Metabolomics And Functional MRI Study Of Obstructive Chronic Pancreatitis

Chronic pancreatitis (CP) is the pancreatic tissue persistent damnification and function change caused by sorts of uncertain pathogenesis, which characterized as pancreatic fibrosis. In recent years, as people’s lifestyle and diet structure change gradually, the prevalence of CP is increasing worldwide. Eventually there will be irreversible pancreatic morphological changes such as pancreatic duct dilation, pancreatic duct stones or calcification as CP progressing, and encompasses a variety of clinical syndromes ranging from abdominal pain, steatorrhea and diabetes mellitus caused by pancreatic exocrine and endocrine function reduction, even can cause pancreatic cancer. Early diagnosis and detection for morphological or functional change will be of great significance in treatment guidance and prognosis improvement for CP.Radiological examinations are important diagnosis methods for CP, especially for moderate and severe CP. The pancreatic atrophy, calcification, pancreatic duct stones and dilation, pseudocyst could be detected and diagnosed accurately. Because of no significant morphological changes, early diagnosis could not be made for mild CP. As an important molecular imaging and metabolomics method, the application of magnetic resonance spectroscopy (MRS) in CP diagnosis is catching more and more attentions. At present, MRS has been widely used in the study for cancer, epilepsy, coronary heart disease and infection diseases, high-resolution magic angle spinning1H nuclear magnetic resonance spectroscopy (HR MAS1H NMR) is the common way for in vitro metabolomic study, of which the study objects include blood, urine, saliva, breast milk, cerebrospinal fluid, pancreatic juice, and biological tissue. Although common MRI equipment could be used for metabolomic study in vivo, enough metabolic information could not be detected because of lower magnetic field intensity and space resolution. By contrast, MRI equipment used for metabolic study in vitro has higher magnetic field intensity, pace resolution, and simpler sample pre process, HR MAS1H NMR could be and has been used for pancreatic diseases by study on blood, pancreatic juice and pancreatic tissue.Diffusion weighted imaging (DWI) is such a kind of functional imaging technology that could reflect the amount and motion state of water molecules, of which the apparent diffusion coefficient (ADC) value is a important parameter. Because of fibrosis and blood perfusion reduction, the ADC value of pancreatic tissue in CP was different from normal pancreas and pancreatic cancer. Recent studies have been proved that ADC value could be helpful for the diagnosis and judgment the severity in CP patients. As a noninvasive and nonradiative method, Magnetic Resonance Cholangiopancreatography (MRCP) can display the pancreatic and biliary duct clearly, which has been widely used in imaging diagnosis for related diseases. MRCP, however, couldn’t display the abnormality of the pancreatic branch duct clearly, which always occurred at the early stage of CP. Secretin-enhanced MRCP could make the branch of the pancreatic duct dilating physiologically, which could improve the show-up rate the abnormality of pancreatic branch duct, and might helpful for the diagnosis of CP at early stage.In our study, we will establish a obstructive chronic pancreatitis model in swine by subtotal ligation of the main pancreatic duct, then DWI sequence and routine MRI, magnetic resonance cholangiopancreatography (MRCP), computed tomography (CT) will be performed on CP swine model, and of which results will be compared with that of HR MAS1H NMR study. All studies are with pathology as the gold standard. Based on results in vivo and in vitro, such methods will be discussed in the evaluation for the severity and early diagnosis of CP. Part One Establishment of an Obstructive Chronic Pancreatitis Model in Swine and CT EvaluationObjective:To establish a swine model with obstructive chronic pancreatitis.Material and Methods:Thirty-nine swine were divided into control group (n=12) and experimental group (n=27). Swine in experiment group were performed an upper midline laparotomy to ligate the main pancreatic duct (MPD) subtotally at the pancreatic head. Swine in the control group underwent the same operation except the MPD ligation process. One third swine of CP group and control group (CP group, n=9; control group, n=4) were randomly performed dynamic CT enhanced scan and then sacrificed at4w,8w and12w after operation, respectively. Pancreatic tissue were obtained to perform pathological examination including HE and VG staining. Based on evaluations of pancreatic fibrosis, acinar atrophy, and inflammatory cell infiltration in tissues, the samples were grouped as normal pancreas, mild, moderate, and severe CP. The relation between pathological degree and MPD ligation time