Transgenic chickens as bioreactors, biotechnologia UP Wrocław losowe pierdoły, Biotechnologia zwierząt i ...

[ Pobierz całość w formacie PDF ]
//-->DDT•Volume 10, Number 3•February 2005REVIEWSTransgenic chickens as bioreactors forprotein-based drugsSimon G.Lillico,Michael J.McGrew,Adrian Sherman and Helen M.SangThe potential of using transgenic animals for the synthesis of therapeutic proteinswas suggested over twenty years ago. Considerable progress has been made indeveloping methods for the production of transgenic animals and specifically in theexpression of therapeutic proteins in the mammary glands of cows, sheep and goats.Development of transgenic hens for protein production in eggs has lagged behindthese systems. The positive features associated with the use of the chicken in terms ofcost, speed of development of a production flock and potentially appropriateglycosylation of target proteins have led to significant advances in transgenicchicken models in the past few years.Simon G. LillicoMichael J. McGrewAdrian ShermanHelen M. Sang*Roslin Institute,Roslin,Midlothian,UK, EH25 9PS*e-mail:helen.sang@bbsrc.ac.ukThe possibility of using transgenic animals for theproduction of therapeutic proteins was raisedshortly after the development of the first methodfor the genetic modification of mice. Significantprogress has been made towards this goal based ontargeting expression of pharmaceutical proteins tosecretory organs of animals, particularly the mammarygland of livestock mammals [1–4]. The predictedadvantages of such production systems comparedwith synthesis in microbial cells or mammalian tissueculture cells include the ability to produce large quan-tities of posttranslationally modified and complexproteins and the possibility of providing a cheaperalternative to the use of large-scale fermentors. Thecost of production of glycosylated proteins fromlarge-scale cell cultures is considerable, and it is pre-dicted that transgenic animal production will be morecost effective [5]. A key requirement for the produc-tion of human proteins for therapeutic purposes isthat the production method should result in a proteinthat incorporates the posttranslational modificationsof the naturally expressed protein [6]. The posttrans-lational modification of many proteins is essentialfor protein function, and if a protein is not appro-priately modified it could have a short half-life inthe patient and poor therapeutic efficacy. Absenceof specific glycosylations can result in a protein thatis immunogenic or unmask peptide epitopes thatmight be antigenic. The requirement for appropriateglycosylation of each therapeutic protein for overallefficacy must be evaluated for each target protein.The basic strategy for targeting expression of aforeign protein to, for example, the mammarygland, has been to identify the regulatory sequencesof milk protein genes required for high-level, tissue-specific expression and subsequently link these tothe coding sequence of a therapeutic protein. Next,various methods can be used to introduce the result-ant transgene into the genome of the host species[2,3]. Over the past decade, various recombinanthuman proteins, for example, proteins that are typ-ically isolated from human blood products andmonoclonal antibodies [7], have been successfullyproduced in the milk of transgenic mammals suchas cows, goats, sheep and rabbits [5]. With the prom-ise of cheaper production cost and ease of purification,it was expected that biopharmaceutical productsproduced in transgenic animals would quickly passfrom the barnyard to the market place. Althoughprogress has been slower than predicted, significantwww.drugdiscoverytoday.com1359-6446/04/$–see front matter ©2005 Elsevier Ltd. All rights reserved. PII: S1359-6446(04)03317-3191Reviews•DRUG DISCOVERY TODAYREVIEWSDDT•Volume 10, Number 3•February 2005TABLE 1Composition of an egg produced by a henConstituentProportion of egg byweight (%)6030Components ofindividualconstituents (%)WaterProteinYolkWaterLipidProteinShellConstituentOvalbuminOvotransferrinOvomucoidLysozymeaintroduced transgene to the wild-type population, mustbe resolved.Why develop chickens as bioreactors?Although there are an increasing number of options forproduction systems for therapeutic proteins, it is recognizedthat the resources of commercial production create abottleneck. Pharmaceutical proteins produced in eggsmight have significant advantages for specific targetdrugs, including appropriate glycosylation, lower coststhan either cell culture or transgenic mammalian systems–and faster scale-up.Reviews•DRUG DISCOVERY TODAY192Albumen8811503020–10aMajor albumen proteins in a 60 g eggWeight (g)2.200.500.500.15Fraction of totalprotein (%)54.012.012.03.4Production of proteins in eggsModern layer hens produce eggs in a 20–24 h cycle, witheach ovulated yolk initially acquiring layers of egg white,followed by shell membranes and eventually a shell duringits passage through the 50–70 cm length of the matureoviduct. At lay, a typical egg weighs around 55–60 g, withthe yolk constituting ~30%, the white 60% and the shell10% of the total weight (Table1)[9]. Albumen (protein)is the main constituent of dried egg white, and isbiochemically relatively simple. Nine different proteinsaccount for 87% of the total protein mass, with ovalbumin,ovotransferrin and ovomucoid being the most abundant(54%, 12% and 12%, respectively). The relatively lowcomplexity of egg white should facilitate purification ofrecombinant proteins from albumen. There is considerablecommercial expertise available in processing of eggs andpurification of some components, for example, lysozyme.The genes encoding egg white proteins are translatedin the secretory cells of the magnum of the oviduct of thelaying hen (Figure1a,c).Secretion is stimulated as the yolkpasses down the oviduct (Figure1b).Transgenic expres-sion of therapeutic proteins in egg white is likely to beachieved by using the regulatory sequences of the genesencoding egg white proteins to drive expression of asequence that encodes a therapeutic protein and that hasbeen modified to promote secretion. The regulatorysequences of two egg white protein genes (ovalbumin andlysozyme) have been characterized in detail at themolecular level and are therefore the most obvious can-didate genes to modify to target expression to the oviduct.Ovalbumin has been less well-characterized because theonly system in which the regulatory elements can be stud-ied is through transient transfection of cells isolated fromthe oviduct of female chicks that have been stimulated byhormone injections to develop prematurely. Significantregulatory elements have been identified, including steroidresponse elements and a negative regulatory region thatinhibits expression of ovalbumin in tissues other than theoviduct [10,11]. There is also evidence that sequenceswithin the transcribed region of the ovalbumin locus con-tribute to the stability of the ovalbumin mRNA, and there-fore the total amount of ovalbumin protein synthesized.It could be important to include such sequences in anFigures expressed as percentages of the total albumen proteins.Reprinted with permission from [21].advances are now being made and several products havereached various stages of clinical trials. Recombinant C1inhibitor (Pharming Group NV) purified from the milk oftransgenic rabbits has recently entered Phase III clinicaltrials. If successful, market launch for this product isexpected in 2005. In addition, recombinant humanantithrombin III protein produced in the milk of trans-genic goats (GTC Biotherapeutics) has completed PhaseIII clinical trials. This product is currently under reviewfor market authorization in Europe, the successful com-pletion of which promises to galvanize the field of animalbiopharming.More recently, the development of genetically modifiedplants for production of pharmaceuticals, a potentiallycompetitive approach to that of using animal bioreactors,has made significant progress, with many companiescurrently in start-up phase [8]. These methods entail theproduction of recombinant human proteins in the leavesor seeds of transgenic plants. The advantages of this systeminclude low production costs and absence of mammalian-derived viral sequences and pathogens. Several recombi-nant proteins produced in plants have entered clinicaltrials and lipase from transgenic maize has been grantedorphan drug status (www.meristem-therapeutics.com).However, there are two key issues that need to be addressedbefore plant-derived biopharmaceuticals will reach themarket place. First, the glycan groups that are added toproteins are not the same in animals and plants. Thisproblem could be circumvented by additional manipula-tion of the production plant species to ‘humanize’ theglycosylation patterns of the proteins produced. Second,and more importantly, the real or perceived issue ofenvironmental biosafety, which involves the risk of foodcrop contamination by the horizontal spread of thewww.drugdiscoverytoday.comDDT•Volume 10, Number 3•February 2005REVIEWS(a)(b)5 cmGlycosylation of proteinsA potential advantage of using chickens as bioreactors isthat some of the oligosaccharide moieties added to nascentpolypeptides in the chicken have greater similarities withthe sugars used by humans than those of other mammals,although the data available on glycosylation in birds arelimited. A study of the glycosylation of IgGs in differentspecies highlighted the species-specific variation in thesialylation ofN-andO-linkedglycans, a major mode ofposttranslational glycosylation [16]. IgGs from somemammals, including cows, sheep and goats, compriseoligosaccharides withN-glycolylneuraminicacid (NGNA),whereas other species, including the rabbit, containNGNA andN-acetylneuraminicacid (NANA). Of thespecies investigated, it was observed that the IgG oligosac-charides of humans and chickens only incorporate NANA.Furthermore, humans produce Hanganutziu–Deicher anti-bodies that recognize NGNA, resulting in potential rejec-tion of proteins carrying this epitope.The absence of some glycosylations of proteins in humansis unusual in comparison with most other mammals. Forexample,α1–3-galactose(α1,3-Gal) epitopes are found ontissues and secreted glycoproteins of mammals, with theexception of humans, apes and Old World monkeys [17].In humans, the gene for the enzymeα1,3-galactosyltrans-ferase is inactive, and 1% of circulating B-lymphocytes pro-duce anti-α-Gal antibodies in response to enteric bacteria,which contributes to the rejection of tissues in transplants(e.g. pig xenografts) and could prove problematic for pro-teins produced in the milk or other body fluids of live-stock species. By contrast, chickens do not produceα1,3-Gal [18], reducing the potential risk of an adverse immuneresponse to pharmaceutical proteins produced in eggs.(c)5 mmDrug Discovery TodayFIGURE 1The oviduct of a laying hen. (a)An oviduct from a hen isolated during passage of ayolk through the magnum.(b)The oviduct opened to show the yolk and theaccumulating egg white.(c)A section through the magnum of a transgenic hen thathas been stained to detect expression of thelacZreporter gene [34].ovalbumin-derived transgene, if high levels of therapeuticprotein expression are to be achieved. Functional elementsof the lysozyme gene have been investigated in more detailthan those of ovalbumin because the chick lysozyme genecan be introduced into transgenic mice, where it is expressedin the macrophages. Analysis of the elements involved andthe effects of deletion of specific sequences indicated thatthe entire lysozyme region of ~20 kb is required for tissue-specific, copy number-dependent expression [12,13].In contrast to egg white, which contains modest con-centrations of lipid, a complex mixture of lipids form thebulk of dried egg yolk [9], the components of which caninclude triglycerides, sterols (mainly cholesterol), phos-pholipids and glycolipids. The protein fraction of the yolkcomprises a variety of products, including vitellogenins andlow-density- and high-density-lipoproteins. Vitellogenins,and some other yolk proteins, are synthesized in the liverand accumulate in the yolk via a receptor-mediatedprocess during the days preceding ovulation. If recombi-nant protein sequestration in the yolk is required, internalrecognition sequences involved in this process wouldneed to be incorporated in a recombinant protein. Theyolk also sequesters up to 400 mg of IgY, a form of IgG.Uptake of a chimeric human IgG has been demonstratedAdditional potential advantages of chicken bioreactorsProduction of human proteins in hens could be themethod of choice for some proteins that are toxic tomammals. For example, although expression of humanerythropoietin in the mammary gland of rabbits had adeleterious effect [19], it is unlikely that human erythro-poietin will be active in chickens [20].The egg is an attractive vehicle for the recovery of thera-peutic proteins because the contents of the egg are sterileand proteins in egg white are stable, suggesting that ther-apeutic proteins could have a long half-life in egg white[21]. Vaccines for human use have been produced in theeggs of hens for many decades, and thus establishedwww.drugdiscoverytoday.com193Reviews•DRUG DISCOVERY TODAY[14] and the sequences important for receptor-mediateduptake into egg yolk identified [15]. Synthesis of recom-binant antibodies in transgenic hens and recovery fromyolk could be exploited specifically for therapeutic anti-body production. Recombinant antibodies that are designedto be sequestered in yolk could be purified using a processalready established for the isolation of polyclonal anti-bodies raised in laying hens.REVIEWSDDT•Volume 10, Number 3•February 2005FIGURE 2(a)Vector pluspackaging plasmidsVirus(b)Production of transgenic birds using a lentiviral vector. (a)Aftercotransfection of the viral vector–transgene construct into tissueculture cells, the virus particles are harvested, concentrated andapproximately 2µlis subsequently injected below the embryonic discof a newly laid egg.(b)Injection of an egg with a virus preparation.Theeggs are then cultured to hatch (three weeks incubation).(c)Hatchedchicks are raised to sexual maturity (approximately 20 weeks) andmales are screened to detect the viral transgene in sperm.They arecrossed to stock hens and hatched chicks are screened to identifyhemizygous transgenic birds.(d)The transgenic birds are raised tomaturity and can be bred to generate a transgenic flock or analysed fortransgene expression and accumulation of therapeutic protein in theeggs of the transgenic hens.Reviews•DRUG DISCOVERY TODAY194(c)XSelecttransgenicchicks(d)production flock can be built up within a comparativelyshort period of time (Figure2).Cockerels can be mated toten hens every day or so and each hen will then lay ~10fertile eggs. Once a transgenic cockerel has been identified,he can be used to breed over 100,000 transgenic offspringin a year. The costs of developing a transgenic productionflock for a specific therapeutic protein have been esti-mated; however, because the companies developing thisapproach are still at an early stage no definite costs arecurrently available. The production of founder transgenicbirds, from which a flock can be developed, could poten-tially cost a few thousand dollars, if the recent advancesin transgenic technology prove to be applicable for thispurpose, which is considerably less than the cost of pro-ducing transgenic cattle. Regulations have yet to beestablished for the conditions for rearing transgenic birdsfor protein production: a convention for rearing birds inspecific pathogen free environments has been establishedbut conforming to these rules would increase the cost ofproduction. The speed at which a production flock couldbe generated and the high rate of production (300 eggsyear−1hen−1) suggest that protein production in henscould prove extremely useful for target proteins that arerequired in large amounts. For example, if each transgenichen deposited 100 mg of therapeutic protein in an egg,then a single hen could produce 30 g of protein each year.Progress towards development of chickenbioreactorsMethods for production of transgenic chickensThere are significant differences between the reproductivephysiology of birds and mammals: avian embryos developfrom a large yolky egg that is enveloped in a hard shellafter fertilization and the embryo then develops in theincubated egg. By the time a fertile egg is laid, the chickembryo has already developed on the yolk to a stage atwhich it consists of ~60,000 cells. Most methods for geneticmodification of mammals involve manipulation of oocytes,fertilized eggs or early embryos recovered from femaledonors, with transfer to recipient females after manipu-lation to introduce transgenes. The large size and fragilityof the egg produced by a hen (i.e. the yolk) and prob-lems in recovering eggs shortly after fertilization makeDrug Discovery Todayregulations are likely to facilitate the development of regu-latory procedures for therapeutic protein production in eggs.A significant advantage of the use of hens as bioreactors,in comparison with the utilization of cattle, sheep or goats,is the short incubation time of three weeks and the rela-tively short generation time of ~20 weeks. A transgenicwww.drugdiscoverytoday.comDDT•Volume 10, Number 3•February 2005REVIEWSmanipulation of the egg for injection of DNA transgenesor nuclear transfer at this early stage difficult. By contrast,it is easy to obtain fertile newly laid eggs but, as the em-bryo develops, attempts to modify embryos genetically atthis stage face different challenges [22].Despite the technical challenges involved, several dif-ferent approaches have been taken to developing amethod for the efficient genetic modification of chickens.These have been directed at manipulation of the avianembryo at three key stages of development: (i) the newlyfertilized egg; (ii) the embryo in newly laid eggs; and (iii)embryos that have reached ~2 days of incubation, whenthe primordial germ cells (PGCs), the precursors of thegametes, can be accessed. The different approaches andtheir relative successes have been discussed extensively[23–27], with the most successful methods described thusfar being based on the use of viruses as gene transfer vectors.Retroviral vector gene transferThe first genetically modified birds were produced usingvectors derived from avian retroviruses, which wereobvious candidates for gene transfer vectors because theyintegrate into the chromosomes of their host as an oblig-atory step in their normal life cycle. Modification of vectorsderived from avian leucosis virus (ALV) and reticuloen-dotheliosis virus resulted in replication-defective vectorsthat only undergo a single cycle of integration within thehost [22,23]. A replication-defective vector comprises acopy of the viral genome from which the genes encodingviral proteins have been removed and the transgene ofinterest has been inserted; this modified copy of the viralgenome retains the sequences required for viral DNA in-tegration and viral genome packaging. Copies of the viralprotein genes, including a gene encoding viral coat pro-tein, are maintained as separate plasmids. The vector andviral protein genes are cotransfected into tissue culturecells to enable the synthesis of complete viral particles.The viral particles are collected and, for introduction intoa chick embryo, injected into the embryo in a newly laidegg. Because the chick embryo at this stage of develop-ment is multicellular, any transgenic birds that hatch willbe chimeras with respect to integration of the viral vec-tor. An ALV vector was used to generate transgenic birds,but the frequency of transgenic cockerels produced waslow (10%) and only one, from a total of 56 cockerelsscreened, produced transgenic offspring. The frequencyof production of transgenic offspring from this male wasalso low – ~0.7% [21]. Although this method is inefficient,it has been used to generate transgenic birds that expresslow levels of a therapeutic protein. An increased germlinetransduction frequency was obtained using a vector de-rived from the avian spleen necrosis virus, with one of 15males demonstrating a germline transmission frequencyof ~0.9% [28].These vectors are derived from oncogenic retrovirusesand the results from their use in birds, and from studiesusing murine retroviral vectors in mice and tissue culturecells, suggest that they have low efficiencies in the pro-duction of transgenic animals and that expression oftransgenes introduced using these vectors is likely to besusceptible to silencing. A novel group of vectors hasrecently been derived from the lentivirus class of retro-viruses [29]. These vectors, which were developed prin-cipally for gene therapy applications, have potentialadvantages over oncoretroviral vectors for production oftransgenic chickens. Moreover, they have been used togenerate transgenic mice, pigs [30,31] and cattle [32] athigh efficiency and, in mice, reliable tissue-specific ex-pression that was maintained after germline transmissionhas been demonstrated [33,34].The efficiency with which lentiviral vectors, which arederived from equine infectious anaemia virus, could beused to transduce the chicken germline was investigated[35]. Vectors carrying a reporter transgene, which wouldfacilitate analysis of expression in transgenic birds, werepackaged using an envelope coat protein from vesicularstomatitis virus. The virus was concentrated to give hightitres and injected into newly laid eggs that were subse-quently cultured to hatch. Twelve cockerels were gener-ated and all identified as chimeric transgenic animals. Tenof these birds were mated and transgenic chicks detectedin offspring from all the cockerels tested at frequenciesfrom 4% to 45%. Analysis of the reporter transgenes inseveral independent transgenic lines showed that the ex-pression pattern was conserved. Moreover, the expressionpattern was maintained after germline transmission tothe next generation with no detectable silencing of trans-gene expression. The key limitation of lentiviral vectorsfor production of transgenic animals is that the size of thetransgene is restricted to ~8 kb. The high efficiency of thissystem will facilitate the testing of many different trans-gene constructs, specifically those designed to expresstherapeutic proteins for incorporation in the egg white oryolk of transgenic hens.Expression of proteins in transgenic chickensTo date, there is limited information in the literaturedescribing the deposition of foreign proteins in eggs.Analysis of eggs from transgenic birds, which were gen-erated using an ALV vector containing a CMV promoterlinked to the coding sequence for human interferonα-2b,detected up to 200µgof human interferon in the whiteof a laid egg [36]. Although the CMV promoter is oftenconsidered ubiquitous, considerable variation in expressionlevels of the reporter genelacZ,which is driven by the CMVpromoter, were detected between tissues in transgenichens, with expression in the oviduct among the lowest(Figure1c)[35]. TheO-linkedglycans of recombinanthuman interferonα-2bexpressed in transgenic hens wereanalysed and an estimated 38% were correctly glycosy-lated with respect to the naturally occurring humanprotein [36].www.drugdiscoverytoday.com195Reviews•DRUG DISCOVERY TODAY [ Pobierz całość w formacie PDF ]