Induction of pancreatic neoplasia in the KRAS/TP53 Oncopig

ABSTRACT The 5-year survival of pancreatic cancer (PC) remains low. Murine models may not adequately mimic human PC and can be too small for medical device development. A large-animal PC model could address these issues. We induced and characterized pancreatic tumors in Oncopigs (transgenic swine containing KRASG12D and TP53R167H). The oncopigs underwent injection of adenovirus expressing Cre recombinase (AdCre) into one of the main pancreatic ducts. Resultant tumors were characterized by histology, cytokine expression, exome sequencing and transcriptome analysis. Ten of 14 Oncopigs (71%) had gross tumor within 3 weeks. At necropsy, all of these subjects had gastric outlet obstruction secondary to pancreatic tumor and phlegmon. Oncopigs with injections without Cre recombinase and wild-type pigs with AdCre injection did not show notable effect. Exome and transcriptome analysis of the porcine pancreatic tumors revealed similarity to the molecular signatures and pathways of human PC. Although further optimization and validation of this porcine PC model would be beneficial, it is anticipated that this model will be useful for focused research and development of diagnostic and therapeutic technologies for PC. This article has an associated First Person interview with the joint first authors of the paper.

of AdCre, all five subjects recovered uneventfully and were tolerating regular feed within 3-4 days. There were no perioperative complications. Subjects were fed ad lib with no specialized husbandry for 4 mo. Average weight gain during the 4 mo observation period was 30.9 ± 8.4 kg (Table S2); all subjects were thriving and otherwise normal appearing. Serum laboratory testing just prior to euthanasia revealed small differences from reference values derived from 3-4 mo old domestic pigs (Table S3). Subjects underwent euthanasia with full necropsy (head, chest, abdomen) at 4 mo post-injection of AdCre (subject age 9-10 mo). Adhesions involving the operative site were present in all subjects, but there were no gross tumors, neither locally nor distantly. The explanted specimen in this panel (4 mo post-injection of AdCre) shows the anterior side of the pancreas. D2 = 2 nd portion of duodenum; DL = duodenal lobe of the pancreas; CL = connecting lobe of the pancreas; SL = splenic lobe of the pancreas; SL* = region of splenic lobe used as negative control tissue for histology; p = proximal end of duodenal specimen; d = distal end of duodenal  Oncopig pancreas from all five subjects was fixed in formalin and then underwent serial slicing at the two AdCre injection sites (site no. 1 = parenchymal, within the duodenal lobe; site no. 2 = into the lumen of the main pancreatic duct) to search for sub-centimeter tumors; none were found. Pancreas from the distal splenic lobe (see panel A) was used for negative control tissue (i.e., pancreas with no AdCre injection). That is, each Oncopig served as its own control. Representative images are shown. No lesions within the epithelium of the main pancreatic duct proper were observed. However, there appeared to be proliferation of accessory ducts around the main pancreatic duct from regions that had been exposed to AdCre (see arrows in panel B). In contrast, the pancreatic duct from the non-injected region (A) had relatively few accessory ducts (arrows) around the main duct. Bar = 1,000 µm; lum = lumen of main pancreatic duct. (C-F) Similar to panels A-B, the noninjected (negative control) tissue derived from the splenic lobe of the pancreas, a region of the organ physically distanced from the injection sites. In the non-injected regions (C, E), the morphology of the parenchyma and ducts was unremarkable. Around the AdCre-treated pancreatic duct, however, there appeared to be proliferation of the  and 1095 received both AdCre and IL8 injections, while 1092 did only receive AdCre but not IL8 injection.
In the control pigs, 1083 and 1091 are wild type pigs who had received both AdCre and IL8 injection, while 1101 was a OCM pig which only received IL8 but not AdCre injection. Each bar represents the mean ± SD of a duplicate array; Two-way ANOVA, ns = not significant; **p <0.01, ***p <0.001, ****p <0.0001.

Supplementary Materials and Methods
Detailed procedure protocols are available from the senior author (MAC) upon request.