was analyzed, too.Result:Three of27CP swine died due to anesthetic events during operation (n=1), pulmonary infection (n=1) and abdominal cavity infection (n=1) after operation. The remaining24CP swine were pathologically diagnosed as having mild (n=10), moderate (n=5), severe (n=7) CP and normal pancreas (n=2). The two normal appearance pancreas were excluded from further analysis. All12control swine were pathologically determined of having normal pancreas. Linear-regression analysis showed that there were a strong relation between MPD ligation time and CP pathological degree (r=0.997, P=0.034). The dynamic enhancement CT scan showed that the peak CT value of pancreatic tissue decreased gradually as CP progresses, and there were significant differences among normal, mild, moderate and severe CP group at30s and35s after contrast agent injection.Conclusions:Subtotal ligation of MPD could induce obstructive chronic pancreatitis in swine successfully and with a high success rates. There was a strong correlation between CP severity and MPD ligation time. The dynamic enhancement CT scan showed that the peak CT value of pancreatic tissue decreased gradually as CP progresses. Part Two Functional MRI and Pathological Correlative Study on a Swine Obstructive Chronic Pancreatitis ModelObjective:To investigate the correlativity between secretin enhanced MRCP, DWI findings and pathological severity in a swine CP model.Material and Methods:Thirty-nine swine were divided into control group(n=12) and experiment group in=27). Upper midline laparotomy was performed on experimental group swine to establish obstructive CP model by subtotal ligation of the main pancreatic duct (MPD). Swine in the control group was sham operation group. One third swine of CP group and control group (CP group, n=9; control group, n=4) were randomly performed DWI and ADC values were measured, then dynamic s-MRCP was performed before and1,3,5,7,9,11min after secretin injection. The change of MPD diameter and duodenum juice filling degree was observed. Then, all swines were sacrificed. Pancreatic tissue was obtained to perform pathological examination including HE and VG staining for histopathological gradings. The imaging and histopathological findings were compared.Results:Three of27CP swines died due to anesthetic events (n=1)and abdominal cavity infection (n=2) after operation. The remaining24CP swine were pathologically diagnosed as having mild (n=10), moderate (n=5), severe (n=7) CP and normal pancreas (n=2).19CP swine (mild CP, n=8; moderate and severe CP, n=11) were performed s-MRCP successfully. All12control swine were pathologically determined of having normal pancreas, and11swines of which underwent s-MRCP successfully. The mean MPD diameter in normal group, mild CP group and moderate to severe CP group before secretin injection were1.56±0.46mm,2.95±1.170mm,7.41±1.91mm, respectively. There were significant differences among three groups (all P<0.05). After secretin injection, the mean MPD diameter of three group were2.39±0.43mm,3.63±1.25mm,7.86±1.87mm,respectively. There was no significant difference between normal group and mild CP group (P=0.056), The mean MPD diameter of mild CP group was significant less than that of moderate to severe CP group (P<0.05). The mean time-to-peak of MPD dilation in mild CP group (2.75±0.71min) was earlier than that (3.36±0.81min) in moderate to severe CP group (P<0.05), but there was no significant difference between normal group (3.00±0.00min)and mild CP group. The maximum change rate (%) of MPD dilation in three groups were59.84±34.76,33.1±14.23,6.92±4.41, respectively. The maximum change rate (%) of MPD in normal group was significant bigger than that in mild and moderate to severe CP group (all P<0.05). The maximum change rate (%) of MPD in mild CP group was significant bigger than that in moderate to severe CP group (P<0.05). Eleven min after secretin injection, the duodenum juice filling degree of mild CP group was significant lower than that in moderate to severe CP group (P=0.744). There was no significant difference between normal and mild CP group (P<0.05). As CP worsens, the mean ADC value of pancreas decreased gradually, the ADC value of three groups were1.68±0.03,1.61±0.03,1.39±0.11, respectively, there were no significant difference between normal and mild CP group, the mean ADC value in mild CP group was significant higher than that in moderate to severe CP group (P<0.05).Conclusions:S-MRCP could be used for in vivo evaluation on the pancreatic external secretion function, the maximum MPD dilation rate might be helpful for the early diagnosis of CP, and could be used for CP pathological grading. DWI and ADC value were less useful for the early diagnosis of CP, but if the CP was confirmed, the ADC value can be used for CP pathological grading. Part Three High-resolution Magic Angle Spinning1H Nuclear Magnetic Resonance Spectroscopy