The Concept Of Molecular Farming Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Different plant species and tissues have now been used as production system for a variety of recombinant proteins including cereals, legumes, leafy crops, oilseeds, higher plant tissue and cell culture, fruits, vegetables, algae, moss and higher aquatic plants (Stoger et al., 2005). All of these production hosts have their own merits and demerits for protein production in terms of time required, cost of end product, safety, scalability, ease of transformation, downstream processing, protein folding and transportation.

I.2. The seed

Seeds are the dispersal and propagation units that allow plants to spread out and grow in new territories, and help plants to survive under unfavourable conditions in a dormant stage. A seed is a small embryogenic plant emerged from an ovule after fertilization and it is composed of three main parts: a seed coat, an endosperm and an embryo with one or two cotyledons (Fig. I.1D). Structurally, plant seeds can be divided into three major categories: monocotyledonous seeds (such as the cereals), endospermic dicotyledonous seeds (like tobacco) and non-endospermic dicotyledonous seeds (beans). The embryo, which represents the next plant generation, is enclosed by the endosperm which serves as a nutritive tissue for the embryo; both the embryo and the endosperm are the product of a double fertilization process unique to flowering plants. The seed coat provides a mechanical barrier whose purpose is to protect the entire seed, including the embryo, enabling the plant to remain dormant for a longer time.

The endosperm is well recognized for its importance as food and feed. Cereal seed endosperm contains 70% starch in terms of dry weight. Starch is made of glucan polymers, amylose and amylopectin, that are packed into semi-crystalline granules in amyloplasts (James et al., 2003). Cereals are the main protein provider in certain regions where rice or maize is the main staple food (Shewry, 2002). In cereals, endosperm provides major space for for storage proteins, which are responsible for the visco-elastic properties of the dough. Hence in the case of wheat the seed protein content is a major quality determinant for bread, pasta and other backed goods.

Figure I.1 Illustration of the seed anatomy of three plant models. A. Cross section of Nicotiana seed coat, cot. cotyledon, en. endosperm. From Tomlinson et al. (2004). B. Light microscopy with toluidine blue of Medicago truncatula mature seed (non-endospermic dicotyledonous) rad. Radical, mes. Mesophyll, col. Cotyledon, en. endosperm. C. Cross section of wheat seed (monocotyledonous) sc. Seed coat, en. Endosperm, cot. Cotyledon. D. Longitudinal section of wheat grain. Figures adapted from Tomlinson et al., 2004 (A), Figure Prof. Eva Stoger Lab (B and C). from Wikipedia (C; Wugo).

I.2.1. Seed storage proteins

Seeds are rich reservoirs of storage proteins (seed storage proteins, SSPs), which constitute up to 90% of the total protein fraction in developing seeds and are crucial for germinating seedlings providing them with carbon, nitrogen and sulphur (Kumamaru et al., 2007). The total mass of seed proteins varies from 10-15% of grain dry weight in cereals to up to 45% (dry weight) in legumes.

Seed proteins are classified into three groups: storage proteins, structural and metabolic proteins and protective proteins. SSPs occur in three different tissues of the grain (embryo, aleurone layer and endosperm) and can be divided into different classes (Galili 2004). A range of criteria have been used to define and classify seed proteins. T. B. Osborne (Shewry 2002) developed a classification based on solubility of plant proteins known as "Osborne fractionation" which is still used today, especially by cereal scientists. Thus, four groups are defined: water soluble albumins (found mainly in dicots), salt soluble globulins (found both in dicots and monocots), alcohol soluble prolamins (found exclusively in cereals) and glutelins (soluble in dilute acids, found also in cereals). Cereal seed proteins can also be classified on the basis of their metabolic properties, deposition sites and allergenic properties, but the latter is not so often in use (Breiteneder et al. 2005).

I.2.1.1. Prolamins

Prolamins are the major endosperm storage proteins in cereals with the exception of oats and rice. They were originally classified as soluble in alcohol/water mixtures but some occur in alcohol-insoluble polymers. In any case, all individual prolamin polypeptides are alcohol-soluble in the reduced state. Prolamins present a wide structural variation and there are indications that the major groups of prolamins in the Triticeae (wheat, barley and rye) and the Panicoideae (maize, sorghum, millets) may have separate evolutionary origins. Nevertheless, most prolamins share two common structural features. One is the presence of distinct regions with different structures and origins. The other is the presence of amino acid sequences consisting of repeated blocks based on one or more short peptide motifs, or enriched in specific amino acid residues. These features are responsible for the high proportions of glutamine, proline and other specific amino acids (such as histidine, glycine, methionine and phenylalanine) in some prolamin groups (Shewry and Halford 2002).

The prolamin superfamily has been reorganised in subgroups according to structural and evolutionary relationships. Thus, the prolamins of the Triticeae are assigned to three different groups: sulphur-rich (S-rich), sulphur-poor (S-poor) and high molecular weight (HMW) prolamins. They all contain extensive repeated sequences based on prolin-rich and glutamine-rich motifs with the repeated motifs of the S-rich and the S-poor groups being clearly related. Sequence similarity is also present between the non-repetitive domains of the S-rich and the HMW prolamins indicating that the S-rich, S-poor and HMW prolamins have a common evolutionary origin. Wider comparisons reveal evolutionary and structural similarities to several groups of zeins, prolamins of oats and rice, 2S albumin storage proteins of dicotyledoneous seeds and other seed proteins which are therefore together defined as the cereal prolamin superfamily of plant proteins (Kreis et al. 1985).

The prolamins of maize (called zeins) constitute more than half of total seed proteins synthesized within endosperm tissue. Alpha-zeins have a relative molecular mass of 19 and 22 kDa and constitute the most important class of zeins, comprising 70% of the total prolamins. Alpha-zeins do not appear to be related to any other prolamins except the alpha-type of the other panicoid cereals. The remaining three classes (beta, gamma and delta) are sulphur rich and contrary to alpha-zeins they have higher cysteine content and form polymers (Kumamaru et al 2007; Shewry and Halford 2002; Woo et al. 2001,).

I.3. Seed as bioreactor

Seed from various plants has been considered as production host for large number of heterologous proteins. Seed crops such as: wheat (Arcalis et al., 2004; Stoger et al., 2000), rice (He et al., 2011), barley (Schunman et al., 2002), arabidopsis (Downing et al., 2006), maize (Ramessar et al., 2008; Rademacher et al., 2008) brassica (Parmenter et al., 1995), pea (Perrin et al., 2000), safflower (Nykiforuk et al., 2010), tobacco (Floss et al., 2008) and soybean (Cunha et al.., 2010) have been used as bioreactor for different recombinant proteins. The structure of seeds provides advantages in terms of post-harvest processing because the small and homogenous size of seeds is beneficial for the concentration and downstream processing of recombinant proteins (Ramessar et al., 2008). Factors influencing protein yield and stability in seeds are crucial and include genotype selection, transgene copy number and zygosity, construct design and protein targeting (Ramessar et al., 2008). Cereal crops such as maize, rice, barley and wheat have been used as hosts for recombinant proteins (Table 2) and the former three of these are being developed commercially (Stoger et al., 2002a).

