Dynamic Monitoring of Cross-Bred Dogs for Translational Research in Ulcerative Colitis

Background: Ulcerative colitis (UC)is a complex, chronic, immu ne-mediated inammatory disease. The preclinical studies and assessment of UC drugs remain a major challenge. Dogs share similar environmental, genomic, anatomical, and intestinal physiologic features with humans. However, the model of UC in dogs is not understood and lacks systematic assessment. Methods: We developed a model for UC in the cross-bred dogs by different dose of acetic acid retention enema. Simultaneously, physiological pathologic indicators, including blood routine examination, inammatory factor and colonoscopy, were monitored continuously for 14 days. Furthermore, olsalazine was used to treatment UC to evaluate the feasibility of the model. Results: The percentage of lymphocytes increases after induction UC by 10% acetic acid; C-reactive protein (CRP) and Cyclooxygenase-2(COX-2) are mainly inammatory markers after induction UC by 10% or 7.5% acetic acid; acetic acid aggravates mucosal damage in a dose-dependent manner, and OLZ prevents 10% acetic acid-induced UC. The acetic acid concentration is controlled between 7.5 % and 10%, and the continuous damage of this model is 10 days. Conclusions: We suggest that cross-bred dogs represent a useful clinical model of UC, and it is suitable for translational research. Finally, we provide a model and dynamic endoscope atlas for UC in dogs.


Background
Ulcerative colitis (UC)is a complex, chronic, immune-mediated in ammatory disease, and is clinically characterized by abdominal cramping, pain, and diarrhea [1]. its incidence is increasing [2]. The potential causes of UC are vague and complicated, including aberrant cytokine secretion [3,4], dysregulated immune responses [5,6], and alterations in barrier function [7] and intestinal microbiota [8]. Although therapeutic strategy for UC has exhibited on anti-in ammatory agents, including glucocoticoids, salicylates and corticosteroids. So far, there is no cure for UC. Animal models in rats or mice which colitis-like symptoms are induced through challenge with toxins or chemicals such as carrageenan [9], 2,4,6-trinitrobenzenesulfonic acid (TNBS), dextran sodium sulfate (DSS), oxazolone [10] or acetic acid [11] have been instrumental in understanding the in ammatory processes of UC. Despite the widespread use of rodent models in biomedical research, there is still controversial in the translational value of mouse studies for human disease [12]. The preclinical studies and assessment of UC drugs remains a major challenge as these models has limitations and drawbacks to compare with clinical diagnosis UC. The internal characteristics of the colon are di cult to analyze dynamically and intuitively through colonoscopy, and the e cacy of drugs in these models cannot be accurately assessed.
Large animal models, such as dogs, are generally more representative than rodents because the anatomical and physiological characteristics of dogs are closer to humans, and they have a relatively large body size, longer life span. and develop spontaneous, similar diseases, such as in ammatory bowel disease (IBD) [13].Previous studies have shown that miniature pigs [14] and dogs [15,16] were induced UC model by acetic acid retention enema and it has been as a new approach for drug-evaluation. In addition, the dog remain considerable to be a superior non-rodent mammalian animal model for preclinical study and is preferred by the FDA for initial safety data of drugs for human application [17], such as colonoscopy. Importantly, it is more conducive to the dynamic study of the mechanism and assessment of treated-UC drugs. However, the model of UC in dogs is not understood and lacks systematic assessment. In an attempt to overcome these problems, we develop a model for UC in the cross-bred dogs by different dose of acetic acid retention enema. Simultaneously, physiological pathologic indicators, including blood routine examination, in ammatory factor and colonoscopy, were monitored continuously for 14 days. Furthermore, olsalazine, a new derivative of 5-aminosalicylic acid, which entered the body as prodrug form, was used to treatment UC to evaluate the feasibility of the model.
These results may provide some evidences for evaluation of treated-UC drug.

Animals
All animals(13dogs) handling including anesthesia; surgical procedures, post-operative care and sacri ce has been approved by the animal care and use Committee (IACUC) of Dali university, China(ethics permission number: 2018-0627, year: 2018). From October 2017, 1 to 2 years old, -fteen male (8-12 kg) small-sized cross-bred dogs were obtained from the Laboratory Animal Center a liated with Dali University, Dali, China. All dogs were housed individually and fed standard food throughout the experiment. The dogs were initially evaluated for any illness by physical examination and laboratory screening. The appropriate amount of dog food is given, and the body temperature, respiratory rate, stool frequency, stool traits, and mental state are recorded every. All sections of this report adhere to the ARRIVE Guidelines for reporting animal research [18,19]. A completed ARRIVE guidelines checklist is included in Checklist S1.

