Recipient Organization
LOUISIANA STATE UNIVERSITY
202 HIMES HALL
BATON ROUGE,LA 70803-0100
Performing Department
Veterinary Clinical Sciences
Non Technical Summary
Expected ResultsIt is anticipated that horses with endoscopic evidence of mild (Grade 1) EGGD will havemild lymphoplasmacytic inflammation, characterized predominantly by T cell infiltration, that isrelated to SI inflammation. It is expected that lymphoplasmacytic gastritis will be associatedwith decreased numbers of regulatory T cells. It is anticipated that horses with severe EGGD(Grade ≥ 2) will have ulceration and severe inflammation, predominantly neutrophilic. It isanticipated that no currently available endoscopic scoring system will correlate to severity ofinflammation as assessed via histopathology. If no presently available scoring system correlateswith histopathologic findings, then a new system will be created. Findings from this study willallow for further investigations into mucosal immune response.Pitfalls and LimitationsDue to the nature of trying to compare endoscopic findings on electively euthanized horses,exact history of horses (e.g., deworming history) may not be collected, and findings ofinflammation within the stomach or small intestine could be associated with medicationadministration or parasitism, rather than spontaneous EGGD or IBD. As horses are presentingfor elective euthanasia, they could have underlying disease that may confound interpretation ofany findings of gastric or intestinal inflammation. Nevertheless, this population should besufficient to allow for initial evaluation of the relationship of glandular inflammatory cellinfiltrate, endoscopic findings, and SI inflammation. Another limitation of this study is that alimited evaluation of T cell immunophenotype is being undertaken. If evidence of T-cellinfiltration is documented, then additional T cell phenotyping will be undertaken.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
Equine gastric ulcer syndrome (EGUS) is a common condition of horses, with describedclinical signs including abdominal pain (colic), loss of appetite, and decreased performance.Recently, because the horse has a compound stomach consisting of non-glandular (squamous)and glandular regions, lesions of the stomach have been further differentiated into equinesquamous gastric disease (ESGD) or equine glandular gastric disease (EGGD) based uponrecognition that causes of the two disorders may differ (1). Although endoscopy is considered tobe the gold standard for diagnosis, available endoscopic scoring systems (1-4) have not beenvalidated for EGGD.A large body of data exists on mechanisms and healing of ESGD; however, thepathogenesis of EGGD remains poorly understood. Recent data suggests that EGGD is poorlyresponsive to standard antiulcer treatment (proton-pump inhibitors [PPIs], e.g. omeprazole)compared to ESGD, suggesting acid injury alone is unlikely to be a primary cause of EGGD (5,6). Two studies in a small number of horses have suggested that inflammatory cell infiltrationmay be associated with lesions found in EGGD (7, 8). Helicobacter bacteria, a common causeof gastritis (inflammation) in dogs and humans, has not been demonstrated to be associated withEGGD in horses. However, a proportion of people with gastritis are Helicobacter-negative. ThisHelicobacter-negative gastritis is primarily lymphocytic in nature, and is strongly associatedwith small intestinal (SI) inflammation, suggesting an immune-mediated component (9, 10).Evaluation of the relationship of endoscopic findings with histopathologic findings is keyto developing accurate EGGD scoring systems. Furthermore, characterizing the inflammatorycell infiltrate in EGGD and its relationship to SI inflammation is critical for development ofeffective, targeted treatment.
Project Methods
Description of the Investigation/Techniques to be EmployedAnimals: This study has been approved by the Louisiana State University Institutional AnimalCare and Use Committee (LSU IACUC #: 18-053). For this study, a power calculation wasperformed using G*Power (http://www.gpower.hhu.de). In a previous study, inflammatory cellinfiltrate score (0= none, mild =1, moderate = 2, severe = 3) was evaluated and found to be(mean ± SD) 0.9 ± 0.6 in horses without evidence of gross lesions and 1.4 ± 0.6 in horses withlesions (8). Using a β=0.80 and ??=0.05, it is estimated that 40 horses will be needed. Therefore,40 horses donated and/or presenting for elective euthanasia will be included. All horses willundergo physical examination to evaluate health status.Endoscopy: Horses will be sedated with xylazine (0.4 mg/kg, IV) following a 16-18 h fast andgastroscopy will be performed using a 3-m endoscope (Karl Storz, Inc., Goleta, CA, USA).Lesions (EGGD) will be scored using the three available scoring systems (1-4) by investigators(FMA, HEB). Horses will be euthanized, and stomachs removed, and rinsed with water. Lesionsize, appearance, and number will be recorded. Since the pylorus is the area that is mostcommonly affected by EGGD (11), samples will be collected from normal-appearing portions ofthe pylorus (2 cm2 from both dorsal and