<%BANNER%>

Porcine Model of Inflammatory-Mediated Visceral Hypersensitivity


PAGE 1

PORCINE MODEL OF INFLAMMA TORY-MEDIATED VISCERAL HYPERSENSITIVITY By LINDA CHRISTINE SANCHEZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Linda Christine Sanchez

PAGE 3

This dissertation is dedicated to the many an imals who have contribut ed to my education.

PAGE 4

iv ACKNOWLEDGMENTS As with most major endeavors, many pe ople helped to make this research and dissertation possible. Despite many direct a nd indirect contributions, these are but a few thanks. My fellow graduate students in the Island Whirl Equine Colic Research Laboratory, Drs. Tamara Widenhouse and Mire ia Lorenzo, were always available for various pig-related duties, a nd I am grateful for their he lp and support. Jim Burrow, Hilken Kuck, and Drs. Chao-Yong Bai and A.M. Merritt provide d truly invaluable assistance as blinded observers for CRD pro cedures. Debi Malcolm and Alex Trapp took outstanding care of my research pigs, and Debi helped each new pig adapt to the system. Tom DeHaan and Joe Bryant provided technical assistan ce surrounding all pig necropsies. Sally O’Connell is an invaluable resource to all VMTH graduate students, and certainly has made my life considerably easier!! The Island Whirl Equine Colic Resear ch Laboratory provided funding for the animal work, and Dr. Nicholas Verne provi ded laboratory space, supplies, and funding for the initial pilot animals and all tissue processing and immunohistochemical analyses. I could not have asked for a better mentor than Dr. Al Merritt. He has continually been a source of enthusiasm, even when things (like my entire first project) were clearly not progressing as planned. Al has a tremendous ability to offer advice without intrusion, and I will always be grateful for his insight and friendship. I would also like to thank the remaini ng members of my committee, Drs. Guy Lester, Nick Verne, Elizabeth Uhl, and Ed Ott. Guy has been an outstanding sounding

PAGE 5

v board and friend, and I am truly grateful that he was willing to fly back to Florida for this defense. Nick has provided continuous in sight into the relevance of IBS in human gastroenterology while also offering the us e of his laboratory for all of the tissue processing and staining. My family and friends, while providing enormous amounts of grief regarding my never-ending education, have also provide d tremendous support. As I said for my mentor, I could also not have asked for better parents. They have always been supportive while also serving as great role models. I truly could not have done this without them.

PAGE 6

vi TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 LITERATURE REVIEW.............................................................................................1 Visceral Hypersensitivity..............................................................................................1 Visceral Afferent Innervation................................................................................2 Mediators of Visceral Hypersensitivity.................................................................4 The Role of Inflammation.....................................................................................8 Animal Models of Visceral Pain.........................................................................10 Animal Models of Inflammatory Bowel Disease.......................................................10 Inflammatory Bowel Disease in Humans............................................................11 Chemical Models.................................................................................................12 Mouse Knock-out Models...................................................................................16 The Pig as a Model.....................................................................................................17 Study Objectives.........................................................................................................17 2 METHODS.................................................................................................................18 IACUC Approval........................................................................................................18 Model Development...................................................................................................18 Main Study Design.....................................................................................................20 Procedures...................................................................................................................22 Tissue Collection, Processing, and Analysis..............................................................26 Statistical Analysis......................................................................................................34 3 RESULTS – ANIMAL STUDIES..............................................................................37 Development of colitis................................................................................................37 Endoscopic Evaluation...............................................................................................37 Visceral Sensitivity.....................................................................................................39 Correlations.................................................................................................................41

PAGE 7

vii 4 RESULTS TISSUE ANALYSIS.............................................................................42 Histological Analysis..................................................................................................42 Immunohistochemical Analysis..................................................................................45 Spinal Cord..........................................................................................................45 Gastrointestinal Tract..........................................................................................48 Correlations.................................................................................................................49 5 DISCUSSION.............................................................................................................56 Model Development...................................................................................................56 Effect of Inflammation on Nociceptive Threshold.....................................................60 Conclusions.................................................................................................................69 APPENDIX A INDIVIDUAL ANIMAL DATA...............................................................................72 B ANOVA TABLES......................................................................................................74 C CORRELATIONS......................................................................................................86 LIST OF REFERENCES...................................................................................................95 BIOGRAPHICAL SKETCH...........................................................................................108

PAGE 8

viii LIST OF TABLES Table page 2-1 Animal grouping......................................................................................................21 2-2 Timeline of study events..........................................................................................21 2-3 Endoscopic lesion scoring........................................................................................23 2-4 Histologic scoring system fo r gastrointestinal tissues..............................................28 4-1 Mean lymphoid aggregate scores.............................................................................44 4-2 Mean edema scores. .................................................................................................45 4-3 Mean dorsal horn Substance P-immunoreactivity...................................................47 4-4 Mean ventral horn Substance P-immunoreactive neurons.......................................48 4-5 Quantitative IHC data for gastrointestinal tissues....................................................49 A-1 Raw data from CRD studies.....................................................................................72 A-2 Raw data from endos copic examinations.................................................................73 B-1 One-way ANOVA analysis for threshold pressure..................................................74 B-2 One-way ANOVA analysis for ABTP.....................................................................75 B-3 One-way ANOVA analysis for endoscopy scores...................................................76 B-4 One-way ANOVA analysis for gastro intestinal histologic scores...........................77 B-5 Tukeys HSD analysis for gast rointestinal histologic scores...................................78 B-6 One-way ANOVA for ventral horn SP-immunoreactive neurons...........................81 B-7 Tukeys HSD analysis for ventral horn SP-immunoreactive neurons.....................82 B-8 One-way ANOVA for SP-immunoreactivity...........................................................83 B-9 Tukeys HSD analysis for spinal SP-Immunoreactivity..........................................84

PAGE 9

ix B-10 Tukey’s HSD for gastrointestinal SP-immunoreactivity.........................................85 C-1 Correlation between weekly AB TP and histological scores....................................86 C-2 Correlation between endosc opic histological scores................................................87 C-3 Correlation between weekly ABTP and endoscopic scores.....................................88 C-4 Correlation between weekly ABTP a nd ventral horn SP-immunoreactive neurons.89 C-5 Correlation between weekly ABTP and dorsal horn SP-immunoreactivity ...........90 C-6 Correlation between histological sc ores and ventral horn SP-immunoreactive neurons.....................................................................................................................91 C-7 Correlation between histological sc ores and dorsal horn SP-immunoreactivity......92 C-8 Correlation between histological scores and gastrointestinal SP-immunoreactivity93 C-9 Correlation between weekly ABTP a nd gastrointestinal SP-immunoreactivity......94 C-10 Correlation between dorsal horn SP-i mmunoreactivity and gastrointestinal SPimmunoreactivity.....................................................................................................94

PAGE 10

x LIST OF FIGURES Figure page 2-1 Pig in crate used for all procedures..........................................................................19 2-2 Polyethylene rectal distention balloon attached to catheter.....................................24 2-3 Close-up of rectal catheter attached to pig’s tail......................................................24 2-4 Typical discomfort response....................................................................................26 2-5 Ventral horn neurons................................................................................................30 2-6 Image selection for spinal sections...........................................................................32 2-7 Pixel square selections for cord image 1..................................................................33 2-8 Pixel square selections for cord image 2..................................................................33 2-9 Pixel square selections for colo nic and rectal myenteric plexus..............................34 3-1 Endoscopy scores.....................................................................................................38 3-2 Normal endoscopy (Grade 0)...................................................................................38 3-3 Abnormal endoscopy (Grade 3)...............................................................................39 3-4 Mean threshold pressures.........................................................................................40 3-5 Mean ABTP..............................................................................................................40 3-6 Correlation between endoscopy scores on weeks 6 and 9 and week 13 ABTP.......41 4-1 Example of edema grade 1.......................................................................................42 4-2 Example of edema grade 2.......................................................................................43 4-3 Example of edema grade 3.......................................................................................43 4-4 Example of edema grade 4.......................................................................................44 4-5 SP-IR in spinal cord.................................................................................................46

PAGE 11

xi 4-6 SP-IR in porcine colon.............................................................................................49 4-7 Correlation between lymphoi d aggregates in the rect al section R2 and ABTP.......50 4-8 Correlation between histological scores in the rectal section R1 and ventral horn SP-immnuoreactive neurons in spinal sections L1 and L2......................................51 4-9 Dorsal horn SP-IR correlation with ABTP..............................................................52 4-10 Ventral horn correla tion with ABTP for the L2 spinal segment..............................53 4-11 Ventral horn correla tion with ABTP for the L6 spinal segment..............................54 4-12 Ventral horn correla tion with ABTP for the L7 spinal segment..............................54 4-13 SP-IR correlation with histological scores...............................................................55

PAGE 12

xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PORCINE MODEL OF INFLAMMA TORY-MEDIATED VISCERAL HYPERSENSITIVITY By Linda Christine Sanchez August 2003 Chair: A.M. Merritt Major Department: Veterinary Medicine Irritable bowel syndrome (IBS) is char acterized by chronic abdominal pain associated with diarrhea and/or constipati on. Visceral hypersensitivity is a biological marker for IBS, but the pathophysiology of this hypersensitivity is unknown. Transient inflammation has been suggested as an inci ting agent, and Substance P (SP) is an important neuropeptide in pain processing. Us ing the pig as an animal model, the major objectives were to 1) develop a model of s ubacute proctitis; 2) validate a model of visceral discomfort; 3) evaluate the eff ect of colorectal inflammation on visceral nociceptive threshold; 4) evaluate the effect of colorectal inflammation on SPimmunoreactivity (SP-IR) in the colon, rectum, and lumbar spinal cord. Nineteen cross-bred male castrated young pi gs (initial bodywei ght 20-30 kg) were used. Colitis was reliably induced (n = 12) by trinitrobenzene sulfonic acid/ethanol (TNBS/EtOH) enema, and resolved within 5 weeks, as documented endoscopically. Control animals (n = 7) received a salin e enema and had no endoscopic abnormalities.

PAGE 13

xiii Half of the animals in each group (6 and 3, respectively) were euthanized 5 weeks following enema administration, and the rema ining animals were euthanized 5 weeks later. SP-IR in the spinal cord, co lon, and rectum was determined via immunohistochemical staining. Colorectal dist ention (CRD) consisted of sequential oneminute barostat-controlled inflations, interspe rsed by five-minute deflations, starting at 15 mmHg and increasing in 10mmHg increments until a discomfort response was induced (considered threshold), up to a maxi mum of 55 mmHg. Mean baseline threshold was 37.1 mmHg and did not differ significantly between TNBS/EtOH and saline groups. However, at weeks 3 and 9 post-enema, the TNBS/EtOH-treated animals had a significantly (p = 0.045 and 0.005 respectivel y) lower CRD threshold, relative to baseline, than the saline-treated animals. SP-IR did not differ statisti cally between groups in any tissues. But, SP-IR in the dorsal horn of spinal segments L1 and L7 had a significant correlation with visceral sensitiv ity. This study demonstrates the potential usefulness of the pig as a large animal m odel for visceral noci ception studies. The correlation between spinal SP-IR and viscer al sensitivity reinforces a relationship between CNS upregulation of SP an d visceral hypersensitivity.

PAGE 14

1 CHAPTER 1 LITERATURE REVIEW Many factors regulate the sensor y function of the gastrointestinal tract. As a result, when alterations of this function occur, a spec ific etiology can be di fficult to ascertain. The experiments described in this dissertati on revolve around using the pig as a model for visceral sensitivity studies. This information could potentially be applicable to other species, including humans, which typically suffe r from numerous clinical problems with associated gastrointestinal pain. Visceral Hypersensitivity Hypersensitivity commonly refers to th e development of either hyperalgesia, a significant upregulation of the magnitude of the response to a given peripheral painful stimulus, and/or allodynia, a nociceptive percep tion of a normally nonpainful stimulus (Willis, 1992). Visceral hypersensitivity (VH) refers to such a response within the abdominal viscera. The Irritable Bowe l Syndrome (IBS) is a common functional gastrointestinal disorder with a wide range of clinical pres entations, includ ing alterations in bowel habits and enhanced visceral se nsitivity (Thompson et al., 1999). VH is a common biologic marker of the Irritable Bowel Syndrome (IBS) and is present in almost all IBS patients (Mer tz et al., 1995). Some IBS patients have also been shown to have altered somatic referral patterns in response to colorectal distention (CRD), wh ich indicates altered spinal processing of visceral sensory information (Kingham a nd Dawson, 1985; Lembo et al., 1994). Some studies have shown cutaneous allodynia in IBS patients (Verne et al., 2001) whereas

PAGE 15

2 others have shown no alteration in somatic pa in tolerance compared to healthy controls (Accarino et al., 1995; Cook et al., 1987; Wh itehead et al., 1990). Many IBS patients also report symptoms of extr aintestinal functional disorders such as irritable bladder, chronic somatic pain, and sleep disturbances (Whitehead et al., 1982; Whorwell et al., 1986). The crossover of IBS with fibromyalgia a chronic somatic pain disorder, appears particularly significant (Chang et al., 2000a; Veale et al., 199 1; Verne and Price, 2002), and may correlate with the severity of IBS symptoms (Lubrano et al., 2001). Patients with IBS and fibromyalgia were found to have somatic hyperalgesia, whereas patients with IBS alone had somatic hypoalgesia with higher pain threshold and lower pain frequency and severity compared to controls (Chang et al., 2 000a). Such alterations may indicate a state of central (with or without a peripheral compon ent) hyperexcitability within the central nervous system (CNS). Hyperexcitability of the CNS has been predominantly characterized through models of cutaneous nociception, but similar mechanisms are now known to play a role in VH as well. Visceral Afferent Innervation Visceral spinal afferent ne rves function similarly to their somatic counterparts; however they allow only limited localization of an offending stimulus.(Hertz, 1911) Vagal afferents also provide important modul ation of nociception (Gebhart and Randich, 1992; Grundy, 1988), and sacral parasympathetic fibers mediate sensory information from the distal colon and rectum (Janig and Morrison, 1986). The receptive fields of upper GI and colonic spinal afferent C fibers normally occur predominantly in the muscularis, serosa, and mese ntery, but increase in size a nd include the mucosa during inflammation (McMahon and Koltzenburg, 1990; Ness and Gebhart, 1990). In the

PAGE 16

3 rectum, sacral A fibers have mucosal receptor fi elds under non-inflamed conditions (Janig and Koltzenburg, 1991; Sengupta and Gebhart, 1994). The current concept of visceral pain perception involves a combination of highthreshold nociceptors as well as low-thresh old mechanoreceptors (Cervero and Janig, 1992; Willis, 1993). In the colon, both C and A fibers can encode a wide range of stimulus intensity, but C fibers are th ought to be primarily slowly-adapting mechanoreceptors, whereas A fibers are predominantly rapidly-adapting mechanoreceptors (Blumberg et al., 1983; Ha bler et al., 1990; Ja nig and Koltzenburg, 1991). Additional silent nociceptors, mechan ically insensitive C fibers innervating areas such as the bladder and colon, only re spond to distending s timuli after chemical irritation in experimental situations (Janig and Koltzenburg, 1990; Janig and Koltzenburg, 1991). Inputs from these neurons converge on wide-dynamic range dorsal horn neurons (Willis and Coggeshal l, 1991). Under normal conditions, lowintensity stimuli activate low-threshold afferents, which do not trigger the nociceptive pathway. Transient high intensity stimuli not only increase the intensity of the lowthreshold afferents, but also recruit the highthreshold afferents, re sulting in nociception (Cervero and Janig, 1992). Inflammation can alter these pathways through persistent stimulation of the peripheral terminals or activation of th e silent nociceptors (Mayer and Gebhart, 1994). Spinal and vagal afferents differ significan tly in the localization of neuropeptides, in that 85-95% of spinal afferents to the st omach and colon, but only 5% of gastric vagal afferents, contain calcitoni n-gene related peptide (CGR P) (Mayer and Raybould, 1990). In the GI tract, Substance P (SP) is f ound predominantly in the muscular layer and

PAGE 17

4 myenteric plexus, coinciding with the previously stated receptor fields for spinal afferents (Otsuka and Yoshioka, 1993). CGRP is pres ent in numerous splancnic afferents. SP is a neuropeptide in the tachykinin family, along with neurokinins A and B (NKA and NKB). Of the th ree tachykinin receptors, NK1 has the greatest affinity for SP which, in turn, preferentially binds to that receptor (Routh and Helke, 1995). In the spinal cord dorsal horn, NK1 is heavily concentrated within Laminae I and II, with decreased density in Laminae III and IV, and very little in Lamina V(Charlton and Helke, 1985a; Helke et al., 1986). In the ventral hor n, motor neurons throughout the spinal cord contain low to moderate levels of NK1 binding(Charlton and He lke, 1985b; Charlton and Helke, 1985a). Antibody microprobe studies document the release of SP following noxious peripheral stimuli primarily within the superficial dorsal horn (Laminae I and II), but extending beyond (Duggan et al., 1992; Schaib le et al., 1992). CGRP and SP co-exist in many primary afferent neurons and may be co-expressed when these nerves are stimulated (Bueno et al., 2000; Hokfelt et al., 1977b; Wiesenfeld-Hallin et al., 1984). CGRP also inhibits SP endopeptidase (SPE) activity, thus potentia ting the biological actions of SP (Le Greves et al., 1985; Woolf and Wiesenfeld-Hallin, 1986). CGRP and SP clearly play a role in th e transmission of nociceptive information within the CNS (Duggan, 1995), but their function in peripheral terminals is less clear. They likely function as neur omodulators or mediators of local tissue responses, which could allow for changes in motility and viscer al sensation at a peripheral level. Mediators of Visceral Hypersensitivity Inflammation is thought to affect the primary affere nts through the action of inflammatory mediators on the peripheral term inals of primary afferents (Cohen and Perl, 1990; Handwerker and Reeh, 2003; Levine et al., 1992; Schaib le and Schmidt, 1988) and

PAGE 18

5 the matrix in which peripheral terminals are imbedded (Janig a nd Morrison, 1986; Mayer and Raybould, 1990). Specifi cally, prostaglandin E2 (PGE2), prostaglandin I2 (PGI2), ATP, bradykinin, serotonin, CGRP, SP, and glut amate can directly mediate sensitivity at the primary afferent (Pur cell and Atterwill, 1995; Tracey and Walker, 1995). Many interleukins (IL-1, IL-8, IL-6), tumor necrosis factor(TNF), nerve growth factor (NGF), bradykinin, leukotriene B4 (LTB4), SP, complement 5a (C5a), and vasoactive intestinal peptide (VIP ) can act indirectly vi a activation of immunoc ytes. Hyperalgesia caused by LTB4 is neutrophil-dependent and extr emely potent (Levine et al., 1984; Levine et al., 1992). Noradrenaline, neuropeptide Y (NPY), and NGF act through mediators released from adrenergic nerve va ricosities (Bueno et al ., 1997; Bueno et al., 2000). While the alteration in re sponsiveness at the peripheral terminal likely subsides with resolution of the initial inflammatory insult, experimentally documented memory persists for hours (Willis, 1993). Altered afferent input to the dorsal horn results in central sensitization, and the associated release of neuropeptid es in the dorsal horn increase s the excitability of spinal neurons and leads to an expansion of the re ceptive fields (Mayer and Gebhart, 1994). The release of CGRP and SP from both cen tral and peripheral te rminals of primary afferent neurons in response to a nociceptiv e stimulus is well recognized, and these peptides are the two most abundant in me dium and small dorsal root ganglia (DRG) neurons (Del Bianco et al., 1991; Hokfelt et al., 1977b; Lundberg et al., 1992; Sternini, 1992). The in-vitro application of the SPantagonist, CP 96345, can prevent capsaicinmediated sensitization of spinothalamic tract cells (Dougherty et al., 1994). This agent has also been shown to produce mild analge sic effects in thermoand chemo-sensitive

PAGE 19

6 models and to prevent noxious responses to carageenan injecti on in the rat paw and thermal skin stimulation in cats (Birch et al., 1992; Lecci et al., 1991; Yamamoto and Yaksh, 1991). Increased spinal cord expr ession of NK1 and NK2 receptors and SP occurs in experimental models of arthritis (Krause et al., 199 5). These findings indicate a role for the neurokinins in central hypere xcitability (Bueno et al., 1997; Bueno et al., 2000). These local effects can include activation of immune cells which will stimulate the local inflammatory response, leading to central and/or peripheral hyp erexcitability. As previously described, mast cells appear to play a major role in the sensitization of primary afferents, and the release of SP is crucial (Bueno et al ., 1997). Essentially, a feedback loop is created whereby SP release can trigger mast cell degranulation, and the subsequent histamine release causes further re lease of SP as well as NGF (Bertrand et al., 1993; Purcell and Atterwill, 1995). The resultant CGRP/SP-immunoreactive network modulates both reflex motor activity and the transmission of sensory information to the CNS. At a spinal level, SP likely contributes to dorsal horn hyperexcitability via a direct action on the postsynaptic cells or by potenti ating the excitatory effects of glutamate (Haley and Wilcox, 1992). Recent studies s howing a reduction in nociceptive response to CRD in rats by intrathecal or intrav enous administration of the CGRP receptor antagonist, hCGRP8-37, provide direct eviden ce for the role of CGRP in visceral nociception (Gschossmann et al., 2001; Plourde et al., 1997). Intrav enous injection of CGRP results in abdominal cramping and d ecreased gastric emptyi ng similar to that induced by peritoneal irritation, and the eff ects are blocked by NK1 antagonists. The

PAGE 20

7 release of SP and CGRP may also result in neurogenic inflammation, although the roles of these peptides in gastrointestinal infl ammation are not yet fully understood (Sharkey and Kroese, 2001). In a rat TNBS-colitis model, SP-immunoreactivity was decreased initially throughout the colon, followed by an in crease in the circular muscle at 7 days (Miampamba and Sharkey, 1998). A similar pattern was shown in the enteric and primary afferent nerves after intraluminal injection of TNBS in the guinea pi g ileum (Miller et al., 1993). Zymosan-induced colitis reduced the number of SP double-labeled (with Fluorogold, to document afferent innervation from the colon) DRG in both the T13-L2 and L6-S2 DRG cells (Traub et al., 1999). Furt her work suggests that the alteration in substance P in response to inflam mation may be regulated through Il-1 (Hurst and Collins, 1993). Indirect effects of neurotransmitters are al so thought to result in enhanced receptor sensitivity. Activation of CGRP, SP, and other non-N-me thyl D-aspartate (NMDA) receptors on the post-synaptic terminal of the central terminal can result in increasing depolarization of the postsynaptic membra ne, leading to activation of the NMDA receptor by removal of the Mg++ block (Mayer et al., 1984; Woolf, 1992). This allows for a further increase in intracellular Ca++, activation of NOS, and availability of NO as an intracellular messenger and diffusible ne urotransmitter (Mayer and Gebhart, 1994). Increased intracellular calcium accumulated by NMDA receptor activation may also lead to neuronal cell death, or exc itotoxicity, which may play a role in disinhibition of second-order sensory neurons, associated wi th the clinical ph enomenon of windup (Bueno et al., 1997; Mayer and Gebhart, 1994).

PAGE 21

8 The Role of Inflammation The timing of events is likely very im portant when considering the role of inflammation in the development of VH. Tr ansient colonic irritation (Al Chaer et al., 2000) or maternal separation (Coutinho et al., 2002) during the neonata l period in rats can lead to provoke chronic vis ceral hyperalgesia that persis ts through adulthood despite the absence of histopathological lesions. Neuroplastic change s, including enlargement of dorsal root ganglia cells, have been document ed within weeks of ch ronic partial urinary bladder obstruction and are associated with lo wered thresholds for urgency to urinate as well as discomfort (Steers et al., 1990; Stee rs et al., 1991). However, the associated changes may result in alterations of motility rather than sensation. Repeated rectal distention in humans to noxious pressures (60 mmHg) can resu lt in alteration of reported sensation, and repetitive CRD in rats to noxious (80 mmHg) but not innocuous (20 mmHg) pressures causes an increase in spinal fos and jun (Traub et al., 1992). These changes are likely a result of alterations in the dorsal horn neurons due to repeated excitability similar to that prev iously described for inflammation. Clinical data to support inflammation as a modulator of VH include the fact that IBS patients and patients with Crohns diseas e in which inflammation is limited to the small bowel have shown similar patterns of abdominal dermatome referral in response to CRD (Bernstein et al., 1996). Patients with mild active ulcerative co litis have attenuated rectal sensitivity responses, which correlate negatively with UC activity index, implying that either a more severe or more chroni c inflammatory insult is needed for the development of hypersensitivity (Chang et al., 2000b). Clinically, most patients suffering from inflammatory-mediated alterations in visceral sensitivity improve in conjunction with resolution of the inflammatory insult, emphasizing the short-term nature of many

PAGE 22

9 changes. The fact that VH associated with inflammation is not limited to the affected site within the GI tract also stresses the role of central hypersensitizat ion in these events. The alteration of dorsal horn neurons such that previously subthreshold stimuli can stimulate a nociceptive response is believed to play a key role in the development of mechanical allodynia. Thus, alterations a ssociated with inflammation cannot only cause short-term VH, but also the alteration of af ferent input to the dor sal horn which likely promotes plasticity within the CNS resulting in central hyperexcitability. For this central hyperexcitability to pers ist, however, additional cofactors, repeated events or specific timing of the insult such as duri ng the neonatal period likely occurs. Stress and other psychosocial factors are al so thought to contribute to VH and/or CNS hyperexcitability. Clinically, IBS pa tients often report an exacerbation of symptoms in conjunction with stressful life events, and the length and severity of the associated exacerbation often reflect the severi ty of the stressful event (Camilleri, 2001; Drossman et al., 1999; Lembo et al., 1999; Sandler, 1990). Psychological stressors can result in many physiologic changes includ ing changes in musc ular tone, immune modulation, and mucosal barrier dysfunction, a lterations in desce nding pain modulating systems, and alterations in sleep (Mayer and Gebhart, 1994; Sant os and Perdue, 2000; Soderholm and Perdue, 2001). Perhaps the most convincing argument for the role of social stressors in VH is the fact that psyc hological treatment for a nxiety and depression is often an effective tool in the treatmen t of functional bowel disorders (Mayer and Gebhart, 1994). Infectious insults to the gastrointestinal tract have also been postulated as contributing events to VH and/or CNS hype rexcitability in the IBS, given that

PAGE 23

10 approximately 33% of patients hospitalized for a bout of infectious diarrhea develop IBS within 3-12 months (Gwee et al., 1996; Mc Kendrick and Read, 1994). The responses to infection are likely mediated through infl ammatory changes or immune modulation within the GI tract as a result of the initi al insult and are likely to act via similar mechanisms as inflammation alone. In acute Nippostrongylus brasiliensis infection, a 2.5-fold increase in nerve content can occur by day 10 post infection, returning to near control values by day 14 (Stead et al., 1991). In summary, a tremendous range of potenti al mediators likely contribute to the development of VH. Genetic, environmenta l, and individual di fferences compound the situation. In addition, other mediators have been proposed as contri buting factors to the development of VH but lie beyond the scope of this review. Animal Models of Visceral Pain Balloon distention of a hollow organ, most not ably within the gast rointestinal tract, is the most widely used stimulus of vis ceral pain experimenta lly (Ness and Gebhart, 1990). CRD in humans produces similar pain ful sensations, both in intensity, quality, and area of somatic referral, to clinically oc curring gastrointestinal-associated pain. CRD has been validated in many animal species to produce brief discomfort with reliable, quantifiable behavioral and physiological responses attenua ted by analgesic drugs, thus fulfilling the criteria for a valid model of visceral pain (Ness and Gebhart, 2001). Animal Models of Inflammatory Bowel Disease Clearly, the main thrust of th is dissertation is to evaluate colitis-mediated visceral hypersensitivity using an animal model. Afte r deciding to further pu rsue the effect of inflammation in the gastrointestinal tract on nociceptive responses, the next step was to choose an inflammatory model. Numerous na turally occurring and inducible models of

PAGE 24

11 inflammatory bowel disease have been documen ted. Most involve rodents, and these are most commonly characterized based upon the specific nature of the insult. Inflammatory Bowel Disease in Humans Although commonly grouped toge ther under the umbrella of Inflammatory Bowel Disease (IBD), Crohns Disease (CD) a nd Ulcerative Colitis (UC) have many differences. The pathophysiological mechanis ms involved in UC and CD have not been fully elucidated. An infectious cause has long been suspected, especially a role for Mycobacterium paratuberculosis in the pathogenesis of CD, but efforts to identify an etiologic agent have thus far been unsuccessful (Ryan et al., 2002; Sh afran et al., 2002). UC usually begins with ulceration of the rect al mucosa and progresses orally to include varying portions of the bowel, potentially the entire colon. Clinical signs can begin with constipation and quickly progress to include rectal bleeding ur gency, diarrhea, and abdominal discomfort. The course of disease is both acute and chronic, and relapse/remission is often unpred ictable. Histologically, UC is predominantly an acute inflammatory process with disease limited to the mucosa and superficial submucosa except in fulminant disease (Fiocchi, 1998). Th e clinical course of CD is much more variable, but consists of acute and chronic inflammation of the small and large intestine and can include extraintestinal symptoms as well. The most common site of initial involvement is the ileocecal region. Histologically, CD has two different presentations indicative of either acute or chronic diseas e. Acutely, focal ap hthoid ulcerations are noted, often in the epitheliu m overlying lymphoid aggregates. These ulcerations can undergo cycles of formation and healing, and the focal ulcerations can progress to more cobblestone-like lesions. Se vere chronic CD can present with transmural inflammation,

PAGE 25

12 inflammation, and fibrosis, often with gra nuloma formation. Often, the histological distinction between the two forms can be difficult (Fiocchi, 1998; Riddell, 2000). Regardless of the initiating event, the chr onicity of IBD suggests some form of immune dysregulation. T-helper lymphocyt es are primarily responsible for the maintenance of an immune steady-state, a nd a tremendous amount of data regarding the association between their related cytokines and the IBD syndromes has recently become available. In general, UC demonstrates a predominantly Th2like (especially IL-5) profile, the increase in cytokine producti on is limited to the involved mucosa, and eicosanoid production is prominent. In CD secretion of Th1-type cytokines (IL-12, TNF IFN ,) predominates, cytokine production is increased in involved and uninvolved mucosa, and eicosanoid production is only moderate (Anand and Adya, 1999; MacDonald et al., 2000). In both forms of IB D, increases of proinflammatory cytokines (IL-1, IL-6, IL-8) are evident, alt hough data are inconsistent and TNF appears to be more important in CD. Although mucosal T ce lls are activated in both CD and UC, their differential response to IL-2 stimulation represents one of the hallmark immunologic differences between these two syndromes CD mucosal T cells demonstrate a hyperreactive response to Il-2 stimulation and have high expression of IL-2R gene products (Fiocchi, 1998). Chemical Models Many chemicals have been used as mucosa l irritants in IBD models (Elson et al., 1995; Wirtz and Neurath, 2000). None of the animal models feature a relapsing/remitting nature of disease characteristic of the hu man problem, and all chemical models have a relatively short duration (1-8 wks) although those that cause disease for 6-8 weeks allow for the characterization of a chronic infl ammatory phase. The extent of colonic

PAGE 26

13 involvement depends upon the method used for chemical instillation, with enema administration resulting in a di stal colitis in most cases. Surgical instillation allows delivery to more orad sites within the GI tract such as the ileum. The four main chemical-induced models of IBD include acetic acid, formalin/immune complex, TNBS/ethanol, and indomethacin. Several polym er/microbial-induced models also exist, including carrageenan and de xtran sodium sulfate (DSS). Acetic acid instillation into the col onic lumen produces predominantly mucosal inflammation (rats, mice, guinea pigs, rabbits) w ith histological sim ilarity to naturally occurring UC (MacPherson and Pfeiffer, 1978; Sharon and Stenson, 1985; Yamada et al., 1992). Depending on the concentration and volume used, the lesion can progress to transmural depth. However, the inflammato ry response remains for only days in mice and 2-3 weeks in rats; thus th e value of this model lies only in the earl y phases of inflammation. The mechanism of injury is prim arily related to a destruction of epithelial cells and subsequent mucosal and submucos al inflammatory response (Elson et al., 1995). Trinitrobenzenesulfonic aci d/ethanol (TNBS/EtOH) inst illation results in acute transmural inflammation, edema, and cryptitis wi th histological simila rity to CD (Torres et al., 1999). TNBS induces a delayed hypersensitivity re sponse to skin contact by haptenating body proteins with trinitrophe nyl (TNP) groups, rendering the resultant proteins immunogenic (Neurath et al., 2000). Ethanol works as a mucosal irritant, allowing for easier access for the TNBS molecule. Both T and B lymphocytes and macrophages predominate initially in an IL-12 driven Th1 T-cell-mediated inflammatory response (Neurath et al., 2000). The TNBS model has been used in numerous species,

PAGE 27

14 including rats, mice, and rabbits with a peak inflammatory response in 2-3 days with an overall duration of 2-3 weeks in mice and up to 8 weeks in rats (Elson et al., 1995). In contrast to acetic acid-induced disease, TN BS/EtOH-induced colitis does appear to have an immunologic component in that susceptibi lity to disease differs among strains of inbred mice and the dosage of TNBS require d to induce lesions varies among species (Beagley et al., 1991; Morris et al., 1989). By eliciti ng oral tolerance via the administration of a TNBS-protein complex or ally simultaneous to TNBS/ethanol enema administration in mice, the predominant Th1 response can be altered to a predominant Th2 response (Neurath et al., 2000; Seder et al., 1998). We have recently documented a TNBS/EtOH-induced ileitis in pigs produced by intraluminal instillation (Merritt et al., 2002a). An immune complex-induced colitis mode l involves the instil lation of a dilute formalin enema, followed by the intravenous injection of preformed immune complexes (Cominelli et al., 1990; Zipser et al., 1987). Th e dosage of formalin is critical to ensure that disease is related to the immune comp lexes rather than the chemical nature of formalin alone. The resultant lesion c onsists of severe mucosal and submucosal inflammation with crypt dist ortion, which resolves within 6-8 weeks. This model obviously involves an immunologi c component, and IL-1 appears to play an important role (Cominelli et al., 1990). The inflam matory cytokine production in this model mirrors that seen in both syndromes of IBD. Because the immune complexes are formed to ubiquitous antigen, normal intestinal flora may be invol ved in either initiation or perpetuation of inflammation (Elson et al., 1995; Fedorak and Madsen, 2000).

PAGE 28

15 Indomethacin administration will induce ente ritis, primarily in the mid-jejunum. Rat strains vary in their response, with ulcer ation resolving by 14 da ys in Fischer rats, lasting at least 14 days in Sprague-Dawley rats and 77 days in inbred Lewis rats (Sartor et al., 1992; Yamada et al., 1993). Lewi s rats develop segmental transmural inflammation throughout the distal jejunum and ileum. Granulomatous inflammation, fibrosis, adhesions, and partial intestinal obstr uction may also result; thus, the histological characteristics of this model more closel y mimic CD (Elson et al., 1995). Length of disease is related to the dosa ge of indomethacin used, but inflammation in most models persists for 1-2 weeks. Because indo methacin is a non-selective cyclooxygenase inhibitor, protective mucosal prostaglandins are depleted. Host susceptibility, normal intestinal flora, and bile and/or enterohepatic circulation can al so play a role in this model given the species and strain differences in susceptibility, the reduction or absence of disease in germ-free rats (Robert and Asa no, 1977), the prevention of lesion formation by bile duct ligation (Yamada et al., 1993), a nd the lesion attenuation with antibiotic administration (Banerjee and Pete rs, 1990; Yamada et al., 1993). The polymer models (DSS and carrageenan ) both induce predominantly mucosal and submucosal lesions with histological similarity to UC (Elson et al., 1995). A definitive role of luminal bacteria has been established in the carrageenan model, but the model is not easily reproducib le in species other than th e guinea pig (Breeling et al., 1988; Onderdonk et al., 1987). DSS can be admini stered easily in drinking water, and results in chronic lesions (C ooper et al., 1993). The colitis is similar to UC; however lymphoid aggregates, fissuring ulceration, a nd focal inflammation seen in the chronic phases more closely resemble CD (Elson et al., 1995).

PAGE 29

16 Mouse Knock-out Models Genetic manipulation, primarily in mice, ha s yielded both transgenic (dominant or dominant-negative expression of a gene pr oduct) and knockout (targe ted gene deletion) animals that develop intestinal inflamma tion (Elson et al., 1995; Wirtz and Neurath, 2000). In general, once lesions develop in these animals, they will persist until the animals death or, in some cases, replacement of the deleted molecule (cytokine, etc.), but a relapsing/remitting course of disease has not yet been duplicated. The number of different knockouts (KO) and transgenic anim als that develop IBDlike disease supports the notion that IBD is a multifactorial syndr ome (Elson et al., 1995; Wirtz and Neurath, 2000). One of the most important findings among the many genetically manipulated animals was that IL-2 and IL-10 knockout mice and HLA-B27 transgenic rats do not develop intestinal inflammation in the abse nce of luminal bacteria (Kuhn et al., 1993; Kundig et al., 1993; Sadlack et al., 1993). The IL-2 KO mice surviving beyond the 10th week of life develop continuous mucosal/submuc osal colitis without small intestinal or major internal organ involvement. I mmunological abnormalitie s include increased numbers of activated T and B cells, a potential Th2-like shift, incr eased IgG1, and anticolon antibodies (Kundig et al., 1993; Sadlack et al., 1993). Thus, these mice develop histological and immunological abnormalities similar to those seen in UC. The IL-10 KO mice develop chronic transmural duodenitis, jejunitis, and proximal colitis with an enhanced Th1 response due to the lack of IL-10 downregulation (Kuhn et al., 1993). This model therefore more closely mi mics CD, although the early lesions are histologically more consiste nt with UC. Rats transg enic for human HLA-B27 and 2microglobulin develop multiorgan disease (incl uding colitis, arthritis, orchitis, and a

PAGE 30

17 psoriasis-like condition), but th e resultant colitis lacks the acute neutrophilic component of IBD (Hammer et al., 1990). T-cell receptor KO and G i2 KO mice also develop colitis, as do SCID mice to which CD4+ T cells expressing high levels of CD45RB have been transferred (Elson et al., 1995; Fedorak and Madse n, 2000; Wirtz and Neurath, 2000). The Pig as a Model The pig was chosen as an animal model fo r these studies for a number of reasons. Because the majority of previous work has been performed in rodent models, a large animal model would be useful to represent larger mammalian species. A larger animal allows for colonic endoscopic examination; thus the inflammato ry status can be documented without sacrificing additional gr oups of animals. From a practical perspective, experimental use of pigs is more economical than other commonly used larger mammals such as dogs and cats. Study Objectives Using the pig as a large animal model, the major objectives of this dissertation were to 1) develop a model of sub acute proctitis; 2) validate an objective evaluation of visceral discomfort; 3) evaluate the effect of colo rectal inflammation on visceral nociceptive threshold; 4) evaluate the effect of co lorectal inflammation on immunoreactivity of Substance P in the colon, rect um, and lumbar spinal cord.