I.3.1. Wheat

Wheat is the major crop grown all over the world. It is the most widely consumed food crop because of its adaptability and high yield in a wide range of environments. Wheat was the first Triticeae species to be stably transformed (Vasil et al. 1992). Among cereals, wheat has been used rarely as a production host for recombinant proteins because the gene transfer in wheat is less efficient and requires more specialized skills for stable transformation (Stoger et al. 2000). Wheat is an attractive candidate for molecular farming because of its higher protein content (>12%) and lower production cost. Wheat is a self-pollinating crop, which implies that the chances of out-crossing to non-transgenic crops and wild relatives are minimized. To make wheat more feasible for molecular farming, more efficient gene transfer and regeneration technology is needed. This may be accomplished through the development of routine methods for high-frequency stable transformation i.e by Agrobacterium-mediated gene transfer. Attempts to express different proteins in wheat seeds have so far produced lower yields of recombinant proteins compared to other cereal species.

I.4. Subcellular protein targeting

The stability and accumulation of the recombinant proteins can be achieved by targeting to appropriate subcellular compartment (Streatfield et al., 2007). The targeting organelle is not the same for all the recombinant proteins because of their diverse structures and properties. Literature provides evidences in terms of appropriate targeting strategies for particular types of proteins, e.g. a range of recombinant antibodies expressed in plants. Proper folding, assembly, the formation of disulfide bonds and glycosylation influence the stability and functionality of antibodies, therefore targeting of such polypeptides to the secretory pathway is recommended. Secretion of proteins to the apoplast is achieved by the addition of N-terminal signal peptide to each antibody chain. An additional C-terminal KDEL or HDEL tetrapeptide will retrieve secreted proteins to the endoplasmic reticulum (Conrad et al., 1998).

Plant seed storage tissues contain more complex endomembrane system than non-seed cells where more complex protein can travel through storage proteins to reach their final destination. The endomembrane system comprises a series of functionally-specialized membrane-bound compartments and organelles including an abundant ER reminiscent of mammalian secretory cells, ER-derived protein bodies, endosomes, the Golgi apparatus and different types of vacuoles. Storage proteins in cereal endosperm accumulate in different compartments, and their abundance and distribution varies according to the species (Arcalis et al., 2004). Proteins with N-terminal signal peptides are translocated into the ER and are normally secreted to the apoplast if there are no further targeting signals, or retrieved to the ER if a C-terminal H/KDEL tetrapeptide is present. In addition to the ER lumen, seeds from different crops offer several alternative subcellular destinations for recombinant proteins including protein bodies derived from the ER, protein storage vacuoles (PSVs), starch granules and the surface of oil bodies. All of these destinations have been tested to see if they increase recombinant protein stability, enhance accumulation, facilitate recovery and purification or provide additional benefits such as increasing the efficacy of oral vaccine delivery.

I.4.1. Endoplasmic reticulum

The endoplasmic reticulum (ER) is considered as the "gateway to the secretory pathway" (Ibl and Stoger 2011). The endoplasmic reticulum (ER) is a part of the plant endomembrane system with highly conserved functions in lower and higher plants. The endoplasmic reticulum is crucial for secretory proteins as they are synthesized on the rough ER and are subjected to the machinery for their processing including glycosylation, disulfide bond formation, proper folding and oligomerization (Hadlington and Denecke 2000). For accumulation in the ER and in ER-derived storage organelles, the constructs are designed to contain a targeting sequence for ER retention such as the C-terminal tetrapeptide H/KDEL, which prevents protein secretion (Takaiwa et al., 2007).

Recombinant proteins have shown higher accumulation and stability when targeted to ER in various plant species. Interestingly, heterologous proteins which contain a KDEL signal and accumulate in the ER of tobacco leaves, tend to secreted or accumulated in protein storage vacuoles in seeds (Petruccelli et al., 2006). In tobacco leaves, the single-chain variable fragment (scFv)-KDEL antibody is expressed with accumulation levels of 1.65% TSP (Fischer et al., 1999). The fusion of hydrophobins HFBI sequence from Trichoderma reesei with Green fluorescent (GFP-HFBI) has successfully induced the formation of protein bodies with higher concentration to 51% (TSP) of fusion proteins in tobacco leaves (Fig. I.2; Joensuu et al., 2010). In another study, the expression and proper accumulation of two protein components of KDEL-tagged spider dragline silk in tobacco leaves has also been described (Menassa et al., 2004). In seeds, an antibody scFv-Fc construct fused to KDEL was produced up to 12% TSP in Arabidopsis where the heterologous protein was detected in the periplasmic space but absent from the ER. In this case the presence of heterologous protein apparently disturbed the ER retention and leading partial secretion of the recombinant protein and endogenous storage proteins in seeds (Van Droogenbroeck et al., 2007). Similarly, an unexpected deposition of KDEL-tagged human serum albumin protein was observed in wheat endosperm, where the recombinant protein was deposited together with prolamin aggregates within the storage vacuole (Arcalis et al., 2004). This mistargeting of KDEL-tagged proteins may be a result of the unique storage properties of seed tissues. While targeting the heterologous proteins to the ER improved their aggregation in plants (Conley et al., 2009c), the over-expression of proteins in the ER tends to be harmful, especially with transient expression systems, which often lead to higher accumulation of target proteins than in stable transgenic plants. Strong over-expression of GFP without fusion partner in transient N. benthamiana leaves induced the formation of necrotic lesions, In contrast, when GFP is expressed as a fusion with HFBI, an increase in the expression up to 2 fold was observed and infiltrated leaves remained healthy up to 10 dpi (Joensuu et al., 2010).

Figure I.2 ER-targeted Green fluorescent protein (GFP). Confocal image of the ER targeted recombinant protein (GFP) expression shows the ER network indicated by the flourecent protein (Joenssu et al., 2010).

I.4.2. Protein bodies and protein storage vacuoles

Cereal endosperm contains two types of protein storage organelles: protein bodies, which are derived from the ER, and protein storage vacuoles (PSVs), which are formed de novo (Jiang et al., 2001). Protein bodies generally form directly within the lumen of the rough ER. If they reach a sufficient size, protein bodies may bud from the ER as discrete spherical organelles and remain as such in the cytosol, or they may be sequestered into PSVs by autophagy, as shown in wheat, barley and oat (Levanony et al., 1992; Herman et al, 1999). The formation of protein bodies is complex and the exact mechanism is unclear. PSVs contain three morphologically distinct regions: the matrix, crystalloid and globoid (Jiang et al., 2001). This coincides with the presence of integral membrane proteins in the PSV crystalloid, strongly arguing for the presence of a distinct storage compartment within PSVs (Jiang et al., 2000).

Both these compartments i.e Protein storage vacuoles (Frigerio et al. 2008; Vitale and Hinz 2005) and protein bodies (Coleman et al. 1996; Müntz 1998) have been used in molecular farming to accumulate recombinant proteins (Takaiwa et al., 2007). Plant seed PBs are ER-derived organelles that stably accumulate high amounts of storage proteins. (Arcalis et al., 2004; Galili 2004; Herman and Larkins 1999; Levanony et al., 1992).