Induction of colitis
Routine blood examination is a basic screening tool for serious pathology, which has reference value for many diseases. To provide a dynamic parameters of blood routine to clinical diagnosis, we observed continuously blood routine for 14 days (Fig.1A). Compared with the 0 day, WBC (Fig.1B), MO (Fig.1E) and granulocyte( Fig.1F) were sharply elevated at 1 day after 10%(P<0.0001), 7.5(P<0.01) or 5% (P<0.01)acetic acid, and remained constantly to 14 days. Comparably, RBC (Fig.1C) and HGB (Fig.1F) mildly exhibited downtrend at 1or 4 days after induction UC by 10% or 7.5% or 5% acetic acid, and not altered much to 14 days. Until 10 days, markedly, there was a dose-dependent manner in LY proliferation after 7.5% or 10% acetic acid-induced UC compare to 0 day (Fig.1D). The routine blood examination of 10%, 7.5% or 5% acetic acid-induced UC have slightly changed after 10day, which have self-healing ability in UC model caused by acetic acid. These observations indicated that UC is associated with LY proliferation, and LY may use an indicator of diagnosis for UC.

Monitor of routine blood parameters and in ammatory cytokines
Blood routine parameters, including granulocyte(GR), hemoglobin(HGB), and lymphocyte(LY), monocyte(MO), red blood cell(RBC)and white blood cell(WBC) , in ammatory cytokines were dynamically monitored. In the procedure for the UC model, blood was collected according to the timeline shown in

Colonoscopy
Timeline of colonoscopy are also illustrated schematically (Fig.1A). Bowel preparation was performed on the day of induction. In detail, before colonoscopy all dogs were orally given 10 % MgSO 4 (5 mL/kg, in distilled water) in order to clean the intestines. All animals were anaesthetized with xylazine hydrochloride injection (0.02 mL/kg), and then normal saline is used to ush the end of the colon. The electronic endoscope was slowly inserted into the colon about 20 cm, and then the camera was moved to transverse colon, descending colon or sigmoid colon to observe colonic mucosa dynamically. Colonic mucosa was assessed using published protocols [20,21] with some modi cations (Table 1). Dense in ammatory cell in ltration Table 1 The variables used for microscopic scoring H&E staining The model we explored decreased the appetite and emotion of animal and resulted in severe and persistent pain. In addition, the colonic mucosa biopsy could not accurately judge the pathological characteristics of tissue. Thus, all animals were euthanized with xylazine hydrochloride injection (0.2 mL/kg) and dissected to further investigated the pathological changes via H&E staining. In detail, Two weeks following post-induction of UC, the dogs were euthanized with xylazine hydrochloride injection (0.2mL/kg), samples were provided from mucosa in 10 cm proximal to the anal verge and moved to 4% paraformaldehyde(4%PFA) (Tianjin Fuchen Chemical Reagents Factory, China, Number: 20150408) for histological studies. Brie y, colons were collected, post-xed(4%PFA,12-24h), and then dehydrated by Automatic dehydrator (Leica ASP-300S, Germany), embedded in para n (Biological tissue embedding machine, Xiaogan Hongye Medical Instrument Co., Ltd., model: BM-VIII). Subsequently, the coronal sections of colon (5µm) were prepared by a rotatory microtome (Leica RM2245, Germany). The sections were subsequently stained with hematoxylin and eosin (H&E). Macroscopic damage of colon was evaluated as described by Mehrabani et al reported [22]( Table 2).
Score Macroscopic morphology 0 The mucosa is pale with normal vascular pattern and without mucosa hyperemia, edema ,1 The mucosa is still smooth but mucosa hyperemia, edema and the refractive index enhanced 2 The mucosa is hyperemia, edema, and granularity. The mucosa is friability with contact bleeding 3 The mucosa becomes congested and rough mucus membrane with Mucosal edema, spontaneous bleeding or contact bleeding. Erosions, ulcers are observed 4 The mucosa becomes extensively congested and rough mucus membrane with mucosal edema, marked and spontaneous bleeding or contact bleeding, erosions, ulcers are obviously observed Table 2 Criteria for scoring of macroscopic damage Drug treatment Olsalazine(OLZ), a new derivative of 5-aminosalicylic acid(Tianjin Lisheng Pharmaceutical Co., Ltd. China, Number: 1608005), is an anti-in ammatory agent was used as positive control to evaluate model. 6 male crossbred dogs were used for induction of UC by 10% acetic acid. The experiment was carried out by self-comparison, OLZ(35.6mg/kg, in distilled saline)were continuously given by intragastric administration (i.g.) for 10 day at the beginning of UC. All indicators are evaluated as described Fig.1A.

