1. Misra S, Wu Y, Venkataraman G, Sopory S, Tuteja N. (2007) Plant J. (Doi: 10.1111/j.1365-313X.2007.03169.x).

2. Oldham WM, Hamm EH. (2006) Q Rev Biophys 1-50

3. Venkataraman G, Goswami M, Tuteja N, Reddy MK, Sopory SK. (2003) Mol Genet Genomics. 270, 378-386

4. Kim HY, Cote GG, Crain RC. (1996) Plant. 198, 279-287

5. Munnik T, Arisz SA, Vrije TD, Musgrave A. (1995) Plant Cell. 7, 2197-2210

WHAT CONTROLS VITAMIN C LEVELS IN PLANTS?
William Laing and Sean Bulley

HortResearch scientists from New Zealand have identified the last remaining unknown enzyme in the major pathway of vitamin C (ascorbic acid) biosynthesis in plants and have shown that this enzyme controls the level of vitamin C in leaves. The results have been published as a cover story in the Proceedings of the National Academy of Sciences (USA),¹ with an accompanying commentary from Professor Jim Giovannoni of Cornell.²

In a near-simultaneous report, scientists from UCLA, working from a different approach based on their experience in the nematode worm Caenorhabditis elegans, have also reported the discovery of this enzyme.³ Both approaches relied on bioinformatic analysis of the predicted protein sequence of a previously identified gene that causes low vitamin C in a mutant of the model plant Arabidopsis thaliana. The enzyme transfers GMP from GDP-L-galactose to an acceptor, producing L-galactose-1-phosphate (see figure). The reaction represents the first totally committed step in the pathway of ascorbic acid biosynthesis, as compounds higher in the pathway (e.g., GDP-D-mannose and GDP-L-galactose) are used in other pathways by plant cells. The acceptor is identified as phosphate (a ‘phosphorylase’) by one group³ and as a sugar-phosphate (a 'transferase'; although the enzyme did transfer GMP to phosphate, but at a lower rate) by the HortResearch scientists.¹ It will be referred to as a transferase in this report. The HortResearch report also showed that transient transformation of tobacco leaves with this gene resulted in a three-fold increase in ascorbic acid levels compared to controls.

The three postulated pathways of ascorbic acid biosynthesis. Only the L-galactose pathway has been confirmed through mutant studies and identification of all the enzyme steps. However, some overexpression studies have provided evidence of other pathways.⁵ Blue arrows and writing represent side pathways utilizing intermediate compounds.

Vitamin C in Plants
Plants, especially leafy vegetables and fruit, are the major source of vitamin C for humans, as we are unable to make our own. However, many traditional fruits such as apples, bananas, and tomatoes are relatively low in this vitamin, having less than 20 mg/100 g fresh weight fruit of vitamin C. Thus, one fruit would provide less than half the recommended US daily intake of 90 mg/day for an adult male. Even an orange would only provide about this amount of vitamin C. Other non-governmental authorities propose much higher intakes of vitamin C than that recommended by the US government.⁴

Kiwifruit are well known for their high vitamin C content, with green kiwifruit (Actinidia deliciosa) containing ~80 mg/100 g fruit fresh weight and yellow kiwifruit (A. chinensis) having ~100 mg/100 g. Thus one 90 g kiwifruit would provide close to the recommended US daily intake. However, there are other species of kiwifruit, such as the less palatable A. eriantha and A. latifolia, with over 800 mg/100 g of this vitamin, so huge genetic variation exists within this genus. Consequently, Actinidia is an excellent genus for studying the control of vitamin C biosynthesis. This work has been enhanced by the large germplasm collections of Actinidia species at HortResearch and in China, as well as HortResearch’s large collection of expressed sequence tags (ESTs) and their associated physical DNA clones.

The pathway to Vitamin C
The major pathway of vitamin C biosynthesis in plants is thought to be through L-galactose,⁵ although other pathways also have been postulated, e.g., through the oxidation of the hexose sugar myo-inositol to glucuronate, and through galacturonate (see figure). These latter two compounds are important constituents in cell walls. The L-galactose pathway was only put on a sound footing in 1998 with the discovery of the enzyme L-galactose dehydrogenase⁶, and since that time, all the other enzymes in the pathway have been identified.^1,3,5

Mutations in genes in the L-galactose pathway reduce ascorbic acid levels in whole plants.⁵ However, in the PNAS paper¹ it was reported for the first time that overexpression of a gene for a committed enzyme in this pathway can increase ascorbic acid levels in leaves. Furthermore, fruit of Actinidia eriantha with high levels of ascorbic acid (800 mg/100g) also have significantly increased levels of the gene for the transferase enzyme (unpublished results).

Overseas work using genetic transformation to modify other proposed pathways of vitamin C biosynthesis has resulted in increased levels of ascorbic acid.⁵ For example, transfer of a gene for the enzyme myo-inositol oxygenase increased ascorbate in Arabidopsis, probably through increased flux through the glucuronate pathway. In strawberry, expression of the enzyme D-galacturonate reductase also increased ascorbate, presumably through the galacturonate pathway (see reference 5 for details).

How to increase plant vitamin C
The worldwide aim in Vitamin C research is to increase vitamin C in fruits and leaves, and knowledge of the biochemistry of the pathways will allow us to provide more natural vitamin C for people’s diets. Furthermore, high vitamin C in leaves may also make the plants more stress resistant, although this has yet to be proved. Certainly reducing vitamin C increases susceptibility to stresses, as this is how low vitamin C mutants were screened.⁷

HortResearch scientists are aiming to produce fruits other than kiwifruit with much higher vitamin C content, using either traditional breeding or genetic engineering. Gene mapping of kiwifruit and apple is aimed at finding markers for high vitamin C in mapping populations, and then using this information to select for fruit with high levels of vitamin C. The transferase is expected to provide a significant lead in identifying high vitamin C markers. This map-based approach is critical in perennial crops such as tree fruit because of long generation times. In this way, it is hoped that high vitamin C apples, for example, may be selected, as most varieties of apple have significantly less than 20 mg/100 g vitamin C.⁴ Using genetic engineering to increase levels in fruit with low vitamin C is also possible (provided regulatory, IP, and consumer issues can be addressed), and it is expected that overexpression of the transferase will increase vitamin C in low vitamin C fruit, as it did in tobacco leaves.

Because of its association with scurvy and long sea voyages, vitamin C was one of the first recognized nutritional/vitamin deficiencies.⁴ However, ascorbic acid was not identified until the 20th century, and today, because of publicity from people such as Nobel Laureate Linus Pauling, vitamin C has strong public recognition. Many people know it is important to maintain a good intake of vitamin C each day, and some may supplement their intake with vitamin C from tablets. There are even people who believe that very high intakes of vitamin C (e.g., greater than 1 g per day) prevent various diseases such as cancer.

Vitamin C in such high doses has not been shown to cause significant harm.⁴ Supplemental vitamin C is likely to be rapidly excreted, because the high levels taken in one pill overwhelm the kidney. Consequently, we consider that taking the same high amount of vitamin C in the form of a fruit may slow uptake and reduce excretion. In addition, we believe this fruit form of vitamin C may also be regarded as more "natural" for those not wanting to take pills, and as pointed out by Jim Giovannoni in his commentary, "strategies to manipulate crop plants for elevated vitamin C accumulation therefore will be important in both developing and developed nations."

Crop-to-crop gene flow has become a greater concern since the advent of genetically engineered crops, and opinions differ widely regarding the potential impacts of gene flow. Impacts on biodiversity, the spread of herbicide tolerance and insect resistance genes to non-target plants, and ecological impacts are some of the concerns. Gene flow can occur between species through hybridization or gene transfer from bacteria or viruses to new hosts. Some scientists concerned about the potential ecological risk of gene flow from GE to non-GE populations attempt to develop models that might mitigate this risk.

In 2006 the number of countries planting biotech crops increased from 21 to 22 when Slovenia planted Bt maize for the first time, bringing the total number of countries planting biotech crops in the EU to six out of 25. Second in abundance to soybean, transgenic maize occupies about 25.2 million hectares, which is around 25% of the global area of maize cultivation.

Various farm scale evaluations (FSEs) of GE corn have been conducted in the EU since 2000 in order to assess the effects of the release and management of herbicide tolerant crops on a range of weeds and invertebrates. One such study was conducted by researchers from the Central Science Laboratory, York, U.K, and the Centre for Ecology and Hydrology, Winfrith, U.K. They studied crop-to-crop gene flow based on the farm scale study sites of fodder maize (Zea mays L) in the UK.

The researchers’ objective was to study pollen mediated gene flow over a distance of up to 200 m from the source to evaluate whether the separation distances for maize stipulated by the Supply Chain Initiative on Modified Agricultural Crops (SCIMAC) are effective. A total of 1055 samples were collected from 55 sites over three years. They used a split field design: half of the field was sown with the Liberty Link^TM Line T₂₅ (containing the pat gene), which is tolerant to the Liberty^TM herbicide, and the other half was sown with an equivalent conventional maize variety.

Seed samples consisting of more than 1000 seeds collected from 3 – 5 cobs, each from a separate plant in the sampling location, were collected from 6 transects in the conventional crop plots. Along each transect, seed samples were collected from 2, 5, 10, 20 or 25, 50, and 150 m distances. DNA extraction and real-time PCR analyses were performed; the percentage of GE DNA was calculated by dividing the amount of GE target by endogenous control. A two-step maize gene flow model was constructed to fit the observed values. Estimates of whole-field mean percentage of GE DNA were made using a dynamic field model, and the 98^th percentile for each field was calculated from the distribution of the modeled percentage of GE DNA for each simulated field. The whole field model was repeated with increasing square field sizes, ranging from 50 × 50 m to 500 × 500 m.

Gene flow from GE to non-GE maize was observed up to 200 m from the source. The frequency of GE to non-GE gene flow decreased rapidly with increasing distance from the GE source. Separation distances did not correlate strongly with field size. Even a 25-fold increase in area reduced the separation distance only by 8 m for a 0.1% GE DNA threshold. The results showed that a field 150 m in length would require a 3 m separation distance for the crop to be below the 0.9% GE threshold. To be below a 0.1% threshold, a 150 m field would require a separation distance of 79 m. The results therefore suggested that the separation distances recommended by earlier researchers are larger than necessary.

The present study thus determined that for compliance with a 98% certainty, 81m is the longest required distance for a maize grain threshold of 0.1% GE in a small (50 m × 50 m) field. However, this model may vary when pollen flow occurs from several sources surrounding a non-GE recipient field. Thus the authors concluded that with effective modeling it is possible to meet current EU thresholds for bulk crops.