PAGE 31

18 CHAPTER 2 METHODS IACUC Approval All procedures were approved by the University of Florida Institutional Animal Care and Use Committee. Pilot animal s were approved under Protocol # A579 Evaluation of visceral discomfort in a chroni c proctitis model in the pig and animals for the major study were approved under Protocol # B167 Porcine Model of InflammatoryMediated Visceral H ypersensitivity. Model Development The primary focus of pilot studies in this project was to establish a reliable method for visceral sensitivity testing, and to determine the appropriate methodology for TNBS/EtOH enema administration. Animals Three animals were used for the initial phases of the pilot study. These animals were used mostly to develop the visceral sensitivity testing protocol. An additional two animals were used to generate preliminar y data regarding differences in visceral sensitivity, if present, between sali ne and TNBS-EtOH enema treatment. Training The first objective of the pilot studies was to find an acceptable method of pig restraint which would allow for their comfor t and easy manipulation, and observation by research personnel. A standard hog transport crate was used, with modifications so that the animal could not escape through a large top opening (Fig.2-1). Otherwise, the

PAGE 32

19 animals were able to move freely in the crate and turn back and forth. They were slowly acclimated to the crate, usi ng corn treats as a training t ool. Once acclimated, the pigs were easily transported from their normal hous ing to and from the laboratory and would remain in the crate without incident during the studies. Figure 2-1. Pig in crate used for all procedures. View from top. Once an acceptable transport device was obtai ned, the pigs were easily accustomed to all laboratory procedures. Food, most not ably whole corn, was used as a tool for positive reinforcement for all training procedures. Procedures The next phase of the pilot studies was to develop a standard protocol for CRD. A standard rectal distention balloon and catheter designed for use in humans (Medtronics, Shoreview, MN) were initially used. Wh ile the balloon volume was appropriate, the

PAGE 33

20 catheter system was too flexible. With a small amount of abdominal strain, the pigs would, in essence, defecate out the balloon. A design modification of the catheter was accomplished such that the catheter would re main in place without causing additional discomfort to the animal or any local damage to the rectal mucosa. The resultant system is described in detail below. Based on the potentially subjective natu re of clinical signs of abdominal discomfort, we chose to use a ramp protocol of CRD in which the threshold of discomfort was used as the response variab le of interest. This also allowed for a minimal amount of discomfort to the animals. Main Study Design The main study was designed based upon pilot data. Animals The study was designed such that eighteen two to three month-old mixed breed swine would be used. Initial bodyweight for all animals was between 20-30 kg. The pigs were divided into four groups (two groups of six and two groups of three) using a random number chart. Animals in groups 1 and 2 were sacrificed at week 9 and those in groups 3 and 4 were sacrificed at week 14. Due to unforeseen complications, one additional animal was added to the 14 week saline c ontrol group, for a total of 19. Except as described below, pigs were meal fed a commer cial swine grower diet (LabDiet) at a rate of 0.2-0.25 kg/kg bodyweight daily. For the first 3 weeks of the study, pigs we re handled 1-2 times daily in order to acclimatize them to human contact. As the animals became increasingly tractable, they were trained to load into and out of a standa rd transport crate and to remain in the crate

PAGE 34

21 for 30-45 minute time periods, followed by acclim ation to rectal thermometer and finally to intra-rectal balloon insertion. Table 2-1. Animal grouping Group # Enema Euthanasia Week 1 Saline 9 2 Saline 14 3 TNBS/EtOH 9 4 TNBS/EtOH 14 Study Timeline The timeline of events did not vary betw een animals and is described below in Table 2-2. Table 2-2. Timeline of study events Week Description Distention studies Other events Animals sacrificed 1 Training None 2 Training None 3 Training One Endoscopy 4 Enema None Monitoring 5 Post-enema None Monitoring 6 Post-enema None Monitoring, Endoscopy 7 Post-enema One 8 Post-enema One 9 Post-enema One Endoscopy Groups 1 & 2 10 Post-enema One 11 Post-enema One 12 Post-enema One 13 Post-enema One 14 Post-enema One Endoscopy Groups 3 & 4

PAGE 35

22 Procedures Enema Administration Enema administration occurred during week 4 of the study. Prior to this procedure, feed was withheld for 12 hours although water was available on a free-choice basis. Each animal was anesthetized with a butor phanol (0.15-0.30 mg/kg)/xylazine (4-8 mg/kg)/ketamine (4-8 mg/kg) combination admi nistered intramuscularly. Pigs in Groups 1 and 3 received 40 ml of 100% EtOH mixed w ith 5 grams of TNBS diluted in 10 ml of water. The enema was retained in a 10-cm portion of the distal colon and proximal rectum for 12 minutes by use of two Foley catheters with 60-ml balloons. For control animals (groups 2 and 4), the retention enema consisted of 50 ml of 0.9% saline. Each pig was observed until fully recovered from an esthesia. They were then monitored 2-3 times daily until vital signs remained within normal ranges for 3 consecutive days and daily thereafter. Clinical evaluation incl uded measurement of rectal temperature, observation of general attitude and fecal output, and monitoring for signs of GI distress such as vomiting, constipation, diarrhea, and an orexia. Three pigs became excitable in response to the initial anesthe tic protocol and were subsequently anesthetized with a combination of xylazine (2.2 mg/kg) and tiletamine/zolazepam (2.2 mg/kg) intramuscularly. Due to a delay in approval of the revised anes thetic protocol, the first of these pigs was re-allocated and became a nontreated control animal for the duration of the study. Endoscopic Evaluation Videoendoscopic examination of the rect um and distal colon was performed (Pentax EFG 1-meter videoendoscope) and r ecorded during weeks 3, 6, 9, and 14 of the study. Prior to the procedure, the rectum wa s evacuated using 1-2 liters of warm soapy

PAGE 36

23 water as an enema, and the endoscopy was performed with the an imals standing in the transport crate previously described. When possible, each endoscopic examination was documented with a series of still images. For each endoscopic procedure, the appearance of the rectum and colon was scored on a 02 scale for each of the following: erythema, edema, granularity, friability, and erosions, with an overall range of 0-10 (Table 2-3) (D'Argenio et al., 2001). Table 2-3. Endoscopic lesion scoring Lesion None Mild Moderate Marked Severe Erythema 0 0.5 1 1.5 2 Edema 0 0.5 1 1.5 2 Granularity 0 0.5 1 1.5 2 Friability 0 0.5 1 1.5 2 Erosions 0 0.5 1 1.5 2 Visceral Sensitivity Evaluation Visceral sensitivity was assessed by mean s of colorectal distention (CRD). Animals of all groups underwent CRD proce dures once during week 3, then once weekly from week 7 until the end of their respective protocol. Rectal catheter Visceral discomfort was stimulated by CRD using a barostat (IsoBar 3, G&J Electronics Corp., Willowdale, Ont.). Th is involved a commercially available 500-ml polyethylene bag (Medtronics, Shor eview, MN) attached to a r ectal catheter with dental floss (Fig. 2-2). The catheter had separate channels dedicated for volume control and pressure transduction, respectively. A thin me tal rod was placed within the lumen of the catheter and secured with sili cone. The distal end of this rod was smoothed, and the silicone filling completely covered the end of the rod such that it would not interfere with

PAGE 37

24 the pressure measurements or volume altera tion functions of the catheter system. The proximal end of the rod extended approximate ly 15 cm from the catheter tip. The purpose of this rod was to provide stiffened support for the catheter such that, when secured in place, it would not be expelled by the abdominal strain demonstrated during an animals discomfort response. Figure 2-2. Polyethylene re ctal distention balloon attached to catheter. Prior to insertion, the cathet er was lubricated and a ro ll of 1 white tape was attached 10 cm from the cathete r tip. After insertion, the ta pe was used to secure the apparatus to the animals tail such that the balloon would remain at a standardized distance from the external an al sphincter. (Fig. 2-3) Figure 2-3. Close-up of rectal ca theter attached to pigs tail.

PAGE 38

25 Ramp Protocol For nociceptive response testing, a stepwi se pattern of inflations from 15 to 55 mmHg (60-second inflation with a 5-minute deflation period between inflations) was used. The inflation pattern continued until the animal displayed a response of discomfort or, if no response was obtained, a maximum pressure of 55 mmHg. Assessment of Response Three observers, blinded as to treatmen t group, monitored each pig throughout each distention protocol. At the conclusion of each inflation period, each observer independently displayed a yes or no respons e to a fourth investig ator responsible for control of the barostat. When at least two of three observers declared a yes response, the fourth investigator discontinued the inflat ion protocol and record ed the pressure at which the response occurred. Based upon pilo t study observations, a discomfort response was considered to include at least three of the following behavior s occurring during one inflation period: abrupt ch ange in behavior (i.e. the animal discontinues previous behavior), arching of the back, abdominal st rain, shifting of the hi ndlimbs. A still image of a typical response is di splayed in Figure 2-4. Each animal was observed for at leas t 5-minutes following the final balloon deflation.

PAGE 39

26 Figure 2-4. Typical discomfort response. Note the arched back and wide-based stance of the hindlimbs. Tissue Collection, Processing, and Analysis Animals were euthanized during the final week of the study after the conclusion of that weeks other events ( CRD, endoscopy). Animals were first anesthetized with the xylazine/butorphanol/ketamine combinati on described previously, followed by a barbiturate overdose (Bea uthanasia D, Shering Pl ough, 0.22ml/kg) administered intravenously. Those pigs re quiring telazol/xylazine anesthes ia for enema administration, received that anesthetic combination prior to Beauthanasia administration. Necropsy examinations were performed immediately afte r euthanasia. Tissue from all animals was processed in similar fashion. Necropsy Complete necropsy examinations were performed on each animal. Any gross abnormalities, if present, were recorded. The gastrointestinal tract was examined in its entirety. Samples taken for la ter histological analysis were cut, rinsed with 0.9% NaCl, and immersed immediately in 10% buffered form alin. Sections of th e rectum were taken

PAGE 40

27 at points 10 and 12 cm orad to the external anal sphincter and, respectively, labeled R1 and R2. Sections of the colon were taken at points 15 and 20 cm orad to the orad-most rectal section and labeled C1 and C2, respectively. A section of ileum was taken 2 cm orad to the ileocecal band (labeled I), a section of cecum was taken along the medial cecal band (CE), and a section of jejunum was taken at a random location within the midjejunum (J). The remaining portions of th e gastrointestinal tract were opened and examined for gross lesions. The thoracic a nd remaining abdominal contents were also examined for the presence of lesions. After completion of this portion of the ex amination, the spinal cord was removed and fixed. The vertebral column was isolat ed from the cranial to mid-thoracic level caudal to its termination and stripped of a ll excess tissue. A dorsal hemilaminectomy was performed via the use of a Stryker saw al ong the length of the column. The spinal cord and associated spinal nerve roots were extricated and immersed in formalin as described above. Tissue Preparation After formalin immersion for a period of 18-24 hours, sections were cut, placed in cassettes, and then dehydrated in ethanol and embedded in paraffin in routine fashion. Sections were cut onto Superfrost Plus slides for staining at a later date. All slides and blocks were stored at room temperature. Histological Analysis For routine histological an alysis, slides were heat-f ixed and deparaffinized in routine fashion then stained with hematoxyl in and eosin. Sections of the colon and rectum of each animal were initially evaluate d by the author and a pathologist (EWU) in order to gauge the range of lesions presen t in the study population. Based upon the initial

PAGE 41

28 analyses, cellular infiltrates in the tissues we re confined to lympocytes, predominantly in mucosal and submucosal aggregates. In a ddition, the tissues had a varying degree of edema. Thus, the degree of edema was scored from 1 (normal) to 4 (severe) and the total number of lymphoid aggregates per slide were counted. (Table 2-3). Spinal sections were evaluated for detectable a bnormalities, but not scored. Table 2-4. Histologic sc oring system for gastrointestinal tissues. Criterion Category Score Edema Normal 1 Mild 2 Moderate 3 Severe 4 Lymphoid aggregates/section 0 1 1 2 2-3 3 4-6 4 Immunohistochemical Analysis Immunostaining for Substance P antigen was performed on the formalin-fixed, paraffin-embedded tissue using a rabbit an ti-human Substance P polyclonal antibody (BYA1145-1, Accurate Chemical and Scientif ic Corp., Westbury, NY). This antibody has been validated in rat, monkey, feline, porcine, and bovine tissu es. Slides were processed in duplicate (antibody and negative control). All sl ides were heat fixed and deparaffinized by immersion in xylene (2 x 5min) followed by decreasing concentrations of ethanol (2 x 3 min at 100%; 2 x 3 min at 95%) and then a rinse in deionized water. Slides were then stained using the D AKO EnVision Peroxidase staining system (DakoCytomation, Inc., Carpinteria, CA). All procedures were performed at room temperature. Briefly, slides were carefully dried and the tissue section was outlined with a hydrophobic pen (PAP pen), leav ing at least a 3 mm margin. Next, they were incubated

PAGE 42

29 with DAKO EnVision Peroxidase Blocking R eagent for 5 min, rinsed with PBS and placed in a PBS bath for 5 min. Slides we re then incubated with primary antibody or control, respectively, for 30 minutes. Based on serial d ilutions of 1:50 to 1:2500, the apparent optimal dilution of anti-substance P antibody was 1:500. This dilution was used for all subsequent staining. Following antibody incubation, slides were placed in a PBS bath (2 x 5 min). They were then incubated with DAKO Envision Peroxidase labeled polymer for 30 min, followed by a PBS rinse and placement in a PBS bath for 5 min. DAKO Envision DAB substrate was applied for 5 min, and then slides were rinsed in deionized water for 1 min. Next, they were counterstaine d with 50% Gills Hematoxylin for 2 min, and then placed in a running deionized water bath until the wate r ran clear. Finally, they were then placed in a bluing solution (deionized water w ith ammonium hydroxide, 10 quick dunks), followed by a deionized water bath (5 mi n), and then dehydrated in increasing concentrations of ethanol (2 min at 95%, 3 x 2 min at 100%). Slides were kept in xylene until coverslips were perman ently affixed with Permount. Ventral Horn Immunohistochemical Analysis Substance P-immunoreactive neurons within the ventral horn (tot al of both right and left hemisections) of each spinal section were counted and recorded. Examples of neurons considered immunoreactive and non-im munoreactive are identified in Figure 25.

PAGE 43

30 Figure 2-5. Ventral horn neurons. Exam ples of SP-immunoreactive (arrows) and nonimmunoreactive (arrowheads) neur ons within the ventral horn. Quantitative Immunohistochemical Analysis Substance P immunoreactivity was evaluate d quantitatively using an algorithm described and validated by Matkowskyj (M atkowskyj et al., 2000) Essentially, the information within a control slide is subtracted from the information within an antibodytreated slide in order to quan titatively identify antibody-ge nerated chromagen content of the slides in question. In orde r to for this to occur effectiv ely, slides were read then and images saved in tagged-image file format (TIFF), which allows for compression without loss of data. The slides were read using a Zeiss Axi oplan 2 Microscope. Corresponding images were captured for antibody-stained and control images for each tissue section using SPOT image capture software. These images were saved in TIFF format and reopened with Adobe Photoshop (Version 6.0, Adobe Systems, Inc., San Jose, CA), where 3 100x100 pixel areas of interest for each slide we re captured and saved as new jpg files. These files were opened with Matlab (Ver sion 6.5, The MathWorks, Inc., Natick, MA),

PAGE 44

31 which calculated the energy contained within each image. The net energy differential (energy within control images subtracted from energy within an tibody images) was used to represent the net chromagen content within th e region of interest as arbitrary units. For spinal sections, image one was taken at the apex of the dorsal horn, and image two was taken along the dorsal margin, approx imately half-way to midline. The 3 100pixel squares were captured as shown. (Fig. 2-6, 2-7, and 2-8) Fo r colon and rectal sections, both original images were taken at arbitrary points al ong the junction between the circular and longitudinal muscle laye rs. The 100-pixel squares were taken at consecutive points along this junction, careful not to incl ude blood vessels. (Fig. 2-9)

PAGE 45

32 Figure 2-6. Image selection for spinal secti ons. Spinal cord, 10x magnification. Note the tip of dorsal horn at upper left of im age, central canal at lower right.

PAGE 46

33 Figure 2-7. Pixel square selections for Co rd Image 1. Spinal cord, 20x magnification. See figure 2-6 for orientation. Figure 2-8. Pixel square selections for Co rd Image 2. Spinal cord, 20x magnification. See figure 2-6 for orientation.

PAGE 47

34 Figure 2-9. Pixel square selections for co lonic and rectal myenteric plexus. Colon, 20x magnification. Statistical Analysis All statistical analyses were perfor med using SPSS 11.0 for Windows (SPSS, Inc., Chicago, IL). Significance was placed at p < 0.05. Colorectal Distention Week 3 threshold pressure s were compared between saline and TNBS groups (combined groups 1 and 3 vs. 2 and 4) using an independent samples t-test. For all subsequent weeks, threshold pressures were compared between saline and TNBS groups (combined groups 1 and 3 vs. 2 and 4) using a one-way analysis of variance (ANOVA). In an effort to gauge a change in each individual animals threshold relative to baseline, an alteration from baseline thresh old pressure (ABTP) for each study from weeks 7-14 by subtracting the baseline threshol d pressure from the threshold pressure for that study. Thus, if the pressure for a part icular week was lower than that animals threshold pressure at week 3, the alteration number for that week would be negative.

PAGE 48

35 ABTP were compared between saline and TN BS groups (combined groups 1 and 3 vs. 2 and 4) using a one-way ANOVA for weeks 7-14. Endoscopy Week 3 endoscopy scores were compared between saline and TNBS groups (combined groups 1 and 3 vs. 2 and 4) using an independent samples t-test. Scores for weeks 6, 9, and 14 were compared between sa line and TNBS groups (combined groups 1 and 3 vs. 2 and 4) using a one-way ANOVA. Histology and Immunohistochemistry Inflammatory scores were compared be tween the four groups using a one-way ANOVA, followed by Tukeys HSD multiple co mparison procedure. Scores for each colonic and rectal section we re evaluated independently. Chromagen content (SP-IR as represen ted by arbitrary energy units/pixel (EU/pixel)) in each section (rectum 1&2, colon 1&2, spinal sections L1, L2, L6, and L7) was compared between the four groups using a one-way ANOVA, followed by Tukeys HSD multiple comparison procedure. Fo r each spinal section, the number of SPimmunoreactive neurons in the ventral horn was similarly compared between groups. Baseline threshold pressures and chromagen content satisfied the Shapiro-Wilk test for normality, thus a parametric approach was justified. A one-way ANOVA was chosen for sequential analyses due to the disparate times of euthanasia for groups 1 and 2 vs. 3 and 4. Essentially, half of the animals were discontinued mid-way through the study, so a one-way analysis was used, though a twoway ANOVA for repeated measures would have been most appropriate had all animals continued through until week 14 of the study.

PAGE 49

36 Correlations The relationship between an individual animals visceral sensitivity, as measured by the weekly alteration from CRD sensitivity threshold, and each of the following were also evaluated using linear regression and Pearsons corr elation: 1) ventral horn SPimmunoreactive neuron count for each evaluated sp inal cord section; 2) dorsal horn SPimmunoreactivity (EU/pixel) for each evaluate d image of each spinal cord section; 3) gastrointestinal pathology for each evalua ted section (edema and lymphoid aggregate counts evaluated independently). In add ition, correlation analysis was similarly performed between the gastro intestinal edema and lymphoi d aggregate counts and the spinal SP-immunoreactivity. Linear regression analysis was subsequently performed on correlations identified as significant in this fashion.

PAGE 50

37 CHAPTER 3 RESULTS ANIMAL STUDIES Animals were well trained and tolerated the distention procedures very well. Pilot studies allowed for refinement of all proced ures and subsequent design of the main study. Development of colitis After TNBS/EtOH enema administrati on, most animals developed mild bloody diarrhea for approximately 24 hours and all became febrile (103-1040F) for 2-5 days. One animal developed signs of sepsis and died despite therapy with in 5 hours following its TNBS/EtOH enema. Necropsy examination reveal ed a perforated rectum. This animal was replaced in the study. None of the othe r pigs displayed any evidence of clinical illness, other than the diarrhea noted previously and occasional mild depression for 12-24 hours. All animals maintained an excelle nt appetite throughout the study. Endoscopic Evaluation Endoscopic evaluations were easily perf ormed using the previously described procedure. Most animals rectum and dist al colon could be easily and fully observed after the first 1-L enema. If further evacuat ion was needed, the enema was repeated. No animal required more than 2 enemas and a ll animals tolerated the procedure well. Raw data from the endoscopic evaluations are presented in Appendix A. Mean baseline endoscopy score (week 3) was 0.33 for the TNBS animals and 0.50 for the saline animals. These values did not differ signi ficantly. Endoscopy scores for subsequent weeks (TNBS animals combined as Group 1 and saline animals combined as Group 2) are presented in Figure 3-1.

PAGE 51

38 Endoscopy ScoresWeek of Study 246810121416 Score 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 TNBS-EtOH Saline Figure 3-1. Endoscopy scores. Data expressed as mean + SEM. TNBS and saline means differed signif icantly at weeks 6 (p=0.000) and 9 (p=0.033) but not week 14 (p=0.134). Examples of normal and abnormal endoscopic evaluations are presented in Fi gures 3-2 and 3-3, respectively. Figure 3-2. Normal endoscopy (Grade 0).

PAGE 52

39 Figure 3-3. Abnormal endoscopy (Grade 3). Visceral Sensitivity CRD studies were completed for all animal s, all weeks, with the exception of one pig (#3179) that developed a small amount of rectal bleeding during balloon insertion for its week 9 (final) trial. The procedure was not repeated for this animal. All other trials were completed successfully. For 116/125 completed trials, the animal displayed a discomfort response as judged by at least 2/ 3 blinded observers. For the remaining 9 trials, a majority opinion discomfort re sponse was not observed (represented by 55* in the raw data). For statistical analyses, these trials were considered to have a threshold pressure of 55 mmHg. For all trials, 55% of the decisions were unanimous, whereas 45% involved a yes response from 2/3 observers. Raw data from the visceral sensitivity st udies are presented in Appendix A. Mean baseline threshold pressure (week 3) was 40.8 mmHg for the TNBS/EtOH animals and 30.7 mmHg for the saline animals. These va lues did not differ significantly. For subsequent weeks, mean values + SEM for TNBS/EtOH (combined as Group 1) and saline (combined as Group 2) anim als are presented in Figure 3-4.

PAGE 53

40 CRD Mean Threshold PressureWeek of Study 246810121416 Threshold pressure (mmHg) 10 15 20 25 30 35 40 45 50 TNBS/EtOH Saline Figure 3-4. Mean threshold pressure s. Data expressed as mean + SEM. TNBS/EtOH and saline groups differed significantly (p<0.05) at week 13. Mean TNBS/EtOH pressures differed significantly from the mean saline pressure for week 13 (p=0.006). There was a trend to wards difference for weeks 7 (p=0.084) and 9 (0.093). Complete ANOVA tables for all an alyses are presented in Appendix B. CRD threshold pressure Mean alteration from baselineWeek of Study 6810121416 Change in threshold from baseline, mmHg -40 -30 -20 -10 0 10 20 TNBS/EtOH Saline Figure 3-5. Mean ABTP. Data expressed as mean + SEM. TNBS/EtOH and saline groups differed significantly (p<0.05) at week 7.

PAGE 54

41 Mean ABTP for TNBS/EtOH (combined as Group 1) and saline (combined as Group 2) animals for all weeks are pres ented in Figure 3-5. Mean ABTP for TNBS/EtOH animals differed significantly from the mean for saline animals for week 7 (p=0.045), although there was a trend for week 13 (p=0.056) When examining data from individual anim als, rather than the groups as whole, only 2/7 saline animals had >2 weeks with threshold pressures be low baseline, whereas 11/12 TNBS/EtOH animals had >2 weeks as such. Correlations Correlations were made between endo scopic scores on weeks 6, 9, and 14and weekly ABTP. Complete Pearsons correlati on data are presented in Appendix C. Of these, endoscopy scores on weeks 6 and 9 had a significant negative correlation with week 13 sensitivity, and the linear regressi on between these factors are presented in Figure 3-8. No other comparisons had signi ficant Pearsons correlations, thus linear regression was not performed. Figure 3-6. Correlation between endoscopy sc ores on weeks 6 and 9 and week 13 ABTP. Dashed lines represent 95% confidence interval.

PAGE 55

42 CHAPTER 4 RESULTS TISSUE ANALYSIS Histological Analysis An initial review of the gastrointestinal sections revealed that the predominant abnormality seen was submucosa l edema. Also, the number of lymphoid aggregates seen in the mucosal and submucosal areas appeared to differ between sections. Thus, these factors were used in order to quantitatively evaluate the gastrointestinal sections as described in the materials a nd methods section. Examples of each histologic edema score are presented below in Figures 4 1 through 4 4. Figure 4 1. Example of edema grade 1. Arrows denote submucosal edema. Porcine rectum, 10x magnification, bar denotes 100 m.

PAGE 56

43 Figure 4 2 Example of edema grade 2. Arrow denotes submucosal edema. Arrowheads denote a mucosal lymphoid aggregate. A dilated lymphatic is also present just below the lymphoid aggregate. Porcine rectum, 10x magnification, bar denotes 100 m. Figure 4 3. Example of edema grade 3. Arrows denote submucosal edema. Mucosal edema is also evident, as noted by an increased space betw een mucosal glands. Porcine rectum, 10x magnification, bar denotes 100 m.

PAGE 57

44 Figure 4 4. Example of edema grade 4. Arrows denote submucosal edema. Porcine colon, 10x magnification, bar denotes 100 m. Mean edema and lymphoid aggregate scores for all sec tions are presented in Tables 4 1 and 4 2, respectively. Statistical analysis did not reveal any significant differences between groups for any of the colonic or rectal sections. Complete ANOVA tables and Tukeys multiple comparison analyses are provide d in Appendix B. Table 4 1. Mean lymphoid aggregate scores. Section Group Mean SD SEM C1 1 2.17 0.75 0.31 2 2.33 0.58 0.33 3 1.67 1.03 0.42 4 2.25 1.50 0.75 C2 1 2.17 0.75 0.31 2 2.00 1.00 0.58 3 2.00 0.89 0.37 4 2.25 0.50 0.25 R1 1 2.17 0 .98 0.40 2 2.67 1.53 0.88 3 1.67 1.03 0.42 4 2.00 1.41 0.71 R2 1 2.17 0.98 0.40 2 1.67 1.15 0.67 3 2.17 0.98 0.40 4 2.50 1.00 0.50

PAGE 58

45 Table 4-2. Mean edema scores. Section Group Mean SD SEM C1 1 1.50 0.55 0.22 2 1.67 0.58 0.33 3 2.33 0.82 0.33 4 2.00 1.41 0.71 C2 1 1.50 0.55 0.22 2 1.33 0.58 0.33 3 2.00 1.10 0.45 4 2.25 1.26 0.63 R1 1 2.33 0.52 0.21 2 2.67 0.58 0.33 3 1.83 1.17 0.48 4 3.25 0.96 0.48 R2 1 2.33 0.82 0.33 2 2.33 1.15 0.67 3 2.17 1.47 0.60 4 3.00 1.15 0.58 Spinal cord sections were also evaluate d. Some sections contained mild gliosis, but this was considered within normal limits thus the spinal cords were not scored. Immunohistochemical Analysis Substance P-immunoreactivity (SP-IR) was seen in all tissues. The immunoreactivity appeared specific, and very little background staining was evident in the spinal cord sections. Gastrointestinal sections had a higher de gree of non-specific chromagen uptake, but specific immunoreactivity was also present. Negative control slides had little to no chro magen uptake for all sections. Spinal cord Substance P-immunoreactivity was most pr ominent in the superficial dorsal horn (lamina I and II) (Figure 4-5). Some large im munoreactive neurons were also detected in the ventral horn of some sections.

PAGE 59

46 Figure 4 5. SP IR in spinal cord. Arrows denote margin of immunoreactivity in dorsal horn. Porcine spinal cord, L7 segment, 2.5x magnification. For details of site selection and immunoreactivity in those regions, see F igures 2 6 through 2 8. Quantitative data describing the Substance P immunoreactivity (EU/pixel) within the dorsal horn are presented in Table 4 3. Statistical analysis did not reveal any significant differences between experimental groups for any spinal s egment. Initially, data from images 1 and 2 were analyzed individually. But, as this did not change any analyses, data were pooled as collective data for the dorsal horn. Complete ANOVA tables and multiple comparison procedural results are presented in Appendix B. Numbers of Substance P immunoreactive neurons within the ventral horn are presented in Table 4 4. Statistical analysis did not reveal any significant differences between experimental groups for any spinal segment. Complete ANOVA tables and mul tiple comparison procedural results are presented in Appendix B.

PAGE 60

47 Table 4-3. Mean dorsal horn Substa nce P-immunoreactivity (EU/pixel) Section Group Mean SD SEM L1 1 150.64 43.67 17.83 2 104.14 10.42 6.02 3 152.38 61.16 24.97 4 110.74 22.53 11.26 L2 1 145.01 32.53 13.28 2 131.65 28.48 16.44 3 158.66 37.05 15.13 4 146.44 64.27 32.13 L6 1 183.18 58.26 23.79 2 180.43 45.04 26.00 3 179.13 48.98 19.99 4 143.30 36.66 18.33 L7 1 138.91 41.26 16.85 2 157.64 32.08 18.52 3 141.95 76.53 31.24 4 202.21 17.41 10.05

PAGE 61

48 Table 4-4. Mean ventral horn S ubstance P-immunoreactive neurons Section Group Mean SD SEM L1 1 1.50 3.67 1.50 2 0.00 0.00 0.00 3 6.83 7.49 3.06 4 4.00 8.00 4.00 L2 1 0.83 1.60 0.65 2 0.00 0.00 0.00 3 8.00 8.67 3.54 4 7.75 8.73 4.37 L6 1 10.50 9.20 3.76 2 6.33 3.79 2.19 3 14.67 16.60 6.78 4 5.50 6.81 3.40 L7 1 3.33 4.18 1.71 2 2.33 4.04 2.33 3 3.83 3.92 1.60 4 2.67 2.52 1.45 Gastrointestinal Tract In the gastrointestinal tissues, the most intense regions of SP-IR were in the submucosal region between circular and longi tudinal muscle layers, corresponding to the location of the myenteric plexus (Fig. 46). Some peripheral chromagen uptake was present in all tissues, likely related to st aining artifact. Quantitative-IHC data for the gastrointestinal tissues are presented in Ta ble 4-5. Values did not differ significantly between experimental groups. Complete ANOVA tables are presen ted in Appendix B.

PAGE 62

49 Figure 4-6. SP-IR in porcine colon. 10x magnification. For pixel square location, see Figure 2-9. Table 4-5. Quantitative IHC data for gastrointestinal tissues. Section Group Mean SD SEM R1 1 92.96 34.20 13.96 2 86.72 46.11 26.62 3 100.13 40.16 17.96 4 140.90 65.47 32.74 R2 1 117.04 65.24 26.63 2 61.06 33.01 19.06 3 98.29 50.79 20.74 4 93.14 52.82 30.49 C1 1 135.69 29.87 13.36 2 94.96 16.00 9.24 3 150.16 89.22 36.42 4 156.23 76.90 38.45 C2 1 137.68 71.19 35.59 2 67.92 45.83 26.46 3 100.83 84.43 37.76 4 203.68 36.68 21.18 Correlations Complete Pearsons correlation data are presented in Appendix C. The following comparisons had a significant negative correlation: R2 lympho id aggregate score and

PAGE 63

50 Week 10 and 12 ABTP (Figure 4-7); R1 hist ological scores a nd Ventral horn SP-IR neurons in the L1 and L2 segment (Figure 48); Dorsal horn SP-IR in the L1 segment and Week 9 ABTP (Figure 4-9) ; Dorsal horn SP -IR in the L7 segment and Week 11, 12, and 14 ABTP (Figure 4-9); Ventra l horn SP-IR neurons in th e L2 segment and Week 11, 12, 13, and 14 ABTP (Figure 4-10); Ventral horn SP-IR neurons in the L6 segment and Week 13 and 14 ABTP (Figure 4-11); Ventral horn SP-IR neurons in the L1 segment and Week 8 ABTP (Figure 4-12); Dorsal horn SP -IR in the L6 segment and R1 lymphoid aggregate score(Figure 4-13). In addition, the following comparisons had a significant positive correlation: SP-IR in the C1 section and R2 lymphoid aggregat e score; SP-IR in the R1 section and C1 lymphoid aggregate score; SP-IR in the R1 section and C2 lymphoid aggregate score (Figure 4-13). All other comparisons did not have significant Pears ons correlations, thus linear regression was not performed. Figure 4-7. Correlation between lymphoid aggregates in the r ectal section R2 and ABTP. Dashed lines represent 95% confidence interval.

PAGE 64

51 Figure 4-8. Correlation between histological scores in the re ctal section R1 and ventral horn SP-immnuoreactive neurons in spinal sections L1 and L2. Dashed lines represent 95% confidence interval.

PAGE 65

52 Figure 4-9. Dorsal horn SP-IR correlation with ABTP. Clockwise from the top left, plots represent linear regression analysis fo r segment L1 vs. Week 9 and segment L7 vs. Weeks11, 14, and 12. Dashed lines represent 95% confidence intervals.

PAGE 66

53 Figure 4-10. Ventral horn correla tion with ABTP for the L2 spinal segment. Clockwise from the top left, plots represent lin ear regression analysis for weeks 11, 12, 14, and 13. Dashed lines represent 95% confidence intervals.

PAGE 67

54 Figure 4-11. Ventral horn correla tion with ABTP for the L6 spinal segment. Plots on the left and right represent linear re gression analysis for weeks 13 and 14, respectively. Dashed lines repres ent 95% confidence intervals. Figure 4-12. Ventral horn correla tion with ABTP for the L7 spinal segment. Dashed lines represent 95% confidence intervals.

PAGE 68

55 Figure 4-13. SP-IR correlation w ith histological scores. Cl ockwise from the top left, plots represent linear regression analysis for spinal segment L6 dorsal horn vs. R1 Lymphoid aggregate scor e; colon section C1 vs. R2 lymphoid aggregate score; rectal section R1 vs .C2 edema score; rectal s ection R1 vs. C1 lymphoid aggregate score. Dashed lines represent 95% confidence intervals.

PAGE 69

56 CHAPTER 5 DISCUSSION Model Development The first two major objectives outlined for this study were clearly met. Animals developed a subacute proctitis after TN BS/EtOH enema administration and showed clinical signs of abdominal di scomfort in response to co lorectal balloon distention. Subacute Proctitis A subacute proctitis developed in the TNBS/EtOH groups. The degree of inflammation was such that the animals be came febrile and most briefly developed bloody diarrhea. These signs were transient a nd did not affect the animals appetite at any point. Some animals appeared slightly depressed for 12-24 hours, but this did not persist. Unfortunately, one pig suffered a perforated rectum shortly after TNBS/EtOH instillation, resulting in spont aneous death. The timing of this event was somewhat unusual in that the peak inflammatory re sponse following TNBS/EtOH instillation is usually 2-3 days (Elson et al., 1995) but th e animal died within 8 hours after the procedure. Due to the transmural nature of the resultant colitis, the effective dose of TNBS is close to the lethal dose in mice (Beagley et al., 1991 ). However, the currently used dose of TNBS/EtOH did not cause systemic illness in any other pig used in the pilot or main portions of the current study or those treated with ileal instillation in another study in this laboratory (Merri tt et al., 2002a). One possi ble confounding factor was the use of inflated Foley catheters for enema rete ntion. Because these catheters were inflated to a set volume, rather than pressure, colonic contracti ons over an inflated balloon,

PAGE 70

57 especially in the presence of a chemical ir ritant, could have resulted in rupture. The timing of such an event more closely coincide s with the animals clinical picture than rupture due to the TNBS alone. In future st udies, use of a pressure transducer with a pop-off valve attached to the Fole y valve could avoid similar problems. Endoscopy was an extremely useful tool for evaluation of ongoing distal colonic and rectal inflammation and a llowed scoring the severity of inflammation without animal sacrifice. The duration of inflammation was similar to previous reports in mice (2-3 weeks) and dogs and rats (up to 8 weeks) (Elson et al., 1995; Shibata et al., 1993). Because animals were not evaluated until on e week after TNBS/EtOH instillation, we likely did not capture the peak inflamma tory response endoscopically. Individual animals varied in both severity and duration of the lesion, typical of TNBS/EtOH colitis (Elson et al., 1995). The development of lesions in all animals was more consistent than that reported in dogs, but endosc opic lesion appearance was simila r (Shibata et al., 1993). Histological analysis at weeks 9 and 14 (5 and 10 weeks post-enema) did not reveal significant differences between groups. Overall, colonic and rectal tissues had little to no cellular infiltrate, and mucosal and submucosal edema was the only prominent abnormality. If the colonic or rectal edem a was related to TNBS /EtOH instillation, one would expect to see the highest scores in th e rectum of animals in that experimental group sacrificed at 9 weeks. Because edema scores did not differ significantly between groups, other contributing factors must be cons idered. Since tissue from all animals was handled similarly, a processing artifact is unlikely. Repeated CRD studies could have resulted in rectal irritation and edema. But, one would therefore expect a difference between colonic and rectal edema and a difference between the 9 and 14 week animals as

PAGE 71

58 a result of additional CRD procedures. These differences were not evident; thus, a specific cause of the rectal edema could not be identified. The presence of edema did not appear to have an impact on remaining study variables. A significant correlation betw een edema and sensitivity among individual animals was not identified, indicating that VH was not related to an ongoing inflammatory process. The other indicator of hist ological differences, number of lymphoid aggregates, could be misleading. This number is depende nt upon section locati on and angle as well as normal variability throughout th e gastrointestinal tract, and cellular infiltrate is more commonly used as an indicator of disease (Kru schewski et al., 2001). But, because very little cellular infiltrate was present in any of the sec tions, and the number of lymphoid aggregates appeared to vary between tissues based on a cursory examination, this count was used. Similar to the edema scoring, the number of lymphoid aggregates did not differ significantly between groups nor did it appear to have a major impact on other aspects of the study. Visceral Discomfort Pigs were easily trained to allow ball oon insertion, and the ramp protocol for balloon distention reliably indu ced a discomfort response in most trials. One of the important factors in developing a pig model of visceral disc omfort was the determination of discomfort. Previous an imal models have used both subjective methods such as an abdominal writhing response in rodents (R eichert et al., 2001) or more objective measurements such quantification of abdomin al muscle contractions (Al Chaer et al., 2000; Coutinho et al., 2002; Ness and Gebhart, 1988) or a partic ular behavior (Merritt et al., 2002b). A conscious passive avoidance behavior, such as pushing a lever to

PAGE 72

59 discontinue the painful stimulus, has also been described in rats (M essaoudi et al., 1999; Ness et al., 1991). For CRD studies, we chose to evaluate di scomfort subjectively, thus utilizing a threshold of discomfort as our primary res ponse variable. Most quantitative measures involve a comparison of a part icular variable between dist ention pressures rather than determination of a threshold response. Su ch a method would have required repeated painful colorectal distentions over the course of many week s (Gschossmann et al., 2001; Ness and Gebhart, 1990). We did not want to cause discomfort above threshold in our subject animals for several reasons, includi ng concern for their well-being. From a physiologic standpoint, repeated pa inful CRD can increase spinal fos and jun protooncogenes which, in turn, have been associated with central hyperexc itability (Traub et al., 1992). Thus, we wanted to minimize th e effect of repeated CRD procedures themselves on visceral sensitivity by subj ecting the animals to the fewest possible number of painful CRD. Similarly, we were concerned that the pigs could develop aversive behavior to the laborat ory, crate, personnel, or testin g procedure in general if we used repeated painful distentions. Because stress alone can influen ce visceral nociception (Bradesi et al., 2002; Couti nho et al., 2002), we wanted to minimize the physical and psychological stress imposed on our animal subjects. We considered using real-time abdominal myoelectrical data to contribute to the threshold response determination. But, due to the nature of the pigs, we felt that movement artifact would make interpretation difficult in a time-dependent situation. Thus, in an attempt to limit the subjective na ture of the behavioral observation technique, strict criteria were used to define a discomfo rt response, and observers were blinded as to

PAGE 73

60 treatment. Most pigs had an obvious discom fort response with ag reement of all three observers in 55% and 2/3 in 45% of trials. So me pigs were difficult to judge, especially for the baseline trials when they were more nervous and less accustomed to the testing procedure. In hindsight, obtaining an averag e baseline result from at least two, if not three trials, may have allowed a more thorough evaluation of the animals true baseline responses. Also in hindsight, equal sample sizes amongst all groups would have been preferable to the current de sign. The pilot animals produced very consistent CRD data. Because the main study was designed based upon their results, we presumed that salinetreated groups would react similarly, thus al lowing for a smaller sample size. Because saline-treated animals exhibited more variati on in CRD threshold response than expected, larger control groups likely would have dimi nished the impact of individual variability and increased the power of the study. Effect of Inflammation on Nociceptive Threshold Once a consistent model was established, a secondary goal of this study was to investigate the effect of subacute inflammation on nociceptive threshold. Weekly Nociceptive Thresholds The two methods used to express visceral sensitivity, threshold pressure and ABTP, describe the same data in slightly different ways. ABTP expresses the weekly thresholds as a variation of the animals own baseline threshold pressure, acc ounting for individual variability in sensitivity. Similar compar isons have been used previously in human visceral sensitivity testing (Sabate et al., 2002) For this reason, the ABTP results were used for all correlations, and will be stre ssed during discussions of sensitivity.