Statistical analysis
The statistical analysis was performed with Graph Pad Prism 7 software. Homogeneity of variance was tested using Levene's, and if not statistically signi cant (P >0.05), one-way ANOVA was used for statistical analysis. Kruskal-Wallis tests were performed if the variance was not consistent (P <0.05). If the Kruskal-Wallis test was statistically signi cant (P <0.05), the Mann-Whitney method was further used to make any pairwise comparison between the means. Data were expressed as mean±S. E. M. (standard error of mean) or SD (standard deviation).

Result
The percentage of lymphocytes increases after induction UC by 10% acetic acid Routine blood examination is a basic screening tool for serious pathology, which has reference value for many diseases. To provide a dynamic parameters of blood routine to clinical diagnosis, we observed continuously blood routine for 14 days (Fig.1A). Compared with the 0 day, WBC(  1F) mildly exhibited downtrend at 1 or 4 days after induction UC by 10% or 7.5% or 5% acetic acid, and not altered much to 14 days. Until 10 days, markedly, there was a dose-dependent manner in LY proliferation after 7.5% or 10% acetic acid-induced UC compare to 0 day (Fig.1D). The routine blood examination of 10%, 7.5% or 5% acetic acid-induced UC have slightly changed after 10 day, which have self-healing ability in UC model caused by acetic acid. These observations indicated that UC is associated with LY proliferation, and LY may use an indicator of diagnosis for UC.
COX-2 and CPR were mainly in ammatory markers after induction UC by 10% or 7.5% acetic acid To explore the effect of acetic acid with different concentrations on the in ammatory response in different periods of ulcerative colitis, the expression pro les of proin ammatory and anti-in ammatory factors were measured. At 0 day, the release of the proin ammatory factors were no statistical difference. At 1 day after UC, the release of the proin ammatory factors in addition to the CRP, including IL-1 , TNFα, iNOS and COX-2 were not elevated in the 10%,7% or 5% acetic acid group.
At 4or 7 days after UC, the secretion level of IL-1 was signi cantly increased in 10% acetic acid group(P < 0.01, P<0.05, respectively, versus BV or 7% or 5% acetic acid group)( Fig.2A). At 7 days after UC, the secretion of TNF-α reached a peak in 10% acetic acid group(P<0.05) (Fig.2B). Compared with the 0 days, there was no signi cance in acetic acid with different dosage (Fig.2C) (Fig.2D). Moreover, besides 5% acetic acid group, CRP was continuously elevated for 14 days (versus 0 day, all P<0.0001) as shown in Fig.2E. However, acetic acid with different dosage did not affect the secretion of anti-in ammatory mediators, such as IL-10. The present observation indicated that COX-2 and CPR were mainly in ammatory markers after induction UC by 10% or 7.5% acetic acid.
Acetic acid aggravates mucosal damage in a dosedependent manner The invasion site UC starts in the rectum and generally extends proximally in a continuous manner through part of, or the entire, colon. Therefore, transverse colon, descending colon or sigmoid were selected to endoscopy as shown in Fig.3A,C. Before inducing UC with acetic acid(at 0 day, all dogs), normal colonic mucosa appears smooth and red with no injury. In contrast, following induction of the UC model, especially at 1,4,7or10 days, colonic mucosa became widely congested and ulcers and edema were observed along with the development of rough mucus membrane with mucosal erosion (Fig.3C), which was markedly in 10% acetic acid group. Besides sigmoid and descending colon, 5% acetic acid group had a less obvious manifestation than 7% or 10% acetic acid group (Fig.3C). Although different doses of acetic acid extensively caused colonic mucosal damage, especially descending colon and sigmoid colon, but recovered at 14 days in 5% or 7% acetic acid group (Fig.3C).
Endoscopy and histopathology are crucially for determination disease activity in UC [23,24].As is shown in Fig.4B, the structure of the submucosa, muscularis, and adventitia of the colonic mucosa is clear, the mucosa epithelium is integrated, and the goblet cells are clearly visible, there is no in ammatory cell in ltration in control group. But, at 14 days after UC, HE staining revealed in ltration of in ammatory cells in mucosa and around the crypts that were polymorphonuclear leukocytes and lymphocytes. Multiple ulcerations were also observed in 10% acetic acid group, and indicated that there was the presence of a crypt abscess. Multifocal areas of ulceration and in ammation were present in the submucosa and it was widely edematous. Group 5%, 7.5% or 10% acetic acid had different degrees of injury. Among them,10% acetic acid group was the most severe(P<0.05) (Fig.4A). The above results indicated that acetic acid aggravates mucosal damage in a dose-dependent manner.
OLZ prevents 10% acetic acid-induced UC In addition to its own comparison, it is also compared with the model group. Same changes of WBC, MO and GR were seen between group model and group OLZ-treatment ( Fig.5A-a, A-d, A-e). As previously mentioned (Fig.1C), the percentage of lymphocytes increases after induction UC by 10% acetic acid, which was reduced by OLZ at 7 days(P<0.0001). Remarkable decreased RBC was also observed in the OLZ group at 1, 4 or 7 days after UC(P < 0.0001, versus10% acetic acid). However, there is no practical signi cance, because base value of the RBC of OLZ group is relatively low (Fig.5A-b).At 7 or 10 days HGB was lower than 10% acetic acid group (Fig.5A-f).
As shown in Fig.5C, following OLZ-treatment after UC, mucosal injury, including ulcers, edema and mucosa hyperemia or bleeding were gradually ameliorated. However, there is no cure for this model. Compared with model group, CMDI was reduced in OLZ-treated group at 10 days after UC(P<0.05).