I.4.3. Oil bodies

Lipid particles are found in seeds, flowers, pollen, fruits and stems of higher plants, but they are especially prevalent in the seeds of 'oil crops', such as sunflower, safflower, rapeseed and mustard. Oil seeds save lipids in subcellular particles as food reserves in form of triacylglycerol (TAG) used to supply energy for germination and post-germinative growth. The TAGs are present in small subcellular spherical oil bodies, being approximately 0.5-1 µm in diameter. Each OB has a matrix of TAGs surrounded by a layer of phospholipids (PLs). Recombinant proteins targeted to oil bodies are relatively stable, do not aggregate or coalesce and are surrounded by a protected layer of unique proteins called oleosins (Huang 1996). Oleosins provide a recognition signal for lipase binding during oil mobilization in seedlings (Huang, 1996; Murphy, 1993). In maturing seeds, TAGs, PLs, and oleosins are synthesized in the presence of the unique enzyme diacylglycerol acyltransferase in the endoplasmic reticulum (ER), from which budding OBs are released (Hsieh and Huang 2004).

Oil bodies are used as storage organelles in plant biotechnology because of their presence in various plant tissues. Oilseeds in particular are exploited for protein expression. Oil body targeting has its own advantages in terms of purification of fused proteins due to a high ratio of triacylglycerides. Using the oleosin-fusion system, the pharmaceutical protein hirudin has been successfully expressed in Brassica species i.e. Brassica napus (Boothe et al., 1997) and Brassica carinata (Chaudhary et al., 1998). Recently, recombinant protein Apoliprotein Al Milano was expressed as a fusion protein in transgenic safflower seed with high levels of expression corresponding to 7g of ApoAlmilano per kilogram of seed in the transgenic line selected for commercialization (Nykiforuk et al., 2011). One of the molecular farming products closest to the market is recombinant human insulin produced in safflower seeds by the Canadian biotechnology company which is currently undergoing phase III clinical trials (Fischer et al., 2011).

The main advantage of oil bodies for protein targeting is the relatively low cost of processing brought about by the convenient purification process, but this is not always a trouble-free process. Several reports describe incomplete separation of recombinant proteins from oleosin (Parmenter et al., 1995) and the need for post-recovery proteolytic cleavage can reduce the overall recovery as well as increase costs (Boothe et al., 1997; Kuhnel et al., 2003). Therefore, it remains to be seen whether this technology can be widely applied.

I.4.4. Starch binding domains

Starch is the primary energy storage polysaccharide in all plants, a main source of calories in the human diet (James et al., 2003) and widely used for food and industrial purposes. Starch is composed of α-1,4-linked glucose residues organised into the essentially linear amylose and the branched amylopectin containing α-1,6-linkages. Starches are deposited as semi-crystalline granules in chloroplasts of leaves (transitory starch) and in amyloplasts of storage organs. Starch consists usually of 20-30% amylase and 70-80% amylopectin (Ji et al., 2003). Starch granule size varies from less than 1 μm to over 100 μm with spherical or elongated shapes (Smith 2001). Cereals such as wheat and barley contain a mixture of small and large granules known as A and B granules (Fig. I.3), which differ in their morphology and chemical composition (Ao and Jane 2007 and Geera et al., 2006). In wheat, A-granules are between 10 and 38 μm in size and disc- or lenticular-shaped, while B-granules are smaller than 10 μm and possess a spherical or polygonal morphology (Wilson et al., 2006).

Starch-binding domains (SBDs) are present in amylolytic enzymes belonging to three different families of glycosidases (Tanaka et al., 1986; Coutinho et al., 1994). They are found in bacterial and fungal α-amylases, bacterial cyclodextrin glucanotransferases, bacterial β-amylases and fungal glucoamylases (Janecek and Sevcek 1999). SBDs consist of ca. 100 amino acids and their sequences are conserved among different enzymes. These enzymes have different catalytic properties and structures but all are multi-domain proteins that show significant SBD sequence homology (Stefin and Josef 1999). SBDs are positioned exclusively at the C-terminal end of enzymes with few exceptions which contain SBDs at the N-terminus such as Rhizopus oryzae glucoamylase (Janecek 1997), the Thermoactinomyces vulgaris a-amylase (Abe et al., 2004) and the Thermotoga maritima pullulanase (Coutinho et al., 1997).

Starch-binding domains (SBD) from bacterial enzymes that catabolise starches have the capability to bind two helices of starch and thus potentially useful for targeting recombinant proteins to starches. SBDs independently retain their function even if fused to another protein and can therefore be used to target recombinant proteins to existing starch granules (Janecek 1999).

Figure I.3 Wheat A-type and B-type granule structure. (Buléon et al., 1998)

I.5. Seed storage protein trafficking

In the seed, storage proteins move along specific routes within the endomembrane system. Seed cells produce vast amounts of SSPs with different subcellular destinations, including complex protein storage vacuoles and protein bodies derived from the ER (Ibl and Stoger, 2011). Storage proteins, such as albumin and globulins, start the journey from ER lumen and are then transported to PSVs, passing through the Golgi apparatus via dense vesicles (Hohl et al., 1996). In contrast to this pathway, 2S albumins and 11S globulins in pumpkin seed accumulate in PSVs, bypassing the Golgi apparatus via precursor-accumulating (PAC) vesicles (Hara-Nishirama et al., 1998).

Storage protein trafficking has been extensively studied because of the nutritional value of such proteins. In cereals, globulins follow the same routes as described above: they accumulate in the cell's PSVs passing through the Golgi apparatus. The deposition of prolamins in cereal endosperm is specific to the species. In cereals such as rice, maize and sorghum, prolamins aggregate into dense protein bodies within the rough-ER lumen (Fig. 1.4) and remain attached to this organelle (Muench et al., 2000). The protein bodies are encapsulated by a membrane protecting them from proteases and desiccation during seed development, until they are used up as an energy source for germinating seedlings (Muntz 1998). Thus, protein bodies are particularly favoured for deposition and storage of large amounts of recombinant proteins which are poorly secreted or toxic to the host (Torrent et al., 2009). Prolamins are exclusively found in cereal endosperm. In wheat and oat, the prolamins accumulate to form aggregates and bud off the ER as in other cereals. But later they are incorporated into PSVs by an autophagy-like process bypassing the conserved mechanism of the Golgi mediated targeting (Arcalis et al., 2004; Galili et al., 1993; Levanony et al., 1992). The role of the Golgi apparatus has been controversial but generally, the Golgi independent pathway is more widely accepted, although a Golgi-dependent pathway has also been proven for small amounts of prolamins (Miflin et al., 1981).

Wheat endosperm contains two types of PBs: the low density type protein bodies (light PBs) where gliadins are present, and the high density type of protein bodies (dense PB), where high molecular weight glutenins are located. Both the high molecular weight glutenins and gliadins are transported and deposited to the PSV by two separate processes, the Golgi-dependent and the Golgi-independent pathway (Kumamaru 2007). Wheat protein bodies do not remain as separate cytosolic structures but are sequestered into provacuoles (Rubin et al., 1992), predominantly in the cells of the subaleurone layer (Fig. I.5A). Surprisingly, cereal prolamins do not contain any known ER or vacuolar targeting signals. Wheat protein bodies are sequestered within a large, central vacuole. In wheat, gliadine and glutenins are the major classes of prolamins in the endosperm, but wheat also contains an 11S globulin homolg and triticin as minor storage proteins, which account for 5% of total seed proteins (Singh et al., 1991). Gliadins accumulate in protein bodies as monomers, while glutenins assemble via non-covalent interactions and intermolecular disulfide bonds. Unlike prolamins, gliadins contain N-terminal signal sequences that determine their passage into lumen of the ER. However, it is not known why some gluten proteins pass and others bypass the Golgi apparatus (Tosi et al., 2009).