PAGE 74

61 The only significant difference in ABTP be tween experimental groups was at week 7, although the difference at week 13 appr oached significance (p=0.056). The mean pressures for week 13 were also significan tly different between groups. These results suggest an interesting biphasic response The initial period of hypersensitivity, manifested by a decrease in threshold pr essure relative to baseline, following TNBS/EtOH instillation is likely related to colonic inflammation and its associated mediators. This response is predictable based on previously documented effects of gastrointestinal inflammation (Messaoudi et al., 1999; Ness and Gebhart, 2000; Sharkey and Kroese, 2001). The second period of hypersensitivity occurred after the endoscopic resolution of inflammation in all animal s and following histologic resolution of inflammation in experiment al group 1. A post-inflamma tory period of visceral hypersensitivity supports the concept of plasticity within the afferent arm of the visceral nociceptive pathway. Some degree of individual variation in noc iceptive threshold is expected (Elmer et al., 1998), and such variability was observed in this study. When examining individual responses, only 2/7 saline animals had >2 week s with threshold pressures below baseline, whereas 11/12 TNBS/EtOH animals had >2 w eeks as such. Within the TNBS/EtOHtreated groups, variabil ity in the resultant inflammatory response may have contributed to this individual variation. Week 6 and 9 e ndoscopy scores had a significant negative correlation with week 13 ABTP (p=0.038 and 0.031, respectively; r2= 0.435 and 0.462, respectively), indicating that those animals with the highest endoscopy scores required less pressure to induce a noci ceptive response. Because the r2 value was less than 0.5 for each, this correlation should be considered a strong trend, rather than a significant

PAGE 75

62 determination. Because a similar correl ation did not exist between ABTP and histological scoring, any effect of inflam mation on visceral sensitivity appears related more to the degree of inflammation initially present in each animal rather than the remaining level of inflammation present at the time of testing (or in this case one week later). This further supports the notion of neuronal plasticity rather than the direct action of inflammatory mediators for the later period of hypersensitivity seen at week 13. The CRD protocol used in this study eval uated threshold pressure, rather than quantifying a particular pain re sponse to a given stimulus. Thus, a reduction in sensory threshold truly indicates allodyn ia, rather than hypersensitivity. Based on information in other species, animals with allodynia were lik ely also hypersensitive, but this cannot be proven given the constraints of the testing system used for this study (Mayer and Gebhart, 1994). The two saline-treated animals which c onsistently demonstrated negative ABTP likely contributed to the lack of a group effect in many week ly sensitivity thresholds. The sensitivity profiles of these two animals appeared to differ from the remaining salinetreated animals. These animals had thre shold pressures approximately 20 mmHg below their initial baseline threshold at week 7 (the first CRD post enema) and then remained at those thresholds (within 10 mmHg) for all subsequent week s. Given the consistency of this response, one potential expl anation is that the reported baseline for those two animals was erroneously high and the remaining weeks represented their normal threshold. Alternatively, these animals became hypersen sitive; possible cause s include the saline retention enema, soapy water enemas used prior to endoscopy, repeated endoscopic examinations, or repeated CRD procedures. To the authors knowledge, none of these

PAGE 76

63 have been reported as causes of visceral or somatic hypersensitivity, nor do they represent noxious procedures As stated previously, CRD to noxious pressures has been reported to cause hypersensitivity, but not non-noxious dist entions (Traub et al., 1992). Normal individual variability in visc eral sensory threshold is another possible explanation. This is the more plausible e xplanation, and would likel y coincide with an erroneously high baseline repor ted in these animals. Animal selection Other investigators have recently describe d models involving various insults during the neonatal period that result ed in visceral hypersensitivity during adulthood (Al Chaer et al., 2000; Coutinho et al., 2002; Ruda et al., 2000). Simila r findings have not, to the authors knowledge, been extensively evalua ted using a slightly older population of animals. At the time of TNBS/EtOH enem a administration, pigs in this study were approximately 8-10 weeks old, corresponding to early adolescence in humans. This stage of development may be important for the development of IBS (Sandler, 1990; Van Ginkel et al., 2001), a concept further s upported by the results of the present study. Castrated male pigs were chosen for this study to avoid the any potential effect of hormonal variation during the es trous cycle on nociceptive thre shold. Gender differences in both the perception and modul ation of pain have been doc umented in humans (Gear et al., 1996), and women are at a higher risk for the development of IBS (Mayer et al., 1999). IBS-related pain can also vary with phase of the menstrual cycle (Heitkemper and Jarrett, 1992). In animal m odels of pain, female rodents display lower nociceptive thresholds to both shock and thermal stimuli (Marks and Hobbs, 1972; Pare, 1969; Romero and Bodnar, 1986).

PAGE 77

64 Furthermore, gonadectomy has been re ported to decrease both nociceptive threshold and response to analgesia in rode nts (Marks and Hobbs, 1972; Romero et al., 1988). Thus, one cannot assume that the resu lts obtained from our castrated animals would be the same as those obtained in intact male pigs or in females. However, since they were castrated at a similar age, any potential effect of gonadectomy should have been uniform across all animals. Substance P When attempting to quantify SP in various tissues, immunohistoc hemical analysis offers several advantages over molecular analyses. Most importantly, immunohistochemistry offers localization of immunoreactivity rather than protein quantification in a particular tissue. For por cine tissue, specific anti-pig SP polyclonal antibody has been raised (Balemba et al., 2001; Balemba et al., 2002), but it is not commercially available. The present study used a rabbit polyclonal SP antibody that had been validated previously at a similar diluti on in porcine gastrointe stinal tissue (KulkarniNarla et al., 1999). Early antibod ies to SP showed some cro ss-reactivity to NK-A or NKB, however more recently purified antibodies have apparently overcome this problem (Duggan, 1995; Hoyle, 1998). In the gastrointestinal tissue, the loca tion chosen for Q-IHC sampling was based upon the location of myenteric plexus (Goyal and Hirano, 1996). SP has previously been identified in the mucosa and submucosal pl exus in the porcine gastrointestinal tract (Balemba et al., 2001; Balemba et al., 2002). During initial microscopic review of the gastrointestinal tissues in th is study, SP-IR was consistently seen in the myenteric plexus, but not consistently in the areas of the other plexes. Some non-specific immunoreactivity was seen in the mucosa, most prominently surrounding the tissue pe riphery. Thus, the

PAGE 78

65 quantitative analysis described in this re port focused on the region of the myenteric plexus. For precise neuroanatomical detail in the enteric nervous system, whole mount preparations of the gastrointestinal ti ssue are preferred (Balemba et al., 2001; Miampamba and Sharkey, 1998). The transverse sections used in th is study provide less specific detail in that most ne urons in the plexus are capture d only in part or in crosssection. However, as the purpose of the study was to evaluate the amount of SP-IR present, rather than to map its distri bution, this method met study requirements. The spinal cord segments were chosen as a representative sample of those receiving afferent input from the colon and rectum. Th e distal colon and rectum have dual sacral and lumbar afferent innervation (Ness a nd Gebhart, 1990). Based on Fluorogold labeling of the descending colon, DRG in the T13-L2 and L6-S2 regions received afferent input, but the number of positive neurons in the T 13-L2 region was greater following colonic inflammation (Traub et al., 1999). Thus, spinal segments L1 and L2 were chosen to represent the proximal extent of innervation, while L6 and L7 were chosen to represent the distal portion. The algorithm used to quantify SP-IR in the gastrointestinal tissues and spinal cord dorsal horn has been previously validated for use with DAB-based immunohistochemistry (Matkowskyj et al., 2000 ). This technique allows for precise documentation of the chromagen content in a specific image by subtracting the cumulative strength of the negative cont rol image from that contained in the corresponding immunostained image. Thus it allowed for a quantitative evaluation of the SP-IR within specific locations in the spin al cord, colon, and rectum. Because the chromagen content is expressed in arbitr ary units, these data cannot be reasonably

PAGE 79

66 compared to other studies. But, this provided a more objective comparison between groups than a subjective scoring system. Direct Relationship between Histopathology and Sensitivity Based on an initial review of the gastro intestinal sections, very little cellular infiltrate was noted. The only apparent abnormalities were mucosal and submucosal edema. Also, the number of lymphoid aggreg ates appeared to vary between sections. One explanation for this varia tion is the inherent variability due to section location and angle of the cut. But, an analysis of the lymphoid aggregate counts was performed to investigate the possibility of a treatment effect. Although the correlation between the R2 lymphoid aggregate score and ABTP in weeks 10 and 12 had significant Pearsons correlations, only the week 10 linear regression had a r2 value >0.5. None of the correlations between histological scoring and SP-IR in either the spinal cord or gastrointestinal tract had an r2 value >0.5, thus these should be considered trends at best. Visual inspection of these plots does not give the impression of a strong linear relationship (Fi g. 4-8 and 4-13). Based on these facts and the lack of correlation between ABTP and a ny edema scores, the histopathologic changes seen within the gastrointestinal tract at the time of euthanasia did not appear to play a significant role in visceral sensitivity. Central versus Peripheral Sensitization The lack of a difference in SP-IR in either the colon or rectum between any of the experimental groups was somewhat unexpecte d. Other investigat ors have shown an initial decrease and subse quent increase in SP-IR throughout the colon following TNBS/EtOH instillation in ra ts (Miampamba and Sharkey, 1998) and in the primary afferent nerves in a guinea pig TNBS ileitis model (Miller et al., 1993). These reported

PAGE 80

67 changes occurred within 2 weeks following th e inflammatory insult, and the long-term effects of inflammation upon SP-IR have not b een fully elucidated. Thus, while this study did not evaluate the ear ly effects of inflammation upon SP-IR, the information presented for the periods 5 and 10 weeks after TNBS/EtOH instillation provide new insight into the pathophysiol ogy of inflammatory-mediated VH. Due to the relatively small sample size and individua l variation in inflammatory response, a small difference between groups may have gone undetected. In addition to a lack of detectable difference between groups, SP-IR in the gastrointestinal tract did not correla te significantly with ABTP for any week. These results, in conjunction with the significant correlation between dorsal horn SP-IR and ABTB, suppor t a central rather than peripheral mechanism of sensitization in these animals. No significant difference wa s detected between experi mental groups in either dorsal horn SP-IR or ve ntral horn SP-IR neurons. Simila r to the reasoning described for gastrointestinal tissues, i ndividual variability and sample size may have prevented detection of a small difference between groups Alternatively, groups truly did not differ in immunoreactivity. Because VH did not devel op in all of the TNBS/EtOH animals, the correlation between weekly ABTP and SP-IR may provide more insight into the relationship between SP in the sp inal cord and the level of vi sceral sensitivity in a given animal. The significant correlation between dorsal horn SP-IR at the L1 and L7 segments with multiple weekly ABTP supports a centr al mechanism of hypersensitivity. This correlation was strongest between L7 and w eek 11, but also significant between L7 and week 14 and between L1 and week 9. The ar gument for a role of SP in the development

PAGE 81

68 of hypersensitivity in these animals would be stronger had SP-IR correlated significantly with other weeks, especially when th e TNBS/EtOH and saline groups differed significantly. But, the correlation between SP -IR in the dorsal horn and ABTP across all experimental groups further validates the im portance of SP within the dorsal horn in the development of central sensitization, re gardless of the inciting cause. SP has been associated with afferent tr ansmission of nociception for years (Hokfelt et al., 1977a; Mayer and Rayboul d, 1990; Otsuka and Yoshioka 1993). Not surprisingly, SP is thought to play a role in central sensitization and inflammatory-mediated VH (Kishimoto, 1994; Miampamba et al., 1992; Pers son et al., 1995; Schne ider et al., 2001; Swain et al., 1992). Zymosan-induced colitis reduced the number of SP-labeled cells in both the T13-L2 and L6-S2 afferent DRG in rats (Traub et al., 1999). Also, TNBS ileitis has been shown to induce hyperexcitability in nociceptive DRG neurons.(Moore et al., 2002) In the present study, SP -IR in the dorsal horn was gr eatest in th e superficial laminae (I and II) which receive afferent input consistent with previous reports (Duggan, 1995; Kawata et al., 1989; Routh and Helke, 1995). The strongest correlation with ABTP in the lumbar spinal cord was at L7, corres ponding to lumbosacral afferent input from the distal colon and rectum. Given the previously suggested importance of SP in the development of central sensitization and the changes previously asso ciated with gastroin testinal inflammation, alterations in SP-IR were expected in this study. However, a specific correlation between visceral sensitivity in indi vidual animals and SP-immunoreactiv ity in the spinal cord dorsal horn has not, to our knowledge, been previously documented. This association between SP in the dorsal horn and the lack of an association with SP -IR in the colon or

PAGE 82

69 rectum highlights the role of SP in central se nsitization and the importa nce of this process in post-inflammatory visceral hypersensitivity. In the ventral horn, the correlation between SP-IR neurons and ABTP, especially at the L2 segment implies an alteration in motor pathways. However, these interpretations should be considered cau tiously because the r2 for all linear regressions were less than 0.51, and visual inspection of the correlat ion does not imply a consistent linear relationship. SP-IR has been documented in large motoneurons of the ventral horn, but an alteration in ventral horn SP-IR has not prev iously been attributed to rectal or colonic inflammation (Charlton and Helke, 1985b; Ch arlton and Helke, 1985a). However, given the appearance of the linear regression plot s and the function of the large ventral horn motoneurons, the true relevance of the co rrelation between vent ral horn SP-IR neurons and visceral sensitivity require s further investigation. Conclusions Current Study The main accomplishment of this study was to integrate models of colonic inflammation and visceral pain in the pig. The use of a large animal subject, such as the pig, allowed for endoscopic scoring of gross mu cosal changes within the distal colon and rectum. This provided an evaluation of each animals individual response to TNBS/EtOH instillation, rather than relying on a group average based on histopathologic changes seen in animals sacrificed at various time points. This information allowed for a more thorough characterization of individual responses while also decreasing the required number of animals. Because of the va riability in response to TNBS/EtOH, the endoscopic information proved useful, in that the resultant colitis severity scores following instillation correlated negatively with ABTP.

PAGE 83

70 TNBS/EtOH instillation resulted in a biphasic pattern of visceral hypersensitivity. The first period occurred in the presence of ongoing inflammation, but the second period occurred after the gross resolution of in flammation in the subject animals and the histological resolution of inflammation in cohor ts euthanized at week 9. These results, combined with a lack of consistent corre lation between ABTP a nd histological scoring, suggest that either central a nd/or peripheral mechanisms of hypersensitization, rather than inflammatory mediators, appear responsible for observed alterations in visceral sensitivity. Although TNBS/EtOH instillation did not ha ve a significant effect on spinal or colonic SP-immunoreactivity, a strong correlation between dorsal horn SP-IR at the L1 and L7 levels and ABTP was evident. Due to the small sample size and variation within each group, this correlation more likely represents the true involvement of spinal SP in a visceral hypersensitivity response, consistent with central sensitization. The inclusion of both salineand TNBS/EtOH-treated animal s in these correlations suggests the importance of SP in VH regardless of the inciting cause, an unexpected but intriguing result. Future Studies Based upon results of this study, the pig ma y serve as an excellent large animal model for future studies c oncerning the pathophysiology and treatment of inflammatorymediated IBS. One of the first options w ould be to determine the immunoreactivity of other mediators of central se nsitization, such as CGRP, NMDA, NGF, and others, within the spinal cords of the pigs in this study. These studies could be easily accomplished as paraffin-embedded tissue from all animals has been retained. Combined with the ABTP, endoscopic scores, histol ogic scores, and SP-IR, fu rther immunohistochemical

PAGE 84

71 characterization of this mode l could contribute to a mo re thorough understanding of the post-inflammatory neuroplastic changes described in this model. The porcine CRD model could be used to evaluate the effect of potential therapeutic agents on visceral sensitivity thre sholds, either with or without a previous inflammatory insult. In such studies endoscopic evaluation would allow for classification of the subjects based on lesion severity. Due to the chronic nature of IBS and the temporal separation between insult and documentation of VH in neonatal rodent mode ls, studies of longer duration are also warranted. For example, a porcine study similar to that described in this report, but that extended to adulthood, would more fully ch aracterize the duration and nature of alterations in visceral sensitivity. If the bi-phasic pattern of VH seen in this study persisted in a relapsing/remitting fashion, the model would offer a similar clinical picture to that seen in IBS. Furthermore, one could evaluate the effect of repeated versus singledose TNBS/EtOH instillations on the visceral sensitivity pattern to determine whether or not a repetitive inflammatory insult further in creases the severity or duration of VH.

PAGE 85

72 APPENDIX A INDIVIDUAL ANIMAL DATA Table A-1. Raw data from CRD studies. An # Group Wk 3 Wk 7 Wk 8 Wk 9 Wk 10 Wk 11 Wk 12 Wk 13 Wk 14 3161 3 45 15 25 25 15 15 25 25 25 3171 1 55 25 15 25 3162 4 25 45 25 25 25 15 15 25 45 3185 1 35 35 25 45 3197 1 55* 45 45 35 3184 4 15 55* 35 35 55 55* 55 25 45 3191 3 45 25 15 25 35 25 15 15 25 John 1 45 25 15 25 Red 1 35 35 15 25 3166 1 35 15 15 25 3198 2 25 25 15 35 3179 2 25 25 25 xx 3182 4 55* 25 25 15 25 25 15 45 45 3189 2 15 35 15 15 3190 3 55* 15 55* 55* 55 35 15 15 35 3193 4 55* 25 15 15 15 25 15 45 35 3192 3 45 25 25 35 55 35 25 15 45 3130 3 25 15 15 25 15 55* 15 15 55* 3126 3 15 15 35 35 45 35 45 15 25

PAGE 86

73 Table A-2. Raw data from endoscopic examinations. Animal # Group Wk 3 Wk 6 Wk 9 Wk 14 3171 1 0.5 3.0 0.5 3185 1 0.5 5.0 2.0 3197 1 0.5 2.0 0.5 John 1 0.5 3.0 Red 1 0.0 2.0 3166 1 0.0 3.0 0.5 3198 2 0.5 0.5 0.5 3179 2 1.0 0.5 0.0 3189 2 0.0 1.5 0.5 3161 3 0.0 3.5 1.0 3191 3 0.5 4.0 1.0 3190 3 0.5 2.0 1.0 0.0 3192 3 0.0 2.0 0.0 0.5 3130 3 0.5 2.0 0.5 0.0 3126 3 0.5 1.0 0.0 0.5 3162 4 0.5 0.0 0.0 3184 4 1.0 0.5 0.0 3182 4 0.5 0.0 0.0 0.0 3193 4 0.5 0.0 0.0

PAGE 87

74 APPENDIX B ANOVA TABLES Table B-1. One-way ANOVA analys is for threshold pressure. Week Groups Sum of Squares df Mean Square F Sig. 3 Between 452.6942 1 452.6942 2.222232 0.154358 Within 3463.095 17 203.7115 Total 3915.789 18 7 Between 391.0401 1 391.0401 3.361862 0.084292 Within 1977.381 17 116.3165 Total 2368.421 18 8 Between 36.09023 1 36.09023 0.261874 0.615419 Within 2342.857 17 137.8151 Total 2378.947 18 9 Between 289.7243 1 289.7243 3.172748 0.09275 Within 1552.381 17 91.31653 Total 1842.105 18 10 Between 106.6667 1 106.6667 0.330323 0.581267 Within 2583.333 8 322.9167 Total 2690 9 11 Between 26.66667 1 26.66667 0.119626 0.73836 Within 1783.333 8 222.9167 Total 1810 9 12 Between 6.666667 1 6.666667 0.028319 0.870539 Within 1883.333 8 235.4167 Total 1890 9 13 Between 806.6667 1 806.6667 13.35172 0.006457 Within 483.3333 8 60.41667 Total 1290 9 14 Between 135 1 135 1.234286 0.298847 Within 875 8 109.375 Total 1010 9

PAGE 88

75 Table B-2. One-way ANOV A analysis for ABTP. Week Groups Sum of Squares df Mean Square F Sig. 7 Between 1687.269 1 1687.269 4.679076 0.04507 Within 6130.179 17 360.5987 Total 7817.447 18 8 Between 226.6931 1 226.6931 0.737567 0.402389 Within 5224.991 17 307.3524 Total 5451.684 18 9 Between 16.34226 1 16.34226 0.048641 0.828073 Within 5711.658 17 335.9799 Total 5728 18 10 Between 86.4 1 86.4 0.114968 0.743287 Within 6012.125 8 751.5156 Total 6098.525 9 11 Between 15.50417 1 15.50417 0.019584 0.892164 Within 6333.396 8 791.6745 Total 6348.9 9 12 Between 13.06667 1 13.06667 0.014309 0.907734 Within 7305.458 8 913.1823 Total 7318.525 9 13 Between 866.4 1 866.4 4.978863 0.05618 Within 1392.125 8 174.0156 Total 2258.525 9 14 Between 160.0667 1 160.0667 0.329147 0.581927 Within 3890.458 8 486.3073

PAGE 89

76 Table B-3. One-way ANOVA anal ysis for endoscopy scores. Week Groups Sum of Squares df Mean Square F Sig. 3 Between 0.122807 1 0.122807 1.252632 0.278616 Within 1.666667 17 0.098039 Total 1.789474 18 6 Between 21.56031 1 21.56031 24.88431 0.000112 Within 14.72917 17 0.866422 Total 36.28947 18 9 Between 1.278151 1 1.278151 5.545698 0.032564 Within 3.457143 15 0.230476 Total 4.735294 16 14 Between 0.125 1 0.125 3 0.133975 Within 0.25 6 0.041667 Total 0.375 7

PAGE 90

77 Table B-4. One-way ANOVA analysis for ga strointestinal histologic scores. Section Group Sum of Squares df Mean Square F Sig. C1LYM Between 1.3640 3 0.4547 0.4377 0.7293 Within 15.5833 15 1.0389 Total 16.9474 18 C1ED Between 2.2895 3 0.7632 0.9954 0.4218 Within 11.5000 15 0.7667 Total 13.7895 18 C1TOT Between 0.8640 3 0.2880 0.1347 0.9379 Within 32.0833 15 2.1389 Total 32.9474 18 C2LYM Between 0.2061 3 0.0687 0.1076 0.9544 Within 9.5833 15 0.6389 Total 9.7895 18 C2ED Between 2.2412 3 0.7471 0.8676 0.4795 Within 12.9167 15 0.8611 Total 15.1579 18 C2TOT Between 2.7895 3 0.9298 0.8204 0.5026 Within 17.0000 15 1.1333 Total 19.7895 18 R1LYM Between 2.1140 3 0.7047 0.5074 0.6831 Within 20.8333 15 1.3889 Total 22.9474 18 R1ED Between 5.0482 3 1.6827 2.1791 0.1330 Within 11.5833 15 0.7722 Total 16.6316 18 R1TOT Between 10.3202 3 3.4401 1.2767 0.3182 Within 40.4167 15 2.6944 Total 50.7368 18 R2LYM Between 1.1930 3 0.3977 0.3890 0.7626 Within 15.3333 15 1.0222 Total 16.5263 18 R2ED Between 1.7982 3 0.5994 0.4316 0.7334 Within 20.8333 15 1.3889 Total 22.6316 18 R2TOT Between 4.7982 3 1.5994 0.6695 0.5838 Within 35.8333 15 2.3889 Total 40.6316 18

PAGE 91

78 Table B-5. Tukeys HSD analysis for gastrointestinal histologic scores. Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95% CI-Lower 95% CI-Upper C1LYM 1 2 -0.1667 0.7207 0.9955 -2.2439 1.9106 3 0.5000 0.5885 0.8300 -1.1961 2.1961 4 -0.0833 0.6579 0.9992 -1.9796 1.8129 2 1 0.1667 0.7207 0.9955 -1.9106 2.2439 3 0.6667 0.7207 0.7921 -1.4106 2.7439 4 0.0833 0.7785 0.9995 -2.1603 2.3270 3 1 -0.5000 0.5885 0.8300 -2.1961 1.1961 2 -0.6667 0.7207 0.7921 -2.7439 1.4106 4 -0.5833 0.6579 0.8118 -2.4796 1.3129 4 1 0.0833 0.6579 0.9992 -1.8129 1.9796 2 -0.0833 0.7785 0.9995 -2.3270 2.1603 3 0.5833 0.6579 0.8118 -1.3129 2.4796 C1ED 1 2 -0.1667 0.6191 0.9929 -1.9511 1.6178 3 -0.8333 0.5055 0.3831 -2.2903 0.6237 4 -0.5000 0.5652 0.8128 -2.1290 1.1290 2 1 0.1667 0.6191 0.9929 -1.6178 1.9511 3 -0.6667 0.6191 0.7084 -2.4511 1.1178 4 -0.3333 0.6687 0.9582 -2.2608 1.5941 3 1 0.8333 0.5055 0.3831 -0.6237 2.2903 2 0.6667 0.6191 0.7084 -1.1178 2.4511 4 0.3333 0.5652 0.9336 -1.2956 1.9623 4 1 0.5000 0.5652 0.8128 -1.1290 2.1290 2 0.3333 0.6687 0.9582 -1.5941 2.2608 3 -0.3333 0.5652 0.9336 -1.9623 1.2956 C1TOT 1 2 -0.3333 1.0341 0.9880 -3.3139 2.6472 3 -0.3333 0.8444 0.9784 -2.7669 2.1003 4 -0.5833 0.9440 0.9248 -3.3042 2.1375 2 1 0.3333 1.0341 0.9880 -2.6472 3.3139 3 0.0000 1.0341 1.0000 -2.9805 2.9805 4 -0.2500 1.1170 0.9959 -3.4694 2.9694 3 1 0.3333 0.8444 0.9784 -2.1003 2.7669 2 0.0000 1.0341 1.0000 -2.9805 2.9805 4 -0.2500 0.9440 0.9932 -2.9709 2.4709 4 1 0.5833 0.9440 0.9248 -2.1375 3.3042 2 0.2500 1.1170 0.9959 -2.9694 3.4694 3 0.2500 0.9440 0.9932 -2.4709 2.9709 C2LYM 1 2 0.1667 0.5652 0.9907 -1.4623 1.7956 3 0.1667 0.4615 0.9833 -1.1634 1.4967 4 -0.0833 0.5159 0.9984 -1.5704 1.4037 2 1 -0.1667 0.5652 0.9907 -1.7956 1.4623 3 0.0000 0.5652 1.0000 -1.6290 1.6290 4 -0.2500 0.6105 0.9760 -2.0095 1.5095 3 1 -0.1667 0.4615 0.9833 -1.4967 1.1634 2 0.0000 0.5652 1.0000 -1.6290 1.6290 4 -0.2500 0.5159 0.9614 -1.7370 1.2370 4 1 0.0833 0.5159 0.9984 -1.4037 1.5704 2 0.2500 0.6105 0.9760 -1.5095 2.0095 3 0.2500 0.5159 0.9614 -1.2370 1.7370

PAGE 92

79 Table B-5. Continued Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95% CI-Lower 95% CI-Upper C2ED 1 2 0.1667 0.6562 0.9940 -1.7245 2.0578 3 -0.5000 0.5358 0.7878 -2.0441 1.0441 4 -0.7500 0.5990 0.6052 -2.4764 0.9764 2 1 -0.1667 0.6562 0.9940 -2.0578 1.7245 3 -0.6667 0.6562 0.7428 -2.5578 1.2245 4 -0.9167 0.7087 0.5807 -2.9594 1.1260 3 1 0.5000 0.5358 0.7878 -1.0441 2.0441 2 0.6667 0.6562 0.7428 -1.2245 2.5578 4 -0.2500 0.5990 0.9746 -1.9764 1.4764 4 1 0.7500 0.5990 0.6052 -0.9764 2.4764 2 0.9167 0.7087 0.5807 -1.1260 2.9594 3 0.2500 0.5990 0.9746 -1.4764 1.9764 C2TOT 1 2 0.3333 0.7528 0.9700 -1.8363 2.5029 3 -0.3333 0.6146 0.9472 -2.1048 1.4381 4 -0.8333 0.6872 0.6286 -2.8139 1.1472 2 1 -0.3333 0.7528 0.9700 -2.5029 1.8363 3 -0.6667 0.7528 0.8123 -2.8363 1.5029 4 -1.1667 0.8131 0.4983 -3.5101 1.1768 3 1 0.3333 0.6146 0.9472 -1.4381 2.1048 2 0.6667 0.7528 0.8123 -1.5029 2.8363 4 -0.5000 0.6872 0.8845 -2.4806 1.4806 4 1 0.8333 0.6872 0.6286 -1.1472 2.8139 2 1.1667 0.8131 0.4983 -1.1768 3.5101 3 0.5000 0.6872 0.8845 -1.4806 2.4806 R1LYM 1 2 -0.5000 0.8333 0.9305 -2.9018 1.9018 3 0.5000 0.6804 0.8816 -1.4611 2.4611 4 0.1667 0.7607 0.9961 -2.0259 2.3592 2 1 0.5000 0.8333 0.9305 -1.9018 2.9018 3 1.0000 0.8333 0.6361 -1.4018 3.4018 4 0.6667 0.9001 0.8792 -1.9276 3.2609 3 1 -0.5000 0.6804 0.8816 -2.4611 1.4611 2 -1.0000 0.8333 0.6361 -3.4018 1.4018 4 -0.3333 0.7607 0.9709 -2.5259 1.8592 4 1 -0.1667 0.7607 0.9961 -2.3592 2.0259 2 -0.6667 0.9001 0.8792 -3.2609 1.9276 3 0.3333 0.7607 0.9709 -1.8592 2.5259 R1ED 1 2 -0.3333 0.6214 0.9488 -2.1242 1.4576 3 0.5000 0.5074 0.7597 -0.9623 1.9623 4 -0.9167 0.5672 0.3996 -2.5515 0.7182 2 1 0.3333 0.6214 0.9488 -1.4576 2.1242 3 0.8333 0.6214 0.5526 -0.9576 2.6242 4 -0.5833 0.6712 0.8205 -2.5177 1.3511 3 1 -0.5000 0.5074 0.7597 -1.9623 0.9623 2 -0.8333 0.6214 0.5526 -2.6242 0.9576 4 -1.4167 0.5672 0.1010 -3.0515 0.2182 4 1 0.9167 0.5672 0.3996 -0.7182 2.5515 2 0.5833 0.6712 0.8205 -1.3511 2.5177 3 1.4167 0.5672 0.1010 -0.2182 3.0515

PAGE 93

80 Table B-5. Continued Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95% CI-Lower 95% CI-Upper R1TOT 1 2 -0.8333 1.1607 0.8884 -4.1786 2.5120 3 1.0000 0.9477 0.7207 -1.7314 3.7314 4 -0.7500 1.0596 0.8924 -3.8038 2.3038 2 1 0.8333 1.1607 0.8884 -2.5120 4.1786 3 1.8333 1.1607 0.4187 -1.5120 5.1786 4 0.0833 1.2537 0.9999 -3.5300 3.6967 3 1 -1.0000 0.9477 0.7207 -3.7314 1.7314 2 -1.8333 1.1607 0.4187 -5.1786 1.5120 4 -1.7500 1.0596 0.3815 -4.8038 1.3038 4 1 0.7500 1.0596 0.8924 -2.3038 3.8038 2 -0.0833 1.2537 0.9999 -3.6967 3.5300 3 1.7500 1.0596 0.3815 -1.3038 4.8038 R2LYM 1 2 0.5000 0.7149 0.8957 -1.5605 2.5605 3 0.0000 0.5837 1.0000 -1.6824 1.6824 4 -0.3333 0.6526 0.9553 -2.2143 1.5476 2 1 -0.5000 0.7149 0.8957 -2.5605 1.5605 3 -0.5000 0.7149 0.8957 -2.5605 1.5605 4 -0.8333 0.7722 0.7070 -3.0589 1.3923 3 1 0.0000 0.5837 1.0000 -1.6824 1.6824 2 0.5000 0.7149 0.8957 -1.5605 2.5605 4 -0.3333 0.6526 0.9553 -2.2143 1.5476 4 1 0.3333 0.6526 0.9553 -1.5476 2.2143 2 0.8333 0.7722 0.7070 -1.3923 3.0589 3 0.3333 0.6526 0.9553 -1.5476 2.2143 R2ED 1 2 0.0000 0.8333 1.0000 -2.4018 2.4018 3 0.1667 0.6804 0.9946 -1.7944 2.1277 4 -0.6667 0.7607 0.8169 -2.8592 1.5259 2 1 0.0000 0.8333 1.0000 -2.4018 2.4018 3 0.1667 0.8333 0.9970 -2.2351 2.5685 4 -0.6667 0.9001 0.8792 -3.2609 1.9276 3 1 -0.1667 0.6804 0.9946 -2.1277 1.7944 2 -0.1667 0.8333 0.9970 -2.5685 2.2351 4 -0.8333 0.7607 0.6976 -3.0259 1.3592 4 1 0.6667 0.7607 0.8169 -1.5259 2.8592 2 0.6667 0.9001 0.8792 -1.9276 3.2609 3 0.8333 0.7607 0.6976 -1.3592 3.0259 R2TOT 1 2 0.5000 1.0929 0.9671 -2.6499 3.6499 3 0.1667 0.8924 0.9976 -2.4052 2.7386 4 -1.0000 0.9977 0.7504 -3.8755 1.8755 2 1 -0.5000 1.0929 0.9671 -3.6499 2.6499 3 -0.3333 1.0929 0.9897 -3.4833 2.8166 4 -1.5000 1.1805 0.5942 -4.9023 1.9023 3 1 -0.1667 0.8924 0.9976 -2.7386 2.4052 2 0.3333 1.0929 0.9897 -2.8166 3.4833 4 -1.1667 0.9977 0.6543 -4.0421 1.7088 4 1 1.0000 0.9977 0.7504 -1.8755 3.8755 2 1.5000 1.1805 0.5942 -1.9023 4.9023 3 1.1667 0.9977 0.6543 -1.7088 4.0421

PAGE 94

81 Table B-6. One-way ANOVA for vent ral horn SP-immunoreactive neurons. Section Groups Sum of Squares df Mean Square F Sig. L1 Between 128.4035 3.0000 42.8012 1.1882 0.3477 Within 540.3333 15.0000 36.0222 Total 668.7368 18.0000 L2 Between 257.0482 3.0000 85.6827 2.0811 0.1457 Within 617.5833 15.0000 41.1722 Total 874.6316 18.0000 L6 Between 253.2895 3.0000 84.4298 0.6434 0.5990 Within 1968.5000 15.0000 131.2333 Total 2221.7895 18.0000 L7 Between 5.6111 3.0000 1.8704 0.1250 0.9438 Within 209.5000 14.0000 14.9643 Total 215.1111 17.0000