Discussion
Acetic acid-induced UC is one of the acknowledged experimental model to study UC [25]. The induction of this animal model is associated with prolonged neutrophil in ltration. The mucosal barrier structure is destroyed by acetic acid administration, and then the in ammation is started, which are similar to the in ammatory properties of arachidonic acid metabolism abnormality in human colitis. Structural weakening of the colonic mucus barrier is an early event in ulcerative colitis pathogenesis [26]. It has been widely used to study the effects of drugs for UC [25,27]. Many reports showed that mice or rats as experimental models of UC [28] [11,29]. But, dogs share similar environmental, genomic, anatomical, and intestinal physiologic features with humans [30].In this paper, we suggest that cross-bred dogs represents a useful clinical model of UC.
Here described investigations, the rst of this study was to the dynastic variation of blood routine and in ammatory response in dogs. In ammation in in ammatory bowel disease (IBD) is sustained by an exaggerated response of lymphocytes [31]. Studies have shown that the intestinal lymphocytes and peripheral blood lymphocytes is increased,which were early activation markers in UC patients [32]. All peripheral immune cells migrated [33] and recirculated from the intestine to the blood,which was an important factor in UC pathogenesis. This is supported by our research, which showed that the percentage of lymphocytes gradually increases in dogs after UC by 10% acetic acid. Lymphocytes, including effector memory T cells are endlessly recirculating by blood, tissues and lymphatics, and surveying for antigens. Subsequently, engagement with their homologous antigens trigger the proliferation and activation of T cells, with quickly transform into cytokine-producing effectors, which sustain recruitment and activation of other immune cells, thereby initiating, spreading,and amplifying in ammation [34][35][36]. Blocking of T cell tra cking is a highly attractive therapeutic strategy in IBD. We detected therapeutic effect of OLZ has found that the percentage of lymphocytes increases after induction UC by 10% acetic acid, which was reduced by at 7 days. Besides at 1 day parameters of blood routine, including WBC, RBC, MO, GR, HGB, were increased sharply after acetic acid-UC, changes were slight at 4,7,10 or 14 days. Previous research has exhibited the presence of a correlation of both peripheral blood mononuclear and neutrophils cells with disease activity. Undoubtedly, neutrophils and macrophages are interestingly associated with IBD pathophysiology [37]. And monocytes also be detected early in IBD as biomarkers of in ammation [38,39]. For this model, we also discovered MO were rapidly elevated at 1 day after different doses of acetic acid retention enema. Taken together, the increased Ly ratio is continuous, but increased MO is only in the early stages of UC.
Cytokine-driven immune networks crucially have a crucial role in the pathogenesis of UC, where they control multiple aspects of the in ammatory response [40]. Intrinsic layer dendritic cells (DCs) and macrophages are the main antigen presenting cells (APCs) found in the in amed mucosa of IBD. Following activation, which occurs in response to combination of Toll-like receptor (TLR) signaling and the commensal microbiota, these cells produce excessive pro-in ammatory factors, such as IL-1β, IL-6, IL-18 and TNF [41].In this paper, the release of TNF-α and IL-1 were continuously measured for 14 days, but only increase their secretion in the middle stage of disease activity for 10% acetic acid-induced UC. In addition, loss of IL-10 signal transduction in severe infant-onset IBD was involved in production of IL-1 by macrophages that resulted in activation of CD4+ T cells [42]. In our present study, there was no changes of IL-10 after UC by different acetic acid doses-induced. Expression of in ammatory proteins which include COX-2 and iNOS are believed to play a key role in regulation in ammation [43,44]. Our ndings agree with reported showing that throughout remarkable