Survival Procedure: Set-Up and Anesthesia
Swine were fasted for 24 h prior to the procedure, with free access to water. On day zero, subjects were weighed, and underwent induction with ketamine (2.2 mg/kg), Telazol® (1:1 w/w tiletamine:zolazepam, 4.4 mg/kg) and xylazine (2.2 mg/kg), given as a single IM (intramuscular) injection. Buprenorphine SR (0.2 mg/kg) was given as an IM injection into the right hip. EKG, pulse oximetry, and lingual end-tidal CO 2 monitors were placed and connected to a BM5 Bionet monitor (bionetus.com). Endotracheal intubation was performed with a 6-7 mm ID (internal diameter) tube. The subject rested on a water-circulated warming blanket that was set at 102˚F. An auricular IV (intravenous) line was placed, and LR (Lactated Ringers solution) was administered at 500 mL/h. Cefovecin sodium (Convenia®; 8 mg/kg IM) and Buprenorphine SR (0.2mg/kg SC or subcutaneous) were administered during the induction period. Anesthesia was maintained with isoflurane (1-2%) and supplemental oxygen (3-5 L/min) using a Matrx® ventilator (midmark.com). The ventilator rate initially was set at 12-15 breaths per minute with a tidal volume of 8 mL/kg, and subsequently adjusted to maintain the EtCO2 at 40-50 mm Hg.
Phlebotomy was performed on a forelimb or auricular vein. The ventral abdomen, groins, and thorax were scrubbed with chlorhexidine soap, and then depilated with electric clippers. Cotton blankets were placed over non-surgical areas to minimize subject heat loss. The final abdominal preparation was performed with alcohol-based chlorhexidine (ChloraPrep™; bd.com), and then the upper midline region was sterilely draped.

Reagents
Ad5CMVCre-eGFP (AdCre) was purchased from the University of Iowa Vector Core (vector-core.medicine.uiowa.edu). Porcine IL-8 was purchased from Novus Biological (NBP2-35234; novusbio.com). General chemicals were purchased from Millipore Sigma (www.sigmaaldrich.com). All the antibodies that were used in this study are listed in Table S10.

Standards, Rigor, Reproducibility, Transparency, Power, and Randomization
To the extent possible, the animal studies of this report were designed, performed, and

Animal Welfare Statement
The animals utilized for this report were maintained and treated in accordance with the Guide for the Care and Use of Laboratory Animals (8 th ed.) 3 Table S11.

Study Termination and Euthanasia
The prescribed post-induction observation period was three months. Criteria for early removal from the study and euthanasia were symptoms of failure to thrive (anorexia, lethargy, decreased movement, abnormal breathing, or other signs of distress) or sepsis (fever, wound disruption or drainage). At the time of euthanasia, each subject received sedation with an IM shot of ketamine/Telazol/xylazine, as described above, and then were endotracheally intubated.
Inhalational isoflurane (5%) was administered via the ventilator. The prior midline incision was reopened. Inferiorly this incision was extended in paramedian fashion to avoid midline structures (such as the urethra in males). The completed necropsy incision extended from xiphoid process to the pelvic inlet. The bilateral thoracic cavity was entered by transversely incising the diaphragm just inferior to the xiphoid process. The intrathoracic portion of the inferior vena cava was easily identified as it emerged from the liver in the posterior mediastinum. Phlebotomy was performed from the cava, and then Fatal-Plus® (pentobarbital sodium, 390 mg/mL; 1 mL per 4.5 kg body weight) was administered by caval injection. Two minutes after administration of Fatal-Plus®, the inferior vena cava was transected just above the diaphragm to exsanguinate the subject. A gross necropsy involving the lungs, heart, liver, kidneys, pancreas, intestines, bladder, and associated peritoneal surfaces then was performed, with biopsy of any suspicious lesions.

Tissue processing, Histology and Immunohistochemistry
Tissue was either formalin fixed and sent for further processing, or fresh tissue was taken and digested for in vitro analysis (see below). Tissue that was cut from formalin-fixed samples was stained with H&E for pathological examination. IHC was performed by deparaffinizing slides and heating using citric acid antigen retrieval buffer (Vector laboratories, H3300).
Following antigen retrieval, endogenous peroxidases were quenched with 3% hydrogen peroxide solution for 5 min. Blocking was done using 2.5% goat serum blocking buffer. Primary antibodies were incubated overnight at 4ºC and are listed in