PAGE 95

82 Table B-7. Tukeys HSD analysis for ventral horn SP-immunoreactive neurons. Section (I) Grp (J) Grp Mean Diff(I-J) SEM Sig. 95% CI-Lower 95% CI-Upper L1 1 2 1.5000 4.2439 0.9843 -10.7317 13.7317 3 -5.3333 3.4652 0.4403 -15.3205 4.6538 4 -2.5000 3.8742 0.9156 -13.6660 8.6660 2 1 -1.5000 4.2439 0.9843 -13.7317 10.7317 3 -6.8333 4.2439 0.4026 -19.0650 5.3984 4 -4.0000 4.5840 0.8188 -17.2117 9.2117 3 1 5.3333 3.4652 0.4403 -4.6538 15.3205 2 6.8333 4.2439 0.4026 -5.3984 19.0650 4 2.8333 3.8742 0.8830 -8.3326 13.9993 4 1 2.5000 3.8742 0.9156 -8.6660 13.6660 2 4.0000 4.5840 0.8188 -9.2117 17.2117 3 -2.8333 3.8742 0.8830 -13.9993 8.3326 L2 1 2 0.8333 4.5372 0.9977 -12.2435 13.9102 3 -7.1667 3.7046 0.2557 -17.8439 3.5105 4 -6.9167 4.1419 0.3724 -18.8542 5.0208 2 1 -0.8333 4.5372 0.9977 -13.9102 12.2435 3 -8.0000 4.5372 0.3279 -21.0769 5.0769 4 -7.7500 4.9007 0.4177 -21.8746 6.3746 3 1 7.1667 3.7046 0.2557 -3.5105 17.8439 2 8.0000 4.5372 0.3279 -5.0769 21.0769 4 0.2500 4.1419 0.9999 -11.6875 12.1875 4 1 6.9167 4.1419 0.3724 -5.0208 18.8542 2 7.7500 4.9007 0.4177 -6.3746 21.8746 3 -0.2500 4.1419 0.9999 -12.1875 11.6875 L6 1 2 4.1667 8.1004 0.9544 -19.1799 27.5133 3 -4.1667 6.6140 0.9208 -23.2291 14.8957 4 5.0000 7.3946 0.9045 -16.3124 26.3124 2 1 -4.1667 8.1004 0.9544 -27.5133 19.1799 3 -8.3333 8.1004 0.7357 -31.6799 15.0133 4 0.8333 8.7494 0.9997 -24.3839 26.0505 3 1 4.1667 6.6140 0.9208 -14.8957 23.2291 2 8.3333 8.1004 0.7357 -15.0133 31.6799 4 9.1667 7.3946 0.6126 -12.1458 30.4791 4 1 -5.0000 7.3946 0.9045 -26.3124 16.3124 2 -0.8333 8.7494 0.9997 -26.0505 24.3839 3 -9.1667 7.3946 0.6126 -30.4791 12.1458 L7 1 2 1.0000 2.7354 0.9826 -6.9505 8.9505 3 -0.5000 2.2334 0.9959 -6.9915 5.9915 4 0.6667 2.7354 0.9947 -7.2838 8.6171 2 1 -1.0000 2.7354 0.9826 -8.9505 6.9505 3 -1.5000 2.7354 0.9455 -9.4505 6.4505 4 -0.3333 3.1585 0.9996 -9.5138 8.8471 3 1 0.5000 2.2334 0.9959 -5.9915 6.9915 2 1.5000 2.7354 0.9455 -6.4505 9.4505 4 1.1667 2.7354 0.9730 -6.7838 9.1171 4 1 -0.6667 2.7354 0.9947 -8.6171 7.2838 2 0.3333 3.1585 0.9996 -8.8471 9.5138 3 -1.1667 2.7354 0.9730 -9.1171 6.7838

PAGE 96

83 Table B-8. One-way ANOVA for SP-i mmunoreactivity (QIHC EU/pixel) Section Groups Sum of Squares df Mean Square F Sig. L1 Between 8487.5138 3 2829.1713 1.4157 0.2772 Within 29975.3478 15 1998.3565 Total 38462.8616 18 L2 Between 1542.5863 3 514.1954 0.2947 0.8286 Within 26170.6220 15 1744.7081 Total 27713.2084 18 L6 Between 4541.2806 3 1513.7602 0.6128 0.6172 Within 37056.2933 15 2470.4196 Total 41597.5739 18 L7 Between 9247.0372 3 3082.3457 1.0665 0.3947 Within 40460.8227 14 2890.0588 Total 49707.8599 17 R1 Between 7141.7124 3 2380.5708 1.1332 0.3695 Within 29411.3091 14 2100.8078 Total 36553.0216 17 R2 Between 6336.3136 3 2112.1045 0.7050 0.5647 Within 41940.1497 14 2995.7250 Total 48276.4633 17 C1 Between 7798.2417 3 2599.4139 0.5905 0.6313 Within 61626.3441 14 4401.8817 Total 69424.5858 17 C2 Between 31907.1178 3 10635.7059 2.3118 0.1326 Within 50607.6520 11 4600.6956 Total 82514.7698 14

PAGE 97

84 Table B-9. Tukeys HSD analysis for spin al SP-Immunoreactivity (QIHC EU/pixel) Section (I) Grp (J) Grp Mean Diff(I-J) SEM Sig. 95% CILower 95% CIUpper L1 1 2 46.5000 31.6098 0.4778 -44.6041 137.6041 3 -1.7317 25.8093 0.9999 -76.1179 72.6545 4 39.9089 28.8557 0.5281 -43.2574 123.0752 2 1 -46.5000 31.6098 0.4778 -137.6041 44.6041 3 -48.2317 31.6098 0.4475 -139.3358 42.8724 4 -6.5911 34.1425 0.9973 -104.9948 91.8126 3 1 1.7317 25.8093 0.9999 -72.6545 76.1179 2 48.2317 31.6098 0.4475 -42.8724 139.3358 4 41.6406 28.8557 0.4936 -41.5257 124.8068 4 1 -39.9089 28.8557 0.5281 -123.0752 43.2574 2 6.5911 34.1425 0.9973 -91.8126 104.9948 3 -41.6406 28.8557 0.4936 -124.8068 41.5257 L2 1 2 13.3519 29.5356 0.9682 -71.7742 98.4780 3 -13.6559 24.1157 0.9406 -83.1610 55.8493 4 -1.4372 26.9622 0.9999 -79.1464 76.2719 2 1 -13.3519 29.5356 0.9682 -98.4780 71.7742 3 -27.0078 29.5356 0.7976 -112.1339 58.1183 4 -14.7892 31.9021 0.9659 -106.7359 77.1575 3 1 13.6559 24.1157 0.9406 -55.8493 83.1610 2 27.0078 29.5356 0.7976 -58.1183 112.1339 4 12.2186 26.9622 0.9680 -65.4905 89.9278 4 1 1.4372 26.9622 0.9999 -76.2719 79.1464 2 14.7892 31.9021 0.9659 -77.1575 106.7359 3 -12.2186 26.9622 0.9680 -89.9278 65.4905 L6 1 2 2.7550 35.1456 0.9998 -98.5397 104.0497 3 4.0541 28.6962 0.9989 -78.6527 86.7609 4 39.8828 32.0834 0.6106 -52.5862 132.3518 2 1 -2.7550 35.1456 0.9998 -104.0497 98.5397 3 1.2991 35.1456 1.0000 -99.9956 102.5938 4 37.1278 37.9615 0.7638 -72.2830 146.5386 3 1 -4.0541 28.6962 0.9989 -86.7609 78.6527 2 -1.2991 35.1456 1.0000 -102.5938 99.9956 4 35.8287 32.0834 0.6852 -56.6403 128.2977 4 1 -39.8828 32.0834 0.6106 -132.3518 52.5862 2 -37.1278 37.9615 0.7638 -146.5386 72.2830 3 -35.8287 32.0834 0.6852 -128.2977 56.6403 L7 1 2 -18.7320 38.0135 0.9594 -129.2209 91.7570 3 -3.0398 31.0379 0.9996 -93.2536 87.1741 4 -63.3003 38.0135 0.3768 -173.7892 47.1886 2 1 18.7320 38.0135 0.9594 -91.7570 129.2209 3 15.6922 38.0135 0.9754 -94.7967 126.1811 4 -44.5683 43.8943 0.7434 -172.1499 83.0133 3 1 3.0398 31.0379 0.9996 -87.1741 93.2536 2 -15.6922 38.0135 0.9754 -126.1811 94.7967 4 -60.2605 38.0135 0.4175 -170.7494 50.2284 4 1 63.3003 38.0135 0.3768 -47.1886 173.7892 2 44.5683 43.8943 0.7434 -83.0133 172.1499 3 60.2605 38.0135 0.4175 -50.2284 170.7494

PAGE 98

85 Table B-10. Tukeys HSD for gastrointestin al SP-Immunoreactivity (QIHC EU/pixel) Section (I) Grp (J) Grp Mean Diff(I-J) SEM Sig. 95% CILower 95% CIUpper R1 1 2 6.2406 32.4099 0.9973 -87.9611 100.4422 3 -7.1699 27.7542 0.9937 -87.8394 73.4995 4 -47.9365 29.5861 0.3994 -133.9305 38.0574 2 1 -6.2406 32.4099 0.9973 -100.4422 87.9611 3 -13.4105 33.4728 0.9774 -110.7015 83.8806 4 -54.1771 35.0067 0.4374 -155.9265 47.5723 3 1 7.1699 27.7542 0.9937 -73.4995 87.8394 2 13.4105 33.4728 0.9774 -83.8806 110.7015 4 -40.7666 30.7468 0.5626 -130.1341 48.6009 4 1 47.9365 29.5861 0.3994 -38.0574 133.9305 2 54.1771 35.0067 0.4374 -47.5723 155.9265 3 40.7666 30.7468 0.5626 -48.6009 130.1341 R2 1 2 55.9847 38.7022 0.4930 -56.5059 168.4754 3 18.7542 31.6002 0.9324 -73.0940 110.6024 4 23.9025 38.7022 0.9248 -88.5881 136.3932 2 1 -55.9847 38.7022 0.4930 -168.4754 56.5059 3 -37.2306 38.7022 0.7726 -149.7212 75.2601 4 -32.0822 44.6895 0.8883 -161.9752 97.8108 3 1 -18.7542 31.6002 0.9324 -110.6024 73.0940 2 37.2306 38.7022 0.7726 -75.2601 149.7212 4 5.1484 38.7022 0.9991 -107.3423 117.6390 4 1 -23.9025 38.7022 0.9248 -136.3932 88.5881 2 32.0822 44.6895 0.8883 -97.8108 161.9752 3 -5.1484 38.7022 0.9991 -117.6390 107.3423 C1 1 2 40.7222 48.4528 0.8344 -100.1090 181.5534 3 -14.4775 40.1749 0.9833 -131.2486 102.2936 4 -20.5413 44.5067 0.9662 -149.9030 108.8205 2 1 -40.7222 48.4528 0.8344 -181.5534 100.1090 3 -55.1997 46.9142 0.6506 -191.5589 81.1595 4 -61.2635 50.6731 0.6315 -208.5483 86.0213 3 1 14.4775 40.1749 0.9833 -102.2936 131.2486 2 55.1997 46.9142 0.6506 -81.1595 191.5589 4 -6.0638 42.8266 0.9989 -130.5421 118.4146 4 1 20.5413 44.5067 0.9662 -108.8205 149.9030 2 61.2635 50.6731 0.6315 -86.0213 208.5483 3 6.0638 42.8266 0.9989 -118.4146 130.5421 C2 1 2 69.7556 51.8048 0.5548 -86.1535 225.6647 3 36.8474 45.5007 0.8486 -100.0891 173.7839 4 -66.0022 51.8048 0.5964 -221.9113 89.9069 2 1 -69.7556 51.8048 0.5548 -225.6647 86.1535 3 -32.9082 49.5349 0.9083 -181.9858 116.1694 4 -135.7578 55.3817 0.1241 -302.4316 30.9160 3 1 -36.8474 45.5007 0.8486 -173.7839 100.0891 2 32.9082 49.5349 0.9083 -116.1694 181.9858 4 -102.8496 49.5349 0.2199 -251.9272 46.2280 4 1 66.0022 51.8048 0.5964 -89.9069 221.9113 2 135.7578 55.3817 0.1241 -30.9160 302.4316 3 102.8496 49.5349 0.2199 -46.2280 251.9272

PAGE 99

86 APPENDIX C CORRELATIONS Table C-1. Correlation between weekly ABTP and histol ogical scores. R1Ed R2Ed C1Ed C2Ed R1Lym R2Lym C1Lym C2Lym WK7 Pearson 0.5748 0.0016 0.5309 0.2071 0.2221 0.3288 0.0819 0.1467 Sig. 0.1055 0.9967 0.1414 0.5929 0.5656 0.3876 0.8342 0.7064 N 9 9 9 9 9 9 9 9 WK8 Pearson 0.3911 0.0343 0.5000 0.3129 0.2786 0.4649 0.1237 0.1789 Sig. 0.2980 0.9301 0.1705 0.4124 0.4678 0.2073 0.7513 0.6451 N 9 9 9 9 9 9 9 9 WK9 Pearson 0.1835 0.3763 0.4941 0.0226 0.1318 0.3527 0.0580 0.3731 Sig. 0.6365 0.3182 0.1764 0.9540 0.7354 0.3519 0.8821 0.3227 N 9 9 9 9 9 9 9 9 WK10 Pearson 0.1127 0.0591 0.2116 0.0964 0.1248 0.7156 0.2305 0.0367 Sig. 0.7566 0.8712 0.5573 0.7911 0.7312 0.0200 0.5217 0.9198 N 10 10 10 10 10 10 10 10 WK11 Pearson 0.0179 0.0062 0.3151 0.0490 0.2922 0.5014 0.4092 0.1950 Sig. 0.9609 0.9865 0.3751 0.8930 0.4126 0.1398 0.2403 0.5892 N 10 10 10 10 10 10 10 10 WK12 Pearson 0.2052 0.1763 0.1177 0.1953 0.1483 0.6532 0.4057 0.4032 Sig. 0.5696 0.6262 0.7461 0.5888 0.6827 0.0405 0.2448 0.2480 N 10 10 10 10 10 10 10 10 WK13 Pearson 0.5357 0.3173 0.0761 0.3515 0.2669 0.2148 0.2030 0.3455 Sig. 0.1105 0.3717 0.8346 0.3193 0.4560 0.5513 0.5738 0.3281 N 10 10 10 10 10 10 10 10 WK14 Pearson 0.2509 0.1209 0.0467 0.1958 0.5130 0.2266 0.2653 0.0735 Sig. 0.4844 0.7394 0.8982 0.5877 0.1294 0.5289 0.4588 0.8402 N 10 10 10 10 10 10 10 10

PAGE 100

87 Table C-2. Correlation between endoscopic histological scores. ENDO6 ENDO9 ENDO14 R1ED Pearson -0.2507 -0.0909 -0.6325 Sig. 0.3006 0.7286 0.1778 N 19 17 6 R2ED Pearson 0.0119 -0.1300 -0.7071 Sig. 0.9613 0.6189 0.1161 N 19 17 6 C1ED Pearson 0.0576 0.1240 0.3162 Sig. 0.8147 0.6354 0.5415 N 19 17 6 C2ED Pearson -0.0819 -0.0731 0.0000 Sig. 0.7389 0.7803 1.0000 N 19 17 6 R1LYM Pearson -0.0310 -0.0473 -0.3162 Sig. 0.8997 0.8571 0.5415 N 19 17 6 R2LYM Pearson 0.0946 0.0711 -0.9258 Sig. 0.7002 0.7864 0.0080 N 19 17 6 C1LYM Pearson -0.1570 -0.0574 -0.5000 Sig. 0.5208 0.8267 0.3125 N 19 17 6 C2LYM Pearson 0.0908 0.1718 -0.6124 Sig. 0.7118 0.5096 0.1963 N 19 17 6

PAGE 101

88 Table C-3. Correlation between week ly ABTP and endoscopic scores. ENDO6 ENDO9 ENDO14 WK7ALT Pearson -0.3344 -0.2093 0.6171 Sig. 0.1617 0.4201 0.1918 N 19 17 6 WK8ALT Pearson -0.3366 -0.1315 0.4796 Sig. 0.1588 0.6148 0.3358 N 19 17 6 WK9ALT Pearson -0.0883 0.1023 0.5377 Sig. 0.7193 0.6961 0.2712 N 19 17 6 WK10ALT Pearson -0.1579 -0.2787 0.8015 Sig. 0.6630 0.4355 0.0552 N 10 10 6 WK11ALT Pearson -0.1981 -0.3012 0.3543 Sig. 0.5833 0.3977 0.4908 N 10 10 6 WK12ALT Pearson -0.2275 -0.3777 0.7039 Sig. 0.5274 0.2820 0.1185 N 10 10 6 WK13ALT Pearson -0.6594 -0.6798 0.0982 Sig. 0.0381 0.0306 0.8532 N 10 10 6 WK14ALT Pearson -0.4252 -0.4966 0.2728 Sig. 0.2206 0.1443 0.6010 N 10 10 6

PAGE 102

89 Table C-4. Correlation between weekly ABTP and ventral horn SP-immunoreactive neurons. L1 L2 L6 L7 WK7 Pearson -0.5814 -0.0524 0.1507 0.2590 Sig. 0.1006 0.8935 0.6987 0.5009 N 9 9 9 9 WK8 Pearson -0.6872 0.1131 0.4095 0.0036 Sig. 0.0408 0.7720 0.2738 0.9927 N 9 9 9 9 WK9 Pearson -0.5938 -0.3538 0.3460 0.4836 Sig. 0.0919 0.3503 0.3617 0.1872 N 9 9 9 9 WK10 Pearson -0.2473 -0.4879 -0.2496 -0.1101 Sig. 0.4909 0.1525 0.4867 0.7781 N 10 10 10 9 WK11 Pearson -0.0777 -0.6905 -0.4310 -0.5102 Sig. 0.8311 0.0271 0.2136 0.1605 N 10 10 10 9 WK12 Pearson -0.4260 -0.7116 -0.5183 -0.4932 Sig. 0.2196 0.0210 0.1248 0.1773 N 10 10 10 9 WK13 Pearson -0.4349 -0.6392 -0.6996 -0.6881 Sig. 0.2091 0.0466 0.0243 0.0405 N 10 10 10 9 WK14 Pearson -0.1321 -0.7111 -0.6323 -0.5071 Sig. 0.7161 0.0211 0.0498 0.1636 N 10 10 10 9

PAGE 103

90 Table C-5. Correlation between weekly ABTP and dorsal horn SP-Immunoreactivity (EU/pixel). L1 L2 L6 L7 WK7 Pearson -0.5990 -0.1063 0.4221 0.6345 Sig. 0.0883 0.7854 0.2578 0.0664 N 9 9 9 9 WK8 Pearson -0.4369 0.0404 0.6071 0.2733 Sig. 0.2396 0.9177 0.0829 0.4767 N 9 9 9 9 WK9 Pearson -0.7536 -0.4110 0.2366 0.4278 Sig. 0.0190 0.2718 0.5400 0.2507 N 9 9 9 9 WK10 Pearson -0.2153 -0.1344 -0.0697 -0.5936 Sig. 0.5502 0.7112 0.8483 0.0920 N 10 10 10 9 WK11 Pearson -0.3204 -0.1908 -0.3231 -0.9589 Sig. 0.3667 0.5974 0.3625 0.0000 N 10 10 10 9 WK12 Pearson -0.3894 -0.3100 -0.3341 -0.6829 Sig. 0.2660 0.3834 0.3455 0.0426 N 10 10 10 9 WK13 Pearson -0.5911 -0.3393 -0.5613 -0.3126 Sig. 0.0719 0.3375 0.0914 0.4128 N 10 10 10 9 WK14 Pearson -0.4596 -0.1198 -0.2605 -0.7598 Sig. 0.1815 0.7417 0.4672 0.0175 N 10 10 10 9

PAGE 104

91 Table C-6. Correlation between histological scores and ventral horn SP-Immunoreactive neurons. L1 L2 L6 L7 R1ED Pearson -0.5195 -0.2684 -0.3009 -0.0790 Sig. 0.0226 0.2666 0.2106 0.7555 N 19 19 19 18 R2ED Pearson -0.4454 -0.3154 -0.3382 -0.3710 Sig. 0.0560 0.1885 0.1567 0.1296 N 19 19 19 18 C1ED Pearson -0.1880 0.0805 0.0469 0.0042 Sig. 0.4409 0.7432 0.8488 0.9868 N 19 19 19 18 C2ED Pearson -0.3785 -0.1591 -0.2429 -0.0721 Sig. 0.1101 0.5154 0.3163 0.7762 N 19 19 19 18 R1LYM Pearson -0.1895 -0.5465 -0.4168 -0.4302 Sig. 0.4372 0.0155 0.0759 0.0748 N 19 19 19 18 R2LYM Pearson 0.4050 0.3388 0.0349 -0.1559 Sig. 0.0854 0.1559 0.8873 0.5367 N 19 19 19 18 C1LYM Pearson -0.0044 0.0048 0.1901 -0.1278 Sig. 0.9856 0.9846 0.4356 0.6133 N 19 19 19 18 C2LYM Pearson 0.1984 0.0774 0.1477 -0.2713 Sig. 0.4155 0.7529 0.5461 0.2761 N 19 19 19 18

PAGE 105

92 Table C-7. Correlation between histologica l scores and dorsal horn SP-immunoreactivity (EU/pixel). L1 L2 L6 L7 R1ED Pearson -0.2147 0.0891 -0.0186 0.4321 Sig. 0.3773 0.7168 0.9398 0.0734 N 19 19 19 18 R2ED Pearson -0.3134 -0.1273 -0.1942 0.2624 Sig. 0.1913 0.6035 0.4256 0.2929 N 19 19 19 18 C1ED Pearson -0.0161 0.1654 0.3380 0.2725 Sig. 0.9479 0.4985 0.1569 0.2740 N 19 19 19 18 C2ED Pearson -0.1937 0.0451 -0.1003 0.2545 Sig. 0.4269 0.8544 0.6829 0.3081 N 19 19 19 18 R1LYM Pearson -0.3227 -0.0766 -0.4582 -0.2263 Sig. 0.1778 0.7554 0.0485 0.3666 N 19 19 19 18 R2LYM Pearson 0.2614 0.3601 -0.1268 0.1989 Sig. 0.2797 0.1299 0.6049 0.4289 N 19 19 19 18 C1LYM Pearson -0.0575 -0.2722 -0.0193 0.3424 Sig. 0.8152 0.2596 0.9375 0.1643 N 19 19 19 18 C2LYM Pearson 0.1942 0.3416 -0.1088 0.0140 Sig. 0.4257 0.1523 0.6574 0.9561 N 19 19 19 18

PAGE 106

93 Table C-8. Correlation between histologi cal scores and gastrointestinal SPimmunoreactivity (EU/pixel). R1 R2 C1 C2 R1ED Pearson 0.2629 0.2498 -0.0916 0.1728 Sig. 0.2919 0.3174 0.7179 0.5380 N 18 18 18 15 R2ED Pearson 0.2812 0.2003 -0.0275 0.2655 Sig. 0.2584 0.4254 0.9137 0.3389 N 18 18 18 15 C1ED Pearson 0.3411 0.3058 -0.0544 0.0361 Sig. 0.1660 0.2172 0.8302 0.8985 N 18 18 18 15 C2ED Pearson 0.5171 0.4543 0.0105 0.0215 Sig. 0.0280 0.0582 0.9669 0.9394 N 18 18 18 15 R1LYM Pearson 0.4296 0.0837 -0.0916 -0.3919 Sig. 0.0752 0.7412 0.7177 0.1485 N 18 18 18 15 R2LYM Pearson 0.3943 0.3397 0.5347 0.4025 Sig. 0.1054 0.1678 0.0222 0.1370 N 18 18 18 15 C1LYM Pearson 0.7106 0.1241 0.2472 0.2224 Sig. 0.0010 0.6236 0.3227 0.4257 N 18 18 18 15 C2LYM Pearson 0.3322 -0.1034 0.0914 0.1333 Sig. 0.1780 0.6832 0.7185 0.6359 N 18 18 18 15

PAGE 107

94 Table C-9. Correlation between weekly ABTP and gastrointestin al SP-immunoreactivity (EU/pixel). R1 R2 C1 C2 WK7 Pearson -0.5907 -0.2644 -0.5939 -0.5692 Sig. 0.0939 0.4918 0.1206 0.1823 N 9 9 8 7 WK8 Pearson -0.5229 -0.2624 -0.6841 -0.3253 Sig. 0.1486 0.4951 0.0613 0.4765 N 9 9 8 7 WK9 Pearson -0.2196 -0.4166 -0.4460 -0.6606 Sig. 0.5702 0.2647 0.2680 0.1062 N 9 9 8 7 WK10 Pearson -0.2325 -0.5033 -0.5639 -0.4152 Sig. 0.5472 0.1672 0.0895 0.3064 N 9 9 10 8 WK11 Pearson -0.4332 -0.1295 -0.2594 -0.6313 Sig. 0.2442 0.7398 0.4693 0.0932 N 9 9 10 8 WK12 Pearson -0.3518 -0.1994 -0.5293 -0.4983 Sig. 0.3531 0.6071 0.1156 0.2088 N 9 9 10 8 WK13 Pearson -0.1143 0.1269 -0.3005 -0.2102 Sig. 0.7697 0.7449 0.3989 0.6173 N 9 9 10 8 WK14 Pearson -0.1242 -0.0231 -0.2996 -0.5600 Sig. 0.7502 0.9529 0.4003 0.1489 N 9 9 10 8 Table C-10. Correlation between dorsal horn SP-immunoreactivity (EU/pixel) and gastrointestinal SP-immunoreactivity (EU/pixel). R1 R2 C1 C2 L1 Pearson 0.0226 0.3396 0.4028 0.4250 Sig. 0.9291 0.1679 0.0975 0.1143 N 18 18 18 15 L2 Pearson -0.0317 0.3874 0.1117 0.3825 Sig. 0.9007 0.1122 0.6591 0.1594 N 18 18 18 15 L6 Pearson -0.1789 0.1665 -0.0111 0.1565 Sig. 0.4776 0.5089 0.9650 0.5774 N 18 18 18 15 L7 Pearson 0.4127 -0.0619 0.2964 0.5232 Sig. 0.0997 0.8133 0.2480 0.0549 N 17 17 17 14

PAGE 108

95 LIST OF REFERENCES Accarino A.M., Azpiroz F., Malagelada J.R. (1995) Selective dysfunction of mechanosensitive intestinal affere nts in irritable bowel syndrome. Gastroenterology 108(3):636-643. Al Chaer E.D., Kawasaki M., Pasricha P.J. (2000) A new model of chronic visceral hypersensitivity in adult rats induced by colon irritation during postnatal development. Gastroenterology 119(5):1276-1285. Anand A.C., Adya C.M. (1999) Cytokine s and inflammatory bowel disease. Trop. Gastroenterol. 20(3):97-106. Balemba O.B., Hay-Schmidt A., Assey R.J ., Kahwa C.K.B., Semuguruka W.D., Dantzer V. (2002) An immunohistoche mical study of the organiza tion of ganglia and nerve fibres in the mucosa of the porcine intestine. Anatomia Histologia EmbryologiaJournal of Veterinary Medicine Series C 31(4):237-246. Balemba O.B., Semuguruka W.D., Hay-Schmid t A., Johansen M.V.,Dantzer V. (2001) Vasoactive intestinal peptide and substan ce P-like immunoreactivit ies in the enteric nervous system of the pig correlate with the severity of pathological changes induced by Schistosoma japonicum. International Journal for Parasitology 31(13):1503-1514. Banerjee A.K., Peters T.J. (1990) Experime ntal non-steroidal an ti-inflammatory druginduced enteropathy in the rat: similarities to inflammatory bowel disease and effect of thromboxane synthetase inhibitors. Gut 31(12):1358-1364. Beagley K. W., Black C.A., Elson C.O. (1991) Strain differences in susceptibility to TNBS-induced colitis. Gastroenterology 100, A560. Bernstein C.N., Niazi N., Robert M., Mert z H., Kodner A., Munakata J., Naliboff B., Mayer E.A. (1996) Rectal afferent func tion in patients with inflammatory and functional intestinal disorders. Pain 66(2-3):151-161. Bertrand C., Geppetti P., Baker J., Yamawaki I., Nadel J.A. (1993) Role of neurogenic inflammation in antigen-induced vascular extravasation in guinea pig trachea. J. Immunol. 150(4):1479-1485. Birch P.J., Harrison S.M., Hayes A.G., Roge rs H., Tyers M.B. (1992) The non-peptide NK1 receptor antagonist, (+/-)-CP-96,345, produces antinociceptive and antioedema effects in the rat. Br. J. Pharmacol. 105(3):508-510.

PAGE 109

96 Blumberg H., Haupt P., Janig W., Kohler W. (1983) Encoding of visceral noxious stimuli in the discharge patterns of viscer al afferent fibres from the colon. Pflugers Arch. 398(1):33-40. Bradesi S., Eutamene H., Garcia-Villar R ., Fioramonti J., Bueno L. (2002) Acute and chronic stress differently affect visceral sensitivity to rectal distension in female rats. Neurogastroenterol. Motil. 14(1):75-82. Breeling J.L., Onderdonk A.B., Ci sneros R.L., Kasper D.L. (1988) Bacteroides vulgatus outer membrane antigens asso ciated with carageenanj-i nduced colitis in guinea pigs. Infect. Immun. 56(7):1754-1759. Bueno L., Fioramonti J., Delvaux M., Frexinos J. (1997) Mediators and pharmacology of visceral sensitivity: from basic to clinical investigations. Gastroenterology 112(5):1714-1743. Bueno L., Fioramonti J., Garcia-Villar R. (2000) Pathobiology of visceral pain: molecular mechanisms and therapeutic implications. II I. Visceral afferent pathways: a source of new therapeutic targets for abdominal pain. Am. J. Physiol Gastrointest. Liver Physiol 278(5):G670-G676. Camilleri M. (2001) Management of the irritable bowel syndrome. Gastroenterology 120(3):652-668. Cervero F., Janig W. (1992) Visceral nociceptors: a new world order? Trends Neurosci. 15(10):374-378. Chang L., Mayer E.A., Johnson T., FitzGerald L.Z., Naliboff B. (2000a) Differences in somatic perception in female patients w ith irritable bowel syndrome with and without fibromyalgia. Pain 84(2-3):297-307. Chang L., Munakata J., Mayer E.A., Schmulson M.J., Johnson T.D., Bernstein C.N., Saba L., Naliboff B., Anton P.A., Matin K. (2000b) Perceptual responses in patients with inflammatory and functional bowel disease. Gut 47(4):497-505. Charlton C.G., Helke C.J. (1985a) Autoradiogr aphic localization and characterization of spinal cord substance P binding sites: hi gh densities in sensory, autonomic, phrenic, and Onuf's motor nuclei. J. Neurosci. 5(6):1653-1661. Charlton C.G., Helke C.J. (1985b) Characte rization and segmental distribution of 125IBolton-Hunter labeled substance P binding sites in rat spinal cord. J. Neurosci. 5(5):1293-1299. Cohen R.H., Perl E.R. (1990) Contributions of arachidonic acid derivatives and substance P to the sensitization of cutaneous nociceptors. J. Neurophysiol. 64(2):457-464.

PAGE 110

97 Cominelli F., Nast C.C., Clark B.D., Schindler R., Lierena R., Eysselein V.E., Thompson R.C., Dinarello C.A. (1990) Interleukin 1 (IL-1) gene expression, synthesis, and effect of specific IL-1 receptor blocka de in rabbit immune complex colitis. J. Clin. Invest 86(3):972-980. Cook I.J., van Eeden A., Collins S.M. (1987) Pa tients with irritable bowel syndrome have greater pain tolerance than normal subjects. Gastroenterology 93(4):727-733. Cooper H.S., Murthy S.N., Shah R.S., Sederg ran D.J. (1993) Clinicopathologic study of dextran sulfate sodium expe rimental murine colitis. Lab Invest 69(2):238-249. Coutinho S.V., Plotsky P.M., Sablad M., Mi ller J.C., Zhou H., Bayati A.I., McRoberts J.A., Mayer E.A. (2002) Neonatal matern al separation alte rs stress-induced responses to viscerosomatic nociceptive stimuli in rat. Am J. Physiol Gastrointest. Liver Physiol 282(2):G307-G316. D'Argenio G., Cosenza V., Riegler G., Della V.N., Deritis F., Mazzacca G. (2001) Serum transglutaminase correlates with endoscopi c and histopathologic grading in patients with ulcerative colitis. Dig. Dis. Sci. 46(3):649-657. Del Bianco E., Santicioli P., Tramontana M ., Maggi C.A., Cecconi R., Geppetti P. (1991) Different pathways by which extracellular Ca2+ promotes calcitonin generelated peptide release from central terminals of capsaicin-sensitive afferents of guinea pigs: effect of capsaicin, high K+ and low pH media. Brain Res. 566(1-2):46-53. Dougherty P.M., Palecek J., Paleckova V ., Willis W.D. (1994) Neurokinin 1 and 2 antagonists attenuate the responses and NK1 antagonists prevent the sensitization of primate spinothalamic tract neurons after intradermal capsaicin. J. Neurophysiol. 72(4):1464-1475. Drossman D.A., Creed F.H., Olden K.W., Svedlund J., Toner B.B., Whitehead W.E. (1999) Psychosocial aspects of the func tional gastrointes tinal disorders. Gut 45 Suppl 2(II25-II30. Duggan A.W. (1995) Release of neur opeptides in the spinal cord. Prog. Brain Res. 104(197-223. Duggan A.W., Schaible H.G., Hope P.J., Lang C.W. (1992) Effect of peptidase inhibition on the pattern of intraspinally released immunoreactive substance P detected with antibody microprobes. Brain Res 579(2):261-269. Elmer G.I., Pieper J.O., Negus S.S., Woods J.H. (1998) Genetic va riance in nociception and its relationship to the potency of mo rphine-induced analgesia in thermal and chemical tests. Pain 75(1):129-140. Elson C.O., Sartor R.B., Tennyson G.S., Ridde ll R.H. (1995) Experimental models of inflammatory bowel disease. Gastroenterology 109(4):1344-1367.

PAGE 111

98 Fedorak R. N., Madsen K.L. (2000). Naturall y occurring and experimental models of inflammatory bowel disease. In: J. B. Ki rsner (Ed.) Inflammatory bowel disease. pp. 113-143. W.B. Saunders, Philadelphia, PA. Fiocchi C. (1998) Inflammatory bowel disease: Etiology and pathogenesis. Gastroenterology 115(1):182-205. Gear R.W., Miaskowski C., Gordon N.C., Paul S.M., Heller P.H., Levine J.D. (1996) Kappa-opioids produce significantly greater analgesia in women than in men. Nat. Med 2(11):1248-1250. Gebhart G.F., Randich A. (1992) Vagal modulation of nociception. Am. Pain. Soc. J. 1(26-32. Goyal R.K., Hirano I. (1996) The enteric nervous system. N. Engl. J. Med 334(17):11061115. Grundy D. (1988) Speculations on the sturct ure/function relationship for vagal and splanchnic afferent endings supplyi ng the gastrointestinal tract. J. Auton. Nerv. Syst. 22(3):175-180. Gschossmann J.M., Coutinho S.V., Miller J.C., Huebel K., Naliboff B., Wong H.C., Walsh J.H., Mayer E.A. (2001) Involvemen t of spinal calcitonin gene-related peptide in the development of acute visceral hyperalgesia in the rat. Neurogastroenterol. Motil. 13(3):229-236. Gwee K.A., Graham J.C., McKendrick M.W., Co llins S.M., Marshall J.S., Walters S.J., Read N.W. (1996) Psychometric scores a nd persistence of ir ritable bowel after infectious diarrhoea. Lancet 347(8995):150-153. Habler H.J., Janig W., Koltzenbu rg M. (1990) Activation of un myelinated afferent fibres by mechanical stimuli and inflammation of the urinary bladder in the cat. J. Physiol 425(545-562. Haley J. E., Wilcox G.L.. (1992). Involvement of excitatory amino acids and peptides in the spinal mechanisms underlying hype ralgesia. In: W. D. Willis (Ed.) Hyperalgesia and allodynia. pp. 281-293. Raven, New York. Hammer R.E., Maika S.D., Richardson J.A., Tang J.P., Taurog J.D. (1990) Spontaneous inflammatory disease in transgenic rats expressing HLA-B27 and human beta 2m: an animal model of HLA-B 27-associated disorders. Cell 63(5):1099-1112. Handwerker H.O., Reeh P.W. (2003). Nocicepto rs: chemosensitivity and sensitization by chemical agents. In: W. D. Willis (Ed.) Hyperalgesia and allodynia. pp. 107-116. Raven, New York. Heitkemper M.M., Jarrett M. (1992) Pattern of gastrointestinal and somatic symptoms across the menstrual cycle. Gastroenterology 102(2):505-513.

PAGE 112

99 Helke C.J., Charlton C.G., Wiley R.G. (1986) Studies on the cellu lar localization of spinal cord substance P receptors. Neuroscience 19(2):523-533. Hertz A.F. (1911) The sensibility of the al imentary tract in health and disease. Lancet 1(1051-1056. Hokfelt T., Elfvin L.G., Schultzberg M., Goldstein M., Nilsson G. (1977a) On the occurrence of substance P-containing fibers in sympathetic ganglia: immunohistochemical evidence. Brain Res 132(1):29-41. Hokfelt T., Ljungdahl A., Terenius L., Elde R., Nilsson G. (1977b) Immunohistochemical analysis of peptide pathways possibly rela ted to pain and analgesia: enkephalin and substance P. Proc. Natl. Acad. Sci. U. S. A 74(7):3081-3085. Hoyle C.H. (1998) Neuropeptide fam ilies: evolutionary perspectives. Regul. Pept. 73(1):1-33. Hurst S., Collins S.M. (1993) Interleukin-1 beta modulation of norepinephrine release from rat myenteric nerves. Am J. Physiol Gastrointest. Liver Physiol 264(1 Pt 1):G30-G35. Janig W., Koltzenburg M. (1990) On the func tion of spinal primary afferent fibres supplying colon and urinary bladder. J. Auton. Nerv. Syst. 30 Suppl(S89-S96. Janig W., Koltzenburg M. (1991) Receptive prop erties of sacral primary afferent neurons supplying the colon. J. Neurophysiol. 65(5):1067-1077. Janig W., Morrison J.F.B. (1986). Functional properties of spinal visceral afferents supplying abdominal and pelvic organs with special emphasis on visceral nociception. In: F. Cervero and J. F. B. Morrison (Eds.) Visceral sensation: Progress in brain research. pp. 87-114. Elsevier, New York. Kawata M., Hirakawa M., Kumamoto K., Minamino N., Kangawa K., Matsuo H., Sano Y. (1989) Brain natriuretic peptide in the porcine spinal cord: An immunohistochemical investigation of its localization and the comparison with atrial natriuretic peptide, substance P, calcitonin gene-rel ated peptide, and enkephalin. Neuroscience 33(2):401-410. Kingham J.G., Dawson A.M. (1985) Origin of chronic right upper quadrant pain. Gut 26(8):783-788. Kishimoto S. (1994) High concentrations of substance P as a possible transmission of abdominal pain in rats with chem ical induced ulcerative colitis. Biomed. Res. 15(133):140. Krause J.E., DiMaggio D.A., McCarson K.E. ( 1995) Alterations in neurokinin 1 receptor gene expression in models of pain and inflammation. Can. J. Physiol Pharmacol. 73(7):854-859.

