 5 minutes into the stress ratsdemonstrated a strong sympathetic response indicated by a significantincrease in heart rate (295.7 ± 2.2 to 444.4 ± 16.5 beats/min), LF power(433.6 ± 6.9 to 487 ± 3.1 ms") and LF:HF ratio(6.2 ± 0.8 to 27.9 ± 4.5).This increase in sympathetic activity was still evident at 2 hr into the stress.Thirty minutes post stress rats demonstrated a rebound increase in para­sympathetic activity indicated by an increase in HF power (76.3 ± 7.1 atbaseline, 20.6 ± 2.7 at 5 min RS, 129.4 ± 23.3 ms? post RS) and adecrease in LF power to 380.2 ± 23.0 ms2 and in the LF:HF ratio to 4.1 ±0.98. Macroscopic examination of the colons revealed gross ulcerationwith an average macroscopic damage score of 4(range 3-5). Microscopicexamination revealed marked mucosal disruption with loss of epithelium,profound hyperemia, and marked infiltration by neutrophils, macrophages,lymphocytes and eosinophils. This study demonstrated the ANS responseto stress, causing colitic relapse, characterized by sympathetic activationfollowed by rebound parasympathetic activation. These data suggest thatchanges in sympathovagal balance are important in the stress response andfurther understanding of these responses will likely lead to new and noveltherapies directed at preventing stress induced relapses of IBD. Supportedby CAGIMRC Canada

4879AUTONOMIC NERVOUS SYSTEM (ANS) DYSFUNCTION EXAC·ERBATES INFLAMMATION IN DEXTRAN SULFATE (DSS)MODEL OF ULCERATIVE COLITIS.Paula C. Miceli, Paul Saunders, Veljko Djuric, Kevan Jacobson, IDRP,McMaster Univ, Hamilton, ON, Canada.

The role of the autonomic nervous system (ANS) in the pathogenesis ofIBD remains unclear. Recent studies in IBD patients have demonstratedANS dysfunction, with sympathetic dysfunction noted in Crohn's diseaseand vagal dysfunction in ulcerative colitis. Previous experiments in ourlaboratory have suggested that the parasympathetic branch of the ANSconfers protection in DNB-colitis, an animal model with histopathologicalsimilarities to Crohn' s disease. In the present study we explored the role ofthe ANS in an animal model of ulcerative colitis using the dextran sulfatesodium salt (DSS) model of colitis. Colitis was induced in Flinders sensi­tive line (FSL) rats, which exhibit ANS dysfunction including enhancedcholinergic sensitivity, and their control counterparts, the Flinders resistantline (FRL) rats via the addition of 5% DSS in drinking water for 7consecutive days ad libitum; controls received tap water only (n ~ 5 pergroup). The rats were observed daily for fluid intake and weight changes.The animals were sacrificed on day 7 and the colon was opened longitu­dinally, assigned a macroscopic damage score, and tissue segments wereremoved for subsequent assay of myeloperoxidase (MPO) activity. All dataare expressed as mean ± SEM. Both DSS-treated FRL and FSL ratsdemonstrated a significant increase in macroscopic damage score com­pared to their respective matched controls. MPO activity levels weresignificantly increased in DSS-treated FSL rats on day 7 compared to FSLcontrols (41.8 ± 11.8 vs. 1.26 ± 0.14 units/mg tissue; p<O.OOl),however,MPO levels in DSS-treated FRLs failed to reach significance. DSS-treatedFSL rats demonstrated significantly greater macroscopic damage score(6.3 ± 0.8 vs. 2.4 ± 0.4; p<0.OO5) and MPO activity (41.8 ± 11.8 vs.0.8 ± 0.22; p<0.OO5) compared to DSS-treated FRL rats. Furthermore, onthe day of sacrifice, DSS-treated FSL rats demonstrated a significantreduction in body weight compared to DSS-treated FRLs and matched FSLcontrols (-34 ± 3.9 vs. 15 ± 3.4 vs. 22 ± 3.9 grams respectively;p<O.OO1). These data demonstrate that the severity of DSS-colitis wassignificantly enhanced in rats exhibiting ANS dysfunction. Furthermore,

GASTROENTEROLOGY Vol. 118. No.4

these results suggest that the modulatory role of the ANS in acute intestinalinflammatory states is dependent upon the nature and characteristics of theinflammation. Supported by MRC/CAG, Canada.

4880EXTRINSIC DENERVATION INHIBITS CLOSTRIDIUM DIFFI­CILE TOXIN A-INDUCED ENTERITIS.Christopher R. Mantyh, Douglas C. McVey, Steven R. Vigna, Duke UnivMed Ctr, Durham, NC.