elevated COX-2 was found in UC using 7% or 10 acetic acid-induced. However, poor activation of iNOS produces in this model. CRP is measure that is used to determine the severity of UC [45][46][47].Additionally, patients exhibiting with acute severe ulcerative colitis (ASUC) usually have fatal systemic in ammation as evidenced by an elevated CRP and conspicuous hypoalbuminemia. Evidence from dynamically measurement of CRP content in our system, signi cant CRP raised was continuously observed in acetic acid UC for 14 days. Activation of CRP and COX-2 produces abundant in ammatory mediators which may conduce to the development of intestinal damage and additionally, COX-2 acts in synergy with CPR to accelerate the in ammatory response. Our data revealed that COX-2 and CPR were mainly in ammatory markers after induction UC by 10% or 7.5% acetic acid. OLZ, as a positive drug, can only reduce the secretion of CRP and IL-1 at 7 or 10 days after UC. Mucosal evaluation is independently essential in UC. In clinical trials of UC, endoscopic evaluation of mucosal disease activity is often used to determine eligibility and response to treatment. However, preclinical trials don't do enteroscopy in rodents result in exiting systems for new drug appaisement. Perhaps this is why the drug for UC is effective in preclianical, not clinically. Endoscopic evaluation is the current gold standard to assess mucosal lesions [48]. In this paper, we provide a model and dynamic endoscope atlas for UC in dogs. We performed dynamic monitoring of enteroscopy on the transverse colon, descending colon or sigmoid colon. Edema, erythema, loss of vascularity, mucosal granularity and friability, ulcers and erosions were observed by endoscopic ndings. At 1 day, these ndings typically begin sigmoid colon, and at 4,7,10 or 14 days, these manifestations extend proximally to transverse colon in a continuous manner with a gradual transition to normal-appearing mucosa. However, there was more seriously pathological phenomena in the 10% acetic acid-induced UC. Additionally, HE staining revealed in ltration of in ammatory cells in mucosa and around the crypts that were polymorphonuclear leukocytes and lymphocytes. Multiple ulcerations were also observed in 10% acetic acid group, and indicated that there was the presence of a crypt abscess. Multifocal areas of ulceration and in ammation were present in the submucosa and it was widely edematous.
The data above demonstrate that 10% acetic acid-induced UC in cross-bred dogs is more suitable for preclinical drug evaluation. First-line therapy in mild to moderate disease is the 5-ASA drugs [49], which can be administered as suppositories, enemas, or oral formulations, such as OLZ. We tested OLZ as a positive drug in our model system to verify the success of the acetic acid-induced UC. Our data con rm that OLZ prevents 10% acetic acid-induced UC.
Finally, the present study has certain limitations. Firstly, we found that The proportion of lymphocytes increased was not classi ed for T cell subtypes. Secondly, only sigmoid colon was stained by pathology. Thirdly, evidence that acetic acid-induced UC may occur self-healing came from colonoscopy. We speculate that the 10% acetic acid-induced model is ASUC.

Conclusion
In summary, we suggest that cross-bred dogs represent a useful clinical model of UC. The acetic acid concentration is controlled between 7.5 % and 10%, and the continuous damage of this model is 10 days. The standard for the success of the model is marked by steadily increased lymphocytes and CRP or COX-2. Finally, we provide a model and dynamic endoscope atlas for UC in dogs. These ndings support drug-evaluation for UC.

Availability of data and materials
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
We declare that there is no nancial con ict of interest in this research work.

Funding
This study is supported by several fundings as the following: the National Natural Science Fund The funding bodies provided nancial support and had no role in the design of the study, in the collection, analysis and interpretation of data or in writing the manuscript.