Mechanisms Of Infection Vir Gene

Agrobacterium tumefaciens Agrobacterium tumefaciens is a ubiquitous soil borne pathogen responsible for Crown Gall disease (right), affecting many higher species of plant. The pathogen is a problem for agriculture all over the world. DNA transfer from Agrobacterium tumefaciens to eukaryotic cells is the only known example of inter-kingdom DNA transfer (Dumaset al., (2001). It is therefore unsuprising that Agrobacterium has evolved its own unique and specialised system to accomplish this task, which is of great interest to plant scientists. The information contained in this website will explore the history and all that we know about this remarkable micro-organism. This website aims to give an insight into the bacterium, with information about the research that has been carried out in the past, and how the knowledge we have gained can be applied to help alleviate problems facing food production around the world. Other potential uses of the technology are also reviewed, such as using our knowledge of Agrobacterium to produce a biocontrol agent and producing effective transformation vectors such as the pGreen binary vector. This website also includes a comprehensive list of references for anyone wishing to read further into any aspect of the research surrounding the bacterium. The study of transgenic or genetically modified plants is currently attracting significant interest because of its potential applications in industry, the environment, in the home and in medicine. Most genetic modification is perceived as being the insertion of foreign DNA into the genome of the target organism, and hence there is much debate as to whether this will have any unseen consequences in the future, and whether it is morally acceptable. There are also many problems associated with current methods of transgenic plant production, most notably the low transformation frequency encountered, or the failure of transgenic plants to grow at all. The use of Agrobacterium tumefaciens in transgenic plant production has arisen from the need to find an effective vector system to successfully integrate the gene of interest into the correct area of the plant genome. In recent years, the subject of uncontrolled cellular growth; Cancer, has been the focus of considerable attention as a result of its life threatening nature in humans. Extensive studies have unlocked information and provided a comprehensive knowledge base of undifferentaiated tissue growth in mammals, however little attention has been paid to cancer found in the plant kingdom. Agrobacterium tumefaciens causes one of the most common plant tumours, commonly known as Crown Gall disease which affects a wide variety of plants. Much research has taken place in an effort to understand the mechanism by which the bacterium infects the plant cell, in the hope that the bacterium

can be used as a vector system to transfer foreign DNA into a plant genome with a higher degree of success than is currently seen with traditional methods. Crown Gall disease is a common plant pathogen, affecting over 600 types of plants. Agrobacterium tumefaciens is a ubiquitous micro-organism that can be found in most soil samples. The disease mostly affects monocotyledonous species, such as woody and herbaceous plants and can be identified by the appearance of tumnours or galls of varying size and shape on the lower stem and main roots of the plant. Crown Gall disease can affect many commercially important and valuable crops such as Grapes, Rice and Sugar Beet.

Current Problems Plant disease is currently a major problem facing the developed world. Currently as much as 30% of the yearly total production of food crops is lost due to plant disease. At the current time, there is enough food produced to feed the population, but only just. If the predicted population increase over the next 2-3 decades take place, it will be necessary to increase food production to meet demand. As such a large proportion of crops are lost to plant pathogens each year, there is currently much interest in developing strategies to increase plants natural resistance to pathogenic attack. A. tumefaciens can remain dormant in the soil over winter, and can live saprophytically for many years. The diagram below gives some of the applications for plant transformation technology.

Characteristics of Agrobacterium tumefaciens Agrobacterium tumefaciens is a gram negative, motile, rod shaped bacterium which is non sporing, and is closely related to the Nfixing rhizobium bacteria which form root nodules on leguminous plants. The bacterium is surrounded by a small number of peritricious flagella. Virulent bacteria contain one or more large plasmids, one of which carries the genes for tumour induction and is known as the Ti (tumour inducing) plasmid. The Ti plasmid also contains the genes that determine the host range and the symptoms, which the infection will produce. Without this Ti plasmid, the bacterium is described as being non virulent and will not be able to cause disease on the plant. Crown Galls first appear as small, white, soft protrusions, initially found at the base of the plant stem. As the tumours enlarge, the surface takes on a mottled dark brown appearance due to the death and decay of the peripheral cells. The tumour usually appears either as a swelling of the plant tissue, or as a separate mass of tissue close to the plant surface, joined only by a narrow neck of tissue. Tumours can either be soft and spongy and may crumble on touch, or can be hard and appear knobbly or knotty. Some tumours can reach up to 30cm in diameter; though 5-10cm is more common (see cross section below). The tumours may rot away in the autumn, only to re-appear again the following spring. When infected with the bacterium, plants may also become stunted, produce small chlorotic leaves, and are more susceptible to extreme environmental conditions such as winter cold and wind. A. tumefaciens is most well known for its ability to integrate a small part of the Ti plasmid into the host plant genome, which causes the plant cells to become cancer cells and produce specific compounds called Opines, which the bacterium utilise as a carbon source. This property means that many textbooks class A. Tumefaciens as a genetic parasite, since the bacterium redirects the metabolic activities of the plant to produce compounds specific to the bacterium. It is this process which gives A. Tumefaciens its potential to be used as a tool for plant transformation.