PAGE 113

100 Kruschewski M., Foitzik T., Perez-Canto A., Hubotter A., Buhr H.J. (2001) Changes of colonic mucosal microcirculation and hi stology in two co litis models: an experimental study using intravital mi croscopy and a new histological scoring system. Dig. Dis. Sci. 46(11):2336-2343. Kuhn R., Lohler J., Rennick D., Rajewsky K., Muller W. (1993) Inte rleukin-10-deficient mice develop chronic enterocolitis. Cell 75(2):263-274. Kulkarni-Narla A., Beitz A.J., Brown D.R. (1999) Catecholaminergic, cholinergic and peptidergic innervation of gutassociated lymphoid tissue in porcine jejunum and ileum. Cell Tissue Res 298(2):275-286. Kundig T.M., Schorle H., Bachmann M.F., He ngartner H., Zinkernagel R.M., Horak I. (1993) Immune responses in in terleukin-2-deficient mice. Science 262(5136):10591061. Le Greves P., Nyberg F., Terenius L., Hokf elt T. (1985) Calcitonin gene-related peptide is a potent inhibitor of substance P degradation. Eur. J. Pharmacol. 115(2-3):309311. Lecci A., Giuliani S., Patacchini R., Viti G ., Maggi C.A. (1991) Role of NK1 tachykinin receptors in thermonociception: effect of (+/-)CP 96,345, a non-peptide substance P antagonist, on the hot plate test in mice. Neurosci. Lett. 129(2):299-302. Lembo T., Munakata J., Mertz H., Niazi N ., Kodner A., Nikas V., Mayer E.A. (1994) Evidence for the hypersensitivity of lumbar splanchnic afferents in irritable bowel syndrome. Gastroenterology 107(6):1686-1696. Lembo T., Naliboff B., Munakata J., Fulle rton S., Saba L., Tung S., Schmulson M., Mayer E.A. (1999) Symptoms and visceral perception in patients with painpredominant irritable bowel syndrome. Am. J. Gastroenterol. 94(5):1320-1326. Levine J.D., Lau W., Kwiat G., Go etzl E.J. (1984) Leukotriene B4 produces hyperalgesia that is dependent on polymorphonuclear leukocytes. Science 225(4663):743-745. Levine J. D., Taiwo Y.O., Heller P.H. ( 1992). Hyperalgesic pain: inflammatory and neuropathic. In: W. D. Willis (Ed.) Hyperalgesia and allodynia. pp. 117-124. Raven Press, New York. Lubrano E., Iovino P., Tremolaterra F., Parsons W.J., Ciacci C., Mazzacca G. (2001) Fibromyalgia in patients with irritable bowel syndrome. An association with the severity of the intestinal disorder. Int. J. Colorectal Dis. 16(4):211-215. Lundberg J.M., Franco-Cereceda A., Alving K., Delay-Goyet P., Lou Y.P. (1992) Release of calcitonin gene-relate d peptide from sensory neurons. Ann. N. Y. Acad. Sci. 657(187-193.

PAGE 114

101 MacDonald T.T., Monteleone G., Pender S. L. (2000) Recent developments in the immunology of inflammatory bowel disease. Scand. J. Immunol. 51(1):2-9. MacPherson B.R., Pfeiffer C.J. (1978) Experi mental production of diffuse colitis in rats. Digestion 17(2):135-150. Marks H.E., Hobbs S.H. (1972) Changes in st imulus reactivity following gonadectomy in male and female rats of different ages. Physiol Behav. 8(6):1113-1119. Matkowskyj K.A., Schonfeld D., Benya R.V. (2000) Quantitative immunohistochemistry by measuring cumulative signa l strength using commercia lly available software photoshop and matlab. J. Histochem. Cytochem. 48(2):303-312. Mayer E.A., Gebhart G.F. (1994) Basic and c linical aspects of vi sceral hyperalgesia. Gastroenterology 107(1):271-293. Mayer E.A., Naliboff B., Lee O., Munakata J ., Chang L. (1999) Review article: genderrelated differences in functiona l gastrointestinal disorders. Aliment. Pharmacol. Ther. 13 Suppl 2(65-69. Mayer E.A., Raybould H.E. (1990) Role of visc eral afferent mechanisms in functional bowel disorders. Gastroenterology 99(6):1688-1704. Mayer M.L., Westbrook G.L., Guthrie P.B. ( 1984) Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 309(5965):261-263. McKendrick M.W., Read N.W. (1994) Ir ritable bowel syndrome--post salmonella infection. J. Infect. 29(1):1-3. McMahon S., Koltzenburg M. (1990) The changi ng role of primary afferent neurons in pain. Pain 43(3):269-272. Merritt A.M., Buergelt C.D.,Sanchez L.C. (2002a) Porcine ileitis model induced by TNBS-ethanol instillation. Dig. Dis. Sci. 47(4):879-885. Merritt A.M., Xie H., Lester G.D., Burro w J.A., Lorenzo-Figueras M., Mahfoud Z. (2002b) Evaluation of a method to experime ntally induce colic in horses and the effects of acupuncture applied at the Gu an-yuan-shu (similar to BL-21) acupoint. Am J. Vet Res 63(7):1006-1011. Mertz H., Naliboff B., Munakata J., Niazi N., Mayer E.A. (1995) Altered rectal perception is a biological marker of patients with irritable bowel syndrome. Gastroenterology 109(1):40-52. Messaoudi M., Desor D., Grasmuck V., J oyeux M., Langlois A., Roman F.J. (1999) Behavioral evaluation of visceral pain in a rat model of colonic inflammation. Neuroreport 10(5):1137-1141.

PAGE 115

102 Miampamba M., Chery-Croze S., Chayvialle J.A. (1992) Spinal and intestinal levels of substance P, calcitonin gene-related pep tide and vasoactive inte stinal polypeptide following perendoscopic injection of formalin in rat colonic wall. Neuropeptides 22(2):73-80. Miampamba M., Sharkey K.A. (1998) Distribu tion of calcitonin gene -related peptide, somatostatin, substance P and vasoactive intestinal polypeptide in experimental colitis in rats. Neurogastroenterol. Motil. 10(4):315-329. Miller M.J., Sadowska-Krowicka H., Jeng A.Y., Chotinaruemol S., Wong M., Clark D.A., Ho W., Sharkey K.A. (1993) Substan ce P levels in experimental ileitis in guinea pigs: effects of misoprostol. Am. J. Physiol 265(2 Pt 1):G321-G330. Moore B.A., Stewart T.M., Hill C., Vanne r S.J. (2002) TNBS ileitis evokes hyperexcitability and change s in ionic membrane properties of nociceptive DRG neurons. Am J. Physiol Gastrointest. Liver Physiol 282(6):G1045-G1051. Morris G.P., Beck P.L., Herridge M.S., Depew W.T., Szewczuk M.R., Wallace J.L. (1989) Hapten-induced model of chronic inflammation and ulceration in the rat colon. Gastroenterology 96(3):795-803. Ness T.J., Gebhart G.F. (1988) Colorectal di stension as a noxious visceral stimulus: physiologic and pharmacologic characterizati on of pseudaffective reflexes in the rat. Brain Res. 450(1-2):153-169. Ness T.J., Gebhart G.F. (1990) Visceral pa in: A review of experimental studies. Pain 41(2):167-234. Ness T.J., Gebhart G.F. (2000) Acute inflamma tion differentially alte rs the activity of two classes of rat spinal visceral nociceptive neurons. Neurosci. Lett. 281(2-3):131134. Ness T. J., Gebhart G.F. (2001) Methods in vi sceral pain researc h. In: L. Kruger (Ed.) Methods in pain research. pp. 93-108. CRC Press, Boca Raton. Ness T.J., Randich A., Gebhart G.F. (1991) Fu rther behavioral evidence that colorectal distension is a 'noxious' vis ceral stimulus in rats. Neurosci. Lett. 131(1):113-116. Neurath M., Fuss I., Strober W. (2000) TNBS-colitis. Int. Rev. Immunol. 19(1):51-62. Onderdonk A.B., Bronson R., Cisner os R.L. (1987) Comparison of Bacteroides vulgatus strains in teh enhancement of experimental ulcerative colitis. Infect. Immun. 55(3):835-836. Otsuka M., Yoshioka K. (1993) Neurotrans mitter functions of mammalian tachykinins. Physiol Rev 73(2):229-308.

PAGE 116

103 Pare W.P. (1969) Age, sex, and strain differen ces in the aversive threshold to grid shock in the rat. J. Comp Physiol Psychol. 69(2):214-218. Persson S., Le Greves P., Thornwall M., Er iksson U., Silberring J., Nyberg F. (1995) Neuropeptide converting and processing enzymes in the spinal cord and cerebrospinal fluid. Prog. Brain Res. 104(111-130. Plourde V., St Pierre S., Quirion R. ( 1997) Calcitonin gene-related peptide in viscerosensitive response to colo rectal distension in rats. Am. J. Physiol 273(1 Pt 1):G191-G196. Purcell W.M., Atterwill C.K. (1995) Ma st cells in neur oimmune function: neurotoxicological and neuropha rmacological perspectives. Neurochem. Res 20(5):521-532. Reichert J.A., Daughters R.S., Rivard R., Si mone D.A. (2001) Peripheral and preemptive opioid antinociception in a m ouse visceral pain model. Pain 89(2-3):221-227. Riddell R.H. 2000. Pathology of Idiopathic In flammatory Bowel Disease. In: J. B. Kirsner (Ed.) Inflammatory Bowe l Disease. pp. 427-447. W.B. Saunders, Philadelphia, PA. Robert A., Asano T. (1977) Resistance of germ free rats to indomethacin-incuced intestinal lesions. Prostaglandins 14(2):333-341. Romero M.T., Bodnar R.J. (1986) Gender differe nces in two forms of cold-water swim analgesia. Physiol Behav. 37(6):893-897. Romero M.T., Cooper M.L., Komisaruk B.R ., Bodnar R.J. (1988) Gender-specific and gonadectomy-specific effects upon swim anal gesia: role of st eroid replacement therapy. Physiol Behav. 44(2):257-265. Routh V.H., Helke C.J. (1995). Tachykinin recepto rs in the spinal cord. In: F. Nyberg, H. S. Sharma, Wisenfeld-Hallin, Z. (Eds.) Pr ogress in brain research: Neuropeptides in the spinal cord. pp. 93-108. Elsevier, Amsterdam. Ruda M.A., Ling Q.D., Hohmann A.G., Pe ng Y.B., Tachibana T. (2000) Altered nociceptive neuronal circuits afte r neonatal peripheral inflammation. Science 289(5479):628-631. Ryan P., Bennett M.W., Aarons S., Lee G., Collins J.K., O'Sullivan G.C., O'Connell J., Shanahan F. (2002) PCR detection of My cobacterium paratuberculosis in Crohn's disease granulomas isolated by laser capture microdissection. Gut 51(5):665-670. Sabate J.M., Gorbatchef C., Flourie B., Ji an R., Coffin B. (2002) Cholecystokinin octapeptide increases rectal sensitiv ity to pain in healthy subjects. Neurogastroenterol. Motil. 14(6):689-695.

PAGE 117

104 Sadlack B., Merz H., Schorle H., Schimpl A., Feller A.C.,Horak I. (1993) Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 75(2):253-261. Sandler R.S. (1990) Epidemiology of irrita ble bowel syndrome in the United States. Gastroenterology 99(2):409-415. Santos J., Perdue M.H. (2000) Stress a nd neuroimmune regulation of gut mucosal function. Gut 47 Suppl 4(iv49-iv51. Sartor R.B., Bender D.E., Holt, L.C. (1992) Susceptibility of inbred rat strains to intestinal inflammation induced by indomethacin. Gastroenterology 102, A690. Schaible H.G., Hope P.J., Lang C.W.,D uggan A.W. (1992) Calcitonin gene-related peptide causes intraspinal spreading of substance P released by peripheral stimulation. Eur. J. Neurosci. 4(8):750-757. Schaible H.G., Schmidt R.F. (1988) Excitation a nd sensitization of fine articular afferents from cat's knee joint by prostaglandin E2. J. Physiol 403(91-104. Schneider J., Jehle E.C., Star linger M.J., Neunlist M., Mich el K., Hoppe S., Schemann M. (2001) Neurotransmitter coding of enteric neurones in the submucous plexus is changed in non-inflamed rectum of patients with Crohn's disease. Neurogastroenterol. Motil. 13(3):255-264. Seder R.A., Marth T., Sieve M.C., Strober W ., Letterio J.J., Roberts A.B., Kelsall B. (1998) Factors involved in the differentia tion of TGF-beta-producing cells from naive CD4+ T cells: IL-4 and IFN-gamma have opposing effects, while TGF-beta positively regulates its own production. J. Immunol. 160(12):5719-5728. Sengupta, J.N., Gebhart, G.F. 1994. Gastrointestinal afferent fi bers and sensation. In: L. R. Johnson (Ed.) Physiology of the gast rointestinal tract. Raven, New York. Shafran I., Piromalli C., Decker J.W., Sandoval J., Naser S.A., El Zaatari F.A. (2002) Seroreactivities against Saccharomyces cerevisiae and Mycobacterium avium subsp. paratuberculosis p35 and p36 an tigens in Crohn's disease patients. Dig. Dis. Sci. 47(9):2079-2081. Sharkey K.A., Kroese A.B. (2001) Conseque nces of intestinal inflammation on the enteric nervous system: ne uronal activation induced by inflammatory mediators. Anat. Rec. 262(1):79-90. Sharon P., Stenson W.P. (1985) Metabolism of ar achidonic acid in acetic acid colitis in rats. Similarity to human inflammatory bowel disease. Gastroenterology 88(1 Pt 1):55-63. Shibata Y., Taruishi M., Ashida T. (1993) Expe rimental ileitis in dogs and colitis in rats with trinitrobenzene sulf onic acid--colonoscopic and histopathologic studies. Gastroenterol. Jpn. 28(4):518-527.

PAGE 118

105 Soderholm J.D., Perdue M.H. (2001) Stress and gastrointestinal tract. II. Stress and intestinal barrier function. Am J. Physiol Gastrointest. Liver Physiol 280(1):G7G13. Stead R.H., Kosecka-Janiszewska U., Oestre icher A.B., Dixon M.F., Bienenstock J. (1991) Remodeling of B-50 (GAP-43)and NSE-immunoreactive mucosal nerves in the intestines of rats infected with Nippostrongylus brasiliensis. J. Neurosci. 11(12):3809-3821. Steers W.D., Ciambotti J., Erdman S., de Groa t W.C. (1990) Morphological plasticity in efferent pathways to the urinary bladde r of the rat following urethral obstruction. J. Neurosci. 10(6):1943-1951. Steers W.D., Ciambotti J., Etzel B., Erdman S., de Groat W.C. (1991) Alterations in afferent pathways from the urinary bladder of the rat in response to partial urethral obstruction. J. Comp Neurol. 310(3):401-410. Sternini C. (1992) Enteric and visceral affe rent CGRP neurons. Targets of innervation and differential expression patterns. Ann. N. Y. Acad. Sci. 657(170-186. Swain M.G., Agro A., Blennerhassett P., Stan isz A., Collins S.M. (1992) Increased levels of substance P in the myenteric plex us of Trichinellainfected rats. Gastroenterology 102(6):1913-1919. Thompson W.G., Longstreth G.F., Drossman D.A., Heaton K.W., Irvine E.J., MullerLissner S.A. (1999) Functional bowel di sorders and functional abdominal pain. Gut 45 Suppl 2(II43-II47. Torres M.I., Garcia-Martin M ., Fernandez M.I., Nieto N., Gil A., Rios A. (1999) Experimental colitis induced by trinitroben zenesulfonic acid: an ultrastructural and histochemical study. Dig. Dis. Sci. 44(12):2523-2529. Tracey D.J., Walker J.S. (1995) Pain due to nerve damage: are inflammatory mediators involved? Inflamm. Res 44(10):407-411. Traub R.J., Hutchcroft K., Gebhart G.F. (1999) The peptide content of colonic afferents decreases following colonic inflammation. Peptides 20(2):267-279. Traub R.J., Pechman P., Iadarola M.J., Gebha rt G.F. (1992) Fos-lik e proteins in the lumbosacral spinal cord following noxious and non-noxious colorectal distention in the rat. Pain 49(3):393-403. Van Ginkel R., Voskuijl W.P., Benninga M.A., Taminiau J.A., Boeckxstaens G.E. (2001) Alterations in rectal sensitivity and motil ity in childhood irritable bowel syndrome. Gastroenterology 120(1):31-38.

PAGE 119

106 Veale D., Kavanagh G., Fielding J.F., Fitzgera ld O. (1991) Primary fibromyalgia and the irritable bowel syndrome: different expres sions of a common pathogenetic process. Br. J. Rheumatol. 30(3):220-222. Verne G.N., Price D.D. (2002) Irritable bow el syndrome as a common precipitant of central sensitization. Curr. Rheumatol. Rep. 4(4):322-328. Verne G.N., Robinson M.E., Price D.D. ( 2001) Hypersensitivity to visceral and cutaneous pain in the irritable bowel syndrome. Pain 93(1):7-14. Whitehead W.E., Holtkotter B., Enck P., Hoel zl R., Holmes K.D., Anthony J., Shabsin H.S., Schuster M.M. (1990) Tolerance fo r rectosigmoid distention in irritable bowel syndrome. Gastroenterology 98(5 Pt 1):1187-1192. Whitehead W.E., Winget C., Fedoravicius A. S., Wooley S., Blackwell B. (1982) Learned illness behavior in patients with irritable bowel syndrome and peptic ulcer. Dig. Dis. Sci. 27(3):202-208. Whorwell P.J., Lupton E.W., Erduran D., Wilson K. (1986) Bladder smooth muscle dysfunction in patients with irritable bowel syndrome. Gut 27(9):1014-1017. Wiesenfeld-Hallin Z., Hokfelt T., Lundbe rg J.M., Forssmann W.G., Reinecke M., Tschopp F.A., Fischer J.A. (1984) Immunor eactive calcitonin gene-related peptide and substance P coexist in sensory neurons to the spinal cord and interact in spinal behavioral responses of the rat. Neurosci. Lett. 52(1-2):199-204. Willis W. D. (1992) Hyperalgesia and all odynia: The Bristol-Myers Squibb Symposium on Pain Research. Raven Press, New York. Willis W.D. (1993) Mechanical allodynia: A ro le for sensitized nociceptive tract cells with convergent input from m echanoreceptors and nociceptors? Am. Pain. Soc. J. 2:23-33. Willis W. D., Coggeshall R.E. (1991). Sensory mechanisms of the spinal cord. Plenum, New York. Wirtz S., Neurath M.F. (2000) Animal models of intestinal inflammation: new insights into the molecular pathogenesis and immunotherapy of inflammatory bowel disease. Int. J. Colorectal Dis. 15(3):144-160. Woolf C., Wiesenfeld-Hallin Z. (1986) Subs tance P and calcitonin gene-related peptide synergistically modulate the gain of the noci ceptive flexor withdrawal reflex in the rat. Neurosci. Lett. 66(2):226-230. Woolf C. J. (1992). Excitability changes in cen tral neurons following peripheral damage: Role of central sensitization in the pat hogenesis of pain. In: W. D. Willis (Ed.) Hyperalgesia and allodynia. pp. 221-243. Raven Press, New York.

PAGE 120

107 Yamada T., Deitch E., Specian R.D., Perry M.A., Sartor R.B., Grisham M.B. (1993) Mecahnisms of acute and chronic in testinal inflammation induced by indomethacin. Inflammation 17(6):641-662. Yamada Y., Marshall S., Specian R.D., Grisha m M.B. (1992) A comparative analysis of two models of colitis in rats. Gastroenterology 102(5):1524-1534. Yamamoto T., Yaksh T.L. (1991) Stereospecif ic effects of a nonpeptidic NK1 selective antagonist, CP96,345: antinociception in the absence of motor dysfunction. Life Sci. 49(26):1955-1963. Zipser R.D., Nast C.C., Lee M., Kao H.W ., Duke R. (1987) In vivo production of leukotriene B4 and leukotriene C4 in rabbi t colitis. Relationship to inflammation. Gastroenterology 92(1):33-39.

PAGE 121

108 BIOGRAPHICAL SKETCH Linda Christine (Chris) Sanchez was born in Clearwater, Florida in 1970. After attending public schools in Pinellas County, she enrolled at the Univer sity of Florida in 1988. Throughout her undergraduate studies, Chris played tenor saxophone in the University of Florida marching and jazz bands She earned her Doctor of Veterinary Medicine degree from the University of Flor ida in 1995, and completed in internship in Equine Medicine and Surgery at a private clinic, Equine Medical Associates in Edmond, Oklahoma in 1996. Chris returned to the UF College of Vete rinary Medicine for a residency program in Large Animal Internal Medicine and was awarded Diplomate status in the American College of Veterinary Internal Medicine, speci alty of Internal Medicine in 1999. She was awarded an Alumni Fellowship to pursue her graduate education in the Island Whirl Equine Colic Research Laboratory follo wing the completion of her residency. Chris has been appointed to the Faculty of the UF College of Veterinary Medicine and will join the Large Animal Medicine service of the Department of Large Animal Clinical Sciences following completion of this degree. Her outside interests include running, cycling, scuba di ving, and snow skiing.


Permanent Link: http://ufdc.ufl.edu/UFE0000892/00001

Material Information

Title: Porcine Model of Inflammatory-Mediated Visceral Hypersensitivity
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000892:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000892/00001

Material Information

Title: Porcine Model of Inflammatory-Mediated Visceral Hypersensitivity
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000892:00001


This item has the following downloads:


Full Text












PORCINE MODEL OF INFLAMMATORY-MEDIATED VISCERAL
HYPERSENSITIVITY















By

LINDA CHRISTINE SANCHEZ


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Linda Christine Sanchez
































This dissertation is dedicated to the many animals who have contributed to my education.















ACKNOWLEDGMENTS

As with most major endeavors, many people helped to make this research and

dissertation possible. Despite many direct and indirect contributions, these are but a few

thanks. My fellow graduate students in the Island Whirl Equine Colic Research

Laboratory, Drs. Tamara Widenhouse and Mireia Lorenzo, were always available for

various pig-related duties, and I am grateful for their help and support. Jim Burrow,

Hilken Kuck, and Drs. Chao-Yong Bai and A.M. Merritt provided truly invaluable

assistance as blinded observers for CRD procedures. Debi Malcolm and Alex Trapp took

outstanding care of my research pigs, and Debi helped each new pig adapt to the system.

Tom DeHaan and Joe Bryant provided technical assistance surrounding all pig

necropsies. Sally O'Connell is an invaluable resource to all VMTH graduate students,

and certainly has made my life considerably easier!!

The Island Whirl Equine Colic Research Laboratory provided funding for the

animal work, and Dr. Nicholas Verne provided laboratory space, supplies, and funding

for the initial pilot animals and all tissue processing and immunohistochemical analyses.

I could not have asked for a better mentor than Dr. Al Merritt. He has continually

been a source of enthusiasm, even when things (like my entire first project) were clearly

not progressing as planned. Al has a tremendous ability to offer advice without intrusion,

and I will always be grateful for his insight and friendship.

I would also like to thank the remaining members of my committee, Drs. Guy

Lester, Nick Verne, Elizabeth Uhl, and Ed Ott. Guy has been an outstanding sounding









board and friend, and I am truly grateful that he was willing to fly back to Florida for this

defense. Nick has provided continuous insight into the relevance of IBS in human

gastroenterology while also offering the use of his laboratory for all of the tissue

processing and staining.

My family and friends, while providing enormous amounts of grief regarding my

never-ending education, have also provided tremendous support. As I said for my

mentor, I could also not have asked for better parents. They have always been supportive

while also serving as great role models. I truly could not have done this without them.
















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .............................. ........... ... .. ...... .............. viii

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT .............. ..................... .......... .............. xii

CHAPTER

1 LITERATURE REVIEW .............. ....................... .. ........................... 1

Visceral Hypersensitivity............. ..................... .
Visceral Afferent Innervation...................................................... 2
M editors of Visceral Hypersensitivity.......................... ............. ..............4
The R ole of Inflam m action ..................................................................... ......... 8
Animal Models of Visceral Pain .......................................................... 10
Animal M odels of Inflammatory Bowel Disease .....................................................10
Inflammatory Bowel Disease in Humans............... ...................... ..................11
Chemical M odels..................... ............................. .. ......12
M house K nock-out M odels ........................................................ ............. 16
The Pig as a M odel .................................... ..... .......... ......... .... 17
S tu d y O bje ctiv e s ................................................................................................... 17

2 M E T H O D S .......................................................................................................1 8

IA CU C A approval ................................................. ...... ................. 18
M odel D evelopm ent ......................................................... .. ............ 18
M ain Study Design .................................... .. .. ..... .. ............20
Procedures ...... ...... ... ....................................................... 22
Tissue Collection, Processing, and Analysis ................................... .................26
Statistical A nalysis................................................... 34

3 RESULTS ANIM AL STUDIES......................................... ......................... 37

D evelopm ent of colitis .................. ............................. ........ .. ............ 37
E ndoscopic E v alu action ....................................................................... ..................37
V isceral Sensitivity ........... ...... ........... ......................... ............... 39
C correlation s ........................................................ ............................. 4 1











4 RESULTS TISSUE ANALYSIS.......................................................... ...............42

H istological A naly sis........... ...... ...................................................... .. .... .. .. 42
Im m unohistochem ical A nalysis............................................ ........... ............... 45
S p in a l C o rd .................................................................. ................................4 5
G gastrointestinal T ract ................................................ .............................. 48
C orrelations ...................................................................................................... ..... 49

5 D ISC U S SIO N ............................................................................... 56

M odel D evelopm ent ............. ............ .......... ................................ .... .. .......... 56
Effect of Inflammation on Nociceptive Threshold................. .............................60
C o n clu sio n s..................................................... ................ 6 9

APPENDIX

A INDIVIDUAL ANIMAL DATA ................................ .......................... 72

B A N O V A T A B L E S ...................................................................... .... ......................74

C C O R R E L A T IO N S ............................................................................. .... ........... 86

L IST O F R E FE R E N C E S ............................................................................. .............. 95

BIOGRAPHICAL SKETCH ............................................................. ............... 108
















LIST OF TABLES

Table p

2-1 A nim al grouping .............................................. .. ........... .... .. .....21

2-2 Tim eline of study events ................................................ .............................. 21

2-3 Endoscopic lesion scoring......................................................... ............... 23

2-4 Histologic scoring system for gastrointestinal tissues................... ................28

4-1 M ean lymphoid aggregate scores. ........................................ ......................... 44

4-2 M ean edem a scores. ....................... ............................ .. ......... ................45

4-3 Mean dorsal horn Substance P-immunoreactivity ............................................. 47

4-4 Mean ventral horn Substance P-immunoreactive neurons.................. ............48

4-5 Quantitative IHC data for gastrointestinal tissues.................................................49

A-i Raw data from CRD studies ........................................................ ............... 72

A-2 Raw data from endoscopic examinations ............... .....................................73

B-l One-way ANOVA analysis for threshold pressure. ............................................74

B-2 One-way ANOVA analysis for ABTP. .........................................................75

B-3 One-way ANOVA analysis for endoscopy scores. .............................................76

B-4 One-way ANOVA analysis for gastrointestinal histologic scores.........................77

B-5 Tukey's HSD analysis for gastrointestinal histologic scores................................78

B-6 One-way ANOVA for ventral horn SP-immunoreactive neurons ..........................81

B-7 Tukey's HSD analysis for ventral horn SP-immunoreactive neurons. ....................82

B-8 One-way ANOVA for SP-immunoreactivity ..................................... .................83

B-9 Tukey's HSD analysis for spinal SP-Immunoreactivity .......................................84









B-10 Tukey's HSD for gastrointestinal SP-immunoreactivity ......................................85

C-1 Correlation between weekly ABTP and histological scores .............. ...............86

C-2 Correlation between endoscopic histological scores.................................... 87

C-3 Correlation between weekly ABTP and endoscopic scores ...............................88

C-4 Correlation between weekly ABTP and ventral horn SP-immunoreactive neurons.89

C-5 Correlation between weekly ABTP and dorsal horn SP-immunoreactivity ..........90

C-6 Correlation between histological scores and ventral horn SP-immunoreactive
neuron s. .............................................................................9 1

C-7 Correlation between histological scores and dorsal horn SP-immunoreactivity ......92

C-8 Correlation between histological scores and gastrointestinal SP-immunoreactivity93

C-9 Correlation between weekly ABTP and gastrointestinal SP-immunoreactivity. .....94

C-10 Correlation between dorsal horn SP-immunoreactivity and gastrointestinal SP-
im m unoreactivity. ............................................... .. .... ...... .. ........... 94
















LIST OF FIGURES

Figure page

2-1 Pig in crate used for all procedures ........................................ ....... ............... 19

2-2 Polyethylene rectal distention balloon attached to catheter. ...................................24

2-3 Close-up of rectal catheter attached to pig's tail. ................................................24

2-4 Typical discom fort response. ............................................................................. 26

2-5 V central horn neurons ........................................................................... 30

2-6 Im age selection for spinal sections.................................... .................................... 32

2-7 Pixel square selections for cord im age 1....................................... ............... 33

2-8 Pixel square selections for cord image 2. ...................................... ...............33

2-9 Pixel square selections for colonic and rectal myenteric plexus............................34

3-1 Endoscopy scores .......................... ....... .. .. .. ...... .. ............38

3-2 N orm al endoscopy (Grade 0) ............................................................................38

3-3 Abnorm al endoscopy (Grade 3) ........................................ .......................... 39

3-4 M ean threshold pressures .............. ..... ............................ ............... 40

3 -5 M e an A B T P ........................................................................................................ 4 0

3-6 Correlation between endoscopy scores on weeks 6 and 9 and week 13 ABTP.......41

4-1 E x am ple of edem a grade 1 ............................................................ .....................42

4-2 Example of edema grade 2 ...................................... ......... ................... 43

4-3 Exam ple of edem a grade 3 ............................................... ............................ 43

4-4 Example of edema grade 4 ...................................... ......... ................... 44

4-5 SP-IR in spinal cord ................................................ ...............46









4-6 SP-IR in porcine colon ............................................ .................................... 49

4-7 Correlation between lymphoid aggregates in the rectal section R2 and ABTP.......50

4-8 Correlation between histological scores in the rectal section R1 and ventral horn
SP-immnuoreactive neurons in spinal sections L1 and L2 ....................................51

4-9 Dorsal horn SP-IR correlation with ABTP ................................... ............... 52

4-10 Ventral horn correlation with ABTP for the L2 spinal segment...........................53

4-11 Ventral horn correlation with ABTP for the L6 spinal segment...........................54

4-12 Ventral horn correlation with ABTP for the L7 spinal segment...........................54

4-13 SP-IR correlation with histological scores ....... ..............................................55















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PORCINE MODEL OF INFLAMMATORY-MEDIATED VISCERAL
HYPERSENSITIVITY

By

Linda Christine Sanchez

August 2003

Chair: A.M. Merritt
Major Department: Veterinary Medicine

Irritable bowel syndrome (IBS) is characterized by chronic abdominal pain

associated with diarrhea and/or constipation. Visceral hypersensitivity is a biological

marker for IBS, but the pathophysiology of this hypersensitivity is unknown. Transient

inflammation has been suggested as an inciting agent, and Substance P (SP) is an

important neuropeptide in pain processing. Using the pig as an animal model, the major

objectives were to 1) develop a model of subacute proctitis; 2) validate a model of

visceral discomfort; 3) evaluate the effect of colorectal inflammation on visceral

nociceptive threshold; 4) evaluate the effect of colorectal inflammation on SP-

immunoreactivity (SP-IR) in the colon, rectum, and lumbar spinal cord.

Nineteen cross-bred male castrated young pigs (initial bodyweight 20-30 kg) were

used. Colitis was reliably induced (n = 12) by trinitrobenzene sulfonic acid/ethanol

(TNBS/EtOH) enema, and resolved within 5 weeks, as documented endoscopically.

Control animals (n = 7) received a saline enema and had no endoscopic abnormalities.









Half of the animals in each group (6 and 3, respectively) were euthanized 5 weeks

following enema administration, and the remaining animals were euthanized 5 weeks

later. SP-IR in the spinal cord, colon, and rectum was determined via

immunohistochemical staining. Colorectal distention (CRD) consisted of sequential one-

minute barostat-controlled inflations, interspersed by five-minute deflations, starting at

15 mmHg and increasing in 10mmHg increments until a discomfort response was

induced (considered threshold), up to a maximum of 55 mmHg. Mean baseline threshold

was 37.1 mmHg and did not differ significantly between TNBS/EtOH and saline groups.

However, at weeks 3 and 9 post-enema, the TNBS/EtOH-treated animals had a

significantly (p = 0.045 and 0.005 respectively) lower CRD threshold, relative to

baseline, than the saline-treated animals. SP-IR did not differ statistically between groups

in any tissues. But, SP-IR in the dorsal horn of spinal segments L1 and L7 had a

significant correlation with visceral sensitivity. This study demonstrates the potential

usefulness of the pig as a large animal model for visceral nociception studies. The

correlation between spinal SP-IR and visceral sensitivity reinforces a relationship

between CNS upregulation of SP and visceral hypersensitivity.














CHAPTER 1
LITERATURE REVIEW

Many factors regulate the sensory function of the gastrointestinal tract. As a result,

when alterations of this function occur, a specific etiology can be difficult to ascertain.

The experiments described in this dissertation revolve around using the pig as a model for

visceral sensitivity studies. This information could potentially be applicable to other

species, including humans, which typically suffer from numerous clinical problems with

associated gastrointestinal pain.

Visceral Hypersensitivity

Hypersensitivity commonly refers to the development of either hyperalgesia, a

significant upregulation of the magnitude of the response to a given peripheral painful

stimulus, and/or allodynia, a nociceptive perception of a normally nonpainful stimulus

(Willis, 1992). Visceral hypersensitivity (VH) refers to such a response within the

abdominal viscera. The Irritable Bowel Syndrome (IBS) is a common functional

gastrointestinal disorder with a wide range of clinical presentations, including alterations

in bowel habits and enhanced visceral sensitivity (Thompson et al., 1999). VH is a

common biologic marker of the Irritable Bowel Syndrome (IBS) and is present in almost

all IBS patients (Mertz et al., 1995).

Some IBS patients have also been shown to have altered somatic referral patterns in

response to colorectal distention (CRD), which indicates altered spinal processing of

visceral sensory information (Kingham and Dawson, 1985; Lembo et al., 1994). Some

studies have shown cutaneous allodynia in IBS patients (Verne et al., 2001) whereas









others have shown no alteration in somatic pain tolerance compared to healthy controls

(Accarino et al., 1995; Cook et al., 1987; Whitehead et al., 1990). Many IBS patients

also report symptoms of extraintestinal functional disorders such as irritable bladder,

chronic somatic pain, and sleep disturbances (Whitehead et al., 1982; Whorwell et al.,

1986). The crossover of IBS with fibromyalgia, a chronic somatic pain disorder, appears

particularly significant (Chang et al., 2000a; Veale et al., 1991; Verne and Price, 2002),

and may correlate with the severity of IBS symptoms (Lubrano et al., 2001). Patients

with IBS and fibromyalgia were found to have somatic hyperalgesia, whereas patients

with IBS alone had somatic hypoalgesia with higher pain threshold and lower pain

frequency and severity compared to controls (Chang et al., 2000a). Such alterations may

indicate a state of central (with or without a peripheral component) hyperexcitability

within the central nervous system (CNS). Hyperexcitability of the CNS has been

predominantly characterized through models of cutaneous nociception, but similar

mechanisms are now known to play a role in VH as well.

Visceral Afferent Innervation

Visceral spinal afferent nerves function similarly to their somatic counterparts;

however they allow only limited localization of an offending stimulus.(Hertz, 1911)

Vagal afferents also provide important modulation of nociception (Gebhart and Randich,

1992; Grundy, 1988), and sacral parasympathetic fibers mediate sensory information

from the distal colon and rectum (Janig and Morrison, 1986). The receptive fields of

upper GI and colonic spinal afferent C fibers normally occur predominantly in the

muscularis, serosa, and mesentery, but increase in size and include the mucosa during

inflammation (McMahon and Koltzenburg, 1990; Ness and Gebhart, 1990). In the









rectum, sacral A6 fibers have mucosal receptor fields under non-inflamed conditions

(Janig and Koltzenburg, 1991; Sengupta and Gebhart, 1994).

The current concept of visceral pain perception involves a combination of high-

threshold nociceptors as well as low-threshold mechanoreceptors (Cervero and Janig,

1992; Willis, 1993). In the colon, both C and A6 fibers can encode a wide range of

stimulus intensity, but C fibers are thought to be primarily slowly-adapting

mechanoreceptors, whereas A6 fibers are predominantly rapidly-adapting

mechanoreceptors (Blumberg et al., 1983; Habler et al., 1990; Janig and Koltzenburg,

1991). Additional "silent nociceptors", mechanically insensitive C fibers innervating

areas such as the bladder and colon, only respond to distending stimuli after chemical

irritation in experimental situations (Janig and Koltzenburg, 1990; Janig and

Koltzenburg, 1991). Inputs from these neurons converge on "wide-dynamic range"

dorsal horn neurons (Willis and Coggeshall, 1991). Under normal conditions, low-

intensity stimuli activate low-threshold afferents, which do not trigger the nociceptive

pathway. Transient high intensity stimuli not only increase the intensity of the low-

threshold afferents, but also recruit the high-threshold afferents, resulting in nociception

(Cervero and Janig, 1992). Inflammation can alter these pathways through persistent

stimulation of the peripheral terminals or activation of the "silent nociceptors" (Mayer

and Gebhart, 1994).

Spinal and vagal afferents differ significantly in the localization of neuropeptides,

in that 85-95% of spinal afferents to the stomach and colon, but only 5% of gastric vagal

afferents, contain calcitonin-gene related peptide (CGRP) (Mayer and Raybould, 1990).

In the GI tract, Substance P (SP) is found predominantly in the muscular layer and









myenteric plexus, coinciding with the previously stated receptor fields for spinal afferents

(Otsuka and Yoshioka, 1993). CGRP is present in numerous splancnic afferents.

SP is a neuropeptide in the tachykinin family, along with neurokinins A and B

(NKA and NKB). Of the three tachykinin receptors, NK1 has the greatest affinity for SP

which, in turn, preferentially binds to that receptor (Routh and Helke, 1995). In the

spinal cord dorsal horn, NK1 is heavily concentrated within Laminae I and II, with

decreased density in Laminae III and IV, and very little in Lamina V(Charlton and Helke,

1985a; Helke et al., 1986). In the ventral horn, motor neurons throughout the spinal cord

contain low to moderate levels of NK1 binding(Charlton and Helke, 1985b; Charlton and

Helke, 1985a). Antibody microprobe studies document the release of SP following

noxious peripheral stimuli primarily within the superficial dorsal horn (Laminae I and II),

but extending beyond (Duggan et al., 1992; Schaible et al., 1992). CGRP and SP co-exist

in many primary afferent neurons and may be co-expressed when these nerves are

stimulated (Bueno et al., 2000; Hokfelt et al., 1977b; Wiesenfeld-Hallin et al., 1984).

CGRP also inhibits SP endopeptidase (SPE) activity, thus potentiating the biological

actions of SP (Le Greves et al., 1985; Woolf and Wiesenfeld-Hallin, 1986).

CGRP and SP clearly play a role in the transmission of nociceptive information

within the CNS (Duggan, 1995), but their function in peripheral terminals is less clear.

They likely function as neuromodulators or mediators of local tissue responses, which

could allow for changes in motility and visceral sensation at a peripheral level.

Mediators of Visceral Hypersensitivity

Inflammation is thought to affect the primary afferents through the action of

inflammatory mediators on the peripheral terminals of primary afferents (Cohen and Perl,

1990; Handwerker and Reeh, 2003; Levine et al., 1992; Schaible and Schmidt, 1988) and









the matrix in which peripheral terminals are imbedded (Janig and Morrison, 1986; Mayer

and Raybould, 1990). Specifically, prostaglandin E2 (PGE2), prostaglandin 12 (PGI2),

ATP, bradykinin, serotonin, CGRP, SP, and glutamate can directly mediate sensitivity at

the primary afferent (Purcell and Atterwill, 1995; Tracey and Walker, 1995). Many

interleukins (IL-1, IL-8, IL-6), tumor necrosis factor-a (TNF-a), nerve growth factor

(NGF), bradykinin, leukotriene B4 (LTB4), SP, complement 5a (C5a), and vasoactive

intestinal peptide (VIP) can act indirectly via activation of immunocytes. Hyperalgesia

caused by LTB4 is neutrophil-dependent and extremely potent (Levine et al., 1984;

Levine et al., 1992). Noradrenaline, neuropeptide Y (NPY), and NGF act through

mediators released from adrenergic nerve varicosities (Bueno et al., 1997; Bueno et al.,

2000). While the alteration in responsiveness at the peripheral terminal likely subsides

with resolution of the initial inflammatory insult, experimentally documented "memory"

persists for hours (Willis, 1993).

Altered afferent input to the dorsal horn results in central sensitization, and the

associated release of neuropeptides in the dorsal horn increases the excitability of spinal

neurons and leads to an expansion of the receptive fields (Mayer and Gebhart, 1994).

The release of CGRP and SP from both central and peripheral terminals of primary

afferent neurons in response to a nociceptive stimulus is well recognized, and these

peptides are the two most abundant in medium and small dorsal root ganglia (DRG)

neurons (Del Bianco et al., 1991; Hokfelt et al., 1977b; Lundberg et al., 1992; Sternini,

1992). The in-vitro application of the SP-antagonist, CP 96345, can prevent capsaicin-

mediated sensitization of spinothalamic tract cells (Dougherty et al., 1994). This agent

has also been shown to produce mild analgesic effects in thermo- and chemo-sensitive









models and to prevent noxious responses to carageenan injection in the rat paw and

thermal skin stimulation in cats (Birch et al., 1992; Lecci et al., 1991; Yamamoto and

Yaksh, 1991). Increased spinal cord expression ofNK1 and NK2 receptors and SP

occurs in experimental models of arthritis (Krause et al., 1995). These findings indicate a

role for the neurokinins in central hyperexcitability (Bueno et al., 1997; Bueno et al.,

2000).

These local effects can include activation of immune cells which will stimulate the

local inflammatory response, leading to central and/or peripheral hyperexcitability. As

previously described, mast cells appear to play a major role in the sensitization of

primary afferents, and the release of SP is crucial (Bueno et al., 1997). Essentially, a

feedback loop is created whereby SP release can trigger mast cell degranulation, and the

subsequent histamine release causes further release of SP as well as NGF (Bertrand et al.,

1993; Purcell and Atterwill, 1995). The resultant CGRP/SP-immunoreactive network

modulates both reflex motor activity and the transmission of sensory information to the

CNS.

At a spinal level, SP likely contributes to dorsal horn hyperexcitability via a direct

action on the postsynaptic cells or by potentiating the excitatory effects of glutamate

(Haley and Wilcox, 1992). Recent studies showing a reduction in nociceptive response

to CRD in rats by intrathecal or intravenous administration of the CGRP receptor

antagonist, hCGRP8-37, provide direct evidence for the role of CGRP in visceral

nociception (Gschossmann et al., 2001; Plourde et al., 1997). Intravenous injection of

CGRP results in abdominal cramping and decreased gastric emptying similar to that

induced by peritoneal irritation, and the effects are blocked by NK1 antagonists. The









release of SP and CGRP may also result in neurogenic inflammation, although the roles

of these peptides in gastrointestinal inflammation are not yet fully understood (Sharkey

and Kroese, 2001). In a rat TNBS-colitis model, SP-immunoreactivity was decreased

initially throughout the colon, followed by an increase in the circular muscle at 7 days

(Miampamba and Sharkey, 1998). A similar pattern was shown in the enteric and primary

afferent nerves after intraluminal injection of TNBS in the guinea pig ileum (Miller et al.,

1993). Zymosan-induced colitis reduced the number of SP double-labeled (with

Fluorogold, to document afferent innervation from the colon) DRG in both the T13-L2

and L6-S2 DRG cells (Traub et al., 1999). Further work suggests that the alteration in

substance P in response to inflammation may be regulated through 11-10 (Hurst and

Collins, 1993).

Indirect effects of neurotransmitters are also thought to result in enhanced receptor

sensitivity. Activation of CGRP, SP, and other non-N-methyl D-aspartate (NMDA)

receptors on the post-synaptic terminal of the central terminal can result in increasing

depolarization of the postsynaptic membrane, leading to activation of the NMDA

receptor by removal of the Mg" block (Mayer et al., 1984; Woolf, 1992). This allows

for a further increase in intracellular Ca+, activation of NOS, and availability of NO as

an intracellular messenger and diffusible neurotransmitter (Mayer and Gebhart, 1994).

Increased intracellular calcium accumulated by NMDA receptor activation may also lead

to neuronal cell death, or "excitotoxicity," which may play a role in disinhibition of

second-order sensory neurons, associated with the clinical phenomenon of "windup"

(Bueno et al., 1997; Mayer and Gebhart, 1994).









The Role of Inflammation

The timing of events is likely very important when considering the role of

inflammation in the development of VH. Transient colonic irritation (Al Chaer et al.,

2000) or maternal separation (Coutinho et al., 2002) during the neonatal period in rats can

lead to provoke chronic visceral hyperalgesia that persists through adulthood despite the

absence of histopathological lesions. Neuroplastic changes, including enlargement of

dorsal root ganglia cells, have been documented within weeks of chronic partial urinary

bladder obstruction and are associated with lowered thresholds for urgency to urinate as

well as discomfort (Steers et al., 1990; Steers et al., 1991). However, the associated

changes may result in alterations of motility rather than sensation. Repeated rectal

distention in humans to noxious pressures (60 mmHg) can result in alteration of reported

sensation, and repetitive CRD in rats to noxious (80 mmHg) but not innocuous (20

mmHg) pressures causes an increase in spinalfos andjun (Traub et al., 1992). These

changes are likely a result of alterations in the dorsal horn neurons due to repeated

excitability similar to that previously described for inflammation.

Clinical data to support inflammation as a modulator of VH include the fact that

IBS patients and patients with Crohn's disease in which inflammation is limited to the

small bowel have shown similar patterns of abdominal dermatome referral in response to

CRD (Bernstein et al., 1996). Patients with mild active ulcerative colitis have attenuated

rectal sensitivity responses, which correlate negatively with UC activity index, implying

that either a more severe or more chronic inflammatory insult is needed for the

development of hypersensitivity (Chang et al., 2000b). Clinically, most patients suffering

from inflammatory-mediated alterations in visceral sensitivity improve in conjunction

with resolution of the inflammatory insult, emphasizing the short-term nature of many









changes. The fact that VH associated with inflammation is not limited to the affected site

within the GI tract also stresses the role of central hypersensitization in these events.

The alteration of dorsal horn neurons such that previously subthresholdd" stimuli

can stimulate a nociceptive response is believed to play a key role in the development of

mechanical allodynia. Thus, alterations associated with inflammation cannot only cause

short-term VH, but also the alteration of afferent input to the dorsal horn which likely

promotes plasticity within the CNS resulting in central hyperexcitability. For this central

hyperexcitability to persist, however, additional cofactors, repeated events, or specific

timing of the insult such as during the neonatal period likely occurs.

Stress and other psychosocial factors are also thought to contribute to VH and/or

CNS hyperexcitability. Clinically, IBS patients often report an exacerbation of

symptoms in conjunction with stressful life events, and the length and severity of the

associated exacerbation often reflect the severity of the stressful event (Camilleri, 2001;

Drossman et al., 1999; Lembo et al., 1999; Sandler, 1990). Psychological stressors can

result in many physiologic changes including changes in muscular tone, immune

modulation, and mucosal barrier dysfunction, alterations in descending pain modulating

systems, and alterations in sleep (Mayer and Gebhart, 1994; Santos and Perdue, 2000;

Soderholm and Perdue, 2001). Perhaps the most convincing argument for the role of

social stressors in VH is the fact that psychological treatment for anxiety and depression

is often an effective tool in the treatment of functional bowel disorders (Mayer and

Gebhart, 1994).

Infectious insults to the gastrointestinal tract have also been postulated as

contributing events to VH and/or CNS hyperexcitability in the IBS, given that









approximately 33% of patients hospitalized for a bout of infectious diarrhea develop IBS

within 3-12 months (Gwee et al., 1996; McKendrick and Read, 1994). The responses to

infection are likely mediated through inflammatory changes or immune modulation

within the GI tract as a result of the initial insult and are likely to act via similar

mechanisms as inflammation alone. In acute Nippostrongylus brasiliensis infection, a

2.5-fold increase in nerve content can occur by day 10 post infection, returning to near

control values by day 14 (Stead et al., 1991).

In summary, a tremendous range of potential mediators likely contribute to the

development of VH. Genetic, environmental, and individual differences compound the

situation. In addition, other mediators have been proposed as contributing factors to the

development of VH but lie beyond the scope of this review.

Animal Models of Visceral Pain

Balloon distention of a hollow organ, most notably within the gastrointestinal tract,

is the most widely used stimulus of visceral pain experimentally (Ness and Gebhart,

1990). CRD in humans produces similar painful sensations, both in intensity, quality,

and area of somatic referral, to clinically occurring gastrointestinal-associated pain. CRD

has been validated in many animal species to produce brief discomfort with reliable,

quantifiable behavioral and physiological responses attenuated by analgesic drugs, thus

fulfilling the criteria for a valid model of visceral pain (Ness and Gebhart, 2001).

Animal Models of Inflammatory Bowel Disease

Clearly, the main thrust of this dissertation is to evaluate colitis-mediated visceral

hypersensitivity using an animal model. After deciding to further pursue the effect of

inflammation in the gastrointestinal tract on nociceptive responses, the next step was to

choose an inflammatory model. Numerous naturally occurring and inducible models of









inflammatory bowel disease have been documented. Most involve rodents, and these are

most commonly characterized based upon the specific nature of the insult.

Inflammatory Bowel Disease in Humans

Although commonly grouped together under the umbrella of "Inflammatory Bowel

Disease" (IBD), Crohn's Disease (CD) and Ulcerative Colitis (UC) have many

differences. The pathophysiological mechanisms involved in UC and CD have not been

fully elucidated. An infectious cause has long been suspected, especially a role for

Mycobacterium paratuberculosis in the pathogenesis of CD, but efforts to identify an

etiologic agent have thus far been unsuccessful (Ryan et al., 2002; Shafran et al., 2002).

UC usually begins with ulceration of the rectal mucosa and progresses orally to include

varying portions of the bowel, potentially the entire colon. Clinical signs can begin with

constipation and quickly progress to include rectal bleeding urgency, diarrhea, and

abdominal discomfort. The course of disease is both acute and chronic, and

relapse/remission is often unpredictable. Histologically, UC is predominantly an acute

inflammatory process with disease limited to the mucosa and superficial submucosa

except in fulminant disease (Fiocchi, 1998). The clinical course of CD is much more

variable, but consists of acute and chronic inflammation of the small and large intestine

and can include extraintestinal symptoms as well. The most common site of initial

involvement is the ileocecal region. Histologically, CD has two different presentations

indicative of either acute or chronic disease. Acutely, focal aphthoid ulcerations are

noted, often in the epithelium overlying lymphoid aggregates. These ulcerations can

undergo cycles of formation and healing, and the focal ulcerations can progress to more

cobblestone-like lesions. Severe chronic CD can present with transmural inflammation,









inflammation, and fibrosis, often with granuloma formation. Often, the histological

distinction between the two forms can be difficult (Fiocchi, 1998; Riddell, 2000).

Regardless of the initiating event, the chronicity of IBD suggests some form of

immune dysregulation. T-helper lymphocytes are primarily responsible for the

maintenance of an immune steady-state, and a tremendous amount of data regarding the

association between their related cytokines and the IBD syndromes has recently become

available. In general, UC demonstrates a predominantly Th2-like (especially IL-5)

profile, the increase in cytokine production is limited to the involved mucosa, and

eicosanoid production is prominent. In CD, secretion of Thl-type cytokines (IL-12,

TNFa, IFNy,) predominates, cytokine production is increased in involved and uninvolved

mucosa, and eicosanoid production is only moderate (Anand and Adya, 1999;

MacDonald et al., 2000). In both forms of IBD, increases of proinflammatory cytokines

(IL-1, IL-6, IL-8) are evident, although data are inconsistent and TNFa appears to be

more important in CD. Although mucosal T cells are activated in both CD and UC, their

differential response to IL-2 stimulation represents one of the hallmark immunologic

differences between these two syndromes. CD mucosal T cells demonstrate a

hyperreactive response to 11-2 stimulation and have high expression of IL-2Ra gene

products (Fiocchi, 1998).

Chemical Models

Many chemicals have been used as mucosal irritants in IBD models (Elson et al.,

1995; Wirtz and Neurath, 2000). None of the animal models feature a relapsing/remitting

nature of disease characteristic of the human problem, and all chemical models have a

relatively short duration (1-8 wks) although those that cause disease for 6-8 weeks allow

for the characterization of a chronic inflammatory phase. The extent of colonic









involvement depends upon the method used for chemical instillation, with enema

administration resulting in a distal colitis in most cases. Surgical instillation allows

delivery to more orad sites within the GI tract such as the ileum. The four main

chemical-induced models of IBD include acetic acid, formalin/immune complex,

TNBS/ethanol, and indomethacin. Several polymer/microbial-induced models also exist,

including carrageenan and dextran sodium sulfate (DSS).

Acetic acid instillation into the colonic lumen produces predominantly mucosal

inflammation (rats, mice, guinea pigs, rabbits) with histological similarity to naturally

occurring UC (MacPherson and Pfeiffer, 1978; Sharon and Stenson, 1985; Yamada et al.,

1992). Depending on the concentration and volume used, the lesion can progress to

transmural depth. However, the inflammatory response remains for only days in mice

and 2-3 weeks in rats; thus the value of this model lies only in the early phases of

inflammation. The mechanism of injury is primarily related to a destruction of epithelial

cells and subsequent mucosal and submucosal inflammatory response (Elson et al.,

1995).

Trinitrobenzenesulfonic acid/ethanol (TNBS/EtOH) instillation results in acute

transmural inflammation, edema, and cryptitis with histological similarity to CD (Torres

et al., 1999). TNBS induces a delayed hypersensitivity response to skin contact by

haptenating body proteins with trinitrophenyl (TNP) groups, rendering the resultant

proteins immunogenic (Neurath et al., 2000). Ethanol works as a mucosal irritant,

allowing for easier access for the TNBS molecule. Both T and B lymphocytes and

macrophages predominate initially in an IL-12 driven Thl T-cell-mediated inflammatory

response (Neurath et al., 2000). The TNBS model has been used in numerous species,









including rats, mice, and rabbits, with a peak inflammatory response in 2-3 days with an

overall duration of 2-3 weeks in mice and up to 8 weeks in rats (Elson et al., 1995). In

contrast to acetic acid-induced disease, TNBS/EtOH-induced colitis does appear to have

an immunologic component in that susceptibility to disease differs among strains of

inbred mice and the dosage of TNBS required to induce lesions varies among species

(Beagley et al., 1991; Morris et al., 1989). By eliciting oral tolerance via the

administration of a TNBS-protein complex orally simultaneous to TNBS/ethanol enema

administration in mice, the predominant Thl response can be altered to a predominant

Th2 response (Neurath et al., 2000; Seder et al., 1998). We have recently documented a

TNBS/EtOH-induced ileitis in pigs produced by intraluminal instillation (Merritt et al.,

2002a).

An immune complex-induced colitis model involves the instillation of a dilute

formalin enema, followed by the intravenous injection of preformed immune complexes

(Cominelli et al., 1990; Zipser et al., 1987). The dosage of formalin is critical to ensure

that disease is related to the immune complexes rather than the chemical nature of

formalin alone. The resultant lesion consists of severe mucosal and submucosal

inflammation with crypt distortion, which resolves within 6-8 weeks. This model

obviously involves an immunologic component, and IL-1 appears to play an important

role (Cominelli et al., 1990). The inflammatory cytokine production in this model

mirrors that seen in both syndromes of IBD. Because the immune complexes are formed

to ubiquitous antigen, normal intestinal flora may be involved in either initiation or

perpetuation of inflammation (Elson et al., 1995; Fedorak and Madsen, 2000).









Indomethacin administration will induce enteritis, primarily in the mid-jejunum.

Rat strains vary in their response, with ulceration resolving by 14 days in Fischer rats,

lasting at least 14 days in Sprague-Dawley rats, and 77 days in inbred Lewis rats (Sartor

et al., 1992; Yamada et al., 1993). Lewis rats develop segmental transmural

inflammation throughout the distal jejunum and ileum. Granulomatous inflammation,

fibrosis, adhesions, and partial intestinal obstruction may also result; thus, the histological

characteristics of this model more closely mimic CD (Elson et al., 1995). Length of

disease is related to the dosage of indomethacin used, but inflammation in most models

persists for 1-2 weeks. Because indomethacin is a non-selective cyclooxygenase

inhibitor, protective mucosal prostaglandins are depleted. Host susceptibility, normal

intestinal flora, and bile and/or enterohepatic circulation can also play a role in this model

given the species and strain differences in susceptibility, the reduction or absence of

disease in germ-free rats (Robert and Asano, 1977), the prevention of lesion formation by

bile duct ligation (Yamada et al., 1993), and the lesion attenuation with antibiotic

administration (Banerjee and Peters, 1990; Yamada et al., 1993).

The polymer models (DSS and carrageenan) both induce predominantly mucosal

and submucosal lesions with histological similarity to UC (Elson et al., 1995). A

definitive role of luminal bacteria has been established in the carrageenan model, but the

model is not easily reproducible in species other than the guinea pig (Breeling et al.,

1988; Onderdonk et al., 1987). DSS can be administered easily in drinking water, and

results in chronic lesions (Cooper et al., 1993). The colitis is similar to UC; however

lymphoid aggregates, fissuring ulceration, and focal inflammation seen in the chronic

phases more closely resemble CD (Elson et al., 1995).









Mouse Knock-out Models

Genetic manipulation, primarily in mice, has yielded both transgenic (dominant or

dominant-negative expression of a gene product) and knockout (targeted gene deletion)

animals that develop intestinal inflammation (Elson et al., 1995; Wirtz and Neurath,

2000). In general, once lesions develop in these animals, they will persist until the

animal's death or, in some cases, replacement of the deleted molecule (cytokine, etc.), but

a relapsing/remitting course of disease has not yet been duplicated. The number of

different knockouts (KO) and transgenic animals that develop IBD-like disease supports

the notion that IBD is a multifactorial syndrome (Elson et al., 1995; Wirtz and Neurath,

2000).

One of the most important findings among the many genetically manipulated

animals was that IL-2 and IL-10 knockout mice and HLA-B27 transgenic rats do not

develop intestinal inflammation in the absence of luminal bacteria (Kuhn et al., 1993;

Kundig et al., 1993; Sadlack et al., 1993). The IL-2 KO mice surviving beyond the 10th

week of life develop continuous mucosal/submucosal colitis without small intestinal or

major internal organ involvement. Immunological abnormalities include increased

numbers of activated T and B cells, a potential Th2-like shift, increased IgG1, and anti-

colon antibodies (Kundig et al., 1993; Sadlack et al., 1993). Thus, these mice develop

histological and immunological abnormalities similar to those seen in UC. The IL-10 KO

mice develop chronic transmural duodenitis, jejunitis, and proximal colitis with an

enhanced Thl response due to the lack of IL-10 downregulation (Kuhn et al., 1993).

This model therefore more closely mimics CD, although the early lesions are

histologically more consistent with UC. Rats transgenic for human HLA-B27 and 32-

microglobulin develop multiorgan disease (including colitis, arthritis, orchitis, and a









psoriasis-like condition), but the resultant colitis lacks the acute neutrophilic component

of IBD (Hammer et al., 1990). T-cell receptor KO and Gai2 KO mice also develop

colitis, as do SCID mice to which CD4+ T cells expressing high levels of CD45RB have

been transferred (Elson et al., 1995; Fedorak and Madsen, 2000; Wirtz and Neurath,

2000).

The Pig as a Model

The pig was chosen as an animal model for these studies for a number of reasons.

Because the majority of previous work has been performed in rodent models, a large

animal model would be useful to represent larger mammalian species. A larger animal

allows for colonic endoscopic examination; thus the inflammatory status can be

documented without sacrificing additional groups of animals. From a practical

perspective, experimental use of pigs is more economical than other commonly used

larger mammals such as dogs and cats.

Study Objectives

Using the pig as a large animal model, the major objectives of this dissertation were

to 1) develop a model of subacute proctitis; 2) validate an objective evaluation of visceral

discomfort; 3) evaluate the effect of colorectal inflammation on visceral nociceptive

threshold; 4) evaluate the effect of colorectal inflammation on immunoreactivity of

Substance P in the colon, rectum, and lumbar spinal cord.














CHAPTER 2
METHODS

IACUC Approval

All procedures were approved by the University of Florida Institutional Animal

Care and Use Committee. Pilot animals were approved under Protocol # A579

"Evaluation of visceral discomfort in a chronic proctitis model in the pig" and animals for

the major study were approved under Protocol # B 167 "Porcine Model of Inflammatory-

Mediated Visceral Hypersensitivity."

Model Development

The primary focus of pilot studies in this project was to establish a reliable method

for visceral sensitivity testing, and to determine the appropriate methodology for

TNBS/EtOH enema administration.

Animals

Three animals were used for the initial phases of the pilot study. These animals

were used mostly to develop the visceral sensitivity testing protocol. An additional two

animals were used to generate preliminary data regarding differences in visceral

sensitivity, if present, between saline and TNBS-EtOH enema treatment.

Training

The first objective of the pilot studies was to find an acceptable method of pig

restraint which would allow for their comfort and easy manipulation, and observation by

research personnel. A standard hog transport crate was used, with modifications so that

the animal could not escape through a large top opening (Fig.2-1). Otherwise, the









animals were able to move freely in the crate and turn back and forth. They were slowly

acclimated to the crate, using corn treats as a training tool. Once acclimated, the pigs

were easily transported from their normal housing to and from the laboratory and would

remain in the crate without incident during the studies.


Figure 2-1. Pig in crate used for all procedures. View from top.

Once an acceptable transport device was obtained, the pigs were easily accustomed

to all laboratory procedures. Food, most notably whole corn, was used as a tool for

positive reinforcement for all training procedures.

Procedures

The next phase of the pilot studies was to develop a standard protocol for CRD. A

standard rectal distention balloon and catheter designed for use in humans (Medtronics,

Shoreview, MN) were initially used. While the balloon volume was appropriate, the









catheter system was too flexible. With a small amount of abdominal strain, the pigs

would, in essence, defecate out the balloon. A design modification of the catheter was

accomplished such that the catheter would remain in place without causing additional

discomfort to the animal or any local damage to the rectal mucosa. The resultant system

is described in detail below.

Based on the potentially subjective nature of clinical signs of abdominal

discomfort, we chose to use a ramp protocol of CRD in which the threshold of discomfort

was used as the response variable of interest. This also allowed for a minimal amount of

discomfort to the animals.

Main Study Design

The main study was designed based upon pilot data.

Animals

The study was designed such that eighteen two to three month-old mixed breed

swine would be used. Initial bodyweight for all animals was between 20-30 kg. The pigs

were divided into four groups (two groups of six and two groups of three) using a random

number chart. Animals in groups 1 and 2 were sacrificed at week 9 and those in groups 3

and 4 were sacrificed at week 14. Due to unforeseen complications, one additional

animal was added to the 14 week saline control group, for a total of 19. Except as

described below, pigs were meal fed a commercial swine grower diet (LabDiet) at a rate

of 0.2-0.25 kg/kg bodyweight daily.

For the first 3 weeks of the study, pigs were handled 1-2 times daily in order to

acclimatize them to human contact. As the animals became increasingly tractable, they

were trained to load into and out of a standard transport crate and to remain in the crate









for 30-45 minute time periods, followed by acclimation to rectal thermometer and finally

to intra-rectal balloon insertion.

Table 2-1. Animal grouping


Group # Enema Euthanasia Week
1 Saline 9
2 Saline 14
3 TNBS/EtOH 9
4 TNBS/EtOH 14


Study Timeline

The timeline of events did not vary between animals and is described below in

Table 2-2.

Table 2-2. Timeline of study events

Week Description Distention Other events Animals
studies sacrificed
1 Training None
2 Training None
3 Training One Endoscopy
4 Enema None Monitoring
5 Post-enema None Monitoring
6 Post-enema None Monitoring, Endoscopy
7 Post-enema One
8 Post-enema One
9 Post-enema One Endoscopy Groups 1 & 2
10 Post-enema One
11 Post-enema One
12 Post-enema One
13 Post-enema One
14 Post-enema One Endoscopy Groups 3 & 4









Procedures

Enema Administration

Enema administration occurred during week 4 of the study. Prior to this procedure,

feed was withheld for 12 hours although water was available on a free-choice basis. Each

animal was anesthetized with a butorphanol (0.15-0.30 mg/kg)/xylazine (4-8

mg/kg)/ketamine (4-8 mg/kg) combination administered intramuscularly. Pigs in Groups

1 and 3 received 40 ml of 100% EtOH mixed with 5 grams of TNBS diluted in 10 ml of

water. The enema was retained in a 10-cm portion of the distal colon and proximal

rectum for 12 minutes by use of two Foley catheters with 60-ml balloons. For control

animals (groups 2 and 4), the retention enema consisted of 50 ml of 0.9% saline. Each

pig was observed until fully recovered from anesthesia. They were then monitored 2-3

times daily until vital signs remained within normal ranges for 3 consecutive days and

daily thereafter. Clinical evaluation included measurement of rectal temperature,

observation of general attitude and fecal output, and monitoring for signs of GI distress

such as vomiting, constipation, diarrhea, and anorexia. Three pigs became excitable in

response to the initial anesthetic protocol and were subsequently anesthetized with a

combination of xylazine (2.2 mg/kg) and tiletamine/zolazepam (2.2 mg/kg)

intramuscularly. Due to a delay in approval of the revised anesthetic protocol, the first of

these pigs was re-allocated and became a non-treated control animal for the duration of

the study.

Endoscopic Evaluation

Videoendoscopic examination of the rectum and distal colon was performed

(Pentax EFG 1-meter videoendoscope) and recorded during weeks 3, 6, 9, and 14 of the

study. Prior to the procedure, the rectum was evacuated using 1-2 liters of warm soapy









water as an enema, and the endoscopy was performed with the animals standing in the

transport crate previously described. When possible, each endoscopic examination was

documented with a series of still images. For each endoscopic procedure, the appearance

of the rectum and colon was scored on a 0-2 scale for each of the following: erythema,

edema, granularity, friability, and erosions, with an overall range of 0-10 (Table 2-3)

(D'Argenio et al., 2001).

Table 2-3. Endoscopic lesion scoring

Lesion None Mild Moderate Marked Severe
Erythema 0 0.5 1 1.5 2
Edema 0 0.5 1 1.5 2
Granularity 0 0.5 1 1.5 2
Friability 0 0.5 1 1.5 2
Erosions 0 0.5 1 1.5 2


Visceral Sensitivity Evaluation

Visceral sensitivity was assessed by means of colorectal distention (CRD).

Animals of all groups underwent CRD procedures once during week 3, then once weekly

from week 7 until the end of their respective protocol.

Rectal catheter

Visceral discomfort was stimulated by CRD using a barostat (IsoBar 3, G&J

Electronics Corp., Willowdale, Ont.). This involved a commercially available 500-ml

polyethylene bag (Medtronics, Shoreview, MN) attached to a rectal catheter with dental

floss (Fig. 2-2). The catheter had separate channels dedicated for volume control and

pressure transduction, respectively. A thin metal rod was placed within the lumen of the

catheter and secured with silicone. The distal end of this rod was smoothed, and the

silicone filling completely covered the end of the rod such that it would not interfere with









the pressure measurements or volume alteration functions of the catheter system. The

proximal end of the rod extended approximately 15 cm from the catheter tip. The

purpose of this rod was to provide stiffened support for the catheter such that, when

secured in place, it would not be expelled by the abdominal strain demonstrated during an

animal's discomfort response.













Figure 2-2. Polyethylene rectal distention balloon attached to catheter.

Prior to insertion, the catheter was lubricated and a roll of 1" white tape was

attached 10 cm from the catheter tip. After insertion, the tape was used to secure the

apparatus to the animal's tail such that the balloon would remain at a standardized

distance from the external anal sphincter. (Fig. 2-3)


Figure 2-3. Close-up of rectal catheter attached to pig's tail.









Ramp Protocol

For nociceptive response testing, a stepwise pattern of inflations from 15 to 55

mmHg (60-second inflation with a 5-minute deflation period between inflations) was

used. The inflation pattern continued until the animal displayed a response of discomfort

or, if no response was obtained, a maximum pressure of 55 mmHg.

Assessment of Response

Three observers, blinded as to treatment group, monitored each pig throughout each

distention protocol. At the conclusion of each inflation period, each observer

independently displayed a "yes" or "no" response to a fourth investigator responsible for

control of the barostat. When at least two of three observers declared a "yes" response,

the fourth investigator discontinued the inflation protocol and recorded the pressure at

which the response occurred. Based upon pilot study observations, a discomfort response

was considered to include at least three of the following behaviors occurring during one

inflation period: abrupt change in behavior (i.e. the animal discontinues previous

behavior), arching of the back, abdominal strain, shifting of the hindlimbs. A still image

of a typical response is displayed in Figure 2-4.

Each animal was observed for at least 5-minutes following the final balloon

deflation.



























Figure 2-4. Typical discomfort response. Note the arched back and wide-based stance of
the hindlimbs.

Tissue Collection, Processing, and Analysis

Animals were euthanized during the final week of the study after the conclusion of

that week's other events (CRD, endoscopy). Animals were first anesthetized with the

xylazine/butorphanol/ketamine combination described previously, followed by a

barbiturate overdose (Beauthanasia D, Shering Plough, 0.22ml/kg) administered

intravenously. Those pigs requiring telazol/xylazine anesthesia for enema administration,

received that anesthetic combination prior to Beauthanasia administration. Necropsy

examinations were performed immediately after euthanasia. Tissue from all animals was

processed in similar fashion.


Necropsy

Complete necropsy examinations were performed on each animal. Any gross

abnormalities, if present, were recorded. The gastrointestinal tract was examined in its

entirety. Samples taken for later histological analysis were cut, rinsed with 0.9% NaC1,

and immersed immediately in 10% buffered formalin. Sections of the rectum were taken









at points 10 and 12 cm orad to the external anal sphincter and, respectively, labeled R1

and R2. Sections of the colon were taken at points 15 and 20 cm orad to the orad-most

rectal section and labeled Cl and C2, respectively. A section of ileum was taken 2 cm

orad to the ileocecal band (labeled I), a section of cecum was taken along the medial

cecal band (CE), and a section of jejunum was taken at a random location within the mid-

jejunum (J). The remaining portions of the gastrointestinal tract were opened and

examined for gross lesions. The thoracic and remaining abdominal contents were also

examined for the presence of lesions.

After completion of this portion of the examination, the spinal cord was removed

and fixed. The vertebral column was isolated from the cranial to mid-thoracic level

caudal to its termination and stripped of all excess tissue. A dorsal hemilaminectomy

was performed via the use of a Stryker saw along the length of the column. The spinal

cord and associated spinal nerve roots were extricated and immersed in formalin as

described above.

Tissue Preparation

After formalin immersion for a period of 18-24 hours, sections were cut, placed in

cassettes, and then dehydrated in ethanol and embedded in paraffin in routine fashion.

Sections were cut onto Superfrost Plus slides for staining at a later date. All slides and

blocks were stored at room temperature.

Histological Analysis

For routine histological analysis, slides were heat-fixed and deparaffinized in

routine fashion then stained with hematoxylin and eosin. Sections of the colon and

rectum of each animal were initially evaluated by the author and a pathologist (EWU) in

order to gauge the range of lesions present in the study population. Based upon the initial









analyses, cellular infiltrates in the tissues were confined to lympocytes, predominantly in

mucosal and submucosal aggregates. In addition, the tissues had a varying degree of

edema. Thus, the degree of edema was scored from 1 (normal) to 4 (severe) and the total

number of lymphoid aggregates per slide were counted. (Table 2-3). Spinal sections

were evaluated for detectable abnormalities, but not scored.

Table 2-4. Histologic scoring system for gastrointestinal tissues.

Criterion Category Score
Edema Normal 1
Mild 2
Moderate 3
Severe 4
Lymphoid aggregates/section 0 1
1 2
2-3 3
4-6 4


Immunohistochemical Analysis

Immunostaining for Substance P antigen was performed on the formalin-fixed,

paraffin-embedded tissue using a rabbit anti-human Substance P polyclonal antibody

(BYA1145-1, Accurate Chemical and Scientific Corp., Westbury, NY). This antibody

has been validated in rat, monkey, feline, porcine, and bovine tissues. Slides were

processed in duplicate (antibody and negative control). All slides were heat fixed and

deparaffinized by immersion in xylene (2 x 5min) followed by decreasing concentrations

of ethanol (2 x 3 min at 100%; 2 x 3 min at 95%) and then a rinse in deionized water.

Slides were then stained using the DAKO EnVision Peroxidase staining system

(DakoCytomation, Inc., Carpinteria, CA). All procedures were performed at room

temperature. Briefly, slides were carefully dried and the tissue section was outlined with

a hydrophobic pen (PAP pen), leaving at least a 3 mm margin. Next, they were incubated









with DAKO EnVision Peroxidase Blocking Reagent for 5 min, rinsed with PBS and

placed in a PBS bath for 5 min. Slides were then incubated with primary antibody or

control, respectively, for 30 minutes. Based on serial dilutions of 1:50 to 1:2500, the

apparent optimal dilution of anti-substance P antibody was 1:500. This dilution was used

for all subsequent staining.

Following antibody incubation, slides were placed in a PBS bath (2 x 5 min). They

were then incubated with DAKO Envision Peroxidase labeled polymer for 30 min,

followed by a PBS rinse and placement in a PBS bath for 5 min. DAKO Envision DAB

substrate was applied for 5 min, and then slides were rinsed in deionized water for 1 min.

Next, they were counterstained with 50% Gill's Hematoxylin for 2 min, and then placed

in a running deionized water bath until the water ran clear. Finally, they were then placed

in a bluing solution deionizedd water with ammonium hydroxide, 10 quick dunks),

followed by a deionized water bath (5 min), and then dehydrated in increasing

concentrations of ethanol (2 min at 95%, 3 x 2 min at 100%). Slides were kept in xylene

until coverslips were permanently affixed with Permount.

Ventral Horn Immunohistochemical Analysis

Substance P-immunoreactive neurons within the ventral horn (total of both right

and left hemisections) of each spinal section were counted and recorded. Examples of

neurons considered immunoreactive and non-immunoreactive are identified in Figure 2-

5.









Qu,'n:titveI' m''muoisoheia An.lysis
t"-: t .% .* *. -* -:






A .
.'*" ". "












Fiubstra P nuro. E s of qu-aitative u(arros a..non
.* ., A V


rid ad vi d b M w j ( w j e a., Esse ,




Figure 2-5. Ventral horn neurons. Examples of SP-immunoreactive (arrows) and non-

immunoreactive (arrowheads) neurons within the ventral horn.
Quantitative Immunohistochemical Analysis

Substance P immunoreactivity was evaluated quantitatively using an algorithm

described and validated by Matkowskyj (Matkowskyj et al., 2000). Essentially, the

information within a control slide is subtracted from the information within an antibody-

treated slide in order to quantitatively identify antibody-generated chromagen content of

the slides in question. In order to for this to occur effectively, slides were read then and

images saved in tagged-image file format (TIFF), which allows for compression without

loss of data.

The slides were read using a Zeiss Axioplan 2 Microscope. Corresponding images

were captured for antibody-stained and control images for each tissue section using

SPOT image capture software. These images were saved in TIFF format and reopened

with Adobe Photoshop (Version 6.0, Adobe Systems, Inc., San Jose, CA), where 3

100x100 pixel areas of interest for each slide were captured and saved as new jpg files.

These files were opened with Matlab (Version 6.5, The MathWorks, Inc., Natick, MA),









which calculated the energy contained within each image. The net energy differential

(energy within control images subtracted from energy within antibody images) was used

to represent the net chromagen content within the region of interest as arbitrary units.

For spinal sections, image one was taken at the apex of the dorsal horn, and image

two was taken along the dorsal margin, approximately half-way to midline. The 3 100-

pixel squares were captured as shown. (Fig. 2-6, 2-7, and 2-8) For colon and rectal

sections, both original images were taken at arbitrary points along the junction between

the circular and longitudinal muscle layers. The 100-pixel squares were taken at

consecutive points along this junction, careful not to include blood vessels. (Fig. 2-9)























Pixel
Squares
PM,








Squares







ImageI 1 [mllage


















Figure 2-6. Image selection for spinal sections. Spinal cord, 10x magnification. Note the
tip of dorsal horn at upper left of image, central canal at lower right.












fly'..,
&&C3..








ISt.r- : i






t"
24i,,'.


4 .. -.*- -

'. t.




1 I



t.

\ ',

-', r .
.

*-. ^ % ,. .*






'. .. '. '- .,
..

B, ,


-^. L*. : '


Figure 2-7. Pixel square selections for Cord Image 1. Spinal cord, 20x magnification.
See figure 2-6 for orientation.






^%.^' .r
-. .* -: .




j. .- : .-,' .


M; i
-. .11,. *.n'.-

. :: '. *l .' + "-'.

'* .. ,





: i +: t,'. '** 4 .. : ;". ,,



Figure 2-8. Pixel square selections for Cord Image 2. Spinal cord, 20x magnification.
See figure 2-6 for orientation.









'P .s.\uar ,l"o," colon-.ic and recal. my.ne" p
m gi- ,at o"
Ci t f. t ,f 4 .-

Chiago ..) Sg,,n w plce ,', 0.05. ": X' "
(combined groups and 3 vs. 2 and 4) using an independent samples t-test. For all


























subsequent weeks, threshold pressures were compared between saline and TNBS groups
(combined groups I and 3 vs. 2 and 4) using a one-way analysis of variance (ANOVA).















In an effort to gauge a change in each individual animal's threshold relative to
v ; ,.-"- s

r. c. -- .+ i 1..
*- r "" -












Figure 2-9. Pixel square selections for colonic and rectal myenteric plexus. Colon, 20x
magnification.

Statistical Analysis

All statistical analyses were performed using SPSS 11.0 for Windows (SPSS, Inc.,

Chicago, IL). Significance was placed at p < 0.05.

