小鼠结肠癌模型

Novel Murine Model for Colon Cancer:Non-OperativeTrans-Anal Rectal Injection 1Melissa Donigan,B.S.,*,†Laurie S.Norcross,M.D.,‡,§John Aversa,D.O.,‡,§Jimmie Colon,M.S.,*Joshua Smith,*Rafael Madero-Visbal,M.D.,*Shuan Li,M.D.,*,§Neal McCollum,M.D.,*,§Andrea Ferrara,M.D.,‡,§Joseph T.Gallagher,M.D.,‡,§and Cheryl H.Baker,Ph.D.*,†,2*Cancer Research Institute of M.D.Anderson Cancer Center Orlando,Orlando,Florida;†Burnett School of Biomedical Sciences,University of Central Florida,Orlando,Florida;‡Colon and Rectal Clinic,Orlando,Florida;and §Orlando Regional Medical Center,Orlando,FloridaSubmitted for publication April 11,2008Background.This study was conducted to develop a modified murine model of colon cancer that is non-operative.Currently,the most accurate orthotopic murine model of colon cancer requires an invasive procedure involving cecal injection of colon cancer cells and therefore limits the ability to perform immu-nological studies subsequent to cecal resections.Materials and methods.Murine colon cancer (CT26)cells were injected submucosally into the distal,pos-terior rectum of BALB/c mice.Care was taken not to pass transmurally into the pelvic cavity.Different magnifications (10؋versus 100؋)were used for injec-tion,and primary tumor growth and metastatic dis-ease were studied.Results.In the initial study,3/7mice injected using 10؋magnifications had notable,large tumor originat-ing from the rectal wall,and histology revealed that all excised tumors were poorly differentiated adeno-carcinoma.In the second study,8/10mice injected us-ing 100؋magnifications had notable tumor originat-ing from the rectal well,and 4/8mice had abnormal lung tissue with pathological evidence of hemorrhagic pulmonary edema.The use of 10؋magnification re-sulted in 43%tumor take.In sharp contrast,80%tumor take was observed with 100؋magnification.The over-all success of tumor take was 65%using the trans-anal rectal injection model.Conclusions.Our modified orthotopic murine model of colon cancer offers an alternative non-operativemurine model for colon cancer and is less invasive than the traditional orthotopic model (i.e.,cecal injec-tion).This model may allow for more accurate inves-tigations of inflammation and immune responses to surgical intervention without the influence of previ-ous abdominal surgery.©2009Elsevier Inc.All rights reserved.Key Words:colon cancer;orthotopic murine model;non-operative injection of cancer cells.INTRODUCTIONIt was estimated that in 2007over 1,500Americans would die of cancer each day.As the third most common cancer in both men and women,colon cancer represents 10%of these cancer-related deaths [1].Despite earlier detection and dropping death rates in colon cancer,112,340new cases were estimated for 2007[1].The most common treatment for colon and rectal cancer is surgical resection,followed by adjuvant therapy with oxaliplatin,5-fluorouricil,and leucovorin.Early detection can provide a 5-y survival rate of up to 90%,and surgery is most often curative.However,if patients present with distant me-tastasis at the time of diagnosis,the 5-y survival rate drops to only 10%[1].Since its inception in the early 1990s,laparoscopy has been an alternative technique to the more tradi-tional open laparotomy for colorectal pa-roscopic surgery incorporates a much smaller incision (ϳ2cm)than either hand-assisted laparoscopic (ϳ5–8cm)or open surgery (ϳ30cm).Some of the benefits of minimally invasive surgery include a shorter hospital stay,less pain,and a quicker return of bowel function [2].However,concerns for safety and adequate onco-logic resection slowed the acceptance of laparoscopy as1Melissa Donigan and Laurie S.Norcross contributed equally to this manuscript.2To whom correspondence and reprint requests should be ad-dressed at Cancer Research Institute of M.D.Anderson Cancer Center Orlando,110Bonnie Loch Court,Orlando,FL 32806.E-mail:cheryl.baker@ .Journal of Surgical Research 154,299–303(2009)doi:10.1016/j.jss.2008.05.0282990022-4804/09$36.00©2009Elsevier Inc.All rights reserved.a safe option in most colorectal settings,most notably in colorectal cancer.Since then,several large-scale ex-periments comparing traditional open surgery(open laparotomy)to minimally invasive surgery(laparo-scopic and hand-assisted laparoscopic)have been con-ducted.With the publication of the COST(Clinical Outcomes of Surgical Therapy)trial[3]published in the New England Journal of Medicine in2004and the “Barcelona Trial”[4]published in Lancet in2002,lapa-roscopic surgery became accepted as a safe alternative to traditional open laparotomy for colorectal resection. Both trials demonstrated that patients undergoing laparoscopy had either equivalent or better cancer-related survival compared with patients with major open laparotomy.Some believe that the possible bene-fits of minimally invasive surgery(i.e.,shorter recovery period,less pain,and earlier return of bowel function) may be related to the patient’s immune and cytological responses,such as serum protein concentration,cyto-kinefluctuations,as well as tumor growth,metastasis, and recurrence.Although minimally invasive surgical techniques have been widely accepted by the medical community,there is still active debate concerning the possible immunological benefit of limited surgical trauma with minimally invasive surgery.Preclinical evaluation and comparison of traditional open laparotomy and minimally invasive surgical tech-niques on the effects on colon cancer and metastasis require animal models.To date,there are multiple mu-rine models for the study of colon cancer[5,6].The orthotopic murine model,which involves injection of tu-mor cells into the cecum of mice,is one of the most accurate representation of human colon cancer[6].This orthotopic model requires a surgical incision for injection of cancer cells,and therefore,an initial immune response following this surgical model would limit the feasibility and reliability of future research goals of measuring im-mune responses following open or minimally invasive cecal resections.Therefore,the traditional cecal injection model must be modified to involve a non-operative ap-proach for injection of colon cancer cells(to establish a primary tumor).Previous studies have demonstrated that colon can-cer can be established orthotopically after the induc-tion of colitis[7].This enema model,however,requires initial induction of colitis to allow invasion of intralu-minal cells,leading to a significant inflammatory re-sponse.In addition,other laboratories developed a mu-rine model of rectal cancer using an intra-rectal injection,which resulted in minor lymph node metas-tasis noted in18%of animals[8].Despite their avail-ability,these models of colon cancer are not widely used and are not reliable models for preclinical evalu-ation of future immunological studies subsequent to cecal resection,most often performed clinically.In this article,we describe a trans-anal injection of colon cancer cells in BALB/c mice.The method is similar to the intra-rectal injection[8]but is performed with a few modifications.This modified orthotopic murine model that does not require abdominal surgery for injec-tion of colon cancer cells makes it a reliable and clinical relevant technique to study the inflammation and im-mune response subsequent to cecal resection.Further-more,in contrast to the colitis and intra-rectal murine models of colon cancer,this novel technique has the po-tential to allow for the study of metastasis and possible tumor recurrence following open laparotomy or the much debated minimally invasive surgical techniques.MATERIALS AND METHODSCell Line and Culture ConditionsThe murine colon cancer CT26cells were obtained from the Amer-ican Type Culture Collection(Manassas,VA).In brief,these cells were initially induced in a BALB/c mouse by chemical carcinogen, and stable cell lines were established[9].The murine colon cancer CT26cells have been demonstrated to be highly metastatic to the liver and lungs[10].Cells for injection were obtained from frozen stocks and maintained in Dulbecco’s Modified Eagle’s Medium (DMEM)supplemented with10%fetal bovine serum(FBS),sodium bicarbonate,and a penicillin–streptomycin mixture(Invitrogen, Carlsbad,CA)for no more than12weeks.To account for the influ-ence of inflammatory cytokine(i.e.,IL-6,IL-1␤)production by the FBS in cell culture,enzyme-linked immunosorbent assays were car-ried out on different FBS concentrations.Results show(unpublished data)that FBS cytokine levels are below the detection limit,indicat-ing that the levels of cytokines present in the FBS are negligible. Adherent monolayer cultures were maintained on plastic at37°C in 5%CO2and95%air.All cultures were free of Mycoplasma and the following pathogenic murine viruses:retrovirus type3,pneumonia virus,K virus,Theiler’s encephalitis virus,Sendai virus,min virus, mouse adenovirus,mouse hepatitis virus,lymphocytic choriomenigi-tis virus,ectromelia virus,and lactate dehydrogenase virus(assayed by Microbiological Associates,Bethesda,MD).Animals and Animal CareMale BALB/c mice were purchased from the National Cancer Institute(NCI)Animal Program of Charles River Laboratory(Fred-erick,MD).The mice were housed and maintained in specific pathogen-free conditions and facilities approved by the American Association for Accreditation of Laboratory Animal Care,and in accordance with current regulations and standards of the United States Department of Agriculture,United States Department of Health and Human Services.The mice were used in accordance with institutional guidelines when they were8–12wk old.Preparation of Cell Suspension for Injection Murine colon cancer CT26cells were harvested from near-confluent cultures by a brief(3-min)exposure to0.5%trypsin and0.02%EDTA (Invitrogen).Trypsinization was stopped with DMEM containing10% FBS,and the cells were concentrated with centrifugation and resus-pended in DMEM containing10%FBS.Trypan blue staining was used to assess cell viability,and only cell suspensions consisting of single cells withϾ90%viability were used for the injections.300JOURNAL OF SURGICAL RESEARCH:VOL.154,NO.2,JUNE15,2009Trans-Anal InjectionMale BALB/c mice were anesthesized with100mg/kg ketamine and50mg/kg xylazine.Mice then received a gentle anal dilation using blunt-tipped forceps at the anal opening.A29-gauge syringe was used to inject2.5ϫ104CT26cells,suspended in DMEM with 10%FBS,submucosally into the distal,posterior rectum.The injec-tion was performed approximately1–2mm beyond the anal canal and into the rectal mucosa,which minimizes the chance of establish-ing anal tumors.In addition,care was taken not to pass transmu-rally into the pelvic cavity.Mice were observed for1h until fully recovered and were then monitored three times weekly for tumor burden.In thefirst experiment,seven mice were injected by using operating loupes(Designs for Vision,Inc.,Ronkonkoma,NY)with 10ϫmagnification and sacrificed on post-injection day17.In a sec-ond experiment,10mice were injected using a100ϫmagnification (Leica MC16F microscope;Leica Microsystems,Bannockburn,IL) and then sacrificed on post-injection day20.Necropsy,Tissue Preparation,and Hemotoxylinand Eosin StainingMice were killed by CO2inhalation on day17(experiment#1)or day20(experiment#2).After dissection and removal of the tumor (rectum),rectal wall,and lungs,the tissue samples were photo-graphed and prepared for histological analysis.For hematoxylin and eosin(H&E)staining procedures,the tissue samples werefixed in formalin(rectum and rectal wall)or bouins(lungs)and embedded in paraffin.All slides were reviewed by a board-certified pathologist.A normal BALB/c mouse,which had not received prior injection or surgical intervention,was dissected for normal anatomy description of the injection site.Images were collected using the100ϫmagnifi-cation(Leica MC16F microscope).RESULTSFormation of Colon Cancer in Distal Posterior RectumIn afirst set of preliminary in vivo injection studies, the investigators performed trans-anal injections on BALB/c mice using trypan blue as an indicator of in-jection location.Due to systemic dissemination of the trypan blue dye,the investigators carried out another set of preliminary in vivo studies,using India Ink.The success of injection was apparent on dissection.Subse-quent to these studies,murine colon cancer CT26cells were injected submucosally into the distal,posteriorrectum of another set of BALB/c mice(Fig.1A).As illustrated in Fig.1B,the injections were performed approximately1–2mm beyond the anal canal and into the rectal mucosa.The short,smooth,white epithelial tissue of the anal canal can clearly be identified as a separate tissue plane from the pink mucosa of the rectum.The submucosal injections were performed into the rectum proximal to the anal canal.In thisfirst set of experiments,7mice were injected with CT26 cells using10ϫmagnification(operating loupes)and killed on post-injection day17.Three of7mice(43%) had notable large tumor originating from the rectal wall.In the second set of experiments,10mice were injected with CT26cells using100ϫmagnification (Leica microscope)and killed on post-injection day20. Eight of10mice(80%)had gross tumor originating from the rectal wall(Fig.2A and B).Even more strik-ing,4of these8mice had the presence of abnormal lung tissue.There was an overall65%successful tumor take using the trans-anal,rectal colon cancer injection model. The discrepancy between successful tumor take in exper-iment#1(3/7)and successful tumor take in experiment #2(8/10)is believed to be due to the limitation of the10ϫmagnification used in experiment#1.Therefore,the use of100ϫmagnification is recommended.Histology Analysis of Rectal Tumors and LungsThe rectal walls and lungs from both experiments were examined by a pathologist for the presence of primary tumor and metastasis,respectively.Rectal tu-mors in both experiments showed poorlydifferentiated FIG.1.Non-operative trans-anal injection of murine colon can-cer CT26cells into BALB/c mice.(A)To demonstrate a non-operative approach to an orthotopic murine model of colorectal cancer,a29-gauge syringe was used to inject2.5ϫ104CT26cells,suspended in DMEM with10%FBS,submucosally into the distal,posterior rec-tum of BALB/c mice.(B)A normal BALB/c mouse,which had not received prior injection or surgical intervention,was dissected for an anatomical description of the trans-anal injections.The short, smooth,white epithelial tissue of the anal canal can clearly be identified as a separate tissue plane from the pink mucosa of the rectum.The submucosal injections were performed into the rectum proximal to the anal canal.301DONIGAN ET AL.:NOVEL MURINE MODEL FOR COLON CANCERadenocarcinoma originating in the submucosal tissue (Fig.3).The lung tissue was grossly abnormal when compared with normal lung (Fig.4A),and pathology showed diffuse,severe,hemorrhagic pulmonary edema (Fig.4B).There were no gross changes noted in the livers of the mice,therefore no histological examina-tion of hepatic tissue was performed.DISCUSSIONMurine models are useful in evaluation of human colon cancer [5,6],and the importance of orthotopic murine models to study the biology and therapy of neoplasm has been demonstrated [6].Nonetheless,these models tradi-tionally involve abdominal surgery for injection of the cancer cells.The non-operative murine model of colon cancer described in this report compliments the most recently reported murine models and may overcome some limitations (induced by surgery)associated with them.First,we verified that the trans-anal injection of colon cancer cells produces large tumor originating from the rectal wall.Second,we determined that 100ϫmag-nification is necessary to produce a more successful tu-mor take than 10ϫmagnification (80%and 65%,respec-tively).The growth of colon cancer cells within the rectal wall is similar to that observed clinically in patients with colorectal cancer,suggesting that the model is clinically relevant.In addition,changes within the lung paren-chyma were observed in four of eight mice.The appear-ance of pulmonary edema in these specimens is concern-ing and may indicate systemic inflammatory changes (i.e.,systemic inflammatory response syndrome).None-theless,these mice did not clinically appear sicker than the others in the series.These findings are currently under investigation in additional trials using the trans-anal model in a larger cohort of mice.The absence of metastatic disease is most likely due to the number of primary colon cancer cells injected and the termination date of the experiment.To overcome this,studies using a larger number of injected cells and longer periods of time between injection and termination will be performed.The most common treatment for colon and rectal cancer is surgery followed by treatment with chemo-therapy agents (with or without neoadjuvant therapy)[1].Furthermore,if the cancer is confined and no met-astatic disease is present,surgery is most often cura-tive and patients may not have to receive chemother-apy.Surgical resection of colon and rectal cancer can be performed using two techniques:a traditionallaparot-FIG.3.Histology of primary tumor growing orthotopically in BALB/c mice.H&E stains were carried out on the submucosal rectal tumor established 17d after 2.5ϫ104CT26cells were injected submucosally into the distal,posterior rectum of BALB/c mice.All slides showed similar findings of poorly differentiated adenocarci-noma arising in the submucosallayers.FIG.2.Gross primary tumor growing orthotopically in BALB/c mice.(A)Evidence of gross tumor protruding from the rectum of mice 17d after a 29-gauge syringe was used to inject 2.5ϫ104CT26cells submucosally into the distal,posterior rectum.(B)Anatomical dissection of gross tumor originating from the rectal wall of mice on day 17after injection of 2.5ϫ104CT26cells submucosally into the distal,posterior rectum of BALB/c mice.The growth of colon cancer cells within the rectal wall is representative of that observed clinically in patients with colorectal cancer,suggesting that the model is clinically relevant.302JOURNAL OF SURGICAL RESEARCH:VOL.154,NO.2,JUNE 15,2009omy with a large abdominal incision or newer mini-mally invasive techniques with significantly smaller incisions and laparoscopic instruments.The clinical benefits of laparoscopic surgery have been repeatedly proven throughout surgical literature.It is accepted that colorectal cancer patients undergoing minimally invasive surgical techniques have a shortened length of hospital stay,require less postoperative pain medica-tions,and have a shorter time to return of bowel func-tion and tolerating a diet.Additionally,some studies have suggested a potential benefit in cancer-free sur-vival,disease progression,recurrence,and metastasis in certain patients undergoing laparoscopic colon re-section for colon cancer,specifically in advanced onco-logic stages [3,4].The proven benefits of these mini-mally invasive surgeries may be related to the patient’s immune response.Therefore,these observed clinical differences need further evaluation,including assess-ment of the immunological differences in patient re-sponse to open and laparoscopic colon resection.Previously reported murine studies have shown that,in mice receiving an abdominal incision,there was a reduced cell mediated immune response as compared with mice receiving anesthesia alone [11].In addition,a significant increase in tumor growth was observed in mice receiving surgical intervention as compared with control [12,13].We have developed a murine model of colon cancer that involves a trans-anal injection of colon cancer cells which does not require abdominal surgery for injection.This model will provide a more accurate,unaltered in vivo tumor growth pattern for colon and rectal cancer.In the end,future in vivo studies involving cecal resections (i.e.,open laparot-omy or minimally invasive)and their associated im-munological benefits are now possible.ACKNOWLEDGMENTSWe thank Donna Schade (M.D.Anderson Cancer Center Orlando)for helping with preparation of this manuscript.REFERENCES1.American Cancer Society.Anonymous Cancer Facts and Fig-ures 2007.2.Lee SW,Whelan RL.Immunologic and oncologic implications of laparoscopic surgery:What is the latest?Clin Colon Rectal Surg 2006;19:5.3.Nelson H,Sargent DJ,Wieand S,et al.A comparison of lapa-roscopically assisted and open colectomy for colon cancer.N Engl J Med 2004;350:2050.4.Lacy AM,Garcia-Valdecasas JC,Delgado S,et paroscopy-assisted colectomy versus open colectomy for the treatment of non-metastatic colon cancer:A randomized ncet 2002;359:2224.5.Rashidi B,Gamagami R,Sasson A,et al.An orthotopic mouse model of remetastasis of human colon cancer liver metastasis.Clin Cancer Res 2000;6:2556.6.Heijstek MW,Kranenburg O,Borel Rinkes IH.Mouse models of colorectal cancer and liver metastases.Dig Surg 2005;22:16.7.Takahashi T,Morotomi M,Nomoto K.A novel mouse model of rectal cancer established by orthotopic implantation of colon cancer cells.Cancer Sci 2004;95:514.8.Kashtan H,Rabau M,Mullen JB,et al.Intra-rectal injection of tumour cells:A novel animal model of rectal cancer.Surg Oncol 1992;1:251.9.Corbett TH,Griswold DP,Roberts BJ,et al.Tumor induction relationships in development of transplantable cancers of the colon in mice for chemotherapy assays,with note of carcinogen structure.Cancer Res 1975;34:2434.10.Schackert HK,Fidler IJ.Development of an animal model to study the biology of recurrent colorectal cancer originating from mesenteric lymph system metastases.Int J Cancer 1989;44:177.11.Allendorf JD,Bessler M,Whelan RL,et al.Better preservation of immune function after laparoscopic-assisted vs.open bowel resection in a murine model.Dis Colon Rectum 1996;39:S67.12.Whelan RL,Allendorf JD,Gutt CN,et al.General oncologic effects of the laparoscopic surgical approach.1997Frankfurt International Meeting of Animal Laparoscopic Researchers.Surg Endosc 1998;12:1092.13.Allendorf JD,Bessler M,Kayton ML,et al.Increased tumor establishment and growth after laparotomy vs laparoscopy in a murine model.Arch Surg1995;130:649.FIG. 4.Histology of murine lungs.(A)H&E-stained slide of normal lung.(B)H&E of lung presenting with diffuse,severe,hem-orrhagic pulmonary edema,consistent with systemic effects.303DONIGAN ET AL.:NOVEL MURINE MODEL FOR COLON CANCER。