BACKGROUND & AIMS: Clostridia difficile enteritis is caused by toxinA (TA), an exotoxin which stimulates substance P (SP) release andsubsequent neurokinin-l (NK-l) receptor activation. This receptor stimu­lation results in secretion, inflammation, and structural damage. There aremultiple neuronal sources of SP in the small intestine including intrinsicenteric neurons and extrinsic primary sensory neurons. We tested thehypothesis that extrinsic neurons are the source of SP released in responseto intraluminal TA. METHODS: Eight rats were anesthetized and an ilealsegment measuring approximately 5 em was chosen for extrinsic denerva­tion. Surgical denervation followed by application of 80% phenol to theneurovascular pedicle was then performed. After 14 days, three ileal loopswere constructed in each rat: the denervated loop was injected intralumi­nally with 5 J.Lg of TA and two intact loops were injected with TA orvehicle respectively. Ileal secretion (mg/cm), myeloperoxidase activity(U/mg), and histology were then assessed. Complete extrinsic denervationwas confirmed by the absence of tyrosine hydroxylase immunoreactivity insurgically denervated segments. RESULTS: Denervated ileal loops in­jected with TA exhibited a 74% reduction in ileal secretion (P<O.OOl),92% reduction in myeloperoxidase activity (P<O.OOl), and a significantreduction in histologic damage (P<O.O1). There was no significant differ­ence between the denervated loops injected with TA and those injectedwith saline. CONCLUSIONS: Extrinsic surgical denervation results inprotection of ileal loops from TA enteritis. Furthermore, these resultsexclude the participation of intrinsic enteric nerves in TA-induced ilealdamage. Finally, this suggests that extrinsic primary sensory neuronsmediate the effects of intraluminal TA in the ileum.

4881

INVERSE COUPLING OF THE SYMPATHETIC AND THE ADRE­NAL STRESS AXES IN PATIENTS WITH INFLAMMATORYBOWEL DISEASE.Rainer H. Straub, Hans H. Herfarth, Werner Falk, Luitpold E. Miller,Juergen Schoelmerich, Tilo Andus, Dept of Internal Medicine I, Univ ofRegensburg, Regensburg, Germany.

Proinflammatory cytokines such as TNF stimulate the hypothalamus di­rectly or via sensory nerves in the periphery, which activates the two mainstress axes: the hypothalamus-autonomic nervous system (HANS) axis andthe hypothalamus-pituitary-adrenal (HPA) axis. The two axes are expectedto be stimulated in a parallel fashion to dampen inflammation in theperiphery. To investigate the parallelism of the two axes, we measured atypical marker of the HANS axis (neuropeptide Y= sympathetic neuro­transmitter) and of the HPA axis (cortisol) in serum of healthy subjects(n=120, age 44.5±1.5 yr), patients with Crohn s disease (CD; 170;31.8±0.9 yr), and patients with ulcerative colitis (UC; 99; 35.4± 1.4 yr)together with TNF. In the range between 20 to 55 yr the measuredparameters were not age-dependent. TNF correlated positively with corti­sol in CD (RRank= 0.319, p=O.OO1) and UC (RRank= 0.245, p=0.012) butnot in healthy subjects. This indicates that TNF-stimulated cortisol secre­tion may be inappropriately low to suppress elevated TNF secretion in theperiphery which was also found in other chronic inflammatory diseases.Serum neuropeptide Y was positively correlated with serum cortisol inhealthy subjects (RRank=0.242, p=O.OII) which is a sign for the parallel­ism of the two axes. However, serum neuropeptide Y was significantlynegatively correlated with serum cortisol in the patients with CD (RRank=­0.234, p=O.OOl) and UC (RRank=-0.385, p=O.OOI). This negative corre­lation remained constant even in patients without prior prednisolone treat­ment (CD: RRank=-0.366, p=0.040; UC: RRank=-0.548, p=0.019). Fur­thermore, after controlling the influence of TNF on the correlation of serumneuropeptide Y and cortisol, the negative correlation remained stable in thetwo patient groups (CD: partial R=-0.239, p=0.016; UC: partial R=­0.337, p=0.OO6) but was positive in healthy subjects (partial R=0.249,p=0.009). In conclusion, this study demonstrates that TNF seems to be astimulatory factor for the activation of HPA axis but cortisol secretion maybe inadequately low to suppress the elevated TNF secretion in IBD pa­tients. The two stress axes seem to act in a parallel fashion in healthysubjects (coupling of the axes) but not in patients with inflammatory boweldisease. Such an inverse coupling of the two antiinflammatory axes may berelevant for the chronicity of the diseases.