Colorectal Distention

Week 3 threshold pressures were compared between saline and TNBS groups

(combined groups 1 and 3 vs. 2 and 4) using an independent samples t-test. For all

subsequent weeks, threshold pressures were compared between saline and TNBS groups

(combined groups 1 and 3 vs. 2 and 4) using a one-way analysis of variance (ANOVA).

In an effort to gauge a change in each individual animal's threshold relative to

baseline, an alteration from baseline threshold pressure (ABTP) for each study from

weeks 7-14 by subtracting the baseline threshold pressure from the threshold pressure for

that study. Thus, if the pressure for a particular week was lower than that animal's

threshold pressure at week 3, the alteration number for that week would be negative.









ABTP were compared between saline and TNBS groups (combined groups 1 and 3 vs. 2

and 4) using a one-way ANOVA for weeks 7-14.

Endoscopy

Week 3 endoscopy scores were compared between saline and TNBS groups

(combined groups 1 and 3 vs. 2 and 4) using an independent samples t-test. Scores for

weeks 6, 9, and 14 were compared between saline and TNBS groups (combined groups 1

and 3 vs. 2 and 4) using a one-way ANOVA.

Histology and Immunohistochemistry

Inflammatory scores were compared between the four groups using a one-way

ANOVA, followed by Tukey's HSD multiple comparison procedure. Scores for each

colonic and rectal section were evaluated independently.

Chromagen content (SP-IR as represented by arbitrary energy units/pixel

(EU/pixel)) in each section (rectum 1&2, colon 1&2, spinal sections L1, L2, L6, and L7)

was compared between the four groups using a one-way ANOVA, followed by Tukey's

HSD multiple comparison procedure. For each spinal section, the number of SP-

immunoreactive neurons in the ventral horn was similarly compared between groups.

Baseline threshold pressures and chromagen content satisfied the Shapiro-Wilk test

for normality, thus a parametric approach was justified. A one-way ANOVA was chosen

for sequential analyses due to the disparate times of euthanasia for groups 1 and 2 vs. 3

and 4. Essentially, half of the animals were discontinued mid-way through the study, so a

one-way analysis was used, though a two-way ANOVA for repeated measures would

have been most appropriate had all animals continued through until week 14 of the study.









Correlations

The relationship between an individual animal's visceral sensitivity, as measured

by the weekly alteration from CRD sensitivity threshold, and each of the following were

also evaluated using linear regression and Pearson's correlation: 1) ventral horn SP-

immunoreactive neuron count for each evaluated spinal cord section; 2) dorsal horn SP-

immunoreactivity (EU/pixel) for each evaluated image of each spinal cord section; 3)

gastrointestinal pathology for each evaluated section (edema and lymphoid aggregate

counts evaluated independently). In addition, correlation analysis was similarly

performed between the gastrointestinal edema and lymphoid aggregate counts and the

spinal SP-immunoreactivity. Linear regression analysis was subsequently performed on

correlations identified as significant in this fashion.














CHAPTER 3
RESULTS ANIMAL STUDIES

Animals were well trained and tolerated the distention procedures very well. Pilot

studies allowed for refinement of all procedures and subsequent design of the main study.

Development of colitis

After TNBS/EtOH enema administration, most animals developed mild bloody

diarrhea for approximately 24 hours and all became febrile (103-1040F) for 2-5 days. One

animal developed signs of sepsis and died despite therapy within 5 hours following its

TNBS/EtOH enema. Necropsy examination revealed a perforated rectum. This animal

was replaced in the study. None of the other pigs displayed any evidence of clinical

illness, other than the diarrhea noted previously and occasional mild depression for 12-24

hours. All animals maintained an excellent appetite throughout the study.

Endoscopic Evaluation

Endoscopic evaluations were easily performed using the previously described

procedure. Most animals' rectum and distal colon could be easily and fully observed

after the first 1-L enema. If further evacuation was needed, the enema was repeated. No

animal required more than 2 enemas and all animals tolerated the procedure well.

Raw data from the endoscopic evaluations are presented in Appendix A. Mean

baseline endoscopy score (week 3) was 0.33 for the TNBS animals and 0.50 for the saline

animals. These values did not differ significantly. Endoscopy scores for subsequent

weeks (TNBS animals combined as Group 1 and saline animals combined as Group 2)

are presented in Figure 3-1.











Endoscopy Scores


2 4 6 8 10 12 14 16
Week of Study


Figure 3-1. Endoscopy scores. Data expressed as mean +SEM.

TNBS and saline means differed significantly at weeks 6 (p=0.000) and 9


(p=0.033) but not week 14 (p=0.134). Examples of normal and abnormal endoscopic


evaluations are presented in Figures 3-2 and 3-3, respectively.

















Figure 3-2. Normal endoscopy (Grade 0).


- TNBS-EtOH
- -0 Saline


-W
-----0




















Figure 3-3. Abnormal endoscopy (Grade 3).

Visceral Sensitivity

CRD studies were completed for all animals, all weeks, with the exception of one

pig (#3179) that developed a small amount of rectal bleeding during balloon insertion for

its week 9 (final) trial. The procedure was not repeated for this animal. All other trials

were completed successfully. For 116/125 completed trials, the animal displayed a

discomfort response as judged by at least 2/3 blinded observers. For the remaining 9

trials, a majority opinion discomfort response was not observed (represented by 55* in

the raw data). For statistical analyses, these trials were considered to have a threshold

pressure of 55 mmHg. For all trials, 55% of the decisions were unanimous, whereas 45%

involved a "yes" response from 2/3 observers.

Raw data from the visceral sensitivity studies are presented in Appendix A. Mean

baseline threshold pressure (week 3) was 40.8 mmHg for the TNBS/EtOH animals and

30.7 mmHg for the saline animals. These values did not differ significantly. For

subsequent weeks, mean values +SEM for TNBS/EtOH (combined as Group 1) and

saline (combined as Group 2) animals are presented in Figure 3-4.






40


CRD Mean Threshold Pressure


15 + TNBS/EtOH
-0 Saline
10 T i T ---
2 4 6 8 10 12 14 16
Week of Study

Figure 3-4. Mean threshold pressures. Data expressed as mean +SEM. TNBS/EtOH and
saline groups differed significantly (p<0.05) at week 13.

Mean TNBS/EtOH pressures differed significantly from the mean saline pressure

for week 13 (p=0.006). There was a trend towards difference for weeks 7 (p=0.084) and

9 (0.093). Complete ANOVA tables for all analyses are presented in Appendix B.

CRD threshold pressure Mean alteration from baseline


TNBS/EtOH
Saline
-40
6 8 10 12 14 16
Week of Study

Figure 3-5. Mean ABTP. Data expressed as mean +SEM. TNBS/EtOH and saline
groups differed significantly (p<0.05) at week 7.


-


I6-







41


Mean ABTP for TNBS/EtOH (combined as Group 1) and saline (combined as

Group 2) animals for all weeks are presented in Figure 3-5. Mean ABTP for

TNBS/EtOH animals differed significantly from the mean for saline animals for week 7

(p=0.045), although there was a trend for week 13 (p=0.056)

When examining data from individual animals, rather than the groups as whole,

only 2/7 saline animals had >2 weeks with threshold pressures below baseline, whereas

11/12 TNBS/EtOH animals had >2 weeks as such.

Correlations

Correlations were made between endoscopic scores on weeks 6, 9, and 14and

weekly ABTP. Complete Pearson's correlation data are presented in Appendix C. Of

these, endoscopy scores on weeks 6 and 9 had a significant negative correlation with

week 13 sensitivity, and the linear regression between these factors are presented in

Figure 3-8. No other comparisons had significant Pearson's correlations, thus linear

regression was not performed.


W0k 6 rndosnopy va Welk 13 Senstit Wek 9 en docopy vs Week 13 SnsMty

2-0435 "... .

P 0 0 o











Figure 3-6. Correlation between endoscopy scores on weeks 6 and 9 and week 13 ABTP.
Dashed lines represent 95% confidence interval.
Dashed lines represent 95% confidence interval.














CHAPTER 4
RESULTS TISSUE ANALYSIS



Histological Analysis

An initial review of the gastrointestinal sections revealed that the predominant

abnormality seen was submucosal edema. Also, the number of lymphoid aggregates seen

in the mucosal and submucosal areas appeared to differ between sections. Thus, these

factors were used in order to quantitatively evaluate the gastrointestinal sections as

described in the materials and methods section. Examples of each histologic edema score

are presented below in Figures 4-1 through 4-4.




















Figure 4-1. Example of edema grade 1. Arrows denote submucosal edema. Porcine
rectum, 10x magnification, bar denotes 100 im.







43


















Figure 4-2 Example of edema grade 2. Arrow denotes submucosal edema. Arrowheads
denote a mucosal lymphoid aggregate. A dilated lymphatic is also present just
below the lymphoid aggregate. Porcine rectum, 10x magnification, bar
denotes 100 im.












P---"
P- M.














Figure 4-3. Example of edema grade 3. Arrows denote submucosal edema. Mucosal
edema is also evident, as noted by an increased space between mucosal
glands. Porcine rectum, 10x magnification, bar denotes 100 rn.




























Figure 4-4. Example of edema grade 4. Arrows denote submucosal edema. Porcine
colon, 10x magnification, bar denotes 100 im.

Mean edema and lymphoid aggregate scores for all sections are presented in Tables

4-1 and 4-2, respectively. Statistical analysis did not reveal any significant differences

between groups for any of the colonic or rectal sections. Complete ANOVA tables and

Tukey's multiple comparison analyses are provided in Appendix B.

Table 4-1. Mean lymphoid aggregate scores.

Section Group Mean SD SEM
Cl 1 2.17 0.75 0.31
2 2.33 0.58 0.33
3 1.67 1.03 0.42
4 2.25 1.50 0.75
C2 1 2.17 0.75 0.31
2 2.00 1.00 0.58
3 2.00 0.89 0.37
4 2.25 0.50 0.25
R1 1 2.17 0.98 0.40
2 2.67 1.53 0.88
3 1.67 1.03 0.42
4 2.00 1.41 0.71
R2 1 2.17 0.98 0.40
2 1.67 1.15 0.67
3 2.17 0.98 0.40
4 2.50 1.00 0.50










Table 4-2. Mean edema scores.

Section Group Mean SD SEM
Cl 1 1.50 0.55 0.22
2 1.67 0.58 0.33
3 2.33 0.82 0.33
4 2.00 1.41 0.71
C2 1 1.50 0.55 0.22
2 1.33 0.58 0.33
3 2.00 1.10 0.45
4 2.25 1.26 0.63
R1 1 2.33 0.52 0.21
2 2.67 0.58 0.33
3 1.83 1.17 0.48
4 3.25 0.96 0.48
R2 1 2.33 0.82 0.33
2 2.33 1.15 0.67
3 2.17 1.47 0.60
4 3.00 1.15 0.58


Spinal cord sections were also evaluated. Some sections contained mild gliosis,

but this was considered within normal limits, thus the spinal cords were not scored.

Immunohistochemical Analysis

Substance P-immunoreactivity (SP-IR) was seen in all tissues. The

immunoreactivity appeared specific, and very little background staining was evident in

the spinal cord sections. Gastrointestinal sections had a higher degree of non-specific

chromagen uptake, but specific immunoreactivity was also present. Negative control

slides had little to no chromagen uptake for all sections.

Spinal cord

Substance P-immunoreactivity was most prominent in the superficial dorsal horn

(lamina I and II) (Figure 4-5). Some large immunoreactive neurons were also detected in

the ventral horn of some sections.














..-













Figure 4-5. SP-IR in spinal cord. Arrows denote margin of immunoreactivity in dorsal
horn. Porcine spinal cord, L7 segment, 2.5x magnification. For details of site
selection and immunoreactivity in those regions, see Figures 2-6 through 2-8.

Quantitative data describing the Substance P-immunoreactivity (EU/pixel) within

the dorsal horn are presented in Table 4-3. Statistical analysis did not reveal any

significant differences between experimental groups for any spinal segment. Initially,

data from images 1 and 2 were analyzed individually. But, as this did not change any

analyses, data were pooled as collective data for the dorsal horn. Complete ANOVA

tables and multiple comparison procedural results are presented in Appendix B.

Numbers of Substance P-immunoreactive neurons within the ventral horn are

presented in Table 4-4. Statistical analysis did not reveal any significant differences

between experimental groups for any spinal segment. Complete ANOVA tables and

multiple comparison procedural results are presented in Appendix B.









Table 4-3. Mean dorsal horn Substance P-immunoreactivity (EU/pixel)


Section Group Mean SD SEM
L1 1 150.64 43.67 17.83
2 104.14 10.42 6.02
3 152.38 61.16 24.97
4 110.74 22.53 11.26
L2 1 145.01 32.53 13.28
2 131.65 28.48 16.44
3 158.66 37.05 15.13
4 146.44 64.27 32.13
L6 1 183.18 58.26 23.79
2 180.43 45.04 26.00
3 179.13 48.98 19.99
4 143.30 36.66 18.33
L7 1 138.91 41.26 16.85
2 157.64 32.08 18.52
3 141.95 76.53 31.24
4 202.21 17.41 10.05









Table 4-4. Mean ventral horn Substance P-immunoreactive neurons

Section Group Mean SD SEM
L1 1 1.50 3.67 1.50
2 0.00 0.00 0.00
3 6.83 7.49 3.06
4 4.00 8.00 4.00
L2 1 0.83 1.60 0.65
2 0.00 0.00 0.00
3 8.00 8.67 3.54
4 7.75 8.73 4.37
L6 1 10.50 9.20 3.76
2 6.33 3.79 2.19
3 14.67 16.60 6.78
4 5.50 6.81 3.40
L7 1 3.33 4.18 1.71
2 2.33 4.04 2.33
3 3.83 3.92 1.60
4 2.67 2.52 1.45


Gastrointestinal Tract

In the gastrointestinal tissues, the most intense regions of SP-IR were in the

submucosal region between circular and longitudinal muscle layers, corresponding to the

location of the myenteric plexus (Fig. 4-6). Some peripheral chromagen uptake was

present in all tissues, likely related to staining artifact. Quantitative-IHC data for the

gastrointestinal tissues are presented in Table 4-5. Values did not differ significantly

between experimental groups. Complete ANOVA tables are presented in Appendix B.










100 ur


~s
5*
cac: ,-
iJ~ -
'I
"T:~


Figure 4-6. SP-IR in porcine colon. 10x magnification. For pixel square location, see
Figure 2-9.

Table 4-5. Quantitative IHC data for gastrointestinal tissues.


Section Group Mean SD SEM
R1 1 92.96 34.20 13.96
2 86.72 46.11 26.62
3 100.13 40.16 17.96
4 140.90 65.47 32.74
R2 1 117.04 65.24 26.63
2 61.06 33.01 19.06
3 98.29 50.79 20.74
4 93.14 52.82 30.49
C1 1 135.69 29.87 13.36
2 94.96 16.00 9.24
3 150.16 89.22 36.42
4 156.23 76.90 38.45
C2 1 137.68 71.19 35.59
2 67.92 45.83 26.46
3 100.83 84.43 37.76
4 203.68 36.68 21.18


Correlations

Complete Pearson's correlation data are presented in Appendix C. The following

comparisons had a significant negative correlation: R2 lymphoid aggregate score and


e^^^-^-8_^;^?Cll; `_IyL^ ^--- ^^^^
i-^, -^-^^*^^






50


Week 10 and 12 ABTP (Figure 4-7); R1 histological scores and Ventral horn SP-IR

neurons in the L1 and L2 segment (Figure 4-8); Dorsal horn SP-IR in the L1 segment and

Week 9 ABTP (Figure 4-9) ; Dorsal horn SP-IR in the L7 segment and Week 11, 12, and

14 ABTP (Figure 4-9); Ventral horn SP-IR neurons in the L2 segment and Week 11, 12,

13, and 14 ABTP (Figure 4-10); Ventral horn SP-IR neurons in the L6 segment and

Week 13 and 14 ABTP (Figure 4-11); Ventral horn SP-IR neurons in the L1 segment and

Week 8 ABTP (Figure 4-12); Dorsal horn SP-IR in the L6 segment and R1 lymphoid

aggregate score(Figure 4-13).

In addition, the following comparisons had a significant positive correlation: SP-IR

in the C1 section and R2 lymphoid aggregate score; SP-IR in the R1 section and C1

lymphoid aggregate score; SP-IR in the R1 section and C2 lymphoid aggregate score

(Figure 4-13). All other comparisons did not have significant Pearson's correlations, thus

linear regression was not performed.


R2Ly-Week 10 R2ym-Week 12





"'"- I: .. 6..




LW LD
l ^ .^ ........ ........ ........




Figure 4-7. Correlation between lymphoid aggregates in the rectal section R2 and ABTP.
Dashed lines represent 95% confidence interval.















L1-R1 Ed


. 9


D 0.3


1.


S 0


L2-RI Lym


I -.
4 ..







00


0'


r2. 0299
Pewon- -0 546
P. 015


0 '0 S2C


n S 10 'S 20
VenUM hom SP~uorsct* w~


Figure 4-8. Correlation between histological scores in the rectal section R1 and ventral

horn SP-immnuoreactive neurons in spinal sections L1 and L2. Dashed lines

represent 95% confidence interval.













L1 vs Week 9


,.a 0 %ea
1 *, Pe.". Q .*/.
p 0 019



I N


-- r ---
1 X0




&jbst*Wce P inEmurjoreacoey EULpal


L7 vs Week 12


I


i


', i


0 919
., .. P ar,.x .-0 95Q
'.. D 0 000


.4
0i
fl


SuMst-e P4- hWorawfc IEEUV|p


L7 vs Week 14


P2

p,


. -


0


0 s 100
SubstaWnc* P-HWM WMMcrv


*0466
S-on -0 683











S 40
200 250 0
Ug-...I
"'"


Figure 4-9. Dorsal horn SP-IR correlation with ABTP. Clockwise from the top left, plots
represent linear regression analysis for segment L1 vs. Week 9 and segment
L7 vs. Weeksl 1, 14, and 12. Dashed lines represent 95% confidence
intervals.


Pea n- -0 760
*"0018


L7 vs Week 11


*t


---...


--


sc 180 158
~ P~ (E~










L2 Wk 11


I"


P4r0.7 0
p;gQW


f

I


I.

i-
] o


a A


2.*0506
Pe ron 712
p. 0021


'* *


- Wem-


4.
i 0.
1 *
-a
I-

I .rc-,


L2W 11 t


pB 0 047 6
B007C


- *- 0


L2-W k 14


I .-


j :1


00- 0 I'
p. 0021


4r- I


Figure 4-10. Ventral horn correlation with ABTP for the L2 spinal segment. Clockwise
from the top left, plots represent linear regression analysis for weeks 11, 12,
14, and 13. Dashed lines represent 95% confidence intervals.













LOW6&k 13


r2 0 489
Persow -0 TDD
p. 0 040


I 4



j


Figure 4-11. Ventral horn correlation with ABTP for the L6 spinal segment. Plots on the
left and right represent linear regression analysis for weeks 13 and 14,
respectively. Dashed lines represent 95% confidence intervals.


L7-Week 13




I .0 473
P PeMron -068
p~ D 040

20









0 5 1o 15 20 25
Vetral horn SPnMnTeO nIuV ns



Figure 4-12. Ventral horn correlation with ABTP for the L7 spinal segment. Dashed
lines represent 95% confidence intervals.


I:

I


*- I


2. 0400
Pea on 0506
p 0050


I


L.-WMk 14


M


S10 20 0 40
Ventm hn SP a -umoeeav neure














ClI v R2Lym


IM i -0 4S
p 0 049
0 s 00 I0 2
SuAsnUIce P kwrunecv0 w (EUefll



R1 v CtLYm


^--'b






e ** .

2 0r 05O


P*arn- 0 711
p. 0 001

0 SO 5 00 150 '00 6 0
S^uble P1 IrnuAracvea (EUal)


SI


*


* -


...- -


pa 0022

0 W 100 1W 200 S 30
Sub stan P-liVmun.drCty (EUDpu



Ri ve C2Ed





4 11


** .. .- *


r27 0 27
Person. 0 517
p- 0 028
so 10 ISO 22 5
Subalane P- rMnunWOaWifr IEUeIl


Figure 4-13. SP-IR correlation with histological scores. Clockwise from the top left,

plots represent linear regression analysis for spinal segment L6 dorsal horn vs.

R1 Lymphoid aggregate score; colon section Cl vs. R2 lymphoid aggregate

score; rectal section R1 vs.C2 edema score; rectal section R1 vs. Cl lymphoid

aggregate score. Dashed lines represent 95% confidence intervals.


L8vs R1Lym


-I














CHAPTER 5
DISCUSSION

Model Development

The first two major objectives outlined for this study were clearly met. Animals

developed a subacute proctitis after TNBS/EtOH enema administration and showed

clinical signs of abdominal discomfort in response to colorectal balloon distention.

Subacute Proctitis

A subacute proctitis developed in the TNBS/EtOH groups. The degree of

inflammation was such that the animals became febrile and most briefly developed

bloody diarrhea. These signs were transient and did not affect the animals' appetite at

any point. Some animals appeared slightly depressed for 12-24 hours, but this did not

persist. Unfortunately, one pig suffered a perforated rectum shortly after TNBS/EtOH

instillation, resulting in spontaneous death. The timing of this event was somewhat

unusual in that the peak inflammatory response following TNBS/EtOH instillation is

usually 2-3 days (Elson et al., 1995) but the animal died within 8 hours after the

procedure. Due to the transmural nature of the resultant colitis, the effective dose of

TNBS is close to the lethal dose in mice (Beagley et al., 1991). However, the currently

used dose of TNBS/EtOH did not cause systemic illness in any other pig used in the pilot

or main portions of the current study or those treated with ileal instillation in another

study in this laboratory (Merritt et al., 2002a). One possible confounding factor was the

use of inflated Foley catheters for enema retention. Because these catheters were inflated

to a set volume, rather than pressure, colonic contractions over an inflated balloon,









especially in the presence of a chemical irritant, could have resulted in rupture. The

timing of such an event more closely coincides with the animal's clinical picture than

rupture due to the TNBS alone. In future studies, use of a pressure transducer with a

"pop-off" valve attached to the Foley valve could avoid similar problems.

Endoscopy was an extremely useful tool for evaluation of ongoing distal colonic

and rectal inflammation and allowed scoring the severity of inflammation without animal

sacrifice. The duration of inflammation was similar to previous reports in mice (2-3

weeks) and dogs and rats (up to 8 weeks) (Elson et al., 1995; Shibata et al., 1993).

Because animals were not evaluated until one week after TNBS/EtOH instillation, we

likely did not capture the peak inflammatory response endoscopically. Individual

animals varied in both severity and duration of the lesion, typical of TNBS/EtOH colitis

(Elson et al., 1995). The development of lesions in all animals was more consistent than

that reported in dogs, but endoscopic lesion appearance was similar (Shibata et al., 1993).

Histological analysis at weeks 9 and 14 (5 and 10 weeks post-enema) did not reveal

significant differences between groups. Overall, colonic and rectal tissues had little to no

cellular infiltrate, and mucosal and submucosal edema was the only prominent

abnormality. If the colonic or rectal edema was related to TNBS/EtOH instillation, one

would expect to see the highest scores in the rectum of animals in that experimental

group sacrificed at 9 weeks. Because edema scores did not differ significantly between

groups, other contributing factors must be considered. Since tissue from all animals was

handled similarly, a processing artifact is unlikely. Repeated CRD studies could have

resulted in rectal irritation and edema. But, one would therefore expect a difference

between colonic and rectal edema and a difference between the 9 and 14 week animals as









a result of additional CRD procedures. These differences were not evident; thus, a

specific cause of the rectal edema could not be identified.

The presence of edema did not appear to have an impact on remaining study

variables. A significant correlation between edema and sensitivity among individual

animals was not identified, indicating that VH was not related to an ongoing

inflammatory process.

The other indicator of histological differences, number of lymphoid aggregates,

could be misleading. This number is dependent upon section location and angle as well

as normal variability throughout the gastrointestinal tract, and cellular infiltrate is more

commonly used as an indicator of disease (Kruschewski et al., 2001). But, because very

little cellular infiltrate was present in any of the sections, and the number of lymphoid

aggregates appeared to vary between tissues based on a cursory examination, this count

was used. Similar to the edema scoring, the number of lymphoid aggregates did not

differ significantly between groups nor did it appear to have a major impact on other

aspects of the study.

Visceral Discomfort

Pigs were easily trained to allow balloon insertion, and the ramp protocol for

balloon distention reliably induced a discomfort response in most trials. One of the

important factors in developing a pig model of visceral discomfort was the determination

of "discomfort". Previous animal models have used both subjective methods such as an

abdominal writhing response in rodents (Reichert et al., 2001) or more objective

measurements such quantification of abdominal muscle contractions (Al Chaer et al.,

2000; Coutinho et al., 2002; Ness and Gebhart, 1988) or a particular behavior (Merritt et

al., 2002b). A conscious passive avoidance behavior, such as pushing a lever to









discontinue the painful stimulus, has also been described in rats (Messaoudi et al., 1999;

Ness et al., 1991).

For CRD studies, we chose to evaluate discomfort subjectively, thus utilizing a

threshold of discomfort as our primary response variable. Most quantitative measures

involve a comparison of a particular variable between distention pressures rather than

determination of a threshold response. Such a method would have required repeated

painful colorectal distentions over the course of many weeks (Gschossmann et al., 2001;

Ness and Gebhart, 1990). We did not want to cause discomfort above threshold in our

subject animals for several reasons, including concern for their well-being. From a

physiologic standpoint, repeated painful CRD can increase spinalfos andjun proto-

oncogenes which, in turn, have been associated with central hyperexcitability (Traub et

al., 1992). Thus, we wanted to minimize the effect of repeated CRD procedures

themselves on visceral sensitivity by subjecting the animals to the fewest possible

number of painful CRD. Similarly, we were concerned that the pigs could develop

aversive behavior to the laboratory, crate, personnel, or testing procedure in general if we

used repeated painful distentions. Because stress alone can influence visceral nociception

(Bradesi et al., 2002; Coutinho et al., 2002), we wanted to minimize the physical and

psychological stress imposed on our animal subjects.

We considered using real-time abdominal myoelectrical data to contribute to the

threshold response determination. But, due to the nature of the pigs, we felt that

movement artifact would make interpretation difficult in a time-dependent situation.

Thus, in an attempt to limit the subjective nature of the behavioral observation technique,

strict criteria were used to define a discomfort response, and observers were blinded as to









treatment. Most pigs had an obvious discomfort response with agreement of all three

observers in 55% and 2/3 in 45% of trials. Some pigs were difficult to judge, especially

for the baseline trials when they were more nervous and less accustomed to the testing

procedure. In hindsight, obtaining an average baseline result from at least two, if not

three trials, may have allowed a more thorough evaluation of the animals' true baseline

responses.

Also in hindsight, equal sample sizes amongst all groups would have been

preferable to the current design. The pilot animals produced very consistent CRD data.

Because the main study was designed based upon their results, we presumed that saline-

treated groups would react similarly, thus allowing for a smaller sample size. Because

saline-treated animals exhibited more variation in CRD threshold response than expected,

larger control groups likely would have diminished the impact of individual variability

and increased the power of the study.

Effect of Inflammation on Nociceptive Threshold

Once a consistent model was established, a secondary goal of this study was to

investigate the effect of subacute inflammation on nociceptive threshold.

Weekly Nociceptive Thresholds

The two methods used to express visceral sensitivity, threshold pressure and ABTP,

describe the same data in slightly different ways. ABTP expresses the weekly thresholds

as a variation of the animal's own baseline threshold pressure, accounting for individual

variability in sensitivity. Similar comparisons have been used previously in human

visceral sensitivity testing (Sabate et al., 2002). For this reason, the ABTP results were

used for all correlations, and will be stressed during discussions of sensitivity.









The only significant difference in ABTP between experimental groups was at week

7, although the difference at week 13 approached significance (p=0.056). The mean

pressures for week 13 were also significantly different between groups. These results

suggest an interesting biphasic response. The initial period of hypersensitivity,

manifested by a decrease in threshold pressure relative to baseline, following

TNBS/EtOH instillation is likely related to colonic inflammation and its associated

mediators. This response is predictable based on previously documented effects of

gastrointestinal inflammation (Messaoudi et al., 1999; Ness and Gebhart, 2000; Sharkey

and Kroese, 2001). The second period of hypersensitivity occurred after the endoscopic

resolution of inflammation in all animals and following histologic resolution of

inflammation in experimental group 1. A post-inflammatory period of visceral

hypersensitivity supports the concept of plasticity within the afferent arm of the visceral

nociceptive pathway.

Some degree of individual variation in nociceptive threshold is expected (Elmer et

al., 1998), and such variability was observed in this study. When examining individual

responses, only 2/7 saline animals had >2 weeks with threshold pressures below baseline,

whereas 11/12 TNBS/EtOH animals had >2 weeks as such. Within the TNBS/EtOH-

treated groups, variability in the resultant inflammatory response may have contributed to

this individual variation. Week 6 and 9 endoscopy scores had a significant negative

correlation with week 13 ABTP (p=0.038 and 0.031, respectively; r2= 0.435 and 0.462,

respectively), indicating that those animals with the highest endoscopy scores required

less pressure to induce a nociceptive response. Because the r2 value was less than 0.5 for

each, this correlation should be considered a strong trend, rather than a significant









determination. Because a similar correlation did not exist between ABTP and

histological scoring, any effect of inflammation on visceral sensitivity appears related

more to the degree of inflammation initially present in each animal rather than the

remaining level of inflammation present at the time of testing (or in this case one week

later). This further supports the notion of neuronal plasticity rather than the direct action

of inflammatory mediators for the later period of hypersensitivity seen at week 13.

The CRD protocol used in this study evaluated threshold pressure, rather than

quantifying a particular pain response to a given stimulus. Thus, a reduction in sensory

threshold truly indicates allodynia, rather than hypersensitivity. Based on information in

other species, animals with allodynia were likely also hypersensitive, but this cannot be

proven given the constraints of the testing system used for this study (Mayer and

Gebhart, 1994).

The two saline-treated animals which consistently demonstrated negative ABTP

likely contributed to the lack of a group effect in many weekly sensitivity thresholds.

The sensitivity profiles of these two animals appeared to differ from the remaining saline-

treated animals. These animals had threshold pressures approximately 20 mmHg below

their initial baseline threshold at week 7 (the first CRD post enema) and then remained at

those thresholds (within 10 mmHg) for all subsequent weeks. Given the consistency of

this response, one potential explanation is that the reported baseline for those two animals

was erroneously high and the remaining weeks represented their normal threshold.

Alternatively, these animals became hypersensitive; possible causes include the saline

retention enema, soapy water enemas used prior to endoscopy, repeated endoscopic

examinations, or repeated CRD procedures. To the author's knowledge, none of these









have been reported as causes of visceral or somatic hypersensitivity, nor do they

represent "noxious" procedures. As stated previously, CRD to noxious pressures has

been reported to cause hypersensitivity, but not non-noxious distentions (Traub et al.,

1992). Normal individual variability in visceral sensory threshold is another possible

explanation. This is the more plausible explanation, and would likely coincide with an

erroneously high baseline reported in these animals.

Animal selection

Other investigators have recently described models involving various insults during

the neonatal period that resulted in visceral hypersensitivity during adulthood (Al Chaer

et al., 2000; Coutinho et al., 2002; Ruda et al., 2000). Similar findings have not, to the

author's knowledge, been extensively evaluated using a slightly older population of

animals. At the time of TNBS/EtOH enema administration, pigs in this study were

approximately 8-10 weeks old, corresponding to early adolescence in humans. This stage

of development may be important for the development of IBS (Sandler, 1990; Van

Ginkel et al., 2001), a concept further supported by the results of the present study.

Castrated male pigs were chosen for this study to avoid the any potential effect of

hormonal variation during the estrous cycle on nociceptive threshold. Gender differences

in both the perception and modulation of pain have been documented in humans (Gear et

al., 1996), and women are at a higher risk for the development of IBS (Mayer et al.,

1999). IBS-related pain can also vary with phase of the menstrual cycle (Heitkemper and

Jarrett, 1992). In animal models of pain, female rodents display lower nociceptive

thresholds to both shock and thermal stimuli (Marks and Hobbs, 1972; Pare, 1969;

Romero and Bodnar, 1986).









Furthermore, gonadectomy has been reported to decrease both nociceptive

threshold and response to analgesia in rodents (Marks and Hobbs, 1972; Romero et al.,

1988). Thus, one cannot assume that the results obtained from our castrated animals

would be the same as those obtained in intact male pigs or in females. However, since

they were castrated at a similar age, any potential effect of gonadectomy should have

been uniform across all animals.

Substance P

When attempting to quantify SP in various tissues, immunohistochemical analysis

offers several advantages over molecular analyses. Most importantly,

immunohistochemistry offers localization of immunoreactivity rather than protein

quantification in a particular tissue. For porcine tissue, specific anti-pig SP polyclonal

antibody has been raised (Balemba et al., 2001; Balemba et al., 2002), but it is not

commercially available. The present study used a rabbit polyclonal SP antibody that had

been validated previously at a similar dilution in porcine gastrointestinal tissue (Kulkarni-

Narla et al., 1999). Early antibodies to SP showed some cross-reactivity to NK-A or NK-

B, however more recently purified antibodies have apparently overcome this problem

(Duggan, 1995; Hoyle, 1998).

In the gastrointestinal tissue, the location chosen for Q-IHC sampling was based

upon the location of myenteric plexus (Goyal and Hirano, 1996). SP has previously been

identified in the mucosa and submucosal plexus in the porcine gastrointestinal tract

(Balemba et al., 2001; Balemba et al., 2002). During initial microscopic review of the

gastrointestinal tissues in this study, SP-IR was consistently seen in the myenteric plexus,

but not consistently in the areas of the other plexes. Some non-specific immunoreactivity

was seen in the mucosa, most prominently surrounding the tissue periphery. Thus, the









quantitative analysis described in this report focused on the region of the myenteric

plexus. For precise neuroanatomical detail in the enteric nervous system, whole mount

preparations of the gastrointestinal tissue are preferred (Balemba et al., 2001;

Miampamba and Sharkey, 1998). The transverse sections used in this study provide less

specific detail in that most neurons in the plexus are captured only in part or in cross-

section. However, as the purpose of the study was to evaluate the amount of SP-IR

present, rather than to map its distribution, this method met study requirements.

The spinal cord segments were chosen as a representative sample of those receiving

afferent input from the colon and rectum. The distal colon and rectum have dual sacral

and lumbar afferent innervation (Ness and Gebhart, 1990). Based on Fluorogold labeling

of the descending colon, DRG in the T13-L2 and L6-S2 regions received afferent input,

but the number of positive neurons in the T13-L2 region was greater following colonic

inflammation (Traub et al., 1999). Thus, spinal segments L1 and L2 were chosen to

represent the proximal extent of innervation, while L6 and L7 were chosen to represent

the distal portion.

The algorithm used to quantify SP-IR in the gastrointestinal tissues and spinal cord

dorsal horn has been previously validated for use with DAB-based

immunohistochemistry (Matkowskyj et al., 2000). This technique allows for precise

documentation of the chromagen content in a specific image by subtracting the

cumulative strength of the negative control image from that contained in the

corresponding immunostained image. Thus it allowed for a quantitative evaluation of the

SP-IR within specific locations in the spinal cord, colon, and rectum. Because the

chromagen content is expressed in arbitrary units, these data cannot be reasonably









compared to other studies. But, this provided a more objective comparison between

groups than a subjective scoring system.

Direct Relationship between Histopathology and Sensitivity

Based on an initial review of the gastrointestinal sections, very little cellular

infiltrate was noted. The only apparent abnormalities were mucosal and submucosal

edema. Also, the number of lymphoid aggregates appeared to vary between sections.

One explanation for this variation is the inherent variability due to section location and

angle of the cut. But, an analysis of the lymphoid aggregate counts was performed to

investigate the possibility of a treatment effect.

Although the correlation between the R2 lymphoid aggregate score and ABTP in

weeks 10 and 12 had significant Pearson's correlations, only the week 10 linear

regression had a r2 value >0.5. None of the correlations between histological scoring and

SP-IR in either the spinal cord or gastrointestinal tract had an r2 value >0.5, thus these

should be considered trends at best. Visual inspection of these plots does not give the

impression of a strong linear relationship (Fig. 4-8 and 4-13). Based on these facts and

the lack of correlation between ABTP and any edema scores, the histopathologic changes

seen within the gastrointestinal tract at the time of euthanasia did not appear to play a

significant role in visceral sensitivity.

Central versus Peripheral Sensitization

The lack of a difference in SP-IR in either the colon or rectum between any of the

experimental groups was somewhat unexpected. Other investigators have shown an

initial decrease and subsequent increase in SP-IR throughout the colon following

TNBS/EtOH instillation in rats (Miampamba and Sharkey, 1998) and in the primary

afferent nerves in a guinea pig TNBS ileitis model (Miller et al., 1993). These reported









changes occurred within 2 weeks following the inflammatory insult, and the long-term

effects of inflammation upon SP-IR have not been fully elucidated. Thus, while this

study did not evaluate the early effects of inflammation upon SP-IR, the information

presented for the periods 5 and 10 weeks after TNBS/EtOH instillation provide new

insight into the pathophysiology of inflammatory-mediated VH. Due to the relatively

small sample size and individual variation in inflammatory response, a small difference

between groups may have gone undetected. In addition to a lack of detectable difference

between groups, SP-IR in the gastrointestinal tract did not correlate significantly with

ABTP for any week. These results, in conjunction with the significant correlation

between dorsal horn SP-IR and ABTB, support a central rather than peripheral

mechanism of sensitization in these animals.

No significant difference was detected between experimental groups in either

dorsal horn SP-IR or ventral horn SP-IR neurons. Similar to the reasoning described for

gastrointestinal tissues, individual variability and sample size may have prevented

detection of a small difference between groups. Alternatively, groups truly did not differ

in immunoreactivity. Because VH did not develop in all of the TNBS/EtOH animals, the

correlation between weekly ABTP and SP-IR may provide more insight into the

relationship between SP in the spinal cord and the level of visceral sensitivity in a given

animal.

The significant correlation between dorsal horn SP-IR at the L1 and L7 segments

with multiple weekly ABTP supports a central mechanism of hypersensitivity. This

correlation was strongest between L7 and week 11, but also significant between L7 and

week 14 and between L1 and week 9. The argument for a role of SP in the development









of hypersensitivity in these animals would be stronger had SP-IR correlated significantly

with other weeks, especially when the TNBS/EtOH and saline groups differed

significantly. But, the correlation between SP-IR in the dorsal horn and ABTP across all

experimental groups further validates the importance of SP within the dorsal horn in the

development of central sensitization, regardless of the inciting cause.

SP has been associated with afferent transmission of nociception for years (Hokfelt

et al., 1977a; Mayer and Raybould, 1990; Otsuka and Yoshioka, 1993). Not surprisingly,

SP is thought to play a role in central sensitization and inflammatory-mediated VH

(Kishimoto, 1994; Miampamba et al., 1992; Persson et al., 1995; Schneider et al., 2001;

Swain et al., 1992). Zymosan-induced colitis reduced the number of SP-labeled cells in

both the T13-L2 and L6-S2 afferent DRG in rats (Traub et al., 1999). Also, TNBS ileitis

has been shown to induce hyperexcitability in nociceptive DRG neurons.(Moore et al.,

2002) In the present study, SP-IR in the dorsal horn was greatest in the superficial

laminae (I and II) which receive afferent input, consistent with previous reports (Duggan,

1995; Kawata et al., 1989; Routh and Helke, 1995). The strongest correlation with ABTP

in the lumbar spinal cord was at L7, corresponding to lumbosacral afferent input from the

distal colon and rectum.

Given the previously suggested importance of SP in the development of central

sensitization and the changes previously associated with gastrointestinal inflammation,

alterations in SP-IR were expected in this study. However, a specific correlation between

visceral sensitivity in individual animals and SP-immunoreactivity in the spinal cord

dorsal horn has not, to our knowledge, been previously documented. This association

between SP in the dorsal horn and the lack of an association with SP-IR in the colon or









rectum highlights the role of SP in central sensitization and the importance of this process

in post-inflammatory visceral hypersensitivity.

In the ventral horn, the correlation between SP-IR neurons and ABTP, especially at

the L2 segment implies an alteration in motor pathways. However, these interpretations

should be considered cautiously because the r2 for all linear regressions were less than

0.51, and visual inspection of the correlation does not imply a consistent linear

relationship. SP-IR has been documented in large motoneurons of the ventral horn, but

an alteration in ventral horn SP-IR has not previously been attributed to rectal or colonic

inflammation (Charlton and Helke, 1985b; Charlton and Helke, 1985a). However, given

the appearance of the linear regression plots and the function of the large ventral horn

motoneurons, the true relevance of the correlation between ventral horn SP-IR neurons

and visceral sensitivity requires further investigation.

Conclusions

Current Study

The main accomplishment of this study was to integrate models of colonic

inflammation and visceral pain in the pig. The use of a large animal subject, such as the

pig, allowed for endoscopic scoring of gross mucosal changes within the distal colon and

rectum. This provided an evaluation of each animal's individual response to

TNBS/EtOH instillation, rather than relying on a group average based on histopathologic

changes seen in animals sacrificed at various time points. This information allowed for a

more thorough characterization of individual responses while also decreasing the required

number of animals. Because of the variability in response to TNBS/EtOH, the

endoscopic information proved useful, in that the resultant colitis severity scores

following instillation correlated negatively with ABTP.









TNBS/EtOH instillation resulted in a biphasic pattern of visceral hypersensitivity.

The first period occurred in the presence of ongoing inflammation, but the second period

occurred after the gross resolution of inflammation in the subject animals and the

histological resolution of inflammation in cohorts euthanized at week 9. These results,

combined with a lack of consistent correlation between ABTP and histological scoring,

suggest that either central and/or peripheral mechanisms of hypersensitization, rather than

inflammatory mediators, appear responsible for observed alterations in visceral

sensitivity.

Although TNBS/EtOH instillation did not have a significant effect on spinal or

colonic SP-immunoreactivity, a strong correlation between dorsal horn SP-IR at the L1

and L7 levels and ABTP was evident. Due to the small sample size and variation within

each group, this correlation more likely represents the true involvement of spinal SP in a

visceral hypersensitivity response, consistent with central sensitization. The inclusion of

both saline- and TNBS/EtOH-treated animals in these correlations suggests the

importance of SP in VH regardless of the inciting cause, an unexpected but intriguing

result.

Future Studies

Based upon results of this study, the pig may serve as an excellent large animal

model for future studies concerning the pathophysiology and treatment of inflammatory-

mediated IBS. One of the first options would be to determine the immunoreactivity of

other mediators of central sensitization, such as CGRP, NMDA, NGF, and others, within

the spinal cords of the pigs in this study. These studies could be easily accomplished as

paraffin-embedded tissue from all animals has been retained. Combined with the ABTP,

endoscopic scores, histologic scores, and SP-IR, further immunohistochemical









characterization of this model could contribute to a more thorough understanding of the

post-inflammatory neuroplastic changes described in this model.

The porcine CRD model could be used to evaluate the effect of potential

therapeutic agents on visceral sensitivity thresholds, either with or without a previous

inflammatory insult. In such studies, endoscopic evaluation would allow for

classification of the subjects based on lesion severity.

Due to the chronic nature of IBS and the temporal separation between insult and

documentation of VH in neonatal rodent models, studies of longer duration are also

warranted. For example, a porcine study similar to that described in this report, but that

extended to adulthood, would more fully characterize the duration and nature of

alterations in visceral sensitivity. If the bi-phasic pattern of VH seen in this study

persisted in a relapsing/remitting fashion, the model would offer a similar clinical picture

to that seen in IBS. Furthermore, one could evaluate the effect of repeated versus single-

dose TNBS/EtOH instillations on the visceral sensitivity pattern to determine whether or

not a repetitive inflammatory insult further increases the severity or duration of VH.
















APPENDIX A
INDIVIDUAL ANIMAL DATA

Table A-1. Raw data from CRD studies.


An# Group Wk3 Wk7 Wk8 Wk9 Wk Wk Wk Wk Wk
10 11 12 13 14
3161 3 45 15 25 25 15 15 25 25 25
3171 1 55 25 15 25
3162 4 25 45 25 25 25 15 15 25 45
3185 1 35 35 25 45
3197 1 55* 45 45 35
3184 4 15 55* 35 35 55 55* 55 25 45
3191 3 45 25 15 25 35 25 15 15 25
John 1 45 25 15 25
Red 1 35 35 15 25
3166 1 35 15 15 25
3198 2 25 25 15 35
3179 2 25 25 25 xx
3182 4 55* 25 25 15 25 25 15 45 45
3189 2 15 35 15 15
3190 3 55* 15 55* 55* 55 35 15 15 35
3193 4 55* 25 15 15 15 25 15 45 35
3192 3 45 25 25 35 55 35 25 15 45
3130 3 25 15 15 25 15 55* 15 15 55*
3126 3 15 15 35 35 45 35 45 15 25












Table A-2. Raw data from endoscopic examinations.


Animal # Group Wk 3 Wk 6 Wk 9 Wk 14
3171 1 0.5 3.0 0.5
3185 1 0.5 5.0 2.0
3197 1 0.5 2.0 0.5
John 1 0.5 3.0
Red 1 0.0 2.0
3166 1 0.0 3.0 0.5
3198 2 0.5 0.5 0.5
3179 2 1.0 0.5 0.0
3189 2 0.0 1.5 0.5
3161 3 0.0 3.5 1.0
3191 3 0.5 4.0 1.0
3190 3 0.5 2.0 1.0 0.0
3192 3 0.0 2.0 0.0 0.5
3130 3 0.5 2.0 0.5 0.0
3126 3 0.5 1.0 0.0 0.5
3162 4 0.5 0.0 0.0
3184 4 1.0 0.5 0.0
3182 4 0.5 0.0 0.0 0.0
3193 4 0.5 0.0 0.0


















APPENDIX B
ANOVA TABLES

Table B-1. One-way ANOVA analysis for threshold pressure.


Week Groups Sum of Squares df Mean Square F Sig.
3 Between 452.6942 1 452.6942 2.222232 0.154358
Within 3463.095 17 203.7115
Total 3915.789 18
7 Between 391.0401 1 391.0401 3.361862 0.084292
Within 1977.381 17 116.3165
Total 2368.421 18
8 Between 36.09023 1 36.09023 0.261874 0.615419
Within 2342.857 17 137.8151
Total 2378.947 18
9 Between 289.7243 1 289.7243 3.172748 0.09275
Within 1552.381 17 91.31653
Total 1842.105 18
10 Between 106.6667 1 106.6667 0.330323 0.581267
Within 2583.333 8 322.9167
Total 2690 9
11 Between 26.66667 1 26.66667 0.119626 0.73836
Within 1783.333 8 222.9167
Total 1810 9
12 Between 6.666667 1 6.666667 0.028319 0.870539
Within 1883.333 8 235.4167
Total 1890 9
13 Between 806.6667 1 806.6667 13.35172 0.006457
Within 483.3333 8 60.41667
Total 1290 9
14 Between 135 1 135 1.234286 0.298847
Within 875 8 109.375
Total 1010 9










Table B-2. One-way ANOVA analysis for ABTP.


Week Groups Sum of df Mean F Sig.
Squares Square
7 Between 1687.269 1 1687.269 4.679076 0.04507
Within 6130.179 17 360.5987
Total 7817.447 18
8 Between 226.6931 1 226.6931 0.737567 0.402389
Within 5224.991 17 307.3524
Total 5451.684 18
9 Between 16.34226 1 16.34226 0.048641 0.828073
Within 5711.658 17 335.9799
Total 5728 18
10 Between 86.4 1 86.4 0.114968 0.743287
Within 6012.125 8 751.5156
Total 6098.525 9
11 Between 15.50417 1 15.50417 0.019584 0.892164
Within 6333.396 8 791.6745
Total 6348.9 9
12 Between 13.06667 1 13.06667 0.014309 0.907734
Within 7305.458 8 913.1823
Total 7318.525 9
13 Between 866.4 1 866.4 4.978863 0.05618
Within 1392.125 8 174.0156
Total 2258.525 9
14 Between 160.0667 1 160.0667 0.329147 0.581927
Within 3890.458 8 486.3073










Table B-3. One-way ANOVA analysis for endoscopy scores.


Week Groups Sum of df Mean F Sig.
Squares Square
3 Between 0.122807 1 0.122807 1.252632 0.278616
Within 1.666667 17 0.098039
Total 1.789474 18
6 Between 21.56031 1 21.56031 24.88431 0.000112
Within 14.72917 17 0.866422
Total 36.28947 18
9 Between 1.278151 1 1.278151 5.545698 0.032564
Within 3.457143 15 0.230476
Total 4.735294 16
14 Between 0.125 1 0.125 3 0.133975
Within 0.25 6 0.041667
Total 0.375 7










Table B-4. One-way ANOVA analysis for gastrointestinal histologic scores.

Section Group Sum of Squares df Mean Square F Sig.
C1LYM Between 1.3640 3 0.4547 0.4377 0.7293
Within 15.5833 15 1.0389
Total 16.9474 18
C1ED Between 2.2895 3 0.7632 0.9954 0.4218
Within 11.5000 15 0.7667
Total 13.7895 18
C1TOT Between 0.8640 3 0.2880 0.1347 0.9379
Within 32.0833 15 2.1389
Total 32.9474 18
C2LYM Between 0.2061 3 0.0687 0.1076 0.9544
Within 9.5833 15 0.6389
Total 9.7895 18
C2ED Between 2.2412 3 0.7471 0.8676 0.4795
Within 12.9167 15 0.8611
Total 15.1579 18
C2TOT Between 2.7895 3 0.9298 0.8204 0.5026
Within 17.0000 15 1.1333
Total 19.7895 18
R1LYM Between 2.1140 3 0.7047 0.5074 0.6831
Within 20.8333 15 1.3889
Total 22.9474 18
R1ED Between 5.0482 3 1.6827 2.1791 0.1330
Within 11.5833 15 0.7722
Total 16.6316 18
R1TOT Between 10.3202 3 3.4401 1.2767 0.3182
Within 40.4167 15 2.6944
Total 50.7368 18
R2LYM Between 1.1930 3 0.3977 0.3890 0.7626
Within 15.3333 15 1.0222
Total 16.5263 18
R2ED Between 1.7982 3 0.5994 0.4316 0.7334
Within 20.8333 15 1.3889
Total 22.6316 18
R2TOT Between 4.7982 3 1.5994 0.6695 0.5838
Within 35.8333 15 2.3889
Total 40.6316 18











Table B-5. Tukey's HSD analysis for gastrointestinal histologic scores.

Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95% CI-Lower 95% CI-Upper
C1LYM 1 2 -0.1667 0.7207 0.9955 -2.2439 1.9106
3 0.5000 0.5885 0.8300 -1.1961 2.1961
4 -0.0833 0.6579 0.9992 -1.9796 1.8129
2 1 0.1667 0.7207 0.9955 -1.9106 2.2439
3 0.6667 0.7207 0.7921 -1.4106 2.7439
4 0.0833 0.7785 0.9995 -2.1603 2.3270
3 1 -0.5000 0.5885 0.8300 -2.1961 1.1961
2 -0.6667 0.7207 0.7921 -2.7439 1.4106
4 -0.5833 0.6579 0.8118 -2.4796 1.3129
4 1 0.0833 0.6579 0.9992 -1.8129 1.9796
2 -0.0833 0.7785 0.9995 -2.3270 2.1603
3 0.5833 0.6579 0.8118 -1.3129 2.4796
C1ED 1 2 -0.1667 0.6191 0.9929 -1.9511 1.6178
3 -0.8333 0.5055 0.3831 -2.2903 0.6237
4 -0.5000 0.5652 0.8128 -2.1290 1.1290
2 1 0.1667 0.6191 0.9929 -1.6178 1.9511
3 -0.6667 0.6191 0.7084 -2.4511 1.1178
4 -0.3333 0.6687 0.9582 -2.2608 1.5941
3 1 0.8333 0.5055 0.3831 -0.6237 2.2903
2 0.6667 0.6191 0.7084 -1.1178 2.4511
4 0.3333 0.5652 0.9336 -1.2956 1.9623
4 1 0.5000 0.5652 0.8128 -1.1290 2.1290
2 0.3333 0.6687 0.9582 -1.5941 2.2608
3 -0.3333 0.5652 0.9336 -1.9623 1.2956
C1TOT 1 2 -0.3333 1.0341 0.9880 -3.3139 2.6472
3 -0.3333 0.8444 0.9784 -2.7669 2.1003
4 -0.5833 0.9440 0.9248 -3.3042 2.1375
2 1 0.3333 1.0341 0.9880 -2.6472 3.3139
3 0.0000 1.0341 1.0000 -2.9805 2.9805
4 -0.2500 1.1170 0.9959 -3.4694 2.9694
3 1 0.3333 0.8444 0.9784 -2.1003 2.7669
2 0.0000 1.0341 1.0000 -2.9805 2.9805
4 -0.2500 0.9440 0.9932 -2.9709 2.4709
4 1 0.5833 0.9440 0.9248 -2.1375 3.3042
2 0.2500 1.1170 0.9959 -2.9694 3.4694
3 0.2500 0.9440 0.9932 -2.4709 2.9709
C2LYM 1 2 0.1667 0.5652 0.9907 -1.4623 1.7956
3 0.1667 0.4615 0.9833 -1.1634 1.4967
4 -0.0833 0.5159 0.9984 -1.5704 1.4037
2 1 -0.1667 0.5652 0.9907 -1.7956 1.4623
3 0.0000 0.5652 1.0000 -1.6290 1.6290
4 -0.2500 0.6105 0.9760 -2.0095 1.5095
3 1 -0.1667 0.4615 0.9833 -1.4967 1.1634
2 0.0000 0.5652 1.0000 -1.6290 1.6290
4 -0.2500 0.5159 0.9614 -1.7370 1.2370
4 1 0.0833 0.5159 0.9984 -1.4037 1.5704
2 0.2500 0.6105 0.9760 -1.5095 2.0095
3 0.2500 0.5159 0.9614 -1.2370 1.7370











Table B-5. Continued


Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95%CI-Lower 95%CI-Upper
C2ED 1 2 0.1667 0.6562 0.9940 -1.7245 2.0578
3 -0.5000 0.5358 0.7878 -2.0441 1.0441
4 -0.7500 0.5990 0.6052 -2.4764 0.9764
2 1 -0.1667 0.6562 0.9940 -2.0578 1.7245
3 -0.6667 0.6562 0.7428 -2.5578 1.2245
4 -0.9167 0.7087 0.5807 -2.9594 1.1260
3 1 0.5000 0.5358 0.7878 -1.0441 2.0441
2 0.6667 0.6562 0.7428 -1.2245 2.5578
4 -0.2500 0.5990 0.9746 -1.9764 1.4764
4 1 0.7500 0.5990 0.6052 -0.9764 2.4764
2 0.9167 0.7087 0.5807 -1.1260 2.9594
3 0.2500 0.5990 0.9746 -1.4764 1.9764
C2TOT 1 2 0.3333 0.7528 0.9700 -1.8363 2.5029
3 -0.3333 0.6146 0.9472 -2.1048 1.4381
4 -0.8333 0.6872 0.6286 -2.8139 1.1472
2 1 -0.3333 0.7528 0.9700 -2.5029 1.8363
3 -0.6667 0.7528 0.8123 -2.8363 1.5029
4 -1.1667 0.8131 0.4983 -3.5101 1.1768
3 1 0.3333 0.6146 0.9472 -1.4381 2.1048
2 0.6667 0.7528 0.8123 -1.5029 2.8363
4 -0.5000 0.6872 0.8845 -2.4806 1.4806
4 1 0.8333 0.6872 0.6286 -1.1472 2.8139
2 1.1667 0.8131 0.4983 -1.1768 3.5101
3 0.5000 0.6872 0.8845 -1.4806 2.4806
R1LYM 1 2 -0.5000 0.8333 0.9305 -2.9018 1.9018
3 0.5000 0.6804 0.8816 -1.4611 2.4611
4 0.1667 0.7607 0.9961 -2.0259 2.3592
2 1 0.5000 0.8333 0.9305 -1.9018 2.9018
3 1.0000 0.8333 0.6361 -1.4018 3.4018
4 0.6667 0.9001 0.8792 -1.9276 3.2609
3 1 -0.5000 0.6804 0.8816 -2.4611 1.4611
2 -1.0000 0.8333 0.6361 -3.4018 1.4018
4 -0.3333 0.7607 0.9709 -2.5259 1.8592
4 1 -0.1667 0.7607 0.9961 -2.3592 2.0259
2 -0.6667 0.9001 0.8792 -3.2609 1.9276
3 0.3333 0.7607 0.9709 -1.8592 2.5259
R1ED 1 2 -0.3333 0.6214 0.9488 -2.1242 1.4576
3 0.5000 0.5074 0.7597 -0.9623 1.9623
4 -0.9167 0.5672 0.3996 -2.5515 0.7182
2 1 0.3333 0.6214 0.9488 -1.4576 2.1242
3 0.8333 0.6214 0.5526 -0.9576 2.6242
4 -0.5833 0.6712 0.8205 -2.5177 1.3511
3 1 -0.5000 0.5074 0.7597 -1.9623 0.9623
2 -0.8333 0.6214 0.5526 -2.6242 0.9576
4 -1.4167 0.5672 0.1010 -3.0515 0.2182
4 1 0.9167 0.5672 0.3996 -0.7182 2.5515
2 0.5833 0.6712 0.8205 -1.3511 2.5177
3 1.4167 0.5672 0.1010 -0.2182 3.0515











Table B-5. Continued


Section (I) Grp (J) Grp Mean Diff (I-J) Std. Error Sig. 95%CI-Lower 95%CI-Upper
R1TOT 1 2 -0.8333 1.1607 0.8884 -4.1786 2.5120
3 1.0000 0.9477 0.7207 -1.7314 3.7314
4 -0.7500 1.0596 0.8924 -3.8038 2.3038
2 1 0.8333 1.1607 0.8884 -2.5120 4.1786
3 1.8333 1.1607 0.4187 -1.5120 5.1786
4 0.0833 1.2537 0.9999 -3.5300 3.6967
3 1 -1.0000 0.9477 0.7207 -3.7314 1.7314
2 -1.8333 1.1607 0.4187 -5.1786 1.5120
4 -1.7500 1.0596 0.3815 -4.8038 1.3038
4 1 0.7500 1.0596 0.8924 -2.3038 3.8038
2 -0.0833 1.2537 0.9999 -3.6967 3.5300
3 1.7500 1.0596 0.3815 -1.3038 4.8038
R2LYM 1 2 0.5000 0.7149 0.8957 -1.5605 2.5605
3 0.0000 0.5837 1.0000 -1.6824 1.6824
4 -0.3333 0.6526 0.9553 -2.2143 1.5476
2 1 -0.5000 0.7149 0.8957 -2.5605 1.5605
3 -0.5000 0.7149 0.8957 -2.5605 1.5605
4 -0.8333 0.7722 0.7070 -3.0589 1.3923
3 1 0.0000 0.5837 1.0000 -1.6824 1.6824
2 0.5000 0.7149 0.8957 -1.5605 2.5605
4 -0.3333 0.6526 0.9553 -2.2143 1.5476
4 1 0.3333 0.6526 0.9553 -1.5476 2.2143
2 0.8333 0.7722 0.7070 -1.3923 3.0589
3 0.3333 0.6526 0.9553 -1.5476 2.2143
R2ED 1 2 0.0000 0.8333 1.0000 -2.4018 2.4018
3 0.1667 0.6804 0.9946 -1.7944 2.1277
4 -0.6667 0.7607 0.8169 -2.8592 1.5259
2 1 0.0000 0.8333 1.0000 -2.4018 2.4018
3 0.1667 0.8333 0.9970 -2.2351 2.5685
4 -0.6667 0.9001 0.8792 -3.2609 1.9276
3 1 -0.1667 0.6804 0.9946 -2.1277 1.7944
2 -0.1667 0.8333 0.9970 -2.5685 2.2351
4 -0.8333 0.7607 0.6976 -3.0259 1.3592
4 1 0.6667 0.7607 0.8169 -1.5259 2.8592
2 0.6667 0.9001 0.8792 -1.9276 3.2609
3 0.8333 0.7607 0.6976 -1.3592 3.0259
R2TOT 1 2 0.5000 1.0929 0.9671 -2.6499 3.6499
3 0.1667 0.8924 0.9976 -2.4052 2.7386
4 -1.0000 0.9977 0.7504 -3.8755 1.8755
2 1 -0.5000 1.0929 0.9671 -3.6499 2.6499
3 -0.3333 1.0929 0.9897 -3.4833 2.8166
4 -1.5000 1.1805 0.5942 -4.9023 1.9023
3 1 -0.1667 0.8924 0.9976 -2.7386 2.4052
2 0.3333 1.0929 0.9897 -2.8166 3.4833
4 -1.1667 0.9977 0.6543 -4.0421 1.7088
4 1 1.0000 0.9977 0.7504 -1.8755 3.8755
2 1.5000 1.1805 0.5942 -1.9023 4.9023
3 1.1667 0.9977 0.6543 -1.7088 4.0421











Table B-6. One-way ANOVA for ventral horn SP-immunoreactive neurons.

Section Groups Sum of Squares df Mean Square F Sig.
L1 Between 128.4035 3.0000 42.8012 1.1882 0.3477
Within 540.3333 15.0000 36.0222
Total 668.7368 18.0000
L2 Between 257.0482 3.0000 85.6827 2.0811 0.1457
Within 617.5833 15.0000 41.1722
Total 874.6316 18.0000
L6 Between 253.2895 3.0000 84.4298 0.6434 0.5990
Within 1968.5000 15.0000 131.2333
Total 2221.7895 18.0000
L7 Between 5.6111 3.0000 1.8704 0.1250 0.9438
Within 209.5000 14.0000 14.9643
Total 215.1111 17.0000











Table B-7. Tukey's HSD analysis for ventral horn SP-immunoreactive neurons.

Section (I) (J) Mean Diff(I-J) SEM Sig. 95%CI-Lower 95%CI-Upper
Grp Grp
L1 1 2 1.5000 4.2439 0.9843 -10.7317 13.7317
3 -5.3333 3.4652 0.4403 -15.3205 4.6538
4 -2.5000 3.8742 0.9156 -13.6660 8.6660
2 1 -1.5000 4.2439 0.9843 -13.7317 10.7317
3 -6.8333 4.2439 0.4026 -19.0650 5.3984
4 -4.0000 4.5840 0.8188 -17.2117 9.2117
3 1 5.3333 3.4652 0.4403 -4.6538 15.3205
2 6.8333 4.2439 0.4026 -5.3984 19.0650
4 2.8333 3.8742 0.8830 -8.3326 13.9993
4 1 2.5000 3.8742 0.9156 -8.6660 13.6660
2 4.0000 4.5840 0.8188 -9.2117 17.2117
3 -2.8333 3.8742 0.8830 -13.9993 8.3326
L2 1 2 0.8333 4.5372 0.9977 -12.2435 13.9102
3 -7.1667 3.7046 0.2557 -17.8439 3.5105
4 -6.9167 4.1419 0.3724 -18.8542 5.0208
2 1 -0.8333 4.5372 0.9977 -13.9102 12.2435
3 -8.0000 4.5372 0.3279 -21.0769 5.0769
4 -7.7500 4.9007 0.4177 -21.8746 6.3746
3 1 7.1667 3.7046 0.2557 -3.5105 17.8439
2 8.0000 4.5372 0.3279 -5.0769 21.0769
4 0.2500 4.1419 0.9999 -11.6875 12.1875
4 1 6.9167 4.1419 0.3724 -5.0208 18.8542
2 7.7500 4.9007 0.4177 -6.3746 21.8746
3 -0.2500 4.1419 0.9999 -12.1875 11.6875
L6 1 2 4.1667 8.1004 0.9544 -19.1799 27.5133
3 -4.1667 6.6140 0.9208 -23.2291 14.8957
4 5.0000 7.3946 0.9045 -16.3124 26.3124
2 1 -4.1667 8.1004 0.9544 -27.5133 19.1799
3 -8.3333 8.1004 0.7357 -31.6799 15.0133
4 0.8333 8.7494 0.9997 -24.3839 26.0505
3 1 4.1667 6.6140 0.9208 -14.8957 23.2291
2 8.3333 8.1004 0.7357 -15.0133 31.6799
4 9.1667 7.3946 0.6126 -12.1458 30.4791
4 1 -5.0000 7.3946 0.9045 -26.3124 16.3124
2 -0.8333 8.7494 0.9997 -26.0505 24.3839
3 -9.1667 7.3946 0.6126 -30.4791 12.1458
L7 1 2 1.0000 2.7354 0.9826 -6.9505 8.9505
3 -0.5000 2.2334 0.9959 -6.9915 5.9915
4 0.6667 2.7354 0.9947 -7.2838 8.6171
2 1 -1.0000 2.7354 0.9826 -8.9505 6.9505
3 -1.5000 2.7354 0.9455 -9.4505 6.4505
4 -0.3333 3.1585 0.9996 -9.5138 8.8471
3 1 0.5000 2.2334 0.9959 -5.9915 6.9915
2 1.5000 2.7354 0.9455 -6.4505 9.4505
4 1.1667 2.7354 0.9730 -6.7838 9.1171
4 1 -0.6667 2.7354 0.9947 -8.6171 7.2838
2 0.3333 3.1585 0.9996 -8.8471 9.5138
3 -1.1667 2.7354 0.9730 -9.1171 6.7838










Table B-8. One-way ANOVA for SP-immunoreactivity (QIHC


Section Groups Sum of Squares df Mean Square F Sig.
L1 Between 8487.5138 3 2829.1713 1.4157 0.2772
Within 29975.3478 15 1998.3565
Total 38462.8616 18
L2 Between 1542.5863 3 514.1954 0.2947 0.8286
Within 26170.6220 15 1744.7081
Total 27713.2084 18
L6 Between 4541.2806 3 1513.7602 0.6128 0.6172
Within 37056.2933 15 2470.4196
Total 41597.5739 18
L7 Between 9247.0372 3 3082.3457 1.0665 0.3947
Within 40460.8227 14 2890.0588
Total 49707.8599 17
R1 Between 7141.7124 3 2380.5708 1.1332 0.3695
Within 29411.3091 14 2100.8078
Total 36553.0216 17
R2 Between 6336.3136 3 2112.1045 0.7050 0.5647
Within 41940.1497 14 2995.7250
Total 48276.4633 17
C1 Between 7798.2417 3 2599.4139 0.5905 0.6313
Within 61626.3441 14 4401.8817
Total 69424.5858 17
C2 Between 31907.1178 3 10635.7059 2.3118 0.1326
Within 50607.6520 11 4600.6956
Total 82514.7698 14


- EU/pixel)











Table B-9. Tukey's HSD analysis for spinal SP-Immunoreactivity (QIHC EU/pixel)


Section (I) (J) Mean SEM Sig. 95% CI- 95% CI-
Grp Grp Diff(I-J) Lower Upper
L1 1 2 46.5000 31.6098 0.4778 -44.6041 137.6041
3 -1.7317 25.8093 0.9999 -76.1179 72.6545
4 39.9089 28.8557 0.5281 -43.2574 123.0752
2 1 -46.5000 31.6098 0.4778 -137.6041 44.6041
3 -48.2317 31.6098 0.4475 -139.3358 42.8724
4 -6.5911 34.1425 0.9973 -104.9948 91.8126
3 1 1.7317 25.8093 0.9999 -72.6545 76.1179
2 48.2317 31.6098 0.4475 -42.8724 139.3358
4 41.6406 28.8557 0.4936 -41.5257 124.8068
4 1 -39.9089 28.8557 0.5281 -123.0752 43.2574
2 6.5911 34.1425 0.9973 -91.8126 104.9948
3 -41.6406 28.8557 0.4936 -124.8068 41.5257
L2 1 2 13.3519 29.5356 0.9682 -71.7742 98.4780
3 -13.6559 24.1157 0.9406 -83.1610 55.8493
4 -1.4372 26.9622 0.9999 -79.1464 76.2719
2 1 -13.3519 29.5356 0.9682 -98.4780 71.7742
3 -27.0078 29.5356 0.7976 -112.1339 58.1183
4 -14.7892 31.9021 0.9659 -106.7359 77.1575
3 1 13.6559 24.1157 0.9406 -55.8493 83.1610
2 27.0078 29.5356 0.7976 -58.1183 112.1339
4 12.2186 26.9622 0.9680 -65.4905 89.9278
4 1 1.4372 26.9622 0.9999 -76.2719 79.1464
2 14.7892 31.9021 0.9659 -77.1575 106.7359
3 -12.2186 26.9622 0.9680 -89.9278 65.4905
L6 1 2 2.7550 35.1456 0.9998 -98.5397 104.0497
3 4.0541 28.6962 0.9989 -78.6527 86.7609
4 39.8828 32.0834 0.6106 -52.5862 132.3518
2 1 -2.7550 35.1456 0.9998 -104.0497 98.5397
3 1.2991 35.1456 1.0000 -99.9956 102.5938
4 37.1278 37.9615 0.7638 -72.2830 146.5386
3 1 -4.0541 28.6962 0.9989 -86.7609 78.6527
2 -1.2991 35.1456 1.0000 -102.5938 99.9956
4 35.8287 32.0834 0.6852 -56.6403 128.2977
4 1 -39.8828 32.0834 0.6106 -132.3518 52.5862
2 -37.1278 37.9615 0.7638 -146.5386 72.2830
3 -35.8287 32.0834 0.6852 -128.2977 56.6403
L7 1 2 -18.7320 38.0135 0.9594 -129.2209 91.7570
3 -3.0398 31.0379 0.9996 -93.2536 87.1741
4 -63.3003 38.0135 0.3768 -173.7892 47.1886
2 1 18.7320 38.0135 0.9594 -91.7570 129.2209
3 15.6922 38.0135 0.9754 -94.7967 126.1811
4 -44.5683 43.8943 0.7434 -172.1499 83.0133
3 1 3.0398 31.0379 0.9996 -87.1741 93.2536
2 -15.6922 38.0135 0.9754 -126.1811 94.7967
4 -60.2605 38.0135 0.4175 -170.7494 50.2284
4 1 63.3003 38.0135 0.3768 -47.1886 173.7892
2 44.5683 43.8943 0.7434 -83.0133 172.1499
3 60.2605 38.0135 0.4175 -50.2284 170.7494











Table B-10. Tukey's HSD for gastrointestinal SP-Immunoreactivity (QIHC EU/pixel)

Section (I) (J) Mean SEM Sig. 95% CI- 95% CI-
Grp Grp Diff(I-J) Lower Upper
R1 1 2 6.2406 32.4099 0.9973 -87.9611 100.4422
3 -7.1699 27.7542 0.9937 -87.8394 73.4995
4 -47.9365 29.5861 0.3994 -133.9305 38.0574
2 1 -6.2406 32.4099 0.9973 -100.4422 87.9611
3 -13.4105 33.4728 0.9774 -110.7015 83.8806
4 -54.1771 35.0067 0.4374 -155.9265 47.5723
3 1 7.1699 27.7542 0.9937 -73.4995 87.8394
2 13.4105 33.4728 0.9774 -83.8806 110.7015
4 -40.7666 30.7468 0.5626 -130.1341 48.6009
4 1 47.9365 29.5861 0.3994 -38.0574 133.9305
2 54.1771 35.0067 0.4374 -47.5723 155.9265
3 40.7666 30.7468 0.5626 -48.6009 130.1341
R2 1 2 55.9847 38.7022 0.4930 -56.5059 168.4754
3 18.7542 31.6002 0.9324 -73.0940 110.6024
4 23.9025 38.7022 0.9248 -88.5881 136.3932
2 1 -55.9847 38.7022 0.4930 -168.4754 56.5059
3 -37.2306 38.7022 0.7726 -149.7212 75.2601
4 -32.0822 44.6895 0.8883 -161.9752 97.8108
3 1 -18.7542 31.6002 0.9324 -110.6024 73.0940
2 37.2306 38.7022 0.7726 -75.2601 149.7212
4 5.1484 38.7022 0.9991 -107.3423 117.6390
4 1 -23.9025 38.7022 0.9248 -136.3932 88.5881
2 32.0822 44.6895 0.8883 -97.8108 161.9752
3 -5.1484 38.7022 0.9991 -117.6390 107.3423
C1 1 2 40.7222 48.4528 0.8344 -100.1090 181.5534
3 -14.4775 40.1749 0.9833 -131.2486 102.2936
4 -20.5413 44.5067 0.9662 -149.9030 108.8205
2 1 -40.7222 48.4528 0.8344 -181.5534 100.1090
3 -55.1997 46.9142 0.6506 -191.5589 81.1595
4 -61.2635 50.6731 0.6315 -208.5483 86.0213
3 1 14.4775 40.1749 0.9833 -102.2936 131.2486
2 55.1997 46.9142 0.6506 -81.1595 191.5589
4 -6.0638 42.8266 0.9989 -130.5421 118.4146
4 1 20.5413 44.5067 0.9662 -108.8205 149.9030
2 61.2635 50.6731 0.6315 -86.0213 208.5483
3 6.0638 42.8266 0.9989 -118.4146 130.5421
C2 1 2 69.7556 51.8048 0.5548 -86.1535 225.6647
3 36.8474 45.5007 0.8486 -100.0891 173.7839
4 -66.0022 51.8048 0.5964 -221.9113 89.9069
2 1 -69.7556 51.8048 0.5548 -225.6647 86.1535
3 -32.9082 49.5349 0.9083 -181.9858 116.1694
4 -135.7578 55.3817 0.1241 -302.4316 30.9160
3 1 -36.8474 45.5007 0.8486 -173.7839 100.0891
2 32.9082 49.5349 0.9083 -116.1694 181.9858
4 -102.8496 49.5349 0.2199 -251.9272 46.2280
4 1 66.0022 51.8048 0.5964 -89.9069 221.9113
2 135.7578 55.3817 0.1241 -30.9160 302.4316
3 102.8496 49.5349 0.2199 -46.2280 251.9272
















APPENDIX C
CORRELATIONS

Table C-1. Correlation between weekly ABTP and histological scores.


R1Ed R2Ed C1Ed C2Ed R1Lym R2Lym C1Lym C2Lym
WK7 Pearson 0.5748 0.0016 0.5309 -
0.2071 0.2221 0.3288 0.0819 0.1467
Sig. 0.1055 0.9967 0.1414 0.5929 0.5656 0.3876 0.8342 0.7064
N 9 9 9 9 9 9 9 9
WK8 Pearson 0.3911 0.5000 -
0.0343 0.3129 0.2786 0.4649 0.1237 0.1789
Sig. 0.2980 0.9301 0.1705 0.4124 0.4678 0.2073 0.7513 0.6451
N 9 9 9 9 9 9 9 9
WK9 Pearson 0.1835 0.4941 0.0226 -
0.3763 0.1318 0.3527 0.0580 0.3731
Sig. 0.6365 0.3182 0.1764 0.9540 0.7354 0.3519 0.8821 0.3227
N 9 9 9 9 9 9 9 9
WK10 Pearson 0.1248 -
0.1127 0.0591 0.2116 0.0964 0.7156 0.2305 0.0367
Sig. 0.7566 0.8712 0.5573 0.7911 0.7312 0.0200 0.5217 0.9198
N 10 10 10 10 10 10 10 10
WK11 Pearson 0.0179 0.0062 0.2922 -
0.3151 0.0490 0.5014 0.4092 0.1950
Sig. 0.9609 0.9865 0.3751 0.8930 0.4126 0.1398 0.2403 0.5892
N 10 10 10 10 10 10 10 10
WK12 Pearson 0.2052 0.1763 0.1953 0.1483 -
0.1177 0.6532 0.4057 0.4032
Sig. 0.5696 0.6262 0.7461 0.5888 0.6827 0.0405 0.2448 0.2480
N 10 10 10 10 10 10 10 10
WK13 Pearson 0.5357 0.3173 0.3515 0.2669 -
0.0761 0.2148 0.2030 0.3455
Sig. 0.1105 0.3717 0.8346 0.3193 0.4560 0.5513 0.5738 0.3281
N 10 10 10 10 10 10 10 10
WK14 Pearson 0.2509 0.1209 0.1958 0.5130 -
0.0467 0.2266 0.2653 0.0735
Sig. 0.4844 0.7394 0.8982 0.5877 0.1294 0.5289 0.4588 0.8402
N 10 10 10 10 10 10 10 10










Table C-2. Correlation between endoscopic histological scores.


ENDO6 ENDO9 ENDO14
R1ED Pearson -0.2507 -0.0909 -0.6325
Sig. 0.3006 0.7286 0.1778
N 19 17 6
R2ED Pearson 0.0119 -0.1300 -0.7071
Sig. 0.9613 0.6189 0.1161
N 19 17 6
C1ED Pearson 0.0576 0.1240 0.3162
Sig. 0.8147 0.6354 0.5415
N 19 17 6
C2ED Pearson -0.0819 -0.0731 0.0000
Sig. 0.7389 0.7803 1.0000
N 19 17 6
R1LYM Pearson -0.0310 -0.0473 -0.3162
Sig. 0.8997 0.8571 0.5415
N 19 17 6
R2LYM Pearson 0.0946 0.0711 -0.9258
Sig. 0.7002 0.7864 0.0080
N 19 17 6
C1LYM Pearson -0.1570 -0.0574 -0.5000
Sig. 0.5208 0.8267 0.3125
N 19 17 6
C2LYM Pearson 0.0908 0.1718 -0.6124
Sig. 0.7118 0.5096 0.1963
N 19 17 6