The gene-deletor technology
Traditionally, site-specific DNA recombination systems, such as bacterial phage Cre/loxP and Saccharomyces cerevisiae FLP/FRT systems, have been used to excise marker genes or short spacer sequences in higher plants¹. FLP/FRT and Cre/loxP function through interactions of a recombinase with its specific recognition sites. Recombinase recognizes these flanking recognition sites and excises any intervening DNA.

Making a case for transplastomic plants
In plants, genetic material is distributed between the nucleus, plastids, and mitochondria. Production of transgenic plants, for basic and applied purposes, has traditionally been mainly through expression of transgenes in the nucleus. Among the concerns raised about genetically engineered organisms is the fear that transgenes may "escape," for example via pollen in corn, "contaminating" non-transformed plants. Such concerns have given birth to a new field of transgene containment. Along these lines, plant biotechnologists have been keen on expressing transgenes in chloroplasts rather than the nucleus. The advantage of these transplastomic plants is that chloroplasts usually exhibit a strict maternal inheritance, and will therefore not be found in, say, pollen grains of corn. Genes expressed in the plastome will therefore not be transferred through pollination.

Chloroplast transformation also carries with it additional advantages. First, the high ploidy number of the plastid genome, the plastome, allows for high levels of expression. Diploid cells bear two copies of the nuclear genome, while they may contain hundreds of chloroplasts. Second, transgene integration in the chloroplast occurs via homologous recombination (site-specific integration), minimizing position effects.

David Wurbs, Stephanie Ruf, and Ralph Bock of the Max Plank Institute in Germany attempted to elevate levels of pro-Vitamin A in tomato via plastid metabolic engineering¹.

Engineering for increased carotenoids
Plant metabolic engineering offers many possibilities for increasing the nutritional characteristics of plants, as well as enhancing production of industrial plant products. Carotenoids, which are naturally occurring organic pigments in plants, can be grouped into two categories: xanthophylls (which contain oxygen) and carotenes (which are oxygen-free). β-carotene, also called pro-vitamin A, is the precursor to vitamin A and plays an important role in vision, growth, and development². It is therefore an important dietary component.

The major storage carotenoid in tomato is lycopene, an antioxidant that gives tomatoes their color. In the tomato fruit, lycopene accumulates in chromoplasts, a type of plastid that differentiates from chloroplasts during fruit ripening. Lycopene can be converted into β-carotene by the lycopene β-cyclase enzyme, as indicated in Figure 1 below. This enzyme is not present in humans and other animals, which explains why we are dependent on our diets for pro-vitamin A.

Figure 1. Conversion of lycopene to either β-carotene or ά-carotene. Modified from The Plant Journal 49, 276-288.

Wurbs et al. used the bacterial lycopene β-cyclase gene to elevate carotenoid biosynthesis in tomato and reported increases in β-carotene in ripe tomato fruit, a non-green plant tissue.

The construct
The basic chloroplast transformation vector pRB96 was used. This vector harbors a transgene expression cassette within the polylinker. The cassette consists of the promoter and 5’ UTR from the plastid atpI gene and the 3’ UTR of the rps16 gene.

Figure 2. Map of the vector and the insert (lycopene β-cyclase from Erwinia LCE).   Source: The Plant Journal 49, 276-288.

Achieving homoplasmy
Heteroplasmic situations (presence of wild type copies of chloroplast genome) are considered genetically unstable⁴. One way to achieve homoplasmy is to allow a sufficient number of cell divisions under selection. Ultimate confirmation of homoplasmy was done by seed assays, and the lack of segregation of antibiotic resistance in the T1 generation demonstrated homoplasmy and maternal inheritance of the transgene.¹

Analyzing transplastomic plants
When transforming plants to increase carotenoids, an increase of only the carotenoids is typically desired. This means that, other than the pathway being engineered, phenotypically the transgenics should resemble wild types. The researchers checked the phenotypes of the transplastomic plants and found they were similar to wild type controls, which is good. Next, they grew the plants in media containing 2-(4-chlorophenylthio)-triethylamine (CPTA), a specific inhibitor for the cyclization of lycopene. CPTA works by inhibiting the lycopene β-cyclase enzyme.

The rationale for the approach is that plants showing increased levels of lycopene β-cyclase (i.e., transplastomic plants) would exhibit increased tolerance to CPTA. Indeed, transplastomic plants tolerated CPTA, indicating that transplastomic expression of bacterial lycopene β-cyclase indeed elevates enzyme levels in the plastids of transformed plants. The increased amount of the enzyme also correlated with increased activity, and therefore increased production of β-carotene (visible as a color change of the fruit as shown in Figure 3 below). Fruits from transplastomic plants had a more orange pericarp and flesh, which could be diminished by adding CPTA

Figure 3. Phenotype of ripe tomato fruits harvested from transplastomic tomato plants expressing the lycopene-cyclase gene from Erwinia. Top row: Wild type; Middle row: Transplastomic tomatoes expressing lycopene b-cyclase; Bottom row: Fruits from transplastomic plants treated with 2.5 mM CPTA.  Source: The Plant Journal 49, 276-288.

Conclusions
This report by Wurbs et al. shows the extension of plastid transformation from the tobacco model plant to tomato, also an important crop plant. The demonstration of successful metabolic engineering via plastid expression of an important pathway is indeed a promising result. It also demonstrates the feasibility of using transplastomic technology for non-green tissues¹. Plastid transformation has also recently been achieved in trees³.

1. The Plant Journal (2007) 49, 276–288

GENETICALLY ENGINEERING FOLATE BIOFORTIFICATION IN TOMATO
Claudia Lerma-Ortiz

Tetrahydrofolate (THF) and its derivatives (folates) are indispensable for the health of humans and other organisms. They participate as cofactors in one carbon transfer reactions in the synthesis of glycine, serine, methionine, purines, and thymidylate^1,2. Humans are unable to synthesize folates. A deficiency of this vitamin can cause a range of serious diseases, including some birth defects (such as spina bifida), megaloblastic anemia, cardiovascular conditions, and some cancers³. Although folate deficiency is still a serious problem all over the world, it has the highest impact in developing countries, causing 200,000 severe birth defects every year.

To help prevent folate deficiency, some western countries, including the United States, have made the fortification of grain products with synthetic folic acid mandatory. Fortification helps people get the recommended allowance—400 µg/day for adults and 600 µg/day for pregnant women—in their regular diet. Unfortunately, in developing countries this kind of fortification presents a series of challenges—distribution inequities, recurrent costs, and lack of a food industry. Furthermore, fortification may itself cause some health problems. When a large amount of this vitamin is ingested, unreduced folic acid can enter into the systemic circulation⁴. Therefore, to achieve a greater margin of safety and lower costs, the consumption of common foods with a high content of natural THF would be a better solution.

Another consideration is that most folate in people's diets comes from plant foods. However, people regularly consume tubers, cereals, and fruits and not green leafy vegetables, which have the highest levels of this vitamin. For these reasons, a good strategy to improve folate intake worldwide would be to genetically engineer common food plants to make more folates. Hanson and Gregory’s laboratories (University of Florida’s Institute of Food and Agricultural Sciences) chose tomatoes (Solanum lycopersicum) for this experimental approach^2,4. Their work is reviewed here.

Biochemistry of folates
Molecularly, folate is comprised of three parts—a p-aminobenzoate (PABA) central motif linked via its carboxyl group to one or more glutamate residues and a pteridine ring bonded to PABA’s amino group. In plants, these molecules are produced in different parts of the cell. Pteridines are synthesized in the cytosol, whereas plastids produce PABA. These building blocks are then joined together inside mitochondria to give dihydropteroate (DHP), which is finally glutamylated to synthesize folates.

In nature, folates can have one-carbon units attached to their N₅ and/or N₁₀ positions at different oxidation levels. Some examples are the 5-formyl, 5-methyl, and 10-formyl derivatives. Alternatively, dihydrofolate (DHF) and other dihydro-derivatives may form by oxidation of the pteridine ring¹. Folic acid per se is absent in nature.

Pteridine overproduction Díaz de la Garza and colleagues initially targeted the first step in pteridine biosynthesis to engineer increased folate production. This reaction is catalyzed by guanosine triphosphate cyclohydrolase I (GCHI) (Fig. 1). Overexpression of this enzyme greatly increased the flux of pteridines in tomato fruit; however, the folate content was increased only 2 – 4 times. A limited supply of the other building blocks that constitute the folate molecule could be a limitation. Consistent with this idea was the observation that PABA pools in the engineered plants were very small⁴.

Figure 1. The tripartite folate biosynthesis pathway in plants. GTP, guanosine triphosphate. DHN, dihydroneopterin; -P, monophosphate; -PP, pyrophosphate -PPP, triphosphate; DHM, dihydromonapterin; HMDHP, hydroxymethyldihydropterin; all these pteridines are in purple. Aminodeoxychorismate (ADC) and p-aminobenzoate (PABA) are in blue. The red arrows represent the engineered GTP cyclohydrolase I (GCHI) and the ADC synthase (ADCS) enzymes.

Enhancement of p-aminobenzoate synthesis The limitation caused by small PABA pools in tomato was overcome by genetically engineering the first enzyme of PABA synthesis, aminodeoxychorismate synthase (ADCS) (Fig. 1). Tomatoes were transformed with the Arabidopsis ADCS coding sequence (AtADCS) (including its native chloroplast targeting peptide) under control of the fruit ripening-specific E8 promoter. RNA levels, measured at different ripening stages in the AtADCS⁺ transformants, were highest in fruit at the red stage. In these transformants, the level of accumulated PABA (56 nmol/g fresh weight) was 19-fold greater than in the empty vector controls. Although a marked increase in PABA was obtained in these transformants, folate production was the same in both controls and AtADCS⁺ tomatoes². This result was consistent with the earlier observation that folate synthesis is also dependent on pteridine production⁴.

Folate biofortification To obtain tomato transformants with both PABA and pteridine overproduction, Diaz de la Garza and colleagues crossed hemizygous T₀ AtADCS⁺ plants with one-insertion, hemizygous T₁ GCHI⁺ plants. The development of plants and fruits was the same for double transgenic and nontransgenic segregants. Genomic DNA from the progenies of plants carrying both transgenes was screened by PCR. Control and transgenic tomato plants were harvested at the late red ripe stage and analyzed for pteridine and PABA content. The double transgenic fruit had 30-fold higher pteridine and PABA levels than controls and produced 25 times more folate (25 nmol/g fresh weight). No DHP, pteroate, or hydroxymethyldihydropterin pyrophosphate (HMDHP-PP) were found. These data suggest the presence of an additional restriction on the synthetic pathway².

Other limitations on folate biofortification
Although the double transgenic tomatoes generated more folates, their pteridine and PABA pools were not depleted. On the other hand, DHP and DHF did not accumulate. These results indicate that there is a constraint in the folate production pathway at the level of DHP synthesis. Díaz de la Garza and colleagues (2007) think that the limitation lies either in the HMDHP- pyrophosphokinase step or in pteridine transport from cytosol to mitochondria.

Types of folate produced by the transformed tomatoes
5-methyl-THF was the main folate derivative found in double transgenics, pteridine overproducers, and wild type controls. DHF and folic acid were not detected. However, the monoglutamate form of 5-methyl-THF was the most abundant in GCHI⁺/AtADCS⁺ fruit, whereas 5-methyl-THF hexaglutamate was predominant in controls and fruit overexpressing GCHI⁺ alone². Because a high concentration of glutamate can be readily found in red stage tomatoes, the low polyglutamylation of folate was not due to a limitation in this amino acid’s availability. Probably, folates were produced in double transgenic tomatoes at a rate that exceeds the capacity of folylpolyglutamate synthase. This was also the case in Lactococcus lactis genetically engineered to overproduce folate⁵. The total folate content in double transgenic tomatoes increased throughout ripening, whereas it didn’t change in the nontransgenic controls. This was expected because the expression of both transgenes was driven by the E8 promoter. Furthermore, in the double transgenic fruit, all types of folate derivatives increased, with polyglutamylation reaching a plateau at red stage followed by accumulation only of monoglutamyl folates⁷.

Safety concerns
As mentioned above, the double transgenic tomatoes had 30-fold higher PABA levels than controls. This amounts to <1 mg of PABA/100g portion. The U.S. Food and Drug Administration classifies the PABA food additive as GRAS (generally recognized as safe) when consumed up to 30 mg/day. Moreover, PABA has been proved in clinical trials to have low toxicities when consumed in quantities up to 12 g/day. Therefore, Díaz de la Garza and colleagues (2007) think it may be considered safe to eat biofortified tomatoes with the low levels of PABA observed.

The double transgenic fruit also contained higher pteridine levels—0.36 mg per portion. Unfortunately, little is known about pteridine in plant foods or its effect in humans. The pericarp of velvet bean (a medicinal legume) contains 470 nmol/g fresh weight of total pteridine, which is 25 times higher than the pteridine content of GCHI⁺/AtADCS⁺ tomatoes. On the other hand, it is well known that mammals catabolize pteridines and synthesize the pteridine tetrahydrobiopterin. Moreover, the total content of pteridines in the double transgenic tomato is <10% of the pteridines excreted/day in human urine. Based on these observations, Díaz de la Garza and colleagues suggest that pteridine levels in double transgenic tomatoes are likely to be harmless; however, they also state that further research is needed to establish the safety of plant foods containing this level of pteridines (2007).

Applications in gas ripened tomatoes and other plant foods
A general procedure in the food industry is to harvest mature green tomatoes and store them. Ripening is induced later by ethylene gas treatment before sending the fruit to the market. Hanson and Gregory’s laboratories applied this gas-ripening procedure to mature green double transgenic tomatoes until they reached the breaker stage. Fruits were allowed to ripen for 12 days. Tomatoes of the same double transgenic plants ripened on the vine were used as controls. The ethylene-ripened fruits contained 79% of the average folate levels of the controls. Also the types of folate and polyglutamylation levels in the gas ripened tomatoes were very similar to those in the controls².

Is it feasible to apply the present strategy to fortify other plants with folates? Indeed, when an overexpression of GCHI was engineered in the small weed Arabidopsis, a three-fold increase in folates was obtained. Furthermore, when supplied with PABA, the transgenic plants accumulated 7 times more folate than the controls⁶. Thus, the technology for biofortification of tomatoes with folates has been well established in Hanson and Gregory’s laboratories, and the prospect of applying it to staple food plants such as tubers and cereal grains is very promising.

Conclusions
In the work reviewed, a marked increase in folate content in plants was demonstrated if both the pteridine and the PABA branches of the synthesis pathway were engineered. The experiments also advanced knowledge of the regulation of this pathway in plants.

The genetic engineering strategy employed was successful in the production of GCHI⁺/AtADCS⁺ transgenic tomato fruits producing 19-fold more folate than the controls. These folate levels are the highest thus far reported in plants. The double transgenic tomatoes contained 840 µg of folates/100 g fruit, which would provide the recommended dietary allowance for a pregnant woman in less than one standard serving. Furthermore, the size and appearance of the fruits did not change with the genetic transformation. Also, the folate content of double transgenic tomatoes was 7-fold higher than that of green leafy vegetables, which as mentioned before, are good folate sources in human diets.

The recent 2007 Biotechnology Industry Organization (BIO) International Convention held in Boston drew a record 22,366 attendees, a nearly 15 percent increase from the previous year. The Convention is the world’s largest event for the biotechnology industry. This year’s Convention drew representatives from 48 states and 64 countries, with one-third of attendees from outside the United States.

Keynote speakers were Queen Noor of Jordan and Michael J. Fox. Queen Noor discussed opportunities presented by biotechnology to address global health and poverty issues. Michael J. Fox, who founded the Michael J. Fox Foundation for Parkinson’s Research since being diagnosed with Parkinson’s disease in 1991, stressed the need for the biotechnology industry to continue to innovate and accelerate the translation of basic science into improved therapies for patients. The best drug for treating Parkinson’s has been around for 40 years, he pointed out. Fox likened treatment discovery to a high stakes poker game—it requires an investment, and sometimes we lose, sometimes we win, playing with the cards that are dealt. With biotechnology, the wild card may help win the game.

Agbiotech Goes Beyond Transgenics
Some people tend to think of biotechnology as only meaning "transgenic," when engineering transgenic traits into plants and animals is really just a small part of the biotech-driven revolution in agriculture. In reality, biotechnology also encompasses advances in genomics, bioinformatics, and molecular biology, and the bio industry needs to communicate the fact that more precise information about the genetic makeup of plants and animals (as well as plant and animal pests) can be used in non-transgenic ways to make better food, fuel, and fiber.

One panel of experts from the United States, New Zealand, and Australia described how information technology and agriculture are converging to create solutions in agriculture. "The convergence of information technology and molecular biology dramatically increases agriculture’s potential to supply fuel and animal feed as well as more nutritious food," said Sano Shimoda, president of California-based BioScience Securities.

Ray Riley, global head of corn and soybean product development for Syngenta, pointed out that agriculture based IT, such as gene sequencing and molecular markers technology, is increasingly becoming focused on consumer attributes "rather than just production by the pound." For example, source verification and DNA-based traceability is a cornerstone of food safety. And a better genetic understanding of cellulosic crops will help create more efficient biofuels.

Gerard Davis, CEO of Catapult Genetics (Australia/New Zealand), noted that global competition for grain drives research aimed at improving the efficiency of cattle and sheep in converting feed to meat, milk, and biofeedstuffs. He described how DNA analysis and genomics is supporting the targeted breeding of more efficient livestock.

Fruit breeding, too, is drawing on a suite of biotechnologies to create more novel, flavorful, and fragrant varieties with increased health promoting qualities. Gavin Ross, from New Zealand fruit developer HortResearch, pointed out that advances in germplasm IT, plant genomics, and flavor biotechnologies are generating tastier fruit varieties and new advances for the human health and flavor and fragrance industries, as well as ecologically better growing practices.

Opening Doors to Seed Banks
A key point made during the session titled 'Yours Mine or Ours; Who Owns the World’s Genetic Resources' was that the fear of big companies owning genetic seed resources is misaligned. In actuality, the problem is a hodge-podge of national policies that can inhibit access to plant genetic resources. Researchers and plant breeders frequently look to foreign sources of germplasm to find new traits for resistance to pest and diseases, yield improvement, and tolerance to environmental stresses. The emergence of new biotechnology-based plant breeding tools only heightens the importance of open access to germplasm, according to David Hegwood, USDA’s minister-counselor for agricultural affairs in Rome.

The International Treaty on Plant Genetic Resources for Food and Agriculture aims to improve international cooperation and the open exchange of crops and their genes that farmers all over the world have developed and exchanged over 10,000 years. Drafted in November 2001 after more than seven years of formal negotiations in the United Nations Food and Agriculture Organization, the ITPGRFA is the first effort to establish international rules governing access to genetic resources.