ventral pylorus), as previously described (8). Inaddition, any area of the glandular mucosa with lesions and a sample of the duodenum, midjejunum,and ileum will be collected and processed as described for normal-appearing mucosa.Histopathology: Samples will be fixed in formalin for 48-72 hours. Samples will be paraffinembedded, sectioned, and stained with hematoxylin and eosin for histopathologic evaluation.Inflammation severity will be scored using a previously described scoring system (8).Additional characteristics, including presence of erosion or ulcer, edema, fibrosis, metaplasia,and glandular atrophy will be recorded. Small intestinal sections (duodenum, jejunum, ileum)will also be evaluated and scored for inflammation and any other intestinal pathology (dilation oflacteals or intestinal crypts, villous blunting, fibrosis). Investigators (FDP, TW) evaluatinghistopathologic appearance will be blinded to the endoscopic appearance.Immunohistochemistry: Lymphocyte phenotyping Among horses with glandular gastriclymphocytic infiltrate (n=10 mild and n=10 moderate-severe), immunohistochemical staining offormalin-fixed samples will be used to evaluate T cell (CD3) and B cell (CD20) populationsusing methods and antibodies previously validated for use in the horse. In brief, samples will beplaced in 10 mM citrate buffer, pH 6, for heat-induced antigen retrieval. Samples will be blockedwith 3% hydrogen peroxide, rinsed, and blocked with goat serum. A 1:400 dilution CD3(Agilent, Santa Clara, CA) or CD20 (ThermoFisher, Waltham, MA) polyclonal antibody will beapplied. After rinsing, a biotinylated IgG (Vector Laboratories, Burlingame, CA) will beapplied. Samples will be detected with avidin/biotin complex (Vector Laboratories, Burlingame,CA) and rinsed. An isotype-matched antibody will be used as a negative control.Samples willbe stained with Nova Red (Vector Laboratories) and counter-stained with hematoxylin. Allrinsing steps will be undertaken with 0.05% TBS-Tween. Number of cells expressing CD3 orCD20 will be averaged over 10 randomly chosen 40x power fields. Small intestinal samplesfrom the horses with mild (n=10) and moderate-severe (n=10) lymphocytic infiltration in theglandular mucosa will similarly have B and T lymphocyte immunohistochemistry performed asdescribed above, in order to evaluate the relationship between lymphocyte subtypes between theglandular gastric mucosa and SI mucosa.Regulatory T cell phenotyping: Immunohistochemistry for regulatory T cell phenotyping will beperformed as previously described (19) in the subset of horses with mild (n=10) or moderatemarkedlymphocytic infiltrate of the glandular mucosa a Foxp3 antibody previously validated foruse in the horse (21). Immunohistochemistry for regulatory T cells will be performed inglandular and SI (duodenum, jejunum, ileum) mucosa of the same subset of horses, to evaluatethe relationship between regulatory T cell populations between the glandular gastric mucosa andjejunal mucosa. In brief, heat-induced antigen retrieval will be undertaken in citrate buffer, pH6.0, 121 °C. Samples will be blocked with 3% hydrogen peroxide. Non-specific binding will beblocked using 5% skim milk in TBS-Tween. Samples will be incubated with a Foxp3monoclonal antibody (Thermo Fisher, Waltham, MA) overnight at 4 °C. Samples will beincubated with a biotin-labeled secondary antibody (Vector Laboratories, Burlingame, CA),followed by HRP-labeled steptavidin (Agilent, Santa Clara, CA). The chromogen 3,3'-diaminobenzidine will be used to reveal Foxp3 immunoreactivity. An isotype-matched antibodywill be used as a negative control. Samples will be stained with Nova Red and counterstainedwith hematoxylin. All rinsing steps will be undertaken with 0.1% TBS-Tween. Number of cellsexpressing Foxp3 will be averaged over 10 randomly chosen 40x power fields.Analysis of ResultsStatistical analysis will be performed using SPSS v. 24 (SPSS Corp, Chicago, IL). Type ofinflammation in glandular mucosa and SI mucosa will be described based upon identified celltype, including characterization of Foxp3 regulatory T-cells. Percentage of horses with and 5without other histopathologic abnormalities of the glandular gastric mucosa (e.g., mucosalulceration, fibrosis, edema) will be described. Relationships between severity ofhistopathological findings and severity of EGGD (using the three available scoring systems) willbe evaluated using a Spearman's coefficient of correlation. Severity of glandular inflammationbetween horses with and without EGGD will be compared with a Mann-Whitney U test.Proportions of regulatory T cells in horses with lymphoplasmacytic EGGD and without EGGDwill be compared using a Chi square test. Relationships between other histologic findings (e.g.,edema, fibrosis) and cell type will be assessed in both the glandular mucosa and SI using a Chisquare test. Relationships between severity of gastric and SI inflammation will be assessed usinga Spearman's correlation coefficient. Proportions of regulatory T cells in the glandular gastricmucosa and SI mucosa of horses with lymphoplasmacytic EGGD and without EGGD will becompared using a Chi square test.