Disease Control As A. tumefaciens is so ubiquitous in the soil, there is currently much interest in ways to try and prevent the bacterium from infecting susceptible plants, either through

biocontrol or through the use of pesticides. As the bacterium can only enter through wounds, the most logical solution would be to prevent the plants from becoming wounded. However, this is not practical in the long term as even wind abrasion can cause enough damage on the surface of the plant for the bacterium to enter. Agrobacterium radiobacter is a non-pathogenic, soil inhabiting bacterium which is related to A. tumefaciens. A. radiobacter produces the antibiotic agrocin, and for many years the K84 strain of this bacterium was used as a biocontrol agent against certain strains of A. tumefaciens. Seedlings of a plant were dipped into a suspension containing K84, which would then colonise the soil surrounding the plant, thus providing a degree of resistance against the gall. However, as the large plasmid of K84 contains a transfer region, there was a concern that the agrocin resistance gene could be transferred to A. tumefaciens, thus leading to agrocin resistance. A new strain, K1026 was thus engineered in the 1980's without this transfer region, and has been sold commercially as a biocontrol agent in Australia, and the USA, with no obvious detrimental effect on the rhizosphere. K1026 is deemed to be entirely safe in the environment, and cannot harm humans or animals as it will not grow at 37 Degrees Centigrade. Crown Gall can also be eradicated using creosote based chemical compounds, copper based solutions and strong oxidants such as sodium hypochlorite. However, these are costly to apply both in terms of labour rate, and the cost of buying the product. They are also very harmful to the surrounding environment, and accumulating large quantities of copper in the soil can have a disastrous impact on other plants in the area. Therefore chemical controls are rarely used against A. tumefaciens. Initial Research As Agrobacterium is such a common plant pathogen, it has been widely studied for many years. Records date back to 1897 when DelDott & Cavara first isolated a bacterium from tumours on infected grape plants. Smith & Townsend, (1907) were the first to discover that plants could be infected using a needle dipped in culture medium. This discovery led to the important conclusion that the bacterium requires a wound site in the plant in order for it to enter and induce a tumorous response. This is the reason why it is present in many soil samples, yet relatively few plants are affected. Jensen (1910), found that he could successfully graft tumours from sugar beet crop onto red beet. The tumours grew in the absence of the bacterium. Plant tumours don't normally kill the plant, but Crown Gall disease can be fatal to the plant if the tumours become too enlarged. When examined under a microscope, the tumours can be seen to develop very small shoots, and are classified as either Teratomas, or Teratomata.

Wood & Braun, (1942) •

• • •

•

Crown Galls have simple nutrient requirements; o Sucrose. o Salts. o Vitamins. Non-transformed tissues require auxins & cytokinins. Tumours produce specific compounds calles Opines, using an arginine precursor. Opines; o Nitrogen base. o Contain a Carbon source. o Some have a closed ring structure. o Several different types, most prevalent;  Octapine.  Nopaline. Opines therefore act as a Carbon and Nitrogen source for the bacterium.

Kado et al., (1984) • •

•

•

Discovered that Auxin (IAA) concentration is 200-500 times higher in Crown Galls than in ordinary plant tissues. Friable, unorganised tumours have; o 10 times more IAA than compact tumours. o 20 times more IAA than teratoma tumours. Nicotiana tabacum Crown Galls were found to contain 1,620 times more cytokinin than non-transformed plant tissues. Also noted that; o Accumulation of Potassium and Phosphorous is more rapid in the galls. o Octapine gall can utilise lactose as a Carbon source. Also found that the shoots produced on the surface of galls will actually grow and produce plants if they are grafted onto another rootstock.