Study on a Swine ObstructiveChronic Pancreatitis ModelObjective:To investigate the metabolic characteristics of obstructive chronic pancreatitis (CP) in a swine model by using high-resolution magic angle spinning’H nuclear magnetic resonance (HR MAS1H NMR) spectroscopy.Material and methods:Thirty-nine miniature swine were randomly divided into eperimental group (n-27)and control group (n=12). Subtotal ligation of the pancreatic main duct were performed to establish the chronic pancreatitis model. The control group was sham group. At the4th,8th and12th week after operation,1/3of the swine both in experimental group (n=9) and control group (n=4) were sacrificed to harvest the pancreas tissue for pathologically examination, which were classified as mild, moderate and severe chronic pancreatitis. The metabolic profiles from the pancreas of normal pancreas, mild CP, moderate CP and severe CP were examined by using HR MAS1H NMR spectroscopy.Results:Three of27CP swine died due to anesthetic accident during operation (n=1), pulmonary infection (n=1) and abdominal cavity infection (n=1) after operation.22CP model was successfully established among the remaining24swine, and pathologically examination showed10was mild,5was moderate and7was severe. Among the metabolites in mild CP. HR-MAS NMR combined with2D MRS and PCA analysis suggested that8metabolites including choline, glycine, betaine, lactate, fatty acid, phosphorylcholine/choline glycerophosphate were the main metabolite marker in different groups. Four changes of metabolic patterns from normal pancreas, to mild, moderate, and severe CP group, PC/GPC, Gly, Bet and Ile/Leu/val decreases, and increase in Lac. Among these changes, the differences between normal or mild CP group with moderate or severe CP group are statistically significant. However, no difference either between normal and mild CP group or between moderate and severe CP group was observed. Furthermore, while Cho and Tau increased sharply in mild CP group but dropped dramatically in moderate and severe CP group, no significant difference of Tau and FA were detected among the tested four groups.Conclusions:HR MAS1H NMR spectroscopy study on normal swine pancreas and different stage CP pancreatic tissue showed that as CP progresses, Gly, Lac, PC/GPC, Bet, Cho, Ile/Leu/Val present trending changes and with a well correlate to the pathological grading of CP. And Cho might be a helpful metabolic marker for the early diagnosis of CP. Part Four Metabolomics Study of Urine and Serum in a Swine Obstructive Chronic Pancreatitis modelObjectives:To investigate the metabolimics characteristics of serum and urine in a swine CP model by UHPLC and Q-TOF/MS.Materials and Methods:39swines were divided into experimental group (n=27) and control group (n=12).27Swines obstructive CP models were established by subtotal ligation of the main pancreatic duct.control group was sham group.At4,8, and12weeks after operation,nine swines in experimental group and4swines in control group were sacrificed at each time. The pancreatic tissue, blood, urine, and panceatic tissue were obtained. Based on the phathological results, blood and urine samples were divided into normal group, mild CP group, and severe CP group. After centrifugation at3000G for10min,The serum samples were seperated and the aliquot were stored at-800degree until for metabolomic study. After special pretreatments, serum and urine samples were analyzed to acquire metabolic profiles by using UHPLC and Q-TOF-MS methods to obtain the total ion chromatography (TIC).Multivariate analysis(MVA) which including partial least squares discriminant analysis(PLS-DA) and orthogonal single collection partial least squares analysis(OPLS) were performed to visualize the metabolic profiles changes of serum and urine from control, mild CP, and severe CP group. Principal component analysis (PCA) was performed to display in score plot and loading plot, and dicover the potential biomarkers related to the outcomes.Results:Based on UHPLC-TOF-MS data of serum, PLS-DA (positive and negative model)showed that there are significant disinctions among control,mild CP,and severe CP groups.OPLS analysis have suggested there are10related biomarkers in the positive and negative model, which include3up-regulating metabolites,2down-regulating metabolites, and5slight down-regulating in mild CP but up-regulating in severe CP group.The urine sample analysis results showed that there were no significant distinction between mild and severe CP group, but all CP group was significant different from control group.OPLS analysis have suggested there were22biomarkers in both positive and negative model.All biomarkers were verified with human metabolome database (http://metlin.scripps.edu).Conclusions:Based on UHPLC and Q-TOF-MS metabolomic methods,the metabolic pattern of CP could be researched and be helpful for the CP diagnosis at the early stage.