Maize γ-zein is able to induce protein body formation in dicot tissues regardless of the presence of other zein subunits (Coleman et al., 1996). Zein constitutes more than half of the total seed proteins in maize (Fig I.3B). Zein proteins can be detected in maize endosperm 10-40 days after pollination (DAP) and constitute approximately 50% of the total proteins in mature seed (Lending and Larkins 1989). N-terminal of -Zein is composed of four domains: eight repetitions of the PPPVHL motif, a hydrophobic cystein rich C-terminal sequence, Pro-X enclosing proline residues and a 19 amino acid signal peptide (Prat et al., 1985). Three types of zein (β, γ, δ) have been successfully produced in leaves of transgenic arabidopsis, where they are deposited to ER derived protein bodies (Bagga et al., 1995, 1997).

Alpha zeins are composed of two major subclasses (19 and 22 kDa zeins), and constitute the most important class of zeins, accounting for 70% of the total prolamins. The remaining minor group of three zein classes include delta, beta and gamma (δ, β and γ) having molecular masses of 10, 15 and 27 kDa respectively, when separated on SDS-PAGE. -zeins are rich in cystein while both δ and β are rich in methionine residues. Beta zeins share some features with γ-zein, such as both being sulphur rich, carrying conserved N and C terminal peptides and requiring a reducing agent for their solubility in alcohol.

Rice endosperm accumulates three types of storage proteins, the glutelins, prolamins and α-globulins (Muench et al., 1999), where globulins (known as glutelins) and prolamins are in majorityand make up 60-80% and 20-30% of total seed proteins, respectively (Kawakatsu et al., 2010). These storage proteins are transported to different sites of the endomembrane system (Krishnan et al., 1986, Tanaka et al., 1980). In rice seeds, more than 80% of the total seed proteins are deposited as storage proteins in protein bodies as protein storage organelles. There are two distinct protein bodies found in the rice endosperm known as PB-I, which originates from the ER and contains prolamins, and vacuolar PB-II, accumulating globulins and glutelins (Fig. I.6; Krishnan et al., 1986; Yamagata and Tanaka 1986). In rice, the chaperone BiP is closely associated with prolamins and is located at the periphery of rice prolamin bodies, which aids the deposition of prolamins on the surface of protein bodies (Li et al., 1993; Muench et al., 1997). BiP is known to be present in cereal PBs, where it is considered to play a role in storage, protein folding and assembly (Li et al., 1993). Okita and co-workers could show that prolamin mRNA plays a crucial role in the deposition of prolamins within the rice endosperm. Evidence has been found that rice prolamins and globulins/glutelins are translated on two separate sub-domains of ER. Prolamin mRNAs are targeted to ER membrane surrounding the protein bodies by a mRNA signal recognition process (Hamada et al., 2003a; Hamada 2003b), whereas mRNA localization on the cisternal ER membrane results in globulin/glutelin transport to the PSV (Li et al., 1993a). The specific mRNA trafficking to protein bodies appears to involve the cytoskeleton (Hamada et al., 2003b).

Figure I.4 Conceptual diagram of the ontogeny of PBs and PSVs. Protein bodies form within the rough endoplasmic reticulum (rER). They can either bud off the ER and remain in the cytoplasm or be sequestered into vacuoles by autophagy. PSVs are formed as the consequence of ER-synthesized storage proteins progressing through the endomembrane secretory system to the vacuole for accumulation (Hermans and Larkins 1999).

Figure I.5 Deposition of wheat protein bodies. Light microscopy. Cross section of a wild type wheat seed stained with methylene blue. The protein bodies are deposited within the large central vacuole (arrow) in the wheat endosperm cells. Figure adapted from Prof. Stoger's Lab at Vienna, Austria.

Figure I.6 Protein bodies deposition in maize and rice. Light microscopy. Semi-thin cross section of wild type maize seed treated with methylene blue. Protein bodies (arrow) are distributed in first few cell layers of maize endosperm tissues. Rice. Light microscopy. Methylene blue stained semi-thin section. Protein bodies Type I and II (arrow) are visible within rice endosperm. Figure adapted from Prof. Stoger's Lab at Vienna, Austria.

I.6. Fluorescent protein markers

Transgenic plants are generally developed through co-insertion of selectable marker genes together with the transgenes. Selection with antibiotics or herbicides is then used to enrich for plants with successful integration of the transgenes in the host cells. The above mentioned reporter gene efficacy may be affected by multiple inserts, escapes of the cell from the selection effect during cell division or diffusion of selection properties to non-expressing neighbouring cells. Selection with selectable marker genes allows transformed cells to survive and multiply at a lethal concentration of a selective agent (antibiotic). In contrast to selectable marker genes, fluorescent reporter genes (GUS; GFP, dsRed) can be easily detected by phenotypic analysis (Miki and McHugh 2004). For the identification of transformed cells, which are very few within a large background of non-transformed cells, reporter genes (cat, uidA, luc, gfp, dsRed) are favoured for plant transformation constructs. To monitor the gene expression, protein localization and intracellular protein trafficking may be visualized with protein markers in situ, without harming the plants (Ziemienowicz 2001). GFP and DsRED are favoured because of their high quantum yields and stability in living cells.

Fluorescence is a phenomenon in which a material absorbs light of one colour (wavelength) and emits light of a different colour (wavelength). Fluorescent reporter proteins, such as green fluorescent protein (GFP) and red fluorescent protein (DsRED), in plant cell biology are essential to track, visualise and quantify gene expression in the host cells. Monitoring of CAT, GUS and LUC activity requires the preparation of protein extracts, addition of suitable substrates and performance of enzyme assays (Jach et al., 2001). Due to intrinsic fluorescence detection, the fluorescent protein offers an advantage over GUS and luciferase marker proteins with respect to being used without any additional specific substrates to monitor gene expression and analysis of transformants in the first generation.

I.6.1. Discosoma red fluorescent protein

Red fluorescent protein (DsRed) has been isolated from coral of the genus Discosoma (Matz et al., 1999) and is used as natural red chromophore. DsRED has the maximum excitation and emission shifted to the red when compared to GFP (excitation/emission maximum of GFPs 488-495 nm/507-510 nm versus 558 nm/583 nm for DsRED), which is just above the excitation peak of chlorophyll (Matz et al., 1999). Discrimination of DsRED and GFP fluorescence can be easily done by using appropriate filter settings; allowing simultaneous multicolour imaging of different genes (Jach et al., 2001). DsRed is frequently used to monitor developmental and spatial gene expression in transgenic plants. Like GFP, it does not require the preparation of protein extracts, additional substrates or enzymes. DsRed allows non-invasive, non-destructive detection at very early stages of plant transformation. DsRed was first expressed in tobacco BY2 protoplast cells by Mas et al. (2000).

The DsRed is visualized by the red phenotype of seeds, seedlings and plant lines, which express the protein in the cytosol, ER or chloroplast. Fertility and germination of seeds were not affected, while seeds and seedlings maintained the red fluorescence phenotype (Jach et al., 2001). Red fluorescent marker proteins are traceable markers and can be easily detected by simple microscopic analysis. The fused protein can be easily detected at very early stages of its development using low tech equipment.

Figure I.7 Red Fluorescent protein as selection marker. Detection of transgenic seeds on a segregating ear of maize using red fluorescent protein (DsRed) as a visible marker (Rademacher et al., 2009)

I.7. Encapsulation of recombinant proteins

Several strategies have been exploited in an attempt to maximize heterologous protein accumulation in plant cells. Despite successful alternative expression systems, foreign protein expression levels are relatively lower in the plant expression system due to major challenges, which include insufficient accumulation levels and lack of efficient purification methods for recovery (Joennsu et al., 2010). Expressing recombinant proteins as fusions to protein - stabilizing partners has a positive impact on the protein accumulation and purification, and has proven itself to be an easy and fast means of purifying recombinant proteins from plants (Witte et al., 2004). To enhance recombinant protein accumulation, a variety of fusion proteins has been used in plants such as cholera toxin B subunit (Molina et al., 2004), viral coat proteins (Canizares et al., 2005), ubiquitin (Mishra et al., 2006), β-glucuronidase (Dus Santos et al., 2002), human immunoglobulin (IgG) α-chains (Obregon et al., 2006). To enhance the purification process, recombinant proteins are fused to small affinity tags or proteins with defined binding characteristics such as His-tag, glutathione S-transferase-tag, FLAG-tag, c-myc-tag, Arg-tag, calmodulin-binding peptide, maltose-binding protein, the cellulose-binding domain (Lichty et al., 2005) and eight-amino acid StrepII epitope tag (Skerra and Schmidt 2000). However, these tags are used for purification of recombinant proteins using affinity chromatography techniques, which are relatively costly and difficult to scale up for industrial use (Menkhaus et al., 2004; Waugh 2005).

Large-scale partitioning of different biological components, including recombinant proteins, is often a bottleneck in modern biotechnology. To increase the accumulation of valuable recombinant proteins in seeds, and to facilitate protein purification, the properties of seed storage proteins and/or the trafficking pathways that lead to storage protein deposition in seeds have been exploited (Table I.1). Thus, several protein fusion strategies have been developed to address these issues (Terpe 2003).

Figure I.8 Using storage organelles for the accumulation and encapsulation of recombinant proteins

Table I.1Fusion partners for creation of artificial storage organelles.

Origin Purification Method Increased yield ER-Signal PB-Formation

Zein Plant Isopycnic sucrose density centrifugation 15-100 fold Protein repeat domain, Pro-X domain, Cystein domain

Elastin-like polypeptides Animal Inverse transition cycling (ITC) 2-100 fold K/HDEL sequence VPGXG

Hydrophobin Fungus Surfactant-based aqueous two phase system (ATPS) Unknown K/HDEL sequence Hydrophobic interation. 8 cysteines

I.7.1. Zein

In the cereal endosperm, zeins are accumulating in the endoplasmic reticulum-derived protein bodies and can accumulate up to 15% of total endospermic proteins (Marzbal 1998). Zeins are among well studied seed storage proteins in maize. Endospermic cells of maize, together with other zein classes (α, β, γ, δ) form large complexes known as protein bodies in endoplasmic reticulum lumen (Fig. I.9). No specific canonical HDEL/KDEL retention/retrieval signals responsible for retaining proteins in the ER have been found in γ-zeins. This phenomenon of protein retention and localization within the ER of endosperm of maize is still to be discovered.

Protein fused to N-terminal proline-rich domain of γ-zein, known as Zera has capability to stably express recombinant proteins in plants and have induced an increase in overall yields (Mainieri et al., 2004; Torrent et al., 2009;). Zera fusions with other protein are able to self-assemble and accumulate the recombinant proteins inside protein bodies. The exact synthesis of protein bodies is still undetermined, but it has been proven in earlier studies that proline rich domains made of hexapeptide PPPVHL are responsible for retention of proteins in the ER. There is evidence that β-zein sharing conserved N- and C-terminals with γ-zeins, which are then capable of inducing protein body formations (Colemann and Larkins 1999). However, expression of only α-zeins fails to induce protein body structures when expressed into different hosts.

Figure I.9 Localization of storage proteins in maize endosperm. Fluorescence microscopy. Two classes of Zein (α, γ) expressed in maize leaves. Figure adapted from Prof. Stoger's Lab at Vienna, Austria.

I.7.2. Elastin-like polypeptide

Elastin like polypeptides (ELP) are synthetic biopolymers made of repeats of amino acid `Val-Pro-Gly-Xaa-Gly` sequences, where residue Xaa is any amino acid except proline (Urry, 1988). Elastin-like polypeptides (ELP) are found in all mammalian elastin proteins and can be purified with temperature based non chromatographic methods. Elastin-like polypeptide (ELP) tags upon expression with particular proteins induce protein body formation of the same size and morphology to natural protein bodies as demonstrated using an ELP fusion to green fluorescent protein (GFP) which significantly increased the stability and accumulation of recombinant proteins in tobacco leaves (Fig. I.10; Conley et al., 2009).

ELP fused protein can be purified with inverse transition cycling (ITC) which is a fast, simple, easily scalable and cheap non-chromatographic method for protein purification. This method is used for purification of antibodies, cytokins and spider milk proteins from transgenic plants. ELP are converted to hydrophobic aggregates and form β-spiral structures when heated around the transition temperature. This thermal response property has also been transferred when ELPs are fused with recombinant proteins, enabling the purification through ITC (Conley et al., 2009). This heat sensitivity is directly related to the size of the proteins in plants, such as that the transition temperature could be 30-39 C for proteins ranging from 30-65 kD in mass (Stibora et al., 2003). However, the heat sensitivity is inversely proportional to the size in E.coli (Meyer and Chilkoti 1999). ELP fusions are up to 40 times more abundant than control protein lacking the ELP when expressed in tobacco seeds (Scheller et al., 2006).

The proteins, once fused with the ELP tags, exhibit an increase in accumulation in plants, but this procedure still needs to be optimized in order to maximize the accumulation levels of these fused proteins in plants. ELP fusion proteins have significantly accelerated protein accumulation and purification in plants, but the impact of ELP fusion proteins on overall factors defining protein quality still needs to be determined.

Figure I.10 Fluorescence microscopic image of tobacco leaf. In the presence of an ELP fusion tag, the ER-targeted GFP was detected in brightly fluorescing spherical-shaped particles distributed throughout the cells of the leaf (Colney et al., 2009)

I.7.3. Hydrophobins

Hydrophobin fusions are originally developed for the purpose of purifying proteins from fungal cultures supernatant (Conley et al., 2011). It has already been demonstrated that this fusion can increase the accumulation of target proteins in plants and fungi (Linder et al., 2004). Endoplasmic reticulum-targeted hydrophobins induce the formation of novel protein bodies (Table.I.1; PBs; Conley et al., 2009b; Torrent et al., 2009).

Purification of macromolecules is the key factor in alteration of the fusion protein partners, through surfactant-based aqueous two-phase system (ATPS; Linder et al., 2004). ATPS is a simple, rapid, and inexpensive procedure providing volume reduction, high capacity and fast separations (Persson et al., 1999). At industrial points of view, hydrophobins are attractive because they require a one-step purification procedure and can be easily scaled up for industrially valuable proteins (Linder et al., 2004; Selber et al., 2004).

Figure I.11 ER-targeted GFP Hydrophobin fusion (GFP-HFBI) promotes the formation of PBs in Nicotiana benthamiana. Leaf epidermal cell accumulation GFP-HFBI fusion protein in protein bodies (Joensuu et al., 2010).

Table I.2 Examples of recombinant proteins produced in cereal seeds.

Crop Recombinant protein Plant Tissue/ Subcellular Localization

Expression Level/Yield Promoter Reference


Barley HIV 2G12



Escherichia coli heat labile enterotoxin (LT-B)

HIV Diagnostic Reagent (HIVDR) Endosperm/embryo

ER-derived Protein Bodies

Endoplasmic Reticulum (ER)

Starch granules

Endoplasmic reticulum

38-75 µg per gram

30μg/g dry weight

5.7% TSP


150 µg of reagent g-1 Endosperm-specific rice glutelin-1

glutelin-1 (gt-1) promoter

Maize ubiquitin promoter

27-kDa γ-zein promoter

BÃ-17 promoter

Ramessar et al., 2008

Rademacher et al., 2008

Kusnadi et al., 1998

Chikwamba et al., 2003

Schünmann et al., 2002



Human serum albumin (HSA)

7Crp peptide

Human interleukin

Cholera toxin B subunit

Human Serum Albumin

Single-chain Fv antibody (ScFvT84.66)


ER-derived Protein Bodies

Endoplasmic reticulum

Protein bodies (PB) I & II

ER-derived Protein Bodies


2.75g/kg of Rice

60 µg/ grain

4.5mg/100g RSP

30 µg/g CBT per Seed

0.5% TSP


Endosperm-specific promoter,

Gt13a promoter,

Glutelin GluB-1

Glutelin GluB-1

Ubiquitin-1 promoter

Maize ubiquitin promoter

He et al., 2011

Takaiwa et al., 2009

Fujiwara et al., 2010

Nochi et al., 2007

Arcalis et al., 2004

Stoger et al., 2000

TSP: Total Seed Protein; RSP: Rice Seed Powder

Objectives of the study

In the context of molecular farming, plant seeds are ideal vehicles for the production of recombinant pharmaceutical proteins because they contain specialized compartments for the stable accumulation of storage proteins. Recombinant proteins and endogenous storage proteins in dry seeds each benefit from the same stable environment, allowing long-term storage and batch processing. Seeds contain several unique types of storage organelles to which recombinant proteins can be directed and in which they accumulate. These include protein bodies derived from the endoplasmic reticulum (ER) and protein storage vacuoles. The maize seed storage protein γ-zein is an endogenous prolamin that can trigger the formation of protein bodies not only in its native environment (maize endosperm) but also in other plant cells and in eukaryotic cells generally. In this project, the repetitive (PPPVHL)8 and PX domains from γ-zein protein were expressed as a fusion with the red fluorescent marker protein DsRed to allow the behavior of the fusion protein to be characterized visually in different wheat tissues.

The main objective of this thesis was to study the expression of recombinant wheat plant and to study trafficking and deposition of the model fusion protein in different wheat tissues, such as the endosperm, embryo and leaves, to investigate its ability to form protein bodies for the storage of recombinant fusion proteins in diverse environments. Our hypothesis was that in wheat cells the model γ-zein-DsRed fusion protein would form a unique and separate population of protein bodies budding from the ER into the cytoplasm, but because endogenous prolamins would be present in the endosperm it would not be possible to predict the behavior of the model protein in this setting. We therefore investigated the precise localization of the fusion protein to find out whether it would form a separate population of ER-derived protein bodies or mix with endogenous prolamins in endogenous glutelin bodies. Autophagy-like processes are well-documented in wheat endosperm cells, so we also set out determine whether the model protein was ultimately incorporated in the protein storage vacuole like wheat glutenins.

One of the bottlenecks in molecular farming is the low yields of recombinant proteins in transgenic plants, often reflecting a combination of low-level accumulation and inefficient extraction and downstream processing. We used the model fusion protein γ-zein-DsRed to investigate strategies for the improvement of recombinant protein accumulation in wheat, based on preventing exposure to proteolytic enzymes in the cytosol. This thesis contributes to the body of knowledge on recombinant protein expression in plants by studying the expression and deposition of γ-zein-DsRed and by extrapolating this knowledge to develop strategies that can be applied to pharmaceutical proteins in commercial processes.

The maize seed storage protein γ-zein is a prolamin that can induce the formation of protein bodies in its native environment, the endosperm. However, the expression of γ-zein in diverse eukaryotic cells has shown that it also induces the formation of novel compartments in many heterologous backgrounds making it potentially useful as a more general approach for recombinant protein expression. We generated an expression construct comprising 90 amino acids sequences i.e repetitive (VHLPPP)8 and Pro-X domain of maize prolamin (γ-zein) fused with Discosoma spp. red fluorescent protein DsRed under the control of a constitutive promoter ubiquitin (Fig. III.1), allowing the protein to be expressed in both seed and non-seed tissues in wheat plants. Transgenic wheat lines produced by particle bombardment (II.4.3) were used to analyze the expression and distribution of the recombinant model proteins, focusing on the formation of fluorescent protein bodies which were investigated using a range of molecular, immunological and imaging methods. Recombinant fusion protein (γ-zeinDsRed) accumulated in the artificial protein bodies in all tissues analyzed. The possible degradation of the fusion proteins was investigated by immunoblot analysis (II.5.4) and their accumulation within novel protein bodies was confirmed by the immunofluorescence labeling (II.6.5) of leaf, endosperm and embryo tissues from transgenic wheat plants.

III.1. Fusion protein expression construct

The fusion protein containing the entire DsRed coding sequence joined to the 90 amino acids of γ-zein comprising proline rich repetitive (PPPVHL)8 and Pro-X domain. The fusion gene was inserted into the pTRA vector downstream of the constitutive ubiquitin promoter, and in frame with a His6 tag to facilitate affinity purification. The vector backbone included an Escherichia coli origin of replication and ampicillin resistance gene for cloning, and the bar resistance gene conferring resistance to the herbicide phosphinothricin (PPT) for selection in plants. Transcription bar gene is also driven by the constitutive promoter ubiquitin. The final expression vector was named DsRedzenH (Fig. III.1).

Figure III.1 Vector map of pTRAbux-DsRed.zen-H (DsRedzenH). E. coli DH5α competent cells (II.2.2) were transformed by heat shock (II.2.2) and single colonies were tested by PCR (II.3.6) to confirm the presence of an insert (II.3.9). The recovered plasmids (II.3.7) were digested with restriction enzymes (II.3.2) to verify the orientation of the insert, and its integrity was confirmed by sequencing (II.3.9).


Figure III.2 Structure of the pTRAbux-DsRed-zenH (DsRedzenH) expression vector.

III.2. Transformation of wheat plants

Approximately 1500 immature wheat embryos were bombarded (II.4.3) with DNA-coated metal particles (Fig. III.4) and the tissues were cultured for 14 days to induce the formation of callus. Individual explants were transferred to selection medium (Table II.2) containing 2 mg/ml PPT and subcultured onto the same medium at regular intervals (Fig. III.5). Following selection (II.4.5), independent transgenic callus lines were placed on regeneration medium (Table II.2) containing 3 mg/ml PPT for 3 weeks in natural light. Transgenic shoots ~15 cm in length were transferred to soil in the greenhouse, and checked for PPT resistance after 2-3 weeks by spraying with 250 mg/L PPT every 2 days. Ten transgenic plants were recovered but only three were selected for further analysis based on highest expression of DsRed in leaves (II.4.3) and were cultivated in the greenhouse until mature. The wheat transformation protocol is summarized in Figure III.3.

Figure III.3 Wheat plant transformation procedure. A. Immature wheat embryos were isolated aseptically 14 days after anthesis (II.4.1) and maintained on induction medium (Table II.2) for 6 days in darkness. B. Six-day-old embryos were placed on osmoticum medium (Table II.2) before bombardment C. Bombardment was carried out using the PDS-1000/HE (II.4.3) D. Fluorescence imaging (II.6.6) confirmed DsRed expression in wheat cells 24 h post-bombardment. E. Transgenic wheat callus pieces began to regenerate under illumination. F. Three-week-old wheat plants continued to flourish on regeneration medium (Table II.2) G. Transgenic plantlets developed on ½MS medium (Table II.2) containing PPT (II.4.4) H. The plantlets were transferred to the greenhouse and maintained under selective conditions (II.4.4). T. Transgenic wheat plants were grown to maturity in the greenhouse.

Figure III.4 Bio-Rad PDS1000/He particle bombardment apparatus.

Figure III.5 Wheat tissue culture: from immature embryos to shoot development.. A. Immature embryos (14 days after anthesis) on induction medium (Table II.2). B. Seven-day-old wheat embryos pre-bombardment (II.4.3) C. Wheat callus 5 days post-bombardment (II.4.3) D. Three-week-old transgenic callus on selection medium (II.4.5.2) E. Wheat callus producing shoots (II.4.5.3) on regeneration medium (Table II.2) under illumination. F. Shoot development after 3 weeks on regeneration medium (Table II.2) under illumination.

III.3. Selection of wheat plants using PPT

Wheat callus derived from immature embryos was bombarded (II.4.3) with plasmid DsRedzenH (Fig. III.1) containing the bar gene for selection with PPT (Fig. III.5). When the primary transformants reached maturity, T1 seeds were harvested to generate T1 plants for segregation analysis under PPT selection (II.4.5.3). We analyzed 20 T1 plants for each line (WA-07, WB-21 and WB-28) by applying an aqueous solution containing 150 mg/L or 250 mg/L PPT on the 15th and 25th days after germination (Fig. III.6). Two weeks later, the appearance of the plants was recorded and the surviving plants were evaluated for the presence of the transgene. All three transgenic lines produced plants that were resistant to PPT whereas the sensitivity of all wild-type plants was apparent (Fig. III.6).

The segregation ratio was determined by harvesting 50 immature seeds at least 15 days post-anthesis from each line, surface sterilizing the seeds and rescuing the embryos (II.4.1). The embryos were placed on ½MS medium (Table II.2) containing 3.0 mg/L of PPT and the proportion of resistant and sensitive embryos was determined (Fig. III.7). Transgenic PPT-resistant wheat plants were used as positive controls and wild-type plants as negative controls (Fig. III.7). We observed the expected 3:1 Mendelian segregation ratio for all three transgenic lines (WB-21, WB-28 and WA-07) after 10 days, indicating single-locus integration.

Figure III.6 Wheats plant after PPT application. The transgenic wheat leaves were examined under fluorescence microscope (II.6) for DsRed expression and were selected for PPT analysis. The plants were sprayed with an aqueous solution containing 150 mg/L or 250 mg/L PPT on the 15th and 25th days after germination. Two weeks later, the appearance of the plants was recorded and the surviving plants were evaluated for the presence of the transgene.

A, B, C. Wild type cv. Bobwhite plants are affected with PPT application. D, E, F. Transgenic wheat lines WB-21, WB-28 and WA-07, respectively, showing resistance to PPT.

Figure III.7 Segregation of transgenic wheat lines: Embryos were removed aseptically from the transgenic wheat seeds and were placed on ½MS medium (Table II.2) containing 3.0 mg/L of PPT. The petri dishes were placed in growth chamber at 25 ˚C for one week and germination of the embryos was observed.

1. Growth of T1 embryos after one week on ½MS medium (Table II.2) with PPT. A, B. Transgenic line WB-21. C, D. Transgenic line WB-28. E, F. Transgenic line WA-7 shows high resistance to PPT with normal germination N. No germination was observed for wild type wheat (cv. Bobwhite) used as negative control 2. Growth of wheat embryos after 10 days on ½MS medium (Table II.2) to confirm growth of positive and negative controls on PPT containing media. A Normal growth was observed for positive control (PPT-resistant wheat line) B. Embryos show no germination for wild type wheat cv. Bobwhite used as negative control.

III.4. Molecular verification of transgenic wheat lines

Transgene integration in the wheat lines was confirmed by testing immature embryos of T1 plants rescued under PPT selection (II.4.6.2) by PCR (II.3.6). Germinating embryos (Fig.III.5) developed into plants after transfer to soil (Fig.III.3T). The total genomic DNA was extracted from leaves (II.3.5) and amplified using forward and reverse primers specific for the DsRed transgene (Table II.1). The anticipated 362-bp product was identified in all three transgenic lines (Fig. III.8) matching the plasmid positive control (II. 3.1). There was no product amplified in the negative control (wild-type wheat cv. Bobwhite).

Figure III.8 Confirmation of transgene integration by PCR. PCR was performed as described (II.3.6) for total genomic DNA isolated (II.3.5) from transgenic wheat leaves. 5µl from 25 µl reaction was applied on to 1.2% TAE agarose gel (II.3.3). PCR products of 362 bp were amplified using primers designed for DsRed (Table II.1). All three transgenic wheat lines (WB21, WB 28 and WA07) yielded a 362-bp amplification product, matching the positive control, DsRedzenH (II.3.1). There was no band in the negative control lane corresponding to wild-type wheat leaves. 100bp ladder was used as marker.

III.5. DsRed as a marker for selection and protein trafficking

Fluorescent marker DsRed was utilized as fusion protein to monitor the expression of recombinant proteins in transgenic wheat tissues. As fusion, DsRed was found very beneficial for transgene selection at very early stages of transformation process. Fluorescence microscopy (II.6.6) revealed strong DsRed fluorescence in transgenic callus 18 h after bombardment (Fig.III.10). A more detailed time-course analysis confirmed DsRed fluorescence in rapidly-dividing callus cells (Fig. II.10 A, B, C, D, E, F), allowing these cells to be selected for regeneration. Transgenic plants from lines WA-7, WB-21 and WB-28 showed a high and consistent level of transgene expression in the leaves, and were selected for further analysis (Fig. III.11A, B,C). Leaves from wild type plants showed no evidence of fluorescence (Fig. III.11D).

DsRed was also used as fluorescence marker for trafficking of recombinant proteins, reflecting the detection of labeled protein bodies by fluorescence (II.6.6) and confocal microscopy (II.6.7) from 8-10 days post anthesis. In addition, the fusion protein could also be detected using DsRed-specific antibodies for immunofluorescence analysis (Fig. III 15, 16 and 18). The γ-zein-DsRed protein was also identified immediately after the isolation of total soluble protein from transgenic wheat leaves and seeds (III.5.1) when the protein pellet was illuminated with green light (Fig. III.9). No DsRed fluorescence was detected in extracts from the wild type cv. Bobwhite (negative control).

Figure III.9 Protein pellets extracted from wheat leaves under green light. Leaf tissues from transgenic wheat line (WA-07) were homogenized with 3 volume of protein extraction buffer (II.1.2). The mixture was centrifuged for 40 min at 13,000 rpm at 4 ËšC and examined under green light with red filter.

A. Red fluorescence was observed in the bottom pellet shows evidence of recombinant proteins accumulation in transgenic wheat leaf. B. Protein extracts (II.5.1) from wild type wheat cv. Bobwhite protein shows no fluorescence.

Figure III.10 DsRed expression during transgenic callus development. Wheat immature embryos (after 14 days post anthesis) were placed on W-ID medium (Table II.2) for six days under dark condition at room temperature. The plasmid DNA (II.3.1) was coated with gold particles (II.4.2) and regenerating wheat calli were bombarded as described (II.4.3).

Fluorescence images (II.6.6) A, B, C, D, E, F taken 1, 2, 6, 12, 16 and 20 days after bombardment (II.4.3). The expression of recombinant protein (γ-zein-DsRed) is visible in growing wheat calli 18 hours after bombardment (II.4.3).

Figure III.11. DsRed as a visual selection marker in transgenic wheat lines. Young transgenic leaves were cut into small pieces in 0.1 M phosphate buffer, pH 7.4 (II.1.2) and examined for the expression of recombinant proteins.

Fluorescence microscopy images (II.6.65) confirms DsRed transgene expression in A. WB-21, B. WB-28 and C. WA-7 transgenic leaves, compared to D. No red fluorescence was observed in wild-type cv. Bobwhite used as negative control. Bars, 500 µm (A, B, C, D).

Figure III.12 Expression of induced protein bodies in wheat endosperm cells. Transgenic wheat seeds were analyzed 17 days post anthesis for recombinant protein production in wheat endosperm cells. The seeds were sliced in 0.1 M phosphate buffer, pH 7.4 (II.1.2) and observed under florescence microscope (II.6.6).

Fluorescence microscopy (II.6.6) images are shown at progressively higher magnifications. Numerous DsRed bodies were observed throughout the endosperm cells confirmed the accumulation of recombinant proteins (γ-zein-DsRed) in wheat storage tissues. Bars 500 µm (A), 200 µm (B), 100µm (C), 50µm (D).

Figure III 13. Expression of DsRed protein bodies in wheat embryo cells. Transgenic wheat seeds were analysed 17 days post anthesis for recombinant protein production in wheat embryo cells. The embryos were isolated and sliced in 0.1 M phosphate buffer, pH 7.4 (II.1.2) and observed under florescence microscope (II.6.6).

Fluorescence microscopy (II.6.6) images are shown at progressively higher magnifications. The fluorescence micrographs confirm the accumulation of recombinant proteins in wheat embryo cells. Higher expressions of recombinant proteins were observed in wheat embryo than endosperm cells. Bars 200 µm (10X), 100 µm (20X), 50µm (40X), 50µm (63X).

III.6. Model fusion protein in wheat tissues

After successful transgenic plant generation, the expression and subcellular localization of the recombinant fusion protein in wheat seed and vegetative tissues was investigated by immunoblot (II.5.4), fluorescence (II.6.6) and confocal microscopy (II.6.7) and immunolocalization experiments (II.6).

III.6.1. Presence and stability of the recombinant protein

The heterologous expression of repeated and Pro-X portion from γ-zein resulted in ubiquitous expression the partner protein (DsReD) in a tissue independent manner in wheat plant. Western blot analysis of wheat leaf, endosperm and embryo tissues extracts (II.5.1) confirmed the integrity of the fused proteins, with an apparent molecular mass of approximately 37 kDa (Fig.III.14 A). Total soluble proteins were extracted from wheat tissues (II.5.1) separated by SDS-PAGE (II.5.2), transferred to a membrane (II.5.4) and probed with a primary (rabbit anti-27kD γ-zein) and secondary antibody (goat anti-rabbit IgG) and visualized using NBT/BCIP substrate. The presence of anticipated 37-kDa band in all three tissues revealed that the recombinant fusion protein has been successfully accumulated in wheat seed and vegetative tissues (Fig. III.14A). There was no equivalent band in total soluble protein extracted (II. 5.1) from corresponding tissues in wild-type wheat cv. Bobwhite. Total protein extracted (II.5.1) from transgenic tobacco leaves expressing DsRed-zenH was used as a positive control.

To find out the possible degradation of the recombinant fusion protein in seed tissues (endosperm, embryo), transgenic wheat seeds were stored for five months at normal room conditions. The total proteins extracted (II.5.1) from wheat endosperm and embryo tissues were subjected to SDS-PAGE (II.5.2) and western blot analysis (II.5.4). The protein extracts were probed with the primary (rabbit anti-27kD γ-zein) and secondary (goat anti-rabbit IgG) antibodies. The presence of 37-kDa bands presented the evidence that no significant degradation in the endosperm and embryo tissues were detected (Fig.III.14B).

Figure III. 14. Immunoblot confirming the presence and stability of the recombinant γ-zein-DsRed fusion protein in wheat tissues. Immunoblot confirms the accumulation of the recombinant γ-zein-DsRed fusion proteins in different wheat tissues. Total soluble protein extracted from three wheat tissues (II, 5.1) was separated by 12% (w/v) SDS-PAGE (II.5.2) and blotted onto a nitrocellulose membrane (II.5.4) for immunodetection with antibodies against γ-zein. Visualization was performed using NBT/BCIP substrate. Transgenic wheat seeds were stored for 5 months at room temperature. To find out possible degradation of recombinant proteins, total protein extract from leaf, endosperm and embryo (II. 5.1) were separated by 12% (w/v) SDS-PAGE (II.5.2) and blotted onto a nitrocellulose membrane (II.5.4) for immunodetection with 27kDa- γ-zein antibody and visualized using NBT/BCIP substrate.

A. Immunoblot micrograph for fresh wheat tissues extracts (II, 5.1).1. Pre-stained plus protein ladder. 2. Positive control extract from transgenic tobacco expressing the DsRedzenH construct (II,5.1). 3. Negative control extract from wild type wheat seed extract (II, 5.1) shows no band. 4. Young transgenic wheat leaf extract (II, 5.1), containing the 37-kDa fusion protein. 5. Transgenic wheat endosperm extract (II, 5.1), containing the recombinant fusion protein. 6. Transgenic wheat embryo extract (II, 5.1), containing the recombinant fusion protein. B. The prescence of the recombinant fusion proteins (DsRed-zeinH) in wheat seed tissues after five months storage. The existence of 37-kDa bands demonstrates that no degradation products were detected in wheat tissues including 1. whole seed 2. embryo 3. endosperm.

III.6.2. Model fusion protein expression and accumulation in transgenic wheat leaves

The detailed expression profile of the γ-zein-DsRed fusion protein in leaves was analyzed by fluorescence (II.6.6) and confocal microscopy (II.6.7) revealing strong and constitutive expression (Figures III.15A and III.1