Current Research A number of scientists are currently interested in being able to combine the best characteristics of A. tumefaciens mediated transformation (high efficiency, low copy number and stable integration) , with the species independant transformation characteristics found whilst using Particle Bombardment. Work by Hansen & Chilton (1996) has uncovered a novel approach for transformation, termed 'Agrolistic' transformation. This technology utilises three plasmids, two separately containing the restriction proteins VirD1 and VirD2, under control of a CaMV promotor. The third plasmid contained the target DNA, incorporating a Neomycin Phosphotransferase (NPTII) gene. The plasmids were then delivered to tobacco cells using direct DNA uptake methods. The theory behind this method is that if the VirD1 and VirD2 proteins could be expressed in the plant, they could nick the T-DNA at the border sequences, which the plant would then integrate into its own genome. Sequence analysis revealed that NPTII was indeed being expressed in the host cells, but also that the entire plasmid

had in some cases been integrated into the plant genome. The authors claimed that this technology could be used to transform any tissue that was successful to transformation by bombardment, and set out to show this by successfully transforming maize (Hansen et al., 1997). The T-DNA and the virulence (vir) region are two distinct regions of all Ti plasmids which are necessary for plant transformation by A. tumefaciens. The binary Ti vectors (able to replicate in Eschrichia coli and A. Tumefaciens) utilised by A. tumefaciens contain an antibiotic resistance gene for selection, whilst some of the more versatile Ti plasmids contain a series of lacZ restriction sites which allow complementation based screening for recombinant plasmids (Norrander et al., 1983). Hellens et al. (2000), set out to try and design a new binary Ti vector system which would overcome some of the current difficulties related to the current plasmids, such as low recombination frequency, interchangable selector markers and the difficulty experienced in transfering large fragments of DNA. The new vector developed by Hellens et al. (2000) was termed the pGreen Ti vector, and overcame the above difficulties by having a reduced plasmid size, transformation selection flexibility and an extensive multiple cloning site. The pGreen Ti vector should also be easily adaptable to future improvements in transformation technology. However, the pGreen vector is unable to replicate in Agrobacterium without the helper plasmid, pSoup, being present, thus improving the biological safety of the plasmid by preventing it from replicating throughout the gene pool. More detailed information can be found at the pGreen internet site. Further interesting discoveries have been made by Tzfira et al. (2002), who found that overexpression of Arabidopsis VIP1 in tobacco plants made them more susceptible to stable genetic transformation by A. tumefaciens, probably due to increased nuclear import of the T-DNA. Research into the mechanism of transfer between the bacterium and the host cell by Dumas et al., (2001) uncovered much information about the role of the VirE2 protein in forming membrane channels. VirE2 interacts with lipids to form a transmembrane channel with a high conductance value, which has been shown to be specific for ssDNA. However, there are still, at the time of writing, many questions which remain unanswered, such as how the pore can open to allow the T-DNA complex through, yet still retain the integrity of the plant cell. To date, much of the research taking place has focused on increasing the transformation efficiency, largely by concentrating on factors within the Agrobacterium cell itself. This has been achieved either by intriducing multiple copies of various vir genes, or by optimizing tissue culture and innoculation techniques (Tzfira et al., (2002). This proves that there is still a large scope for improvement of the new technology before we can fully appreciate its potential usefulness as a gene delivery system.

Mechanisms of Infection Sensing Cell Signals Agrobacterium Tumefaciens, and its related species A. rhizogenes, & A. vitris are the only known bacterial pathogens that invade plans by transferring their DNA to the plant, hence they have evolved as a major tool for plant genetic engineering. There are many different biological processes involved in this inter-kingdom DNA transfer, such as intercellular signalling, cell-to-cell DNA transport and DNA integration into the host cell nucleus (Tzfira & Citovsky, 2000). The genetic modification of the plant occurs as a result of the integration to the plant genome of a specific fragment of DNA, termed the T-DNA, or Transfer DNA, which originates from the bacterial Ti (tumor inducing) plasmid. Expression of these genes leads to the formation of Opines; specific oligosaccharides used purely by A. tumefaciens as a source of carbon. From then on, expression of several oncogenic (onc) genes leads to the formation of tumours (Gaudin et al., 1994). The entire process is regulated and controlled by a set of genes known as Vir genes, which are activated by the detection of wounded plant phenolic compounds. The transfer process can be divided into 2 steps; the bacterial cell step and the plant cell step (Tinland, 1996).

Vir Genes and their Function Vir Gene

Function Sense phenolic compounds from wounded plant cells and induce expression Vir A, Vir G of other virulence genes VirD2 Endonuclease; cuts T-DNA at right border to initiate T-strand synthesis Topiosomerase; Helps Vir D2 to recognise and cleave within the 25bp Vir D1 border sequence Covalently attaches to the 5I end of the T-strand, thus forming the Vir D2 T-DNA Complex. Also guides the T-DNA complex through the nuclear pores Binds to the 'overdrive' region to promote high efficiency T-strand Vir C Synthesis Binds to T-strand protecting it from nuclease attack, and intercalates Vir E2 with lipids to form channels in the plant membranes through which the T-complex passes Vir E1 Acts as a chaperone which stabilises Vir E2 in the Agrobacterium Assemble into a secretion system which spans the inner and outer bacterial Vir B & Vir membranes. Required for Export of the T-complex and Vir E2 into the D4 plant cell

Attachment & Penetration

The initial pre-penetration event in the soil rhizosphere is the conjugal transfer of the Ti plasmid, therefore increasing the number of pathogenic isolates in the soil. Quorum sensing proteins TraI and TraR induce the expression of genes required for bacterial cell mating and mobilisation of the plasmid. This response is also affected by opines produced by infected plants, which either suppress or activate a repressor of the TraR gene. The response is determined by the strain of A. tumefaciens present, and the opines being produced. A. tumefaciens posess swimming motility which is mediates by flagella. The precise mechanism controlling chemotaxis is so far unknown, but it is thought that migration occurs towards sugars and amino acids which accumulate around plant roots in the rhizosphere. Some strains may also be attracted to specific plant compounds released from wounded plants such as acetosyringone, and also to opines. Mutants which are non motile are still virulent (disease causing), but are unable to infect a plant unless directly innoculated, which suggests that motility is an important part of the infection process. Attachment to the plant is a two stage process, firstly involving a weak initial adhesion, then the bacteria synthesise cellulose fibrils which anchor them to the wounded plant cell surface. Some of the bacterial genes required for this process have been identified, namely chvA, chvB, pscA and att, as a mutation in any of these genes leaves the bacterium unable to attach to the plant. There are also molecules within the plant which are thought to be involved in the attachment process. One such molecule is vitronectin; an adhesive glycoprotein which is a component of the plant extracellular matrix (ECM). Vitronectin is more commonly associated with the cohesion of plant cells, thus having a role in plant structure and rigidity. The diagram below opens in a new window, and shows the mechanism by which the bacterium enters the plant and proleferates to form a gall.

B rial Step

acte Cell

The bacterial step involves all the processes leading to the production and export of the T-DNA complex, containing the information necessary to cause infection. The Tcomplex is made up of nucleoprotein; a single stranded T-DNA coated by VirE2 proteins, with a VirD2 protein attached at the 5I end (Howard & Citovsky, 1990). The VirD2 acts as a site specific endonuclease which recognises and cuts the left and right T-DNA borders with the VirD1. The VirE2 acts as a coat protein which envelopes the T-DNA, thus protectecting it from nuclease attack when it enters the plant cell (Tinland, 1996). These VirD2 and VirE2 proteins are also believed to have a role in targeting the T-DNA to the cell nucleus, once it has entered the plant cell. Once the T-DNA has been enveloped to prevent it from being degraded once outside the bacterial cell, it is referred to as the Tcomplex.

The T-complex requires a specific export system to deliver it across both the bacterial envelope and the plant cell membrane and into the plant cell cytoplasm. Tcomplex export occurs via a type IV secretion mechanism (right), comprising of a filamentous pilus and a transporter complex that translocates substances through the cell membranes (Salmond, 1994). In A. tumefaciens, the type IV secretion system is assembled from proteins encoded by the virD4 gene and the virB operon (Tzfira and Citovsky, 2000). Eleven VirB proteins are encoded by the VirB operon, all of which have a role to play in the transport of the T-complex across the membrane. VirB1 initiates the assembly, and VirB2 is the main structural protein in the pillus. The pilus was proposed by Zupan et al. (1998) to sense contact with a plant cell and relay this information back to the transporter complex and initiate the export of the T-complex.

Plant Cell Step It is believed that the T-DNA complex passes into the plant cell nucleus by active nuclear uptake, as the size of the T-DNA complex (12.6nm diameter) (Citovsky et al. 1997) exceeds that of the diameter of the nuclear pores (9nm) (Forbes, 1992), although the size of the nuclear pore can increase to 23nm during nuclear uptake (Forbes, 1992). Unlike other mobile genetic elements such as retroviruses, T-DNA does not encode

functions for transport and integration, therefore the DNA sequence is non-specific. It is this property of the T-DNA which makes it so useful; any DNA sequence inserted between the T-DNA borders will be transferred into the plant genome, enabling the efficient production of transgenic plants. Integration into the plant cell genome occurs essentially at random, a process which is believed to be controlled by the host factors. As a result of research into Arabidopsis mutants, a VIP1 basic zipper transcription factor has been identified which appears to facilitate the entry of the T-complex into the cell nucleus. A second protein, VIP2, acts to target the T-complex to transcriptionally active DNA. Once integrated into the plant genome, the auxin and cytokinin biosynthetic genes are expressed, resulting in uncontrolled proliferation and growth of the gall. Opine biosynthetic genes are also expressed, and these opines are used only by the gall as its sole carbon source, making it almost independent of the plant.

Sensing Plant Signals As A. tumefaciens is so ubiquitous in the soil, it needs a specific mechanism to enable it to assess whether a particular plant is suitable for infection or not. A. tumefaciens has adopted a two component regulatory system which is common in many prokaryotes, and involves the VirA and VirG proteins (Tzfira & Citovsky, 2000). VirA acts as a membrane sensor protein, whereas VirG regulates the cytoplasmic response to wounded plant cell phenolic compounds and promotes activation of all the Vir genes. VirG specifically interacts with the vir box; a conserved 12 base pair sequence located in the promotor sequence of all the vir genes. VirG is also able to induce its own expression, as it is produced from mRNA both in the presence and absence of plant phenolic compounds (Stachel & Zambryski, 1986). Compounds known to induce Vir gene expression include lignin, flavanoid precursors and acetosyringone (Stachel et al. 1985). The signalling pathway is initiated when wounded plant phenolics interact with VirA, which can either be through direct or indirect interaction, depending on the nature of the signal intercepted (Tzfira & Citovsky, 2000). Through mutational analysis of VirA, Tzfira and Citovsky (2000) concluded that during signal transduction, VirA functions both as a protein kinase and a phosphotransferase. Plant phenolics are known to be bacteriostatic at higher concentrations, a well known plant defense mechanism. To overcome this, a VirH protein is expressed as a result of VirG, which is believed to detoxify these harmful compounds. However, VirH expression will only occur after VirG has played its role in signalling during the infection process (Tzfira & Citovsky, 2000).

Agrobacterium tumefaciens - Uses

Usefulness as a gene delivery system Agrobacterium tumefaciens is seen as such a useful gene delivery system because it is able to carry any gene of interest within the T-complex, and insert the gene into the target plants DNA with a high degree of success. The reason for this is because unlike other mobile genetic elements such as transposons and retroviruses, the T-DNA strand does not encode functions required for movement and integration of the DNA. Therefore the T-DNA strand can be replaced by a gene of interest which will be inserted automatically into the host plant nucleus with a high degree of success and with little human intervention. This process is usually much more efficient than traditional methods of genetic modification. Follow the link below to the Purdue Agricultural Biotechnology Website, to see an animated demonstration of how A. tumefaciens can be used to genetically modify plants. A. tumefaciens mediated transformation is relatively efficient for many species, and a low copy number of intact, unmonified transgenes are frequently integrated successfully into the plant genome. However, transformation of many crop species has, in the past, been relatively ineficient, although recent advances in transformation technology is set to change that. There are however a few species of dicotyledonous plants and most species of monocotyledonous which are recalcitrant to transformation by A. tumefaciens. Ke et al. (2001) investigated whether synthesis at a high level of a T-strand DNA intermediate could improve the transformation efficiency of plants. It was found that a mutation in the gene regulator virG & VirGN54D when combined to produce a strain producing high level of T-strand DNA did indeed have a positive effect on the efficiency of transformation. Despite the many recent developments in the world of plant genetic manipulation, A. tumefaciens still remains a major method of choice for transforming plant cells, despite the development of sophisticated alternative gene transfer methods. Work is still ongoing to try and improve our understanding of the gene transfer mechanism. A number of economically important cereals have now been transformed using A. tumefaciens (Newell, 2000), working alongside other, more traditional gene transfer methods.

Applications of the Technology Agrobacterium tumefaciens may prove to be the breakthrough needed in order to successfully insert foreign DNA into plant genomes for genetic modification. Bacterial vectors such as Eschericia coli have already been used successfully as vectors in microbiology (Kikkert et al., 1999) ; this same technology can now be applied to the field of botany. Several different plant species have already been successfully transformed, including Lettuce (Curtis, 1995), Rice (Hiei, 1997) and Tomato (Tzfira et al., 2002). This proves that direct gene transfer methods are no longer the only avenue of approach for transforming important crop plants (Newell, 2000). One of the main reasons for favouring transformation by A. tumefaciens is that it allows delivery of a well defined

piece of DNA into the plant genome, although the success rate is not 100% (Gheysen et al., 1998). There are however some valid arguments against the validity of A. tumefaciens mediated transformation.

The Alternatives Other Factors Affecting Transformation There are many different approaches to plant genetic manipulation, some of which are more successful than others, and some have been superseded by more up to date methods. Two classes of plant transformation technology currently exist. These are non natural or in vitro methods, or natural methods. Non natural methods include microinjection and direct DNA uptake, whilst natural methods include technologies such as the use of viral vectors and A. tumefaciens. Some of the main methods of in vitroplant transformation are listed below.

Alternative Methods of Transformation with Examples Method of Transformation Species Used Agrobacterium Transformation Wounding with Glass Beads Sunflower Wounding by Bombardment Carnation Floral Dip Arabidopsis thaliana Germinating seed imbibition Arabidopsis thaliana Direct DNA Delivery PEG Fusion Tobacco Electroporation Carrot, Tobacco, Maize Microinjection Tobacco Gunpowder Charge Onion Electric Discharge Soybean, Cotton, Rice Compressed Air Gun Maize, Rice Other Methods Tissue Electroporation Pea Laser Rice Silicon Carbide Whiskers Chlamydomonas, Maize Adapted from Newell (2000).

Protoplasts

Reference Grayburn and Vick (1995) Zuker et al. (1999) Clough and Bent (1998) Feldman and Marks (1987) Krens et al. (1982) Fromm et al. (1985) Crossway et al. (1986) Klein et al. (1987) McCabe and Christou (1993) Oard (1993) Chowrira et al. (1995) Guo et al. (1995) Wang et al. (1995)

Plant cell protoplasts are simply a plant cell in which the outer cellulose cell wall has been removed, to leave a cell membrane layer that is easier to work with and is more amenable to cell fusion. Several methods have been successfully used to introduce DNA into plant protoplasts, such as fusion using PEG (polyethylene glycol), electroporation and microinjection. Although this method can and does work, regeneration of plants from protoplasts is a time consuming exercise and isn't always successful, as the process is very much a 'hit and miss' affair.

Particle Bombardment One method that avoids the host range restriction of agrobacterium and the regeneration problems when using protoplast techniques is the use of particle bombardment. The DNA is coated onto many small micro projectiles, which are then accelerated into the plant cells, usually either in culture or whole plant parts. This process is much simpler than agrobacterium mediated transformation as it avoids the complex TDNA transfer process thus eliminating the need for DNA sequences necessary for Tstrand replication and transfer. Follow the link below to the Purdue Agricultural Biotechnology Website, to see an animated demonstration of how particle bombardment can be used to genetically modify plants. There are many different methods of particle bombardment available, with many scientific labs adapting the same basic theory to suit their own particular needs and interests. Traditionally, DNA is coated onto inert particles such as gold and tungsten, but biological projectiles such as E.coli, yeast and phage have been complexed with tungsten and used as particles with some success (Kikkert et al., 1999). The method by which particles are accelerated into the plant also differs. Some devices use compressed air, others use a magnetic field to accelerate the particles, whilst some of the earlier methods actually used a controlled explosion for propulsion. Although this method has been shown to work successfully (Luthra et al. 1995) care must to be taken to ensure that the process does not irreparably damage the plant tissue. Therefore the technology is regarded as being relatively inefficient as relatively few numbers of cells are stably transformed (Gelvin, 1998).

Tissue Electroporation This technique uses an electrical pulse of high voltage but low amperage to create pores in the cell membrane through which DNA can pass. The cells must be in solution for this to take place, therefore no-one has actually seen the pores form so the knowledge is theoretical, but it does appear to work. This method was previously only used for protoplast transformation, but also works on whole cells, even tissues. This method has been successfully used to produce transgenic maize (D'Halluin et al., 1992) and transgenic legumes (Chowrira et al., 1995). However, the main drawback of this technology is that it depends on protoplasts being regenerated to form whole plants, a process which is difficult to carry out, and has a low success rate for many species (Gelvin, 1998).

Silicon Carbide Whiskers These are very small hollow silicon tubes, which are agitated in a solution of the target cells and a supply of DNA. The theory is that the whiskers will pierce the plant cell, thus allowing the DNA to enter. However, care must be taken when handling the whiskers as they are very small and are very similar to asbestos fibres. Whilst not widely used, they have been proven to work by Wang & Colleagues (1995). The risk of causing tissue damage however is great, so transformation efficiency is relatively low.

Direct Injection Direct injection of DNA has the potential to be species independent (Newell, 2000) and has been successful at producing transgenic plants. However, it is a very difficult process to carry out as it has to be done under a microscope, and the person carrying out the procedure needs to be very highly skilled with lots of experience. Therefore it is not a routinely used procedure as it involves large amounts of expenditure and effort for very limited returns (Jones-Villeneuve et al., 1995).

Other Factors Affecting the Efficiency of Plant Transformation Selectable markers allow plants that are expressing the new foreign DNA to be selected for quickly and efficiently. One of the most common marker system used is antibiotic resistance, whereby a gene for antibiotic resistance is also inserted into the plant genome, and the resultant plant cells are cultured in the presence of the antibiotic. The most widely used system for dicotyledonous plants is NPTII (Neomycin Phosphotransferase II) and kanamycin, although there are currently doubts as to the effects these proteins may have further down the line, especially where the consumer is concerned (Fuchs et al., 1993). Scorable markers have evolved from the early days of plant transformation to confirm whether or not transformation has really taken place. Recently, utillisation of the green fluorescent protein (GFP) from the jellyfish Aequorea Victoria has allowed visualisation of those plants expressing the new genes by excitation with light, without the need to supply any additional substrates to the plant (Sheen et al., 1995). Many improvements have also been made to the vectors which aid the integration of DNA into the plant genome during transformation. Some examples include pBECKS, an updated version of pBIN19, development of smaller, simpler vectors and the use of yeast (YAC) and bacterial (BAC) artificial chromosomes to increase the size of DNA fragments that can be integrated. Another suggested improvement is to couple the required transgene to a gene that would render the hybrid less able to compete and cross with wild species, thus reducing the spread of the transgenes into the gene pool (Gressel, 1999).

Conclusions

Since the identification of Agrobacterium tumefaciens as the causitive agent of Crown Gall disease, the interaction of this species with the host plant has been of great fascination to many botanists. However, it was nut until recently when it was apparant just how useful A. tumefaciens could be as a gene delivery system. During the last 15 years, improvements in biotechnology have come a long way since the realisation that plants can be genetically modified to give desirable phoentypic variations. Now that we are able to make transgenic plants, the main questions facing plant scientists are how to regulate gene expression, how can transformation be made more efficient and consistent, and perhaps most importantly, what are the environmental implications of this technology. There has been much speculation in the past as to whether the world is able to increase food production to cope with the projected population increase. In the past, this increase has been met by a combination of growers gain through breeding for highyielding varieties, and by improvements in growing technology and crop protection. However, the gap between food production and demand is narrowing much faster, and already there are civil wars breaking out between countries in Africa over the right to use land for food production. One of the main drawbacks of A. tumefaciens is its inability to effectively transform many monocotyledons, although current research by Ke et al. (2001) suggest that genetically engineered "supervirulent" strains may be effective in transforming many different plant species. In a study carried out in 1994 by Hiei et al., it was found that almost all of the transgenic Japonica rice plants had normal morphology, and 70% were fully fertile. Similar results were found when Indica varieties were also investigated. Delivery of foreign DNA into rice by A. tumefaciens is becoming standard practice in a growing number of laboratories, thus allowing the genetic improvement of many ovarieties of this fundamentally important crop plant. Important problems facing plant transformation which still remain to be solved include regulation of the DNA integration, and achieving the holy grail of plant transformation technology, that is targeted gene disruption and gene replacement hy homologous recombination. Recent reports of efficient targeting in Arabidopsis thaliana suggest that this breakthrough is closer than we might think (Gelvin, 1998). It seems probable that Agrobacterium mediated transfer techniques will soon be extended to other recalcitrant species of commercially important plants as soon as the methodologies are optimized.