Entertainment & Humor

A novel endothelial cell-based gene therapy platform for the in vivo delivery of apolipoprotein E

A novel endothelial cell-based gene therapy platform for the in vivo delivery of apolipoprotein E
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  GT (MAC )MASTER/EDITOR/AUTHOR/MAC Photo·graphics  . Gene Therapy (1999) 6 , 1153–1159 ©  1999 Stockton Press All rights reserved 0969-7128/99 $12.00 http://www.stockton-press.co.uk/gt A novel endothelial cell-based gene therapy platform for the in vivo delivery of apolipoprotein E  L Cioffi 1 , FG Sturtz 1 , S Wittmer 1 , B Barut 1 , J Smith-Gbur 1 , V Moore 1 , T Zupancic 1 , B Gilligan 2 ,R Auerbach 2 , F Gomez 3 , F Chauvin 3 , M Antczak 4 , D Platika 5 and HR Snodgrass 1 1 Progenitor, Inc., Menlo Park, CA;  2 Department of Zoology, University of Wisconsin, Madison, WI, USA;  3 BiostatisticsDepartment, Centre Anticance´ reux Le´ on Be´ rard, Lyon, France;  4 University of Colorado at Boulder, Boulder, CO; and  5 Ontogeny,Inc., Cambridge, MA, USA A major focus in gene therapy has been the use of recom- and lesions at a young age. After transplantation of the binant viruses to deliver genes in vivo. Although this apoE secreting Pro 175 endothelial cells into apoE- approach shows much promise, there are many safety deficient mice, serum cholesterol levels were measured at concerns associated with the use of viral materials in the 2 week intervals. During the 3 months after the initiation of treatment of human diseases. Our alternative cell-based these experiments, levels of cholesterol in the animals hav- gene therapy approach utilizes endothelial cells (Pro 175) ing received the apoE secreting endothelial cells were stat- isolated from the murine embryonic yolk sac. These endo- istically lower compared with the levels of age-matched thelial cells were evaluated for their potential use in gene controls having received non-secreting endothelial cells.therapy as a gene delivery platform. As a test model, we Concomitant with cholesterol reduction, atherosclerotic used these cells to deliver apolipoprotein E (apoE) in the aortic plaques were noticeably reduced in the experimental murine apoE knockout atherosclerosis model. The lack of apoE  +  animals. These results highlight the potential of apoE protein in these animals results in high levels of these unique endothelial cells as an efficient delivery serum cholesterol and formation of severe aortic plaques platform for somatic gene therapy. Keywords:  atherosclerosis; gene therapy; endothelial cells; yolk sac; ApoE; cardiovascular disease  Introduction  Cell-based gene therapy is a rapidly growing area of research. This approach offers several advantages overother types of viral and nonviral gene delivery systemscurrently being utilized to treat human diseases. Cells can be grown  in vitro  and genetically engineered to secretehigh levels of a therapeutic protein. Long-termexpression of biologically relevant proteins can be main-tained, thus making successive injections and continualinvasive procedures unnecessary. Safety concerns withthe use of adenoviral vectors is a major considerationwhen designing gene therapy treatments of disease. Byusing cell-based gene therapy, this is not a critical issue,as cells can be carefully screened and characterized before injection.Various cells have been reported as potential vehiclesfor use in gene therapy, but there have been many road- blocks in the way of utilizing their potential, includinginstability and lack of long-term viability. 1–10 The ulti-mate cell to be used for cell-based gene therapy has therequirements of being easily cultured, easily transfectedwith genetic material, able to secrete high levels of anexogenous protein, able to be transplanted  in vivo , non-malignant, and no observed host-versus-graft effects. Correspondence: L Cioffi, Progenitor Inc., 4040 Campbell Avenue, MenloPark, CA 94025, USAThe two first authors have equally contributed to this study Received 7 April 1998; accepted 18 December 1998 Atherosclerotic cardiovascular disease is the leadingcause of death in the Western world today. Various gen-etic and environmental factors have been implicated inthe manifestation of atherosclerosis. 11 Among the geneticfactors, lipoprotein abnormalities seem to play a key rolein the pathogenesis of this disease. Apolipoprotein E(apoE) is one of the best characterized plasma apolipo-proteins. 12–15 ApoE is a glycoprotein with a molecularsize of approximately 34 kDa. 16 The major site of syn-thesis of apoE in humans and normal mice is the liveralthough other tissues, such as the brain, spleen, adrenalgland, lung, testis, macrophages and ovary, are knownto synthesize this protein. 17–21 In lipoprotein metabolism,apoE plays a major role in facilitating the cellular uptakeof low density lipoproteins (VLDL) and chylomicronremnants by serving as a ligand on lipoprotein particlesfor the VLDL receptor and the putative remnant recep-tor. 22–27 In humans, apoE is one of two known ligandsfor the low density lipoprotein (LDL) receptor. 28 Type IIIhyperlipoproteinemia and atherosclerotic coronary heartdiseases have been found to be associated with geneticvariation of apolipoprotein E resulting in defective recep-tor binding. 29 Normal inbred strains of mice do not develop athero-sclerotic lesions spontaneously on a standard diet,although different strains of mice exhibit heritable differ-ences in the levels of their circulating lipids. 30,31 An apoE-deficient mouse model (apoE − / − ) has been described bytwo groups of independent investigators. 32,33 The apoE-deficient mice have been found to have five times normal  Endothelial cells as a gene delivery platform L Cioffi  et al  1154 plasma cholesterol levels. These mutant mice developfoam cell-rich depositions in their proximal aortas within3 months, which progress to cause severe occlusion of the coronary arteries by age 7–8 months. This animalmodel yields a severe yet viable phenotype, which isideal for studying the effect of new strategies on thedevelopment of atherosclerosis.Recently, bone marrow transplantation was used tocorrect the apoE deficiency in apoE-deficient mice. 34,35 ApoE expressing macrophages or blood cells appeared tocorrect the metabolic defect, and to inhibit the athero-sclerotic symptoms in these animals. In an alternativegene therapy approach, apoE deficiency was successfullycorrected utilizing recombinant adenovirus vectorsexpressing human apoE. 36 Systemic delivery of adeno-viruses expressing the human apoE to mouse liver wasachieved and resulted in normalized plasma apoE levels.Significant reduction of aortic atherosclerotic mean lesionarea was obtained when compared with control apoE-deficient mice. These findings indicate the great potentialof gene replacement. However, both of these strategieshave limitations. Bone marrow transplantation and sub-sequent injection of transfected cells is labor intensive,requires histocompatibility matching, irradiation of thehost, and has significant morbidity and mortality associa-ted with the procedure. The use of adenoviruses todeliver genes is subject to all the safety concerns associa-ted with adenoviral gene therapy. Furthermore, adeno-viral vectors provide only a transient therapy andare immunogenic which hinders the success of laterreinjections and long-term treatment. 37 An approach using a novel endothelial cell-based genetherapy for the expression of apoE in an atherosclerosismodel is described in this study. An endothelial cell linederived from murine yolk sacs was stably transfectedwith a plasmid construct containing the human apoEgene and a high expressing apoE +  cell line was selected(Pro 175/apoE + ). These engineered cells were injectedinto apoE-deficient mice as an  in vivo  mini-organ system.A decrease in cholesterol levels as well as in plaque for-mation was observed in treated animals for over 3months. In addition, the stable yolk sac (YS) cell linesexpress classic endothelial markers, are stable, easilytransfected with genetic material, express no detectablelevels of MHC I under normal growth conditions, andhave been maintained untransformed for over 50 pass-ages  in vitro . These results support the continued evalu-ation of these cells for long-term cell-based gene therapy. Results  Characterization of YS endothelial cells  Cells were isolated from the YS tissues of mice and cul-tured for a long period of time, some cultures having been expanded now for over 100 passages with no appar-ent change in morphology, growth characteristics, orendothelial cell surface marker phenotype. The YS cellclone Pro175 was found to be positive for the followingendothelial markers: VCAM, flk-1, ACE, and acetylatedLDL receptor (Table 1). Pro 175 cells also formed tube-like structures when cultured on Matrigel (CollaborativeBiomedical Products) (Figure 1), which is a characteristicof endothelial cells such as HUVECs (Clonetics). Interest-ingly, these cells do not express MHC I markers  in vitro Table 1  Endothelial characteristics of yolk sac-derived endo-thelial cells  Marker  +  /  −  Marker % Positive ac-LDL-r  +  Mac-1 0Tek  +  F4/80 0flk-1  +  MHC I and II 0‘tube’ formation  +  VCAM 55Histology was used to determine the presence of ac-LDL-r. RT-PCR was used to assay for expression of ACE, Tek and  flk-1 .Flow cytometric analysis was used for the expression of VCAM,Mac-1, F4/80, MHC I and II. The matrigel assay was used todetermine tube formation.  Figure 1  Yolk sac endothelial tube formation. Phase microscopy (a) andconfocal microscopy (b) of murine yolk sac cells grown on matrigel demon-strates the hollow lumen of the tube formed from these cells. under normal culture conditions. To determine if theMHC I antigens could be up-regulated, Pro175 cells weretreated with interferon which is known to up-regulateMHC I proteins in mature endothelial cells. The MHC Iantigens in Pro175 cells were found to be up-regulated by only 50% when treated with interferon, in which onlyhalf of the cells were labeled with the MHC-I antibodywhich caused a shift in those positively labeled cells inthe FACS scan. The positive control endothelial cells,MAE (murine aortic endothelial cells), up-regulatedMHC I antigens by 100% (data not shown) when treatedwith interferon. This up-regulation was determined byFACS analysis, in which flow cytometry is used to meas-ure the number of cells which are stained positively bythe MHC-I antibody, as determined by a shift in the pos-  Endothelial cells as a gene delivery platform L Cioffi  et al  1155 ition of the positively labeled cells on the FACS scan.Lack of MHC I expression of these YS cells under normalconditions may facilitate engraftment and reducerejection. Treatment of atherosclerosis with genetically engineered YS cells expressing apoE protein  To study the ability of Pro175 cells to secrete biologicallyrelevant proteins, these cells were transfected with ahuman  apoE3  gene expression plasmid. After G418 selec-tion, the level of apoE expression was measured in 30clones by an apoE ELISA. The highest apoE expressingclone secreted 150 ng/24 h/10 6 cells and was used in thefollowing animal experiments.ApoE + /Pro175 cells were embedded in Matrigel andinjected intradermally in the lower abdomen of apoE-deficient mice at age 6–8 weeks. Because the Matrigel sol-idifies at body temperature, the injected cells remainedclustered and formed ‘mini-organs’ in the animals. Thecholesterol levels in the negative control animals (fiveanimals injected with untransfected Pro175 cells) and inthe five apoE + /Pro175 animals were monitored bi-monthly before and after the cell injection. A statisticalanalysis was performed on the data of the cholesterol lev-els from week 0 until week 16. Owing to the death of some of the animals on days 12,14 and 16, some groupsdo not contain the same amounts of animals per groupthroughout the study. The levels of cholesterol of theApoE animals were statistically lower compared with thelevels of their nontreated counterparts ( P    0.03). As istypical in this animal model, the cholesterol levels con-tinued to rise in the control animals reaching an absolutelevel of 391 mg/dl which represents 231% of pre-injectionlevels 16 weeks after injection (Figure 2). 33 In contrast, adecrease in cholesterol levels was observed in the animalshaving received the apoE secreting mini-organ as soonas 2 weeks after the injection and remained under the baseline level until week 16. After 3 months, the mini-organs were examined and found to be vascularized withno sign of tumorigenicity at the site of injection orelsewhere. Qualitative analysis of plaque formation in the heart  Consistent with the reported pathology in this model,untreated animals (five) developed severe aortic plaques(Figure 3c and d). Large lesions were found throughoutthe aorta, some of which showed extensive calcification.In contrast, the experimental animals (five) receiving theapoE secreting mini-organs showed considerably lesslesion and plaque formation (Figure 3a and b). Ex vivo mini-organ YS apoE expression  Despite a sensitive apoE ELISA method, no detectablelevels of apoE were found in the serum of any of theexperimental animals. Single cell cultures were estab-lished from several mini-organs and evaluated for thesecretion of apoE. The  ex vivo  expression of apoE in theapoE + /Pro175 cells was even higher (up to 1000  versus 150 ng/24 h/10 6 cells) than seen in the primary culturesof these cells. This continued high expression of apoEcells after these cells have been removed from the ani-mals indicates that the cells were functional in the animaland have probably enhanced their secretion capability.The lack of apoE-detection in the treated animal may be due to a rapid cycling due to the high levels of cholesterol.  Figure 2  Cholesterol levels of apoE-deficient mice grafted with mini-organs containing apoE secreting Pro175 cells or non-secreting controlcells. Cholesterol levels were measured in each animal before the mini-organ injection and thereafter every 2 weeks for 16 weeks. Bars representmeans of the relative variation of the cholesterol levels  ±  s.e.m. for treatedand control groups at each time of measurements. Discussion  The current study was undertaken to determine the effi-cacy of using a novel endothelial cell line (Pro175) iso-lated from the murine yolk sac as a potential candidatefor cell-based gene therapy. There have been various celltypes which have been studied for their potential useful-ness in gene therapy, including epithelial cells, endo-thelial cells, skin fibroblasts and myoblasts, but each of these cell lines has had problems with viability, secretion,long-term culture or rejection. 1,3,7–10 The use of endo-thelial YS cells in gene therapy offers several advantagesover currently used systems. These cells have a long life-span even after being grafted in animals, they retain theirphenotype over an extended period of time, they aregood secretors of exogenous foreign proteins, and theyare not tumorigenic when grafted in immunocompro-mised animals. Furthermore, these YS endothelial cellsform hollow capillary-like structures when incubatedunder the appropriate conditions. These endothelialtubes may anastomose with the host vascular system,leading to a more efficient delivery of proteins. Lastly,these cells do not spontaneously express MHC I antigens.Other studies utilizing murine YS cells have shown themnot to be immunogenic in immunocompetent animals. 1  Endothelial cells as a gene delivery platform L Cioffi  et al  1156  Figure 3  Cardiac sections from control and experimental apoE-deficient animals. H&E staining of sections (a, b) from experimental animals receivingapoE secreting YS mini-organs and (c, d) from animals killed after 3 months receiving control YS mini-organs. Note severe lesion formation in thecontrol aortic region indicated by arrows, whereas there is little plaque formation seen in the experimental apoE animal. This immunological property may be an important fea-ture, should these cells be used in allogeneic or xeno-geneic settings. Future studies are planned to study theseimmunological properties of YS cells in appropriatemodel systems.We have chosen the atherosclerosis disease model as amethod in which to study the potential of our YS cells tosecrete a biologically relevant protein, apoE3. In thepresent study, engineered YS endothelial cells were usedto secrete apoE using a mini-organ strategy with the goalof averting atherosclerotic lesions in apoE-deficient mice.Several animal models have been proposed to study themechanisms of atherosclerosis and to develop new treat-ments. Among these animal models, mice that lack theapoE3 gene are readily available. In this animal model,the phenotype features no detectable apoE3 secretion,high total serum cholesterol levels (five times normallevels), high VLDL and chylomicron levels, and severeaortic lesions which can be seen as early as 3 monthsof age.The delivery of apoE by a variety of methods has beendescribed. 34–36,38–40 Although adenovirus-based genetransfer strategies are powerful, they raise major safetyconcerns when envisioned for treating long-term humandiseases such as atherosclerosis. 37 Also, gene expressionfrom these vectors is only transient and strong immuno-logical responses increase the risk of further injections.Alternatively, bone marrow transplantation afterirradiation has also been used for gene delivery of apoE,however, this method has high morbidity and mortalityassociated with it. 34,35,40 Transformed cell lines such asBHK expressing the apoE gene may be tumsrcenic andfibroblastic lines such as NIH/3T3 cells may have only alimited life-span.In the current study, Pro175 cells were engineered witha plasmid bearing the human apoE3 cDNA driven by theCMV promoter. High levels of apoE expression wereobtained  in vitro  before grafting these cells. Similarly, thecells were still secreting high levels of apoE after having been in the animals for 12 weeks. The  in vivo  tests showedthat animals containing mini-organs secreting apoE werefound to have significantly lower cholesterol levels thancontrol animals. Even though there was a dramatic effectobserved, apoE secretion did not reach detectable levelsin the serum of the experimental animals. One possibleexplanation for this is that the apoE protein was beingquickly linked to the lipids, taken up by the cells andtransported to tissues, thereby having a low serum con-centration in these animals. This explanation is corrobor-ated by the fact that Bellosta  et al 35 found only 3–8% of normal apoE levels after irradiation and injection of apoE +  macrophages. Despite the lack of detectable levelsof apoE in the ApoE + /Pro 175 animals, apoE was beingsecreted by the cells as indicated by the  in vitro  cultureof the cells. In addition, it is difficult to explain the effectson cholesterol levels and on lesion area other than by asignificant apoE secretion. Experiments are beingplanned to investigate tissues such as the liver to deter-mine if there are detectable levels of apoE sequestered inthe individual tissues in the animals.In accordance with the apoE  in vitro  assays and thecholesterol  in vivo  assays, a significant reduction in theaortic lesion areas was observed in the experimental ani-mals when compared with the control animals 3 months  Endothelial cells as a gene delivery platform L Cioffi  et al  1157 after the mini-organ implantation. Contrasting with theexperimental animals, the control apoE-deficient animalsshowed severe plaque formation containing calcifiedareas as well as infiltration of a large population of foam cells.In conclusion, the use of endothelial YS cells in genetherapy offers several advantages over currently usedsystems which are being utilized to deliver missing pro-teins. Bone marrow transplantation results in significantmorbidity and mortality associated with this procedureand adenovirus- and retrovirus-based gene transfer stra-tegies raise major safety issues in the treatment of humandiseases. Alternatively, transformed cell lines may betumsrcenic and fibroblastic lines may have only a lim-ited life-span. Thus, a well established YS endothelial cellline such as Pro175 may be a good platform for long-term in vivo  protein production. Indeed, these cells have a longlife-span even after being grafted, they retain theirphenotype over an extended period of time, they aregood secretors of exogenous foreign proteins, and theyare not transformed cells. Furthermore, these YS endo-thelial cells may develop vascular structures that mayanastomose with the host vascular system, leading to amore efficient delivery of proteins. This first study usingthe mini-organ strategy and YS-derived endothelial cellsfor the treatment of atherosclerosis in apoE-deficient micehas shown efficacy and the potential of the system. Cell- based therapy may have numerous applications in dis-eases other than cardiovascular diseases such as diabetesand endocrinopathies. Future studies are necessary toevaluate the transplantation issues, the mode of implan-tation and/or the encapsulation devices. Materials and methods  Endothelial yolk sac cell lines  Murine yolk sacs were isolated from day 8–9 pregnantfemales (C57BL/6J) and dispersed by gentle aspiration in0.025% trypsin. They were then placed on 0.1% gelatin-coated dishes in culture medium containing alpha MEM(Gibco/BRL, Gaithersburg, MD, USA), 18% fetal bovineserum (FBS, Hyclone), 10% conditioned medium contain-ing 40 ng/ml LIF, 70   M  MTG (Sigma), and 50   g/mlgentamicin (Gibco/BRL). These cells are mildly adherentand require treatment with mild trypsinization (0.25%) orPBS/EDTA (1 m M ) for passage or for single cell prep-arations. The stable YS cell line used in this study isnamed Pro175. For some experiments, Pro 175 c-ells wereplaced on dishes pre-coated with Matrigel (CollaborativeBiomedical Products) at 2  ×  10 5 cells/35 mm. The cellswere cultured under standard conditions at 37 ° C, 5%CO 2 , and their morphology and amount of endothelialcell network formation was photographed. PCR analysis  Total cellular RNA was isolated and extracted usingUltraspec-II RNA Isolation System (Biotecx Laboratories,Houston, TX, USA). cDNA was generated using RNAPCR core reagents and a 9600 gene amp thermocycler(Perkin Elmer). All reactions were standardized based onthe relative expression of the   -actin gene. The cDNA pri-mers for   -actin were 5  -GTGACGAGGCCCAGAGCAAGA-3   (forward) and 5  -AGGGGCCGGACTCATCGTACTC-3   (reverse). The primers for  flk-1  were 5  -AGCTGTCGCTCTGTGGTTCT-3   (forward) and 5  -TGGGGAGAGTAAAGCCTATC-3   (reverse). The ACE pri-mers were 5  -TGC ACTGGTGGTATCTTCGAACCA-3  (forward) and 5  -CATACTCTTCCACGAACCTGTCAG-3   (reverse). Each reaction was amplified using the fol-lowing PCR conditions: 94 ° C (30 s), 53 ° C (30 s) and 72 ° C(30 s), the total number of cycles was 40. Ethidium bro-mide staining following agarose gel electrophoresis con-firmed the expected 934 bp product for   -actin, the551 bp product for  flk-1 , and a 500 bp product for  ace . Flow cytometry LDL receptor analysis  The rapid uptake of a fluorogenic derivative of acetylatedlow density lipoprotein (dilAcLDL) (Biomedical Technol-ogies, Palatine, IL, USA) was measured by incubatingcells for 4 h at 37 ° C with alpha-MEM containing10   g/ml of dilAcLDL. After washing three times, thecells were fixed in 3% paraformaldehyde and visualizedunder a fluorescent microscope for characteristic punc-tate fluorescence of receptor-mediated update of dilAcLDL. To identify cell surface markers, the cells wereremoved from tissue culture flasks by PBS/EDTA treat-ment. Cells (10 6 ) were then stained for 40 min in PBS con-taining 0.1% BSA and 0.1% NaN 3  with 1   g/100   l of anti-mouse CD105 (VCAM-1), Mac-1, F4/80, MHC I(H2K  b ), MHC II (I-A h ), or rat IgG2A isotype control(Pharmingen). After washing, cells were incubated with1   g/ml anti-rat IgG FITC-labeled secondary antibody(Southern Biotechnology Associates, Birmingham, AL,USA) and fixed with 1% freshly prepared paraformal-dehyde in PBS. The yolk sac cells were quantified by flowcytometry analysis (EPICS XL; Coulter). Parallel experi-ments were run on unfixed cells which, following stain-ing, were also characterized by flow cytometric analysis. MHC induction with     -interferon  Yolk sac cells were plated (800/cm 2 ) and cultured inprimary endothelial medium with 400 U/ml   -IFN(Genzyme, Cambridge, MA, USA) for 96 h. 41 The primaryendothelial medium consists of the following compo-nents: DMEM (Gibco/BRL), 10% FBS (Hyclone), 1% BMEvitamins (Gibco/BRL), 25   g/ml heparin (Sigma),50   g/ml ECGS (Collaborative Biomedical Products),20% EOMA-conditioned medium, and 50   g/ml genta-micin. The cells were harvested, blocked with 1   g/mlanti-Fc receptor antibody (Pharmingen) and incubatedwith 1   g/ml FITC-conjugated anti-MHC I (H2K  b ) anti- bodies (Pharmingen) for 30 min at 4 ° C. The endothelialYS cells were analyzed by flow analysis (EPICS XL;Coulter), with the MHC I-positive stained cells showinga shift in fluorescence in comparison to the negativelystained cells. Plasmid construction and transfection procedures  The human apoE3 cDNA fragment was obtained by an Eco RI/  Hin dIII digestion from the pHE53 plasmid. 43 Theisolated 1200 bp fragment was cloned into the pcDNA3expression plasmid (Invitrogen, Carlsbad, CA, USA). Forcell transfection, plasmid DNA was obtained usingQiagen columns (Santa Clarita, CA, USA). Following anethanol precipitation and wash step, plasmid DNA wasresuspended in sterile endotoxin free water at a concen-tration of 500 ng/  l.For transfection and selection, Pro175 YS cells (2.5  ×  10 5 cells/35 mm dish) in 2.5 ml YS complete medium (alpha-
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks