This page has been archived and is no longer updated
Genetic modification: The production of recombinant pharmaceutical proteins in plants
Author: Julian K. Ma
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"794 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS Proteins are widely used in research, medicine and industry, but the extraction of proteins from their nat- ural sources can be difficult and expensive. Also, the use of pharmaceutical proteins from natural sources can pose risks. For example, many people have contracted diseases from contaminated blood products or hor- mones. Other proteins, such as SINGLE-CHAIN FV FRAGMENTS (scFvs), are not found naturally. So, a simple and inex- pensive system that allows the large-scale production of safe recombinant proteins would be highly desirable. Traditional production systems that use microbial fer- mentation, insect and mammalian cell cultures, and transgenic animals have drawbacks in terms of cost, scalability, product safety and authenticity 1?3 (TABLE 1). Recent studies have shown that MOLECULAR FARMING in plants has many practical, economic and safety advan- tages compared with more conventional systems, and so the use of plants for large-scale protein synthesis is gain- ing wider acceptance 4,5 . In this review, we discuss the technological basis of molecular farming in plants, with a focus on proteins that can be used for diagnostic, therapeutic and prophy- lactic applications. We provide a broad account of the types of pharmaceutical protein that can be produced on a commercial scale and examine the different expres- sion systems that are being developed. We consider the advantages and limitations of each system, with a focus on the biochemical constraints that need to be addressed for the technology to reach its full potential. Recombinant proteins expressed in plants Plants have provided humans with useful molecules for many centuries, but only in the past 20 years has it become possible to use plants for the production of specific HETEROLOGOUS proteins. The first pharmaceuti- cally relevant protein made in plants was human growth hormone, which was expressed in transgenic tobacco in 1986 (REF. 6).In this study, the hormone was expressed as a fusion with the Agrobacterium nopaline synthase enzyme. Since then, many other human pro- teins have been produced in an increasingly diverse range of crops. In 1989, the first antibody was expressed in tobacco 7 ,which showed that plants could assemble complex functional glycoproteins with sev- eral subunits. The structural authenticity of plant- derived recombinant proteins was confirmed in 1992, when plants were used for the first time to produce an experimental vaccine: the hepatitis B virus (HBV) sur- face antigen 8 .In a further report, the same group showed that the vaccine produced in tobacco plants induced the expected immune response after it had been injected into mice 9 .More recently, the range of THE PRODUCTION OF RECOMBINANT PHARMACEUTICAL PROTEINS IN PLANTS Julian K-C. Ma*,Pascal M. W. Drake* and Paul Christou ? Imagine a world in which any protein, either naturally occurring or designed by man, could be produced safely, inexpensively and in almost unlimited quantities using only simple nutrients, water and sunlight. This could one day become reality as we learn to harness the power of plants for the production of recombinant proteins on an agricultural scale. Molecular farming in plants has already proven to be a successful way of producing a range of technical proteins. The first plant-derived recombinant pharmaceutical proteins are now approaching commercial approval, and many more are expected to follow. SINGLE-CHAIN FV FRAGMENTS (scFvs). Monoclonal antibody derivatives that comprise a single polypeptide in which the variable regions of the heavy and light immunoglobulin chains are joined together by a flexible linker. scFvs are advantageous because only one transgene is required, and the molecules themselves are small and lack the effector functions of normal antibodies; however, a disadvantage is that they are univalent, whereas serum antibodies are divalent. *Department of Infectious Diseases, St. George?s Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK. ? Fraunhofer Institute for Molecular Biology and Applied Ecology (IME), Grafschaft, Auf dem Aberg 1, 57392 Schmallenberg, Germany. Correspondence to J.K-C.M. e-mail: jma@sghms.ac.uk doi:10.1038/nrg1177 GENETIC MODIFICATION NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 795 REVIEWS MOLECULAR FARMING The large-scale production of recombinant proteins in living cells or organisms; frequently applied to the use of crop plants or domestic animals as expression hosts because of the allusion to agriculture. HETEROLOGOUS In the context of this article, a gene or protein that is not derived from the species in which it is expressed. TRANSPLASTOMIC A transgenic plant in which the transgene is found in the plastid genome rather than the nuclear genome. DIABODY A recombinant antibody that comprises the heavy- and light-chain variable regions joined by a flexible peptide linker. The linker is long enough to allow separation of the domains so that two of the polypeptides can assemble into a dimer, making the antibody divalent. MINIBODY A recombinant antibody in which the heavy- and light-chain variable regions are part of the same polypeptide chain, which also includes the heavy-chain hinge region and one heavy-chain constant domain. AGROINFILTRATED LEAVES Usually leaves of tobacco (although many other species can be used) that are transiently transformed with Agrobacterium tumefaciens,which results in the transient expression of recombinant proteins. This is a useful strategy for testing expression constructs and obtaining small amounts of protein for analysis before going to the expense of transgenics. antibody and include two extra polypeptide chains. Two different cell types are required to assemble such anti- bodies in mammals, but plants that express four dif- ferent transgenes can assemble these antibodies in a single cell 18 . As well as full size immunoglobulins, further anti- body derivatives have been expressed successfully in plants, including Fab fragments, scFvs, bispecific Fvs, DIABODIES, MINIBODIES, single variable domains, antibody- fusion proteins, large single-chain antibodies and camelid heavy-chain antibodies (BOX 1).Unlike the pro- teins discussed above, antibodies have been expressed in many different plant systems and extensive comparisons have been carried out to determine optimal parameters for expression. Therefore, yields in excess of 1% of the total soluble protein are routine and antibodies are set to become the first generation of plant-derived therapeutic proteins to be produced on a commercial basis. Although tobacco was used in the early studies, many antibodies are now produced in cereal seeds instead, as protein accumulation in dry seeds allows long-term storage at ambient temperatures without notable degradation or loss of activity. Antibodies have also been produced in less established plant-based expression platforms such as AGROINFILTRATED LEAVES,cell- suspension cultures and virus-infected plants 4 .A partic- ular advantage of these systems is the short development time and rapid onset of protein production, which can be counted in weeks rather than the several months that are required to establish a production line of transgenic or transplastomic plants. At least six types of plant-derived recombinant anti- body have progressed to the preclinical testing stage, with the most advanced product now undergoing phase II clinical trials. These front-runners are discussed briefly in BOX 2. Recombinant subunit vaccines.Since the HBV vaccine was produced and tested (see above), the concept of oral vaccination with raw fruit, vegetables, leaves and seeds has risen in popularity. Edible plants, rather than tobacco, are now the focus of research into HBV vaccine production in plants 19 , and clinical trials have been car- ried out with the surface antigen that is expressed in potato and lettuce 20 .Two further vaccine candidates have reached the clinical trials stage, both of which are recombinant proteins made in plants has extended to include industrial enzymes 10 ,technical proteins that are used in research 11 , milk proteins that are suitable nutritional supplements 12 and new protein polymers with both medical and industrial uses 13 .Examples of pharmaceutical proteins that have been produced in plants are listed in TABLE 2. Proteins as pharmaceuticals. Many pharmaceutical pro- teins of mammalian origin have been synthesized in plants. These range from blood products, such as human serum albumin for which there is an annual demand of more than 500 tonnes, to cytokines and other signalling molecules that are required in much smaller amounts. Until recently, most plant-derived proteins have been produced in transgenic tobacco and extracted directly from leaves. Generally, these proteins are produced at low levels, typically less than 0.1% of the total soluble protein. This low level of production prob- ably reflects a combination of factors, with poor protein folding and stability among the most important. In the past few years, the tobacco chloroplast system has been used to express human proteins at much higher levels. Human growth hormone was produced in TRANSPLASTOMIC tobacco leaves at levels exceeding 7% of the total soluble protein 14 and human serum albumin was produced at levels greater than 11% of the total sol- uble protein 15 .Higher expression levels have also been obtained in other plant species. For example, hirudin, which is expressed as a fusion to the oil-body protein oleosin, has been produced in transgenic canola at 0.3% of the total seed protein 16 . Recombinant antibodies. Antibodies are complex gly- coproteins that recognize and bind to target antigens with great specificity. This individual and specific bind- ing activity allows antibodies to be used for a range of applications, including the diagnosis, prevention and treatment of disease 17 . The production of antibodies in plants represents a special challenge because the molecules must fold and assemble correctly to recognize their cognate antigens. Typical serum antibodies are tetramers of two identi- cal heavy chains and two identical light chains; how- ever, there are more complex forms, such as secretory antibodies, which are dimers of the typical serum Table 1 | Comparison of production systems for recombinant human pharmaceutical proteins System Overall Production Scale-up Product Glycosylation Contamination Storage cost timescale capacity quality risks cost Bacteria Low Short High Low None Endotoxins Moderate Yeast Medium Medium High Medium Incorrect Low risk Moderate Mammalian High Long Very low Very high Correct Viruses, prions and Expensive cell culture oncogenic DNA Transgenic High Very long Low Very high Correct Viruses, prions and Expensive animals oncogenic DNA Plant cell Medium Medium Medium High Minor Low risk Moderate cultures differences Transgenic Very low Long Very high High Minor Low risk Inexpensive plants differences 796 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS showed that the consumption of raw potato tubers that contained 0.3?10 mg of LT-B produced high titres of mucosal and systemic antibodies 21 . Other proteins of medical relevance. Plants have been used to produce several other proteins with direct or indirect medical applications. These include the milk proteins ?-casein and lysozyme, which could be used to improve child health 12,23 , and protein polymers that expressed in potato: the heat-labile toxin B subunit (LT-B) of ENTEROTOXIGENIC Escherichia coli (ETEC) and the capsid protein of Norwalk virus (NVCP) 21,22 .These antigens, from two important enteric pathogens, might be ideal oral-subunit vaccine candidates, as both are multimeric structures that survive in the extreme condi- tions of the human gut. Each protein accumulated to high levels in potato tubers and was correctly assembled into oligomers. Clinical trials with the LT-B vaccine ENTEROTOXIGENIC Producing toxins in the gut that specifically affect the intestinal mucosa. Table 2 | Important pharmaceutical proteins that have been produced in plants Protein Host plant system Comments References Human biopharmaceuticals Growth hormone Tobacco, sunflower First human protein expressed in plants; initially expressed as fusion protein 6,14 with nos gene in transgenic tobacco; later the first human protein expressed in chloroplasts, with expression levels ~7% of total leaf protein Human serum albumin Tobacco, potato First full size native human protein expressed in plants; low expression levels in transgenics (0.1% of total soluble protein) but high levels (11% of total leaf protein) 15,98 in transformed chloroplasts ?-interferon Rice, turnip First human pharmaceutical protein produced in rice 99 Erythropoietin Tobacco First human protein produced in tobacco suspension cells 100 Human-secreted alkaline Tobacco Produced by secretion from roots and leaves 59,60 phosphatase Aprotinin Maize Production of a human pharmaceutical protein in maize 101 Collagen Tobacco First production of human structural-protein polymer; correct modification 13,26 achieved by co-transformation with modification enzyme ?1-antitrypsin Rice First use of rice suspension cells for molecular farming (see REF. 102 103 for discussion of antibody production in rice cell culture) Recombinant antibodies IgG1 (phosphonate ester) Tobacco First antibody expressed in plants; full length serum IgG produced 7 by crossing plants that expressed heavy and light chains IgM (neuropeptide hapten) Tobacco First IgM expressed in plants and protein targeted to chloroplast for accumulation 104 SIgA/G Tobacco First secretory antibody expressed in plants; achieved by sequential 89,90, (Streptococcus mutans crossing of four lines carrying individual components; at present the 105 adhesin) most advanced plant-derived pharmaceutical protein scFv-bryodin 1 Tobacco First pharmaceutical scFv produced in plants; first antibody produced 106 immunotoxin (CD 40) in cell-suspension culture IgG (HSV) Soybean First pharmaceutical protein produced in soybean 72 LSC (HSV) Chlamydomonas First example of molecular farming in algae 107 reinhardtii Recombinant subunit vaccines Hepatitis B virus Tobacco First vaccine candidate expressed in plants; third plant-derived 8,19,20 envelope protein vaccine to reach clinical trials stage Rabies virus glycoprotein Tomato First example of an ?edible vaccine? expressed in edible plant tissue 77 Escherichia coli heat-labile Tobacco, potato First plant vaccine to reach clinical trials stage 21,108 enterotoxin Norwalk virus capsid Potato Second plant vaccine to reach clinical trials stage 22 protein Diabetes autoantigen Tobacco, potato First plant-derived vaccine for an autoimmune disease 109 Cholera toxin B subunit Tobacco, potato First vaccine candidate expressed in chloroplasts 65 Cholera toxin B and A2 Potato First plant-derived multivalent recombinant antigen designed for 110 subunits, rotavirus protection against several enteric diseases enterotoxin and enterotoxigenic E. coli fimbrial antigen fusions Porcine transmissible Tobacco, maize First example of oral feeding inducing protection in an animal 111 gastroenteritis virus glycoprotein S HSV, herpes simplex virus; IgG, immunoglobulin G; IgM, immunoglobulin M; LSC, long single chain; nos, nopaline synthase; scFv, single-chain FV fragment; SIgA, secretory immunoglobulin A. NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 797 REVIEWS Genetic aspects of molecular farming in plants Gene constructs. One of the main aims in molecular farming is the production of recombinant proteins at high yields. To achieve high yields, expression-construct design must optimize all stages of gene expression, from transcription to protein stability. Expression constructs are chimeric structures in which the transgene is brack- eted by various regulatory elements that are known to be active in plants. For high-level transcription, the two most important elements are the promoter and the polyadeny- lation site, which are often derived from the 19S and 35S transcripts of the cauliflower mosaic virus (CaMV) 28,29 . The CaMV 35S promoter is now the most popular choice in DICOTYLEDONOUS PLANTS (dicots). It is a strong constitutive promoter that can be made even more active by duplicat- ing the enhancer region 30 .However, this promoter has a lower activity in MONOCOTYLEDONOUS PLANTS (monocots), so alternatives such as the maize ubiquitin-1 promoter are preferred 31 .The presence of an intron in the 5? untrans- lated region of the expression construct has also been could be used in surgery and tissue replacement. Early experiments with artificial polymers that were based on bovine elastin provided disappointing yields, even though mRNA levels were high, which indi- cated inefficient protein synthesis 24,25 .More recently, it has been shown that human collagen can be produced in transgenic tobacco plants and that the protein is spontaneously processed and assembled into its typical triple-helical conformation 13 .The original plant-derived collagen had a low thermal stability owing to the lack of hydroxyproline residues, but this was remedied by co- expressing the enzyme proline-4-hydroxylase 26 .A syn- thetic spider silk has also been expressed in transgenic plants 27 .Genes that were modelled on the endogenous silk protein genes of the spider Nephila clavipes were synthesized in the laboratory and introduced into tobacco and potato. Proteins up to 100 kDa in size and with 90% identity to the genuine silk protein were pro- duced in tobacco leaves, potato leaves and potato tubers, at up to 2% of the total soluble protein. DICOLYLEDONOUS PLANTS (Dicots). Broad-leaf flowering plants the seeds of which contain two cotyledons (embryonic seed leaves that either remain in the seed when the plant germinates or emerge and become green). Examples include potato, tomato, tobacco and all peas and beans. MONOCOTYLEDONOUS PLANTS Narrow-leaf plants the seeds of which contain one cotyledon. Examples include cereals, grasses, orchids and lilies. Box 1 | Recombinant antibodies that are expressed in plants Plants have been used to express many different antibody forms, including full-length immunoglobulins and artificially constructed derivatives 86,87 .Several different types of immunoglobulin (Ig) have been produced successfully in plants, including IgG (various subclasses), IgA and a chimeric IgA/G. These have ranged in sequence and structure from completely murine to fully humanized. All these molecules comprise two identical heavy chains and two identical light chains, with each chain encoded by a separate transgene. Typically, the genes are introduced into separate plant lines and then stacked by creating hybrids of the transgenic parents. However, multiple gene transfer can achieve the same aim in one transformation procedure, and the correct assembly of immunoglobulins has also been shown after co-infection of tobacco plants with two viruses that each express a separate chain 88 .A variation on this theme is the camelid heavy-chain antibody, which lacks a light-chain component and can therefore be expressed as a single transgene. Secretory IgAs are dimers of the typical serum-type immunoglobulins and include two extra components: the secretory component and the joining chain. Four separate transgenes are required to produce such molecules 89 . Smaller engineered antibody derivatives, which for convenience have both the heavy- and light-chain variable regions on the same polypeptide chain, have also been expressed in plants 86,87 . These include derivatives that form spontaneous dimers (minibodies, diabodies and large single chains) and those that bind univalently to the antigen (single-chain Fv fragments (scFvs) and single variable regions). More specialized derivatives include bispecific scFvs, which contain the variable regions from two parent immunoglobulins and recognize two unrelated antigens, and scFv fusion proteins in which the scFv is genetically fused to a toxin, cytokine or enzyme. Serum antibody (IgG, IgM, IgA or chimaera) Minibody Fab fragment scFv fusion Bispecific scFvDiabody scFv Single variable domain Camelid heavy chain Large single chain Secretory IgA Heavy-chain domain Light-chain domain Fusion partner Variable region Secretory component Joining chain Disulphide bond Hinge region 798 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS tissue-specific promoters, such as those from the maize zein, rice glutelin, wheat glutenin and pea legumin genes. The advantages of such promoters include the increased stability of the protein and the avoidance of protein accumulation in vegetative organs, so preventing toxicity to the host plant and contact with non-target organisms. Inducible promoter systems, which respond to external chemical 35,36 and physical 37 stimuli, might also be used to restrict transgene expression on a temporal basis. shown to enhance transcription in monocots 32 .Widely used polyadenylation sites include those from the CaMV 35S transcript, the Agrobacterium tumefaciens nos gene and the pea ssu gene. Promoters that allow the expression of a transgene in a particular environmental, developmental or tissue- specific manner might also be useful. For example, there are several advantages to the restriction of transgene expression to cereal seeds 33 and potato tubers 34 using EPITOPE A single antigenic determinant on a protein that is recognized by an antibody. A single protein can have many epitopes. LEYDIG CELLS Interstitial cells in the testis that are responsible for the production of male sex hormones, such as testosterone, and are important in male sexual differentiation. Box 2 | Plant-derived antibodies: the front-runners Many antibodies and antibody derivatives have been produced in plant systems, but only seven have reached advanced stages of product development. Some of these will probably represent the first plant-derived recombinant pharmaceutical proteins to achieve commercial status. Avicidin This is a full size immunoglobulin G (IgG) that recognizes the anti-epithelial cellular-adhesion molecule (EpCAM) antigen, which is a marker of colorectal cancer. The antibody has been produced in both animal and plant systems, and the two variants behave in the same way, despite some biochemical differences. The plant-derived antibody, which is produced in corn, was the first to be administered to humans and showed some anti-cancer activity. However, the molecule was withdrawn completely by the developers (NeoRx and Monsanto) because of diarrhoea and other side effects in phase II trial patients, which were probably caused by cross-reaction with related EPITOPES on the intestinal epithelium. These effects were not specific to the plant-derived antibody. CaroRx CaroRx is a chimeric secretory IgA/G that is produced in transgenic tobacco plants through the expression of four separate transgenes. These were stacked by the sequential crossing of independent transgenic plants that each expressed a different component 89,90 . The antibody recognizes the main adhesion protein of Streptococcus mutans,which is the oral pathogen that is responsible for tooth decay in humans. Phase II clinical trials have shown that topical application, after bacteria have been removed from the mouth, helps to prevent recolonization by S. mutans for several months. T84.66 T84.66 is a monoclonal antibody that recognizes carcinoembryonic antigen (CEA), which is a well-characterized marker of epithelial cancers (carcinomas). It has been widely tested for cancer imaging and therapy. T84.66 and recombinant derivatives such as single-chain Fv fragment (scFv) 84.66 have been produced in many different plant systems and used as objective comparators for expression efficiency and product quality 33,68,74,91 .A fusion protein that combines scFv84.66 and interleukin-2 has also been produced and could be used to stimulate lymphokine-activated killer cells and tumour-infiltrating lymphocytes near tumour masses. Anti-HSV and anti-RSV A full-length humanized IgG1 that recognizes herpes simplex virus (HSV)-2 glycoprotein B has been expressed in transgenic soybean and Chinese hamster ovary (CHO) cells 72 .Antibodies from both sources have been shown to prevent vaginal HSV-2 transmission in mice after topical application and would represent an inexpensive preventative for sexually transmitted diseases if they worked similarly in humans. This antibody, along with an IgG that recognizes the R9 protein of respiratory syncitial virus, is being developed by the United States biotechnology company Epicyte Pharmaceutical. 38C13 This scFv antibody is based on the idiotype of malignant B lymphocytes in the well-characterized mouse lymphoma cell line 38C13. Administration of the antibody to mice resulted in the production of anti-idiotype antibodies that are able to recognize 38C13 cells, which help to protect the mice against a lethal challenge with injected lymphoma cells 88 . This system could be adapted to produce antibodies that recognize unique markers on the surface of any malignant B cell and could therefore be an effective therapy for human diseases such as non-Hodgkin lymphoma. The antibodies were produced using virus-infected plants rather than transgenic plants, which is a strategy that is well suited to the rapid and small-scale production that is required to treat individual patients with unique antibodies. Large Scale Biology Corp. has completed phase I trials. PIPP PIPP is a monoclonal antibody that recognizes human chorionic gonadotropin (hCG). The full-length monoclonal antibody, as well as scFv and diabody derivatives, has been produced in transgenic plants and by agroinfiltration in transiently transformed tobacco 92 .Each of the antibodies was able to inhibit the hCG-stimulated production of testosterone in cultured LEYDIG CELLS and to delay uterine weight gain in mice, which are the standard tests for hCG activity. The antibodies could be used for the diagnosis and/or therapy of tumours that produce hCG, pregnancy detection and (emergency) contraception. NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 799 REVIEWS In the absence of further targeting information, pro- teins in the endomembrane system are secreted to the APOPLAST,where they might be retained or secreted into the environment. However, antibody yields can be increased even further if the protein is retained in the ER lumen using an H/KDEL C-terminal tetrapeptide tag, as this compartment has a stabilizing influence 41 .Yields are generally twofold to tenfold greater for ER-retention compared with secretion. Proteins that are retained in this manner are not modified in the Golgi apparatus, which means that they have high-mannose glycans but no plant-associated xylose and fucose residues (see below). Although the measures discussed above will help to achieve high intrinsic yields, the actual amount of recom- binant protein that is then obtained ultimately depends on the processing and purification methods (BOX 3). Other factors that influence transgene expression. Transgene expression is influenced by several factors that cannot be controlled precisely through construct design, which lead to variable transgene expression and, in some cases, its complete inactivation 42 .Such factors include the position of transgene integration, the structure of the transgenic locus, gene-copy num- ber and the presence of truncated or rearranged trans- gene copies. Several strategies have been adopted in an attempt to minimize variation in transgene expression, including, most recently, the use of viral genes that sup- press gene silencing 43 .Preliminary studies have indi- cated that the co-transformation of plants with a primary transgene and a viral-silencing suppressor notably increases transgene expression level 44 .The abil- ity to integrate single-copy transgenes into precise loca- tions in the plant nucleus would eliminate POSITION EFFECTS and the problems that are associated with vari- able locus structure. Several laboratories are therefore investigating ways to improve the efficiency of gene targeting in plants 45 .In practice,however,commer- cially developed transgenic plants undergo an enor- mous amount of screening to identify phenotypic, yield and agronomic variation. The screening includes For example, the mechanical gene activation (MeGA) system that was developed by Cramer (CropTech Corp., Virginia, United States) uses a tomato hydroxy-3- methylglutaryl CoA reductase 2 (HMGR2) promoter, which is inducible by mechanical stress. Transgene expression is activated when harvested tobacco leaves are sheared during processing, which leads to the rapid induction of protein expression, usually within 24 hours. Many other inducible promoters have been developed ? for example, those that use ethanol, dexamethasone and the insecticide methoxyfenozide ? and have recently been reviewed 35 . Transgenes from heterologous species often have a different codon bias to the host plant, which might result in pausing at disfavoured codons and truncation, misin- corporation or frameshifting. Such effects can be avoided by introducing silent mutations into the coding region of the transgene by SITE-DIRECTED MUTAGENESIS,which brings transgene codon usage in line with that of the host. One of the most important factors governing the yield of recombinant proteins is subcellular target- ing,which affects the interlinked processes of fold- ing, assembly and post-translational modification. Comparative experiments with recombinant antibodies have shown that the secretory pathway is a more suit- able environment for folding and assembly than the cytosol, leading to higher yields 38 .Proteins are targeted to the secretory pathway through the inclusion of an N-terminal SIGNAL PEPTIDE in the expression construct. Although most antibodies accumulate to higher levels in the secretory pathway compared with the cytosol, there are some notable exceptions 39,40 ,which indicate that intrinsic features of each antibody might also influence their overall stability. The oxidizing environ- ment of the endoplasmic reticulum (ER), the lack of proteases and the abundance of MOLECULAR CHAPERONES are important factors for correct protein folding and assembly. Also, protein glycosylation occurs only in the endomembrane system and this modification is required for the correct function of many proteins of human origin. SITE-DIRECTED MUTAGENESIS An in vitro mutagenesis procedure that is often carried out using the polymerase chain reaction in which specific mutations are introduced into a DNA molecule. SIGNAL PEPTIDE A short sequence of mainly hydrophobic amino acids at the N-terminus of secreted proteins. This peptide is captured by a signal-recognition particle as it emerges from the ribosome, which allows the ribosome to be transported to the endoplasmic reticulum. MOLECULAR CHAPERONES Proteins the function of which is to ensure correct folding of other proteins during or after synthesis, or the refolding of denatured proteins. APOPLAST The extracellular space. In plants, this is a large and continuous network of cavities under the cell wall. Proteins that are secreted from the cell often remain trapped here. POSITION EFFECTS When transgenes integrate into genomic DNA, the expression level is often influenced by the surrounding chromatin. Local regulatory elements, such as enhancers, also influence transgene expression. Position effects lead to wide variations in transgene expression levels, even in plants that are transformed with identical constructs. AFFINITY TAGS Short peptide sequences added to recombinant proteins, which bind strongly to particular affinity matrices and can be used to purify recombinant proteins. Box 3 | Production issues Highly efficient purification schemes are a prerequisite for the conventional use of recombinant proteins as pharmaceuticals, and this requirement must be built into strategies for molecular farming in plants 93?95 .Although AFFINITY TAGS,such as His 6 or the FLAG epitope, can be used for purification, this type of modification alters the primary structure of the protein and might adversely affect its properties. So, it might be necessary to avoid affinity tags and devise specific purification schemes for individual proteins on the basis of their native structures. Protocols are available for the purification of antibodies from whole plants, plant cell-suspension cultures, leaves and seeds, using protein-A and protein-G-based affinity matrices 96 , although this approach is not suitable for some sub-classes of antibodies and most small recombinant antibody derivatives (including scFvs). In general, the costs of processing are reduced when the product is more concentrated in the starting material and this is one of the advantages of expressing recombinant proteins in the seeds of transgenic plants, in which high levels of the product can accumulate in a small volume 33 .Ifconventional extraction from seeds is too expensive, further strategies to assist purification might be used. One example is the oleosin-fusion platform that was developed by SemBioSys Genetics Inc., in which the target recombinant protein is expressed in oilseed crops as a fusion with oleosin. The fusion protein can be recovered from oil bodies using a simple extraction procedure, and the recombinant protein separated from its fusion partner by endoprotease digestion 16 .Similarly, Schillberg and colleagues have devised a strategy in which recombinant proteins are expressed as fusion constructs that contain an integral membrane-spanning domain that is derived from the human T-cell receptor 97 . The recombinant protein accumulates at the plasma membrane and can be extracted in a small volume using appropriate buffers and detergents. 800 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS case of particle bombardment, superfluous DNA trans- fer occurs because whole plasmids are generally used to coat the microprojectiles 47 .Superfluous DNA transfer is a regulatory problem under the strict new guidelines for the release of genetically modified organisms into the environment. Therefore, several strategies have been developed to avoid the transfer of vector sequences during transformation (see also the article by Stewart et al. in this issue). Incorporation of the barnase gene outside the T-DNA border sequences is one approach that works during Agrobacterium-mediated transformation. This ensures that all plant cells that contain vector sequences linked to the T-DNA are killed, as barnase expression is lethal 50 .Clean-DNA techniques, in which only the necessary transgenes but no vector backbone or superfluous marker genes are introduced into the plant, were also developed for par- ticle bombardment, after it was shown that the micro- projectiles can be coated with minimal cassettes (essen- tially the promoter, transgene and polyadenylation site), without compromising transformation efficiency. Furthermore, transgenic loci in clean-DNA plants are considerably simpler than those of whole-plasmid transformants, and the plants show a notable reduction in the frequency of transgene silencing 51 . Post-translational modification. The protein-synthesis pathway is highly conserved between plants and ani- mals, so human transgenes that are expressed in plants yield proteins with identical amino-acid sequences to their native counterparts. However, there are some important differences in post-translational modifica- tion. One example, as discussed above, concerns the inability of transgenic plants to correctly modify human collagen unless a gene that encodes proline-4-hydroxylase is also expressed 26 .The main difference between pro- teins that are produced in animals and plants, however, concerns the synthesis of glycan side chains. All eukary- otes add glycan chains to proteins as they pass through the secretory pathway, but owing to differences in the levels of different modification enzymes, the glycan- chain structures vary widely across different taxa. Plant- derived recombinant proteins tend to lack the terminal galactose and sialic-acid residues that are normally found in mammals, but have the carbohydrate group ?(1,3)fucose, which has a (1,6) linkage in animal cells, and ?(1,2)xylose, which is absent in mammals although present in invertebrates (FIG. 1). These minor differences in glycan structure could potentially change the activity, biodistribution and longevity of recombinant proteins compared with the native forms. The possibility of plant-specific glycans inducing allergic responses in humans has been con- sidered 52 and the finding that human serum contains antibodies that are reactive against these residues has been interpreted as evidence that the ?(1,3)fucose and ?(1,2)xylose residues might lead to adverse reactions 53 . However, carbohydrate epitopes are rarely allergenic. Moreover, the presence of antibodies in serum is not indicative of an adverse reaction. Finally, these glycan residues are also associated with every normal plant identification of the site of transgene insertion, which allows a rational risk assessment to be made of the likeli- hood of adverse unintentional effects that result from the transformation process. Transformation methods. Two general methods are used to generate transgenic plant lines for molecular farming: AGROBACTERIUM-MEDIATED TRANSFORMATION and particle bom- bardment, in which DNA-coated microprojectiles are accelerated into plant tissue. Each method has advan- tages and disadvantages, and the choice depends on a combination of factors, including the selected host species, local expertise and intellectual property issues. Other methods, such as WHISKER TRANSFORMATION, ELECTROPORATION and PROTOPLAST TRANSFORMATION,have not so far been used for molecular-farming applications. The soil pathogen A. tumefaciens provides a simple method for the transformation of most dicot species and is commonly used for molecular farming in tobacco, alfalfa, pea, tomato and potato 44,46 .Monocots can also be transformed by Agrobacterium,but in most cases the technology has been optimized for selected model varieties. Particle bombardment shows less geno- type dependence and might be the preferred transfor- mation method for cereals, such as rice, wheat and maize, as well as soybean and other legumes 47 .Particle bom- bardment is also necessary for plastid transformation, as the Agrobacterium T-DNA complex is targeted to the nucleus and is therefore unsuitable for gene transfer to chloroplasts 44 .Transformation can also be achieved using Agrobacterium rhizogenes,but this organism is preferred for the production of transgenic root cultures. These transformation methods generally lead to the introduction of superfluous DNA sequences into the nuclear genome 48 .In the case of Agrobacterium- mediated transformation, this is because inefficient processing of the T-DNA BORDER SEQUENCES often results in the co-transfer of flanking vector sequences that might sometimes correspond to the entire plasmid 49 .In the AGROBACTERIUM-MEDIATED TRANSFORMATION Transformation that is achieved using the natural gene-transfer mechanism of Agrobacterium tumefaciens. WHISKER TRANSFORMATION Transformation that is achieved by mixing walled plant cells with silicon carbide fibres that penetrate the cell wall and membrane, which generate pores through which DNA can be taken up into the cell. ELECTROPORATION Transformation that is achieved by exposing cells or protoplasts to a brief pulse of electricity, which results in the formation of transient membrane pores through which DNA can be taken up into the cell. PROTOPLAST TRANSFORMATION Any technique for introducing DNA into unwalled plant cells (protoplasts), such as calcium phosphate transfection, PEG transfection or electroporation. T-DNA BORDER SEQUENCES Imperfect 25 bp direct repeat sequences that flank the piece of DNA that is transferred to the plant genome by Agrobacterium tumefaciens.These sequences are recognized by the bacterial VIRD1 and VIRD2 proteins, which form an endonuclease complex. Cleavage of the border sequences initiates T-DNA transfer. Protein ?(1,4) ?(1,4) ?(1,4) ?(1,4) ?(1,2) ?(1,2) ?(1,2) ?(1,6) ?(1,6) ?(1,3) ?(1,3) GlcNAc Gal NeuAC Man Xyl Fuc Figure 1 | Complex long-chain glycan structure in plants and humans. To ?humanize? the recombinant proteins that are made in plants, ?(1,3) fucose and ?(1,2) xylose residues must be removed (dotted lines), whereas galactose and sialic-acid residues must be added. Blue residues are common to plants and humans. Red residues are found in humans and not plants, so they need to be added. Yellow residues are found in plants but not humans and need to be removed. Fuc, fucose; Gal, galactose; GlcNAc, N-acetylglucosamine; Man, mannose; NeuAC, acetylneuraminic acid (sialic acid); Xyl, xylose. NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 801 REVIEWS expression, the high biomass yield (more than 100,000 kg per hectare for close-cropped tobacco), the potential for rapid scale-up owing to prolific seed production, and the availability of large-scale infrastructure for processing. Although many tobacco cultivars produce high levels of toxic alkaloids, there are low-alkaloid varieties that can be used for the production of pharmaceutical proteins 4 . In most cases, nuclear transgenic plants have been used for production and the proteins have been extracted from leaf tissue. Targeting proteins to the secretory path- way in tobacco can result in them being exuded from the roots or leaves (rhizosecretion and phyllosecretion, respectively) 59,60 .Although not widely adopted so far, this strategy is potentially useful because no cropping or har- vesting is necessary. The technology is under commercial development for the production of human secreted alka- line phosphatase (see Phytomedics Inc. in online links box). Perhaps surprisingly, even large molecules can be rhizosecreted from transgenic plants. For example, a monoclonal antibody was secreted into hydroponic cul- ture medium resulting in a yield of 11.7 �g antibody per gram of dry root mass per day 61 . Plants can reasonably be expected to survive in a hydroponic system for many months, and proteins are relatively easily purified from culture medium compared with extraction from leaves, so rhizosecretion represents an attractive option for antibody production. As an alternative to nuclear transgenics, transplas- tomic plants are produced by introducing DNA into the chloroplast genome rather than the nuclear genome, a process that is generally achieved by particle bombardment 62,63 .The advantages of chloroplast trans- formation include the high transgene-copy number (there can be several thousand chloroplasts in a photo- synthetic cell) and the absence of position effects and transgene silencing. In combination, these properties can lead to astonishing levels of expression, in the best cases exceeding 25% of the total soluble protein 64 . Further advantages of chloroplast engineering include the ability to express several genes as operons and the accumulation of recombinant proteins in the chloro- plast (which reduces toxicity to the host plant). As dis- cussed above, both human growth hormone and serum albumin have been produced at high levels in tobacco chloroplasts, and each protein was found to be structurally authentic and biologically active 14,15 .More recently, a tetanus toxin fragment has been expressed in tobacco chloroplasts and was shown to induce protec- tive levels of anti-tetanus antibodies 64 .The cholera toxin B subunit has also been expressed in chloroplasts, which shows that plastids can fold and assemble oligomeric proteins correctly 65 . One disadvantage of the chloroplast transgenic system is that plastids do not carry out glycosylation. It is therefore unlikely that chloroplasts could be used to synthesize human glyco- proteins in cases in which the glycan-chain structure is crucial for protein activity. Another limitation is that chloroplast transformation outside the SOLANACEAE still presents a formidable technical challenge 66 . Recombinant proteins can also be produced in plant cell cultures. Tobacco suspension cells are generated by glycoprotein that is found in our diet. So, it is highly unlikely that they will be associated with adverse reactions. Indeed, studies in which mice were a dministered a recombinant antibody that contained plant-specific glycans showed no evidence of an anti- glycan immune reaction 54 .Nevertheless, the perceived negative effect of ?foreign? glycan structures is one of the most important issues that affect the use and acceptance of plant-derived recombinant proteins. Therefore, recent attention has focused on the development of strategies to ?humanize? the glycosyla- tion patterns of recombinant proteins. Warner 55 provides an overview of the biochemistry of glycan- chain synthesis in different expression hosts, and lists the changes that are required to produce proteins with typical human glycan structures in plants. In the moss Physcomitrella patens,gene targeting has been used to disable the plant-specific fucosyltransferase and xylosyltransferase enzymes. Strategies that have been attempted in transgenic plants include the use of purified human ?(1,4)-galactosyltransferase and sialyltransferase enzymes to modify plant-derived recombinant proteins in vitro 56 , and the expression of human ?(1,4)-galactosyltransferase in transgenic tobacco plants to produce recombinant antibodies with galactose-extended glycans 57 .In the latter case, ~30% of the recovered antibody was galactosylated. This is similar to the proportion of galactosylated anti- bodies that are produced by HYBRIDOMA CELLS. In vivo sialylation is unlikely to be achieved in the near future because plants seem to lack the metabolic pathway for the precursors of sialic acid, so several new enzymes would need to be introduced and coordinately expressed. To place the issue in perspective, it should be remem- bered that there is natural variation in glycan structures, with many proteins having several glycosylation sites and, even in mammalian cells, a range of glycoforms. There are recognizable differences in glycan structure even when comparing native human proteins to those produced in rodent cell lines. For example, human antibodies contain only the sialic-acid residue N-acetylneuraminic acid (NANA), whereas rodents produce a mixture of NANA and N-glycosylneuraminic acid (NGNA) 58 .At this stage, it is hard to generalize about how crucial the humanized glycosylation of plant-derived pharmaceuticals is, and whether it might be more important for some classes of proteins than for others. Similarly, there are still not enough data to address whether the method of administration of recombinant proteins (oral versus injection) could make a difference in terms of the immune response that might occur. Plant-expression hosts Tobacco production systems. To bacco has an established history as a model system for molecular farming and is the most widely used species for the production of recombinant pharmaceutical proteins at the research- laboratory level 4,37 .The main advantages of tobacco include the mature technology for gene transfer and HYBRIDOMA CELLS A hybrid cell line that is created by fusing a mortal antibody- producing B-lymphocyte with an immortalized myeloma line. The hybridoma line is immortal and produces a continuous supply of a particular monoclonal antibody. SOLANACEAE A family of flowering plants (order Solanales) that comprise ~100 genera and ~2,500 species, many of which are economically important as food or medicinal crops. Examples include tobacco, potato and tomato. 802 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS such as high biomass yield, ease of transformation and in vitro manipulation, and ease of scale-up. These factors, as well as intellectual property issues, prompted Prodigene to choose maize as the crop for the first com- mercial molecular-farming venture, which involved the production of the technical proteins avidin and ?-glucuronidase 11,69 .Maize is also being used for the production of recombinant antibodies 70 and further technical/pharmaceutical enzymes, such as laccase, trypsin and aprotinin 71 . Alfalfa and soybean are legumes. They each pro- duce lower amounts of leaf biomass than tobacco, but have the advantage of using atmospheric nitro- gen through nitrogen fixation, thereby reducing the need for chemical inputs. Both species have been used to produce recombinant antibodies 72,73 . One of the potential advantages of alfalfa is the recent find- ing that recombinant antibodies are produced as a single glycoform rather than the heterogeneous col- lection of different glycoforms that is found in other plant systems (see Medicago in online links box). Grain legumes are also useful production crops because of the high protein content in the seeds. For example, pea is being developed as a production sys- tem, although at present the yields that are possible with this species are low 74 . Fruit and vegetables. The main benefit of fruit, vegetable and leafy salad crops is that they can be consumed raw or partially processed, which makes them particularly suit- able for the production of recombinant subunit vac- cines, food additives and antibodies for topical passive immunotherapy. As discussed above, potatoes have been widely used for the production of plant-derived vaccines and have been administered to humans in most of the clinical trials carried out so far. The potential of potato tubers for antibody production was first shown by Artsaenko and colleagues 34 , and recently this crop has been investigated as a possible bulk-production system for antibodies 75 .Potatoes have also been used for the production of diagnostic antibody-fusion proteins 76 and human milk proteins 12,23 .Tomatoes, which were used to produce the first plant-derived rabies vaccine 77 ,are more palatable than potatoes and offer other advantages including high biomass yields (~68,000 kg per hectare) and the increased containment that is offered by growth in greenhouses. Lettuce is also being investigated as a production host for edible recombinant vaccines, and has been used in one series of clinical trials for a vaccine against HBV 20 .Bananas have been considered as hosts for the production of recombinant vaccines, as they are widely grown in the countries in which vaccines are most needed and can be consumed raw or as a puree by both adults and children 78 . Acceptability Molecular farming involves the use of genetically enhanced plants to produce pharmaceuticals, and is therefore covered by a myriad of established and emerg- ing regulations. Environmental biosafety issues, such as the potential for transgene spread and the possible the continuous agitation of FRIABLE CALLUS TISSUE,which results in a homogeneous suspension of single cells and small clumps. The cultures can be maintained in conventional microbial fermenters with only minor technical modifications, and various different culture modes can be used, including BATCH, FED-BATCH, PERFUSION AND CONTINUOUS FERMENTATION. Fischer and colleagues have reported the expression of several recombinant pro- teins, including several antibody derivatives, in a sus- pension cell line that was derived from the tobacco strain BY-2 (REF. 67).This approach is particularly advan- tageous when defined and sterile production conditions are required together with straightforward purifica- tion protocols. Recombinant proteins that are secreted into the culture medium are more easily purified, although proteins that are larger than 20?30 kDa tend to be retained in the apoplast and must be released by mechanical or enzymatic disruption. Cereals and legumes. One of the disadvantages of recombinant-protein production in tobacco is the instability of the product, which means that the leaf tis- sue must be frozen or dried for transport, or processed at the farm. By contrast, the accumulation of recombi- nant antibodies in seeds allows long-term storage at ambient temperatures because the proteins amass in a stable form. Seeds have the appropriate biochemical environment for protein accumulation, and achieve this through the creation of specialized storage com- partments, such as protein bodies and storage vacuoles, which are derived from the secretory pathway. Seeds are also desiccated, which reduces the exposure of stored proteins to non-enzymatic hydrolysis and protease degradation. It has been shown that antibodies that are expressed in seeds remain stable for at least three years at room temperature with no detectable loss of activ- ity 33 .Cereal seeds also lack the phenolic substances that are present in tobacco leaves, so increasing the effi- ciency of downstream processing. The main concern about the use of cereals and other established food crops for the production of pharmaceuticals relates to the potential for genes to spread into crops that are grown for food purposes, and the possibility of inad- vertent contamination during seed collection and stor- age (see the article by Stewart et al. in this issue). Important variables that must be considered when choosing a cereal production crop include the grain yield per hectare, the yield of recombinant protein per unit biomass, the ease of transformation and the speed of scale-up. The same single-chain variable-fragment antibody has been expressed in rice, wheat and tobacco to compare production levels in leaves and seeds 68 . With the optimal promoter system (the enhanced CaMV35S promoter for tobacco and the ubiquitin-1 promoter for rice) it was found that rice plants showed the highest yields per unit biomass, and levels were lowest in wheat. However, the wheat system is still under development and improved construct design will probably give rise to higher yields. Maize is now the main commercial production crop for recombinant proteins, which reflects advantages FRIABLE CALLUS TISSUE Callus tissue is undifferentiated plant tissue, which grows when seeds or explants are cultured on media that contains an appropriate balance of plant hormones. Friable callus tissue is easily broken into fragments. BATCH, FED-BATCH, PERFUSION AND CONTINUOUS FERMENTATION Batch fermentation is a closed system in which all of the substrate is added at the beginning, whereas in the fed-batch process the substrate is added in increments as fermentation proceeds. Continuous fermentation is an open system in which substrate is added continuously at a steady rate.Perfusion fermentation is a continuous process that allows cells to be grown at high density, and so results in increased biomass and product yields. NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 803 REVIEWS Another strategy is to express the recombinant protein as an inactive precursor that must be proteolytically cleaved before it shows biological activity. Recently publicized incidents in which genetically enhanced crops have been inadvertently mixed with those destined for human consumption have highlighted the need for mechanisms to ensure the segregation of plants that express pharmaceuticals (for example, see Pew Initiative on Food and Biotechnology in online links box). Although greenhouse containment would address many of these biosafety issues, this negates the advantages of the technology in terms of low cost and large-scale production. If the main goal is to provide affordable medicines to developing countries, then these are important considerations for regulatory authorities. In any case, if food and feed crops are cho- sen for molecular farming, many other levels of safety can be built into the system. These include geographical isolation, differential planting seasons, the use of male- sterile plants and transplastomic plants, and the use of inducible promoters, as discussed above. It would also be helpful to use identity-preserved varieties, such as white tomatoes or maize, which are easily identified by their pigmentation. Identity preservation and tracking are important parts of the regulatory procedure for the production of pharmaceuticals in transgenic plants 85 . Conclusions Plants have many advantages compared with tradi- tional systems for the molecular farming of phar- maceutical proteins. These include the low cost of production, rapid scalability, the absence of human pathogens and the ability to fold and assemble com- plex proteins accurately. The commercial success of plant-derived avidin and ?-glucuronidase, which are both available from Sigma Inc. at a lower cost than the native proteins, shows the potential of molecular farming in plants to compete in an already estab- lished market. Other important examples that show the potential of the technology include the synthesis, folding and assembly of multimeric proteins, such as serum and secretory antibodies, and subunit vac- cines, which need to assemble into multimeric com- plexes to raise an immune response. Plants might one day surpass other production systems, particularly cultured animal cells, for the production of pharma- ceutical proteins because of these economic and safety benefits. Several challenges remain to be met in terms of increasing yields, improving glycoprotein authenticity, removing processing bottlenecks and addressing biosafety and acceptability issues, as well as industry inertia. The difficulties that must be over- come as part of the existing drug regulatory process are not the least of these challenges; however, this complex issue is beyond the scope of this review. If these hurdles can be overcome, we might soon be in the position to produce any pharmaceutical protein on a scale that meets market demand. Ultimately, it might be possible to make protein-based pharmaceu- ticals available to everyone who needs them, at a cost that everyone can afford. toxicity of the recombinant proteins to non-target organisms, need to be addressed on a case-by-case basis depending on the location, production host and product, taking into account measures that have been used to reduce biosafety risks 79 (see the article by Stewart et al. in this issue). The rigorous scientific eval- uation of each application for release will help in the development of effective risk-management strategies that will facilitate future decision-making processes. The main perceived risks are transgene spread by pollen dispersal, seed dispersal and horizontal gene transfer, and the effects of potentially toxic recombi- nant proteins on herbivores, pollinating insects and microorganisms in the RHIZOSPHERE.There is also con- cern that plant material that contains recombinant proteins could inadvertently enter the food chain. Although it is apparent that pharmaceutical crops do not suffer the same acceptability problems as genetically modified food crops, risk assessment and environmental- impact studies must be carried out to the same level, to ensure the highest standards of responsibility and regula- tory compliance. The removal of selectable marker genes 80 or the use of innocuous plant-derived markers for metabolic selection 81 is required to limit the incor- poration of superfluous DNA sequences. The risk of transgene spread by pollen dispersal can be addressed by several physical and genetic barrier techniques, as well as by the choice of a suitable production crop that does not outcross with wild plants near the production site 82 . The risk of horizontal gene transfer is thought to be extremely low. The potential for the horizontal spread of DNA from transgenic plants to bacterial populations has been considered. No such gene flow has been shown with nuclear transgenic plants, despite many attempts to create conditions in the laboratory that would simulate such an occurrence. Recently, Kay et al. 83 showed the horizontal transfer of an antibiotic-resistance marker from the chloroplasts of transplastomic tobacco plants to opportunistic strains of Acinetobacter spp. However, transfer was achieved only under idealized conditions in which the bacteria were modified to contain a sequence that was homologous to the transgene. No transfer was observed to wild-type bacteria. The potential for trans- fer is not surprising in light of the prokaryotic origin of plastids, but the transferred gene would need to offer a selective advantage to the recipient bacteria to persist in natural populations. All plants, both natural and trans- genic, have been shown to be covered with antibiotic- resistant bacteria, and these are a much more likely source of resistance genes that could jump to human pathogens 84 . The possible negative effect of recombinant proteins on non-target organisms can be addressed by the use of regulated promoters to restrict transgene expression to particular organs (for example, seeds) or to induce protein expression at particular times. The retrieval of proteins to the ER lumen, or the direction of proteins to other compartments such as the vacuole or chloroplast, can also improve the containment of proteins by pre- venting secretion to the apoplast and possible leaching into the LEAF GUTTATION FLUID or the root exudates. RHIZOSPHERE The soil zone that surrounds plant roots, which is rich in microorganisms and in which interactions occur between plants and microbes. LEAF GUTTATION FLUID Fluid that seeps from the apoplast onto the leaf surface. In plants with large leaves, such as tobacco, large amounts of guttation fluid can be produced each day. 804 | OCTOBER 2003 | VOLUME 4 www.nature.com/reviews/genetics REVIEWS 1. Schwartz, J. R. Advances in Escherichia coli production of therapeutic proteins. Curr. Opin. Biotechnol. 12, 195?201 (2001). 2. Chu, L. & Robinson, D. K. Industrial choices for protein production by large-scale cell culture. Curr. Opin. Biotechnol. 12, 180?187 (2001). 3. Houdebaine, L. M. Transgenic animal bioreactors. Transgenic Res. 9, 305?320 (2000). 4. Fischer, R. & Emans, N. Molecular farming of pharmaceutical proteins. Transgenic Res. 9, 279?299 (2000). 5. Giddings, G. Transgenic plants as protein factories. Curr. Opin. Biotechnol. 12, 450?454 (2001). 6. Barta, A. et al. The expression of a nopaline synthase human growth hormone chimaeric gene in transformed tobacco and sunflower callus tissue. Plant Mol. Biol. 6, 347?357 (1986). 7. Hiatt, A., Cafferkey, R. & Bowdish, K. Production of antibodies in transgenic plants. Nature 342, 76?78 (1989). 8. Mason, H. S., Lam, D. M. K. & Arntzen, C. J. Expression of hepatitis B surface antigen in transgenic plants. Proc. Natl Acad. Sci. USA 89, 11745?11749 (1992). 9. Thanavala, Y., Yang, Y.-F., Lyons, P., Mason, H. S. & Arntzen, C. J. Immunogenicity of transgenic plant-derived hepatitis B surface antigen. Proc. Natl Acad. Sci. USA 92, 3358?3361 (1995). 10. Hood, E. E. et al. Criteria for high-level expression of a fungal laccase gene in transgenic maize. Plant Biotechnol. J. 1, 129?140 (2003). 11. Hood, E. E. et al. Commercial production of avidin from transgenic maize: characterization of transformant, production, processing, extraction and purification. Mol. Breeding 3, 291?306 (1997). 12. Chong, D. K. X. et al. Expression of the human milk protein ?-casein in transgenic potato plants. Transgenic Res. 6, 289?296 (1997). 13. Ruggiero, F. et al. Triple helix assembly and processing of human collagen produced in transgenic tobacco plants. FEBS Lett. 469, 132?136 (2000). 14. Staub, J. M. et al. High-yield production of a human therapeutic protein in tobacco chloroplasts. Nature Biotechnol. 18, 333?338 (2000). This report shows that the secreted human protein somatotropin is soluble and biologically active when expressed in tobacco chloroplasts and has correctly formed disulphide bonds. 15. Fernandez-San Millan, A., Mingo-Castel, A., Miller, M. & Daniell, H. A chloroplast transgenic approach to hyper- express and purify human serum albumin, a protein highly susceptible to proteolytic degradation. Plant Biotechnol. 1, 77?79 (2003). 16. Moloney, M., Boothe, J. & Van Rooijen, G. Oil bodies and associated proteins as affinity matrices. US Patent 6,509,453 (2003). 17. Chadd, H. E. & Chamow, S. M. Therapeutic antibody expression technology. Curr. Opin. Biotechnol. 12, 188?194 (2001). 18. Ma, J. K. et al. Generation and assembly of secretory antibodies in plants. Science 268, 716?719 (1995). This study shows for the first time that secretory antibodies, with 10 polypeptide chains that represent the products of four genes, can be assembled correctly in transgenic plants. Two rounds of crossing, which involved four singly-transgenic lines, were required to stack all four transgenes in the same plant. 19. Richter, L. J., Thanavala, Y., Arntzen, C. J. & Mason, H. S. Production of hepatitis B surface antigen in transgenic plants for oral immunization. Nature Biotechnol. 18, 1167?1171 (2000). 20. Kapusta, J. et al. A plant-derived edible vaccine against hepatitis B virus. FASEB J. 13, 1796?1799 (1999). 21. Tacket, C. O. et al. Immunogenicity in humans of a recombinant bacterial-antigen delivered in a transgenic potato. Nature Med. 4, 607?609 (1998). 22. Tacket, C. O. et al. Human immune responses to a novel Norwalk virus vaccine delivered in transgenic potatoes. J. Infect. Dis. 182, 302?305 (2000). 23. Chong, D. K. X. & Langridge, W. H. R. Expression of full-length bioactive antimicrobial human lactoferrin in potato plants. Transgenic Res. 9, 71?78 (2000). 24. Zhang, X., Urry, D. W. & Daniell, H. Expression of an environmentally friendly synthetic protein-based polymer in transgenic tobacco plants. Plant Cell Reps. 16, 174?179 (1996). 25. Guda, C., Lee, S. B. & Daniell, H. Stable expression of a biodegradable protein-based polymer in stable tobacco chloroplasts. Plant Cell Reps. 19, 257?262 (2000). 26. Merle, C. et al. Hydroxylated human homotrimeric collagen I in Agrobacterium tumefaciens-mediated transient expression and in transgenic tobacco plant. FEBS Lett. 515, 114?118 (2002). 27. Scheller, J., Guhrs, K. H., Grosse, F. & Conrad, U. Production of spider silk proteins in tobacco and potato. Nature Biotechnol. 19, 573?577 (2001). 28. O?Dell, J. T., Nagy, F. & Chua, N. H. Identification of DNA sequences required for activity of the cauliflower mosaic virus 35S promoter. Nature 313, 810?812 (1985). 29. Lawton, M. A. et al. Expression of a soybean ?-conclycinin gene under the control of the cauliflower mosaic virus 35S and 19S promoters in transformed petunia tissues. Plant Mol. Biol. 9, 315?324 (1987). 30. Kay, R., Chan, A., Daly, M. & McPherson, J. Duplication of CaMV-35S promoter sequences creates a strong enhancer for plant genes. Science 236, 1299?1302 (1987). 31. Christensen, A. H. & Quail, P. H. Ubiquitin promoter-based vectors for high-level expression of selectable and/or screenable marker genes in monocotyledonous plants. Transgenic Res. 5, 213?218 (1996). 32. Vain, P., Finer, K. R., Engler, D. E., Pratt, R. C. & Finer, J. J. Intron-mediated enhancement of gene expression in maize (Zea mays L.) and bluegrass (Poa pratensis L.). Plant Cell Rep. 15, 489?494 (1996). 33. Stoger, E. et al. Cereal crops as viable production and storage systems for pharmaceutical scFv antibodies. Plant Mol. Biol. 42, 583?590 (2000). 34. Artsaenko, O., Kettig, B., Fiedler, U., Conrad, U. & D�ring K. Potato tubers as a biofactory for recombinant antibodies. Mol. Breeding 4, 313?319 (1998). 35. Padidam, M. Chemically regulated gene expression in plants. Curr. Opin. Plant Biol. 6, 169?177 (2003). 36. Padidam, M., Gore, M., Lu, D. L. & Smirnova, O. Chemical- inducible, ecdysone receptor-based gene expression system for plants. Transgenic Res. 12, 101?109 (2003). 37. Cramer, C. L., Boothe, J. G. & Oishi, K. K. Transgenic plants for therapeutic proteins: linking upstream and downstream technologies. Curr. Top. Microbiol. Immunol. 240, 95?118 (1999). 38. Schillberg, S., Zimmermann, S., Voss, A. & Fischer, R. Apoplastic and cytosolic expression of full-size antibodies and antibody fragments in Nicotiana tabacum. Transgenic Res. 8, 255?263 (1999). This paper compares the stability of identical scFv antibodies that are targeted to different compartments, and shows that the secretory pathway is generally much more suitable for antibody accumulation than the cytosol. 39. De Jaeger, G. et al. High-level accumulation of single-chain variable fragments in the cytosol of transgenic Petunia hybrida. Eur. J. Biochem. 259, 1?10 (1998). 40. Schouten, A., Rossien, J., Bakker, J. & Schots, A. Formation of disulfide bridges by a single-chain Fv antibody in the reducing ectopic environment of the plant cytosol. J. Biol. Chem. 277, 19339?19345 (2002). 41. Conrad, U. & Fiedler, U. Compartment-specific accumulation of recombinant immunoglobulins in plant cells: an essential tool for antibody production and immunomodulation of physiological functions and pathogen activity. Plant Mol. Biol. 38, 101?109 (1998). 42. Plasterk, R. H. A. & Ketting, R. F. The silence of the genes. Curr. Opin. Genet. Dev. 10, 562?567 (2000). 43. Anandalakshmi, R. et al. A viral suppressor of gene silencing in plants. Proc. Natl Acad. Sci. USA 95, 13079?13084 (1998). 44. Gelvin, S. B. Agrobacterium-mediated plant transformation: the biology behind the ?gene-jockeying? tool. Microbiol. Mol. Biol. Rev. 67, 16?23 (2003). 45. Britt, A. B. & May, G. D. Re-engineering plant gene targeting. Trends Plant Sci. 8, 90?95 (2003). 46. Veluthambi, K., Gupta, A. K. & Sharma, A. The current status of plant transformation technologies. Curr. Sci. India 84, 368?380 (2003). 47. Christou, P. Transformation technology. Trends Plant Sci. 1, 423?431 (1996). 48. Twyman, R. M., Stoger, E., Kohli, A. & Christou, P. in Genetic Engineering: Principles and Practice Vol. 24 (ed. Setlow, J. K.) 107?136 (Kluwer-Plenum, New York, 2002). 49. Martineau, B., Voelker, T. A. & Sanders, V. A. On defining T-DNA. Plant Cell 6, 1032?1033 (1994). 50. Hanson, B. et al. A simple method to enrich an Agrobacterium-transformed population for plants containing only T-DNA sequences. Plant J. 19, 727?734 (1999). 51. Fu, X. et al. Linear transgene constructs lacking vector backbone sequences generate low-copy-number transgenic plants with simple integration patterns. Transgenic Res. 9, 11?19 (2000). 52. Lerouge, P., Bardor, M., Pagny, S., Gomord, V. & Faye, L. N-glycosylation of recombinant pharmaceutical glycoproteins produced in transgenic plants: towards humanisation of plant N-glycans. Curr. Pharmaceutical Biotech. 1, 347?354 (2000). 53. Bardor, M. et al. Immunoreactivity in mammals of two typical plant glyco-epitopes: core ?(1,3)-fucose and core xylose. Glycobiology 13, 427?434 (2003). 54. Chargelegue, D., Vine, N. D., van Dolleweerd, C. J., Drake, P. M. & Ma, J. K. A murine monoclonal antibody produced in transgenic plants with plant-specific glycans is not immunogenic in mice. Transgenic Res. 9, 187?194 (2000). The first published paper that discusses the immunogenicity of a plant-derived glycosylated recombinant protein. 55. Warner, T. G. in Carbohydrates in Chemistry and Biology (eds Ernst, B., Hart, G. W. & Sanay, P.) 1043?1064 (Wiley, New York, 2000). 56. Blixt, O., Allin, K., Pereira, L., Datta, A. & Paulson, J. C. Efficient chemoenzymatic synthesis of O-linked sialyl oligosaccharides. J. Am. Chem. Soc. 124, 5739?5746 (2002). 57. Bakker, H. et al. Galactose-extended glycans of antibodies produced by transgenic plants. Proc. Natl Acad. Sci. USA 98, 2899?2904 (2001). In this study, a transgenic tobacco line that expressed the heavy and light chains of a murine antibody was crossed with a line that expressed human ?-1,4-galactosyltransferase. The progeny produced antibodies ~30% of which had partially galactosylated N-glycans, which provided a useful approach for the ?humanization? of plant glycans. 58. Raju, T. S., Briggs, J., Borge, S. M. & Jones, A. J. S. Species-specific variation in glycosylation of IgG: evidence for the species-specific sialylation and branch-specific galactosylation and importance for engineering recombinant glycoprotein therapeutics. Glycobiology 10, 477?486 (2000). In this study, cell-specific glycosylation of immunoglobulins was studied by mass spectrometry and capillary electrophoresis/laser-induced fluorescence in 13 different animal systems. The glycan patterns were found to be unique in different species, which indicated that some might be more suitable than others for the production of human therapeutic proteins. The same might apply to plant systems, which were not considered in this paper. 59. Borisjuk, N. V. et al. Production of recombinant proteins in plant root exudates. Nature Biotechnol. 17, 466?469 (1999). 60. Komarnytsky, S., Borisjuk, N. V., Borisjuk, L. G., Alam, M. Z. & Raskin, I. Production of recombinant proteins in tobacco guttation fluid. Plant Physiol. 124, 927?933 (2000). 61. Drake, P. M. W. et al. Rhizosecretion of a monoclonal antibody protein complex from transgenic tobacco roots. Plant Mol. Biol. 52, 233?241 (2003). 62. Maliga, P. Engineering the plastid genome of higher plants. Curr. Opin. Plant Biol. 5, 164?172 (2002). 63. Daniell, H., Khan, M. S. & Allison, L. Milestones in chloroplast genetic engineering: an environmentally friendly era in biotechnology. Trends Plant Sci. 7, 84?91 (2002). 64. Tregoning, J. S. et al. Expression of tetanus toxin fragment C in tobacco chloroplasts. Nucleic Acids Res. 31, 1174?1179 (2003). 65. Daniell, H., Lee, S. B., Panchal, T. & Wiebe, P. O. Expression of the native cholera B toxin subunit gene and assembly as functional oligomers in transgenic tobacco chloroplasts. J. Mol. Biol. 311, 1001?1009 (2001). 66. Khan, M. S. & Maliga, P. Fluorescent antibiotic resistance marker for tracking plastid transformation in higher plants. Nature Biotechnol. 17, 910?915 (1999). 67. Fischer, R., Emans, N., Schuster, F., Hellwig, S. & Drossard, J. Towards molecular farming in the future: using plant-cell-suspension cultures as bioreactors. Biotechnol. Appl. Biochem. 30, 109?112 (1999). 68. St�ger, E. et al. Practical considerations for pharmaceutical antibody production in different crop systems. Mol. Breeding 9, 149?158 (2002). This paper considers in detail the factors that should be evaluated when choosing a crop system for the production of pharmaceutical proteins. The same scFv is expressed in many species to compare intrinsic yields, and features such as storage, distribution and biosafety are discussed, as well as economic factors. 69. Witcher, D. et al. Commercial production of ?-glucuronidase (GUS): a model system for the production of proteins in plants. Mol. Breeding 4, 301?312 (1998). 70. Hood, E. E., Woodard, S. L. & Horn, M. E. Monoclonal antibody manufacturing in transgenic plants ? myths and realities. Curr. Opin. Biotechnol. 13, 630?635 (2002). 71. Hood, E. E. From green plants to industrial enzymes. Enzyme Microbial Technol. 30, 279?283 (2002). 72. Zeitlin, L. et al. A humanized monoclonal antibody produced in transgenic plants for immunoprotection of the vagina against genital herpes. Nature Biotechnol. 16, 1361?1364 (1998). 73. Khoudi, H. et al. Production of a diagnostic monoclonal antibody in perennial alfalfa plants. Biotechnol. Bioeng. 64, 135?143 (1999). 74. Perrin, Y. et al. Transgenic pea seeds as bioreactors for the production of a single-chain Fv fragment (scFV) antibody used in cancer diagnosis and therapy. Mol. Breeding 6, 345?352 (2000). NATURE REVIEWS | GENETICS VOLUME 4 | OCTOBER 2003 | 805 REVIEWS 75. De Wilde, C., Peeters, K., Jacobs, A., Peck, I. & Depicker, A. Expression of antibodies and Fab fragments in transgenic potato plants: a case study for bulk production in crop plants. Mol. Breeding 9, 2871?282 (2002). 76. Schunmann, P. H. D., Coia, G. & Waterhouse, P. M. Biopharming the Simpli-RED TM HIV diagnostic reagent in barley, potato and tobacco. Mol. Breeding 9, 113?121 (2002). 77. McGarvey, P. B. et al. Expression of the rabies virus glycoprotein in transgenic tomatoes. Biotechnology 13, 1484?1487 (1995). 78. Sala, F. et al. Vaccine antigen production in transgenic plants: strategies, gene constructs and perspectives. Vaccine 21, 803?808 (2003). 79. Commandeur, U., Twyman, R. M. & Fischer, R. The biosafety of molecular farming in plants. AgBiotechNet 5, ABN 110 (2003). 80. Hare, P. D. & Chua, N.-H. Excision of selectable marker genes from transgenic plants. Nature Biotechnol. 20, 575?579 (2002). 81. Zuo, J. R., Niu, Q. W., Ikeda, Y. & Chua, N. H. Marker-free transformation: increasing transformation frequency by the use of regeneration-promoting genes. Curr. Opin. Biotechnol. 13, 173?180 (2002). 82. Eastham, K. & Sweet, J. Genetically Modified Organisms (GMOs): the Significance of Gene Flow through Pollen Transfer. Environment Issue Report No. 28 (European Environment Agency, Copenhagen, 2002). 83. Kay, E., Vogel, T. M., Bertolla, F., Nalin, R. & Simonet, P. In situ transfer of antibiotic resistance genes from transgenic (transplastomic) tobacco plants to bacteria. Appl. Environ. Microbiol. 68, 3345?3351 (2002). 84. Smalla, K. et al. in Proceedings of the 6th International Symposium on the Biosafety of Genetically Modified Organisms 146?154 (Univ. Extension Press, Univ. of Saskatchewan, Canada, 2000). 85. Smyth, S. & Phillips, P. W. B. Product differentiation alternatives: identity preservation, segregation and traceability. AgBioForum 5, 30?42 (2002). 86. Schillberg, S., Fischer, R. & Emans, N. Molecular farming of recombinant antibodies in plants. Cell. Mol. Life Sci. 60, 433?445 (2003). This is a comprehensive discussion of the technical issues concerning the production of antibodies in plants, which is treated in much more detail than is possible in the present review. 87. Stoger, E., Sack, M., Fischer, R. & Christou, P. Plantibodies: applications, advantages and bottlenecks. Curr. Opin. Biotechnol. 13, 161?166 (2002). 88. McCormick, A. A. et al. Rapid production of specific vaccines for lymphoma by expression of the tumor-derived single-chain Fv epitopes in tobacco plants. Proc. Natl Acad. Sci. USA 96, 703?708 (1999). 89. Larrick, J. W., Yu, L., Naftzger, C., Jaiswal, S. & Wycoff, K. Production of secretory IgA antibodies in plants. Biomolecular Eng. 18, 87?94 (2001). This paper presents a useful summary of recent advances in the plant-based production of secretory IgAs with a discussion of purification methods and production costs. 90. Ma, J. K. et al. Characterization of a recombinant plant monoclonal secretory antibody and preventive immunotherapy in humans. Nature Med. 4, 601?606 (1998). 91.Vaquero, C. et al. A carcinoembryonic antigen-specific diabody produced in tobacco. FASEB J. 16, 408?410 (2002). 92. Kathuria, S. et al. Efficacy of plant-produced recombinant antibodies against HCG. Human Reproduction 17, 2054?2061 (2002). 93. Miele, L. Plants as bioreactors for pharmaceuticals: regulatory considerations. Trends Biotechnol. 15, 45?50 (1997). 94. Emlay, D. in Plants as Factories for Protein Production (eds Hood, E. E. & Howard, J.) 175?180 (Kluwer Academic, New York, 2002). 95. Lloyd-Evans, M. & Nair, A. in Biopharming: the Emerging World Market of Plant-based Therapeutics. Theta Report No. 1214 63?80 (PJB Medical Publications Inc., New York, 2002). 96. Stoger, E., Schillberg, S., Twyman, R. M., Fischer, R. & Christou, P. in Methods in Molecular Biology. Antibody Engineering: Protocols and Methods 2nd edn (ed. Lo, B. K. C.) (Humana Press Inc., New Jersey, in the press). 97. Schillberg, S., Zimmermann, S., Findlay, K. & Fischer, R. Plasma membrane display of anti-viral single chain Fv fragments confers resistance to tobacco mosaic virus. Mol. Breeding 6, 317?326 (2000). 98. Sijmons, P. C. et al. Production of correctly processed human serum albumin in transgenic plants. Biotechnology 8, 217?221 (1990). 99. Zhu, Z. et al. Expression of human ?-interferon in plants. Virology 172, 213?222 (1994). 100. Matsumoto, S., Ikura, K., Ueda, M. & Sasaki, R. Characterisation of a human glycoprotein (erythropoetin) produced in cultured tobacco cells. Plant Mol. Biol. 27, 1163?1172 (1995). 101. Delaney, D. et al. in Plant Biotechnology: 2002 and Beyond. Proceedings of the 10th IAPTC&B Congress, Orlando, Florida. (ed. Vasil, I.) 393?394 (Kluwer Academic, Dordrecht, The Netherlands, 2002). 102. Torres, E. et al. Rice cell culture as an alternative production system for functional diagnostic and therapeutic antibodies. Transgenic Res. 8, 441?449 (1999). 103. Terashima, M. et al. Production of functional human ?1?antitrypsin by plant cell culture. Appl. Microbiol. Biotechnol. 52, 516?523 (1999). 104. D�ring, K., Hippe, S., Kreuzaler, F. & Schell, J. Synthesis and self-assembly of a functional monoclonal antibody in transgenic Nicotiana tabacum. Plant Mol. Biol. 15, 281?293 (1990). 105. Ma, J. K.-C. et al. Generation and assembly of secretory antibodies in plants. Science 268, 716?719 (1995). 106. Francisco, J. A. et al. Expression and characterization of bryodin 1 and a bryodin 1-based single-chain immunotoxin from tobacco cell culture. Bioconjug. Chem. 8, 708?713 (1997). 107. Mayfield, S. P. et al. Expression and assembly of a fully active antibody in algae. Proc. Natl Acad. Sci. USA 100, 438?442 (2003). 108. Streatfield, S. J. et al. Plant-based vaccines: unique advantages. Vaccine 19, 2742?2748 (2001). 109. Ma, S. W. et al. Transgenic plants expressing autoantigens fed to mice to induce oral immune tolerance. Nature Med. 3, 793?796 (1997). 110. Yu, J. & Langridge, W. H. A plant-based multicomponent vaccine protects mice from enteric diseases. Nature Biotechnol. 19, 548?552 (2001). 111. Lamphear, B. J. et al. Delivery of subunit vaccines in maize seed. J. Control Release 83, 169?180 (2002). Acknowledgements The authors are grateful to R. Twyman for critical assessment and help with manuscript preparation. Online links DATABASES The following term in this article is linked online to: BioCyc: http://www.biocyc.org nos FURTHER INFORMATION Epicyte Pharmaceutical: http://www.epicyte.com Large Scale Biology Corp.: http://www.lsbc.com Medicago: http://www.medicago.com Monsanto: http://www.monsanto.com NeoRx: http://www.neorx.com Pew Initiative on Food and Biotechnology: http://www.pewagbiotech.org Phytomedics, Inc.: http://www.phytomedics.com Prodigene: http://www.prodigene.com SemBioSys Genetics Inc.: http://www.sembiosys.ca Sigma Inc.: http://www.sigmaaldrich.com Access to this interactive links box is free online. "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment