Genetic control based on the sterile insect technique (SIT) uses the release of mass-reared and sterilized insects to cause infertile matings that reduce the pest population level. The concept of fighting insect pests by a repeated inundation with sterile individuals of its own kind was already promoted in the first half of last century.¹ The power of the technique lies in the simplicity of the biological principle on which it is founded and the lack of negative ecological effects following its application. Due to its species specificity, SIT is considered an environmentally friendly alternative to insecticides and has been successfully employed in area-wide approaches to suppress or eradicate pest insects like the pink bollworm Pectinophora gossypiella in California, the tsetse fly Glossina austeni in Zanzibar, the new world screwworm Cochliomyia hominivorax in North and Central America, and several tephritid fruit fly species in various regions of the world.

Additional programs indicate that these approaches would be valuable for a much more widespread use, but conventional sterilization by ionizing radiation decreases the competitiveness of sterilized insects. Thus, high quantities are required to inundate the pest population. This is a problem especially in lepidopteran pest species, where even highly irradiated males can still produce viable progeny. Moreover, SIT releases often require only males, but both sexes are needed for the rearing process. However, not only is it expensive to rear large numbers of potentially `useless' females, but it is detrimental to release any females, sterile or not, in the case of species that sting fruit with ovipositors or transmit diseases by biting domestic animals and humans. In addition, for the Mediterranean fruit fly (medfly), Ceratitis capitata, male-only releases have been shown to increase effectiveness of the SIT.

Recent advances in insect transgenesis² have encouraged the idea of transgenically manipulating pest insects in a way that will improve SIT approaches and widen its applicability. At least three different traits could be transgenetically introduced into insect strains to improve their use in the SIT: first, a marker gene could enable discrimination of released and naturally occurring insects; second, a female-specific lethality gene could allow for efficient genetic sexing; and third, a gene that causes lethality after transmission to the progeny could replace the irradiation procedure.

Transgenic marking. Discrimination between released sterile and wild insects is critical for monitoring the effectiveness of an ongoing SIT program. Currently, released insects are labeled with a fluorescent dye powder, which is expensive, labor intensive, and error-prone. The transgenic introduction of a fluorescent transformation marker, which does not compromise survival or fitness, would enable the identification of released insects in a simple way.

Sexing strains. In the medfly, separation of undesirable females has so far been based on genetic sexing strains that cannot be transferred to other species. Recently, transgene-based methods for sex-separation that are based on the female-specific expression of a conditional dominant lethal gene have been examined in the model insect Drosophila melanogaster and might be transferable to other insect pest species.^3,4 In both studies, conditionality of female lethality was established by the use of the binary expression system based on the tetracycline-controlled transactivator (tTA), which can be suppressed by supplementing the food with tetracycline during insect rearing.

Replacement of irradiation. Currently, ionizing radiation is used to cause the required genetic damage that results in the death of sired progeny and thereby causes sterility. However, in medflies irradiation reduces the mating competitiveness by approximately 30%, and many lepidopteran pest species require especially high doses of irradiation, which impairs fitness severely. In order to generate sterile but vigorous insects, my Ph.D. student, Carsten Horn, followed a strategy that interferes neither with the adult phase of the insect life cycle nor with gametogenesis. The sterility is based on the transmission of a transgene combination that causes dominant embryo-specific lethality.⁵ This allows for the generation of vigorous and potent sterile insects, with males being able to transfer competitive sperm. The embryonic lethality is caused by the expression of a lethal factor under the control of a promoter that is active at early blastoderm stages only. If the male has been homozygous for the transgene combination, each fertilization event will lead to embryonic lethality. The advantage of this approach lies in the proposed high competitiveness of such males, since none of their reproductive organs is affected and matings actually lead to sperm transfer. For this, it is very important that the promoter is active only in early stages of development. Then the lethal phase can be passed while growing up under permissive conditions in the rearing facilities, whereas after release, non-permissive conditions will not affect the males themselves but only their progeny.

To cause organismal lethality, the proapoptotic gene head involution defective (hid) was chosen as effector gene, which induces cell death when expressed ectopically. To avoid down-regulation of HID, the phospho-acceptor-site mutant allele hid^Ala5 was employed. In order to limit the detrimental effect of the transgenes to the embryo, enhancer/promoters of genes that encode structural components of the microfilament network specifically required for blastoderm cellularization were used, such as serendipity a and nullo, which are absolutely specific to the blastoderm stage but are expressed then at very high levels. To establish conditionality of the embryonic lethality we also employed the suppressible binary expression system based on the tetracycline-controlled transactivator tTA.

The system was successfully tested in D. melanogaster; the lethality occurred efficiently and was restricted to the embryo. The progeny died as embryos and only rarely (1:10,000) did a larva escape. This is important, because for many insects, the larvae cause most of the economic damage. In laboratory experiments, the competitiveness of the male flies was not greatly affected by the transgene combination when a nine-fold excess of sterile males was used in competitive matings. Correspondingly, the progeny was reduced by about 85% in this situation.

Since broad range transposon vectors and widely applicable transformation markers were employed, the examination of this system should be straightforward in pest species for which germ line transformation protocols have been established. Currently, my group is testing to see if the embryonic lethality system⁵ can be combined with a female lethality system,⁴ which would then make it possible to raise vigorous but sterile males only.

Nevertheless, before thinking of any applications involving the release of transgenic insects, great care must to be taken to employ as many safety mechanisms as possible to prevent an undesired spread of the transgenes. Laboratory studies will first have to assess transgene stability and fitness constraints in large populations. To achieve transgene stability, transposons should be non-autonomous and chosen so that no endogenous or related transposon activities are present in the species of choice. To further avoid rare cross-mobilization of the introduced transgenes, vectors that enable effective immobilization by deletion or rearrangement of transposon ends should be developed. Moreover, the introduced transgenes must not contain positively-selectable drug resistance markers.

For first evaluations of the environmental impact of transgenic insects, SIT programs will minimize the ecological concerns that the release of transgenic organisms might bring about. The sterility of the released insects will serve as a biological safety mechanism that impedes vertical transmission of the transgenes, which will be removed from the ecosystems with the cessation of the SIT program. Transgene constructs containing fluorescent transformation markers only will be suitable for first field trials, since they will improve SIT applications by simplifying the monitoring, but should not provide advantages to the carrier organism and actually allow the identification of carriers at later stages. This will minimize the risk of a rare but potential horizontal gene transfer. Given the understandably intense public scrutiny and the general lack of knowledge on potential risks, all projects that require the release of transgenic insects into the environment need to be planned with utmost care. Already at the initial stages of this methodology, molecular and population geneticists, entomologists, ecologists, as well as pest management specialists need to coordinate their efforts along with regulatory agencies to establish a safe use of the great potential transgenic insects have to offer.

References

1. Knipling EF. 1955. Possibilities of insect control or eradication through the use of sexually sterile males. J Econ Entomol 48: 459-462.

2. Wimmer EA. 2003. Applications of insect transgenesis. Nature Reviews Genetics 4, (March issue, in press).

4. Heinrich JC and Scott MJ. 2000. A repressible female-specific lethal genetic system for making transgenic insect strains suitable for a sterile-release program. Proc Natl Acad Sci USA 97: 8229-8232.

5. Horn C and Wimmer EA. 2003. A transgene-based embryo-specific lethality system for insect pest management. Nature Biotech 21: 64-70.

SILKWORMS PRODUCE HUMAN TYPE III PROCOLLAGEN
Eric A. Wong

The silkworm, Bombyx mori, has been treasured by the Chinese for centuries because of its value in producing silk fiber cocoons that are the basis for fine fabrics and textiles. Now the biomedical community may have a reason to also treasure these silkworms. In the January 2003 issue of Nature Biotechnology, a Japanese research group reported the generation of transgenic silkworms that produce cocoons containing recombinant human type III procollagen.

Collagen is used in many medical applications such as tissue engineering because of its strength and compatibility with tissues. The current main source of collagen is extraction from cow skin. Because of the potential risk of transferring viruses between species, an alternative source of collagen would be desirable. Thus a recombinant collagen would serve as an alternative non-mammalian source for this important therapeutic molecule.

The silkworm, which is actually a caterpillar, is the larva of the common, domesticated silkworm moth. Silkworms possess a pair of modified salivary glands called silk glands, which secrete a clear, viscous fluid through spinnerets located on the mouthparts of the larva. This fluid hardens as it comes into contact with air. The resultant silk fiber is used to spin a cocoon around the larva. One cocoon is made of a single thread about 900 meters long, and about 3000 cocoons are needed to make a pound of silk.

The generation of transgenic silkworms requires an efficient method for transferring foreign genes into silkworms. The piggyBac transposon, which was originally discovered in the lepidopteran Trichoplusia ni, can mediate stable germline transformation in B. mori. Because collagen is a long molecule, a truncated version of the type III procollagen gene was used. A fusion gene consisting of the procollagen mini gene fused to fibroin light chain, a major protein component of silk, and enhanced green fluorescent protein (EGFP) was injected into pre-blastodermal silkworm embryos. In addition, a red fluorescent protein gene was coinjected as a marker gene to allow rapid identification of transgenic and non-transgenic worms. The overall rate of successful transgenesis was approximately 25%.

Southern blot analysis confirmed that the transgene was integrated into the B. mori chromosomes. The majority of the insertions were single insertions, but double and triple insertions were seen at a frequency of 21% and 2.2%, respectively. The fibroin-procollagen-EGFP fusion protein was synthesized specifically in the silk glands and secreted into cocoons. Both the silk gland and cocoons showed fluorescence due to the presence of EGFP.

Photo courtesy Paolo Mazzei, http://www.leps.it

Fusion proteins were extracted from cocoons and analyzed biochemically. The fusion protein concentration of one transgenic line was 36.7 µg/mg of total extracted protein or 8.4 µg/mg of dried cocoon. As expected, the fusion protein could be cut with collagenase, which recognizes specific cleavage sites within the collagen molecule. The recombinant protein was readily purified to homogeneity in a single step method because the cocoons are greater than 95% protein and consist of only a few different proteins.

Silkworm cocoons represent an excellent source of recombinant protein because the silk gland is highly active in protein synthesis and the protein composition of cocoons is simple, which facilitates protein purification. The strain of silkworm used in these studies produces about 70 mg of dry weight protein per cocoon. According to the authors' calculations, a scale-up facility of 300 square meters with five workers could care for 1.5 million silkworms. These worms would produce a total of about 600 kg of cocoon, which would translate into 5 kg of collagen.

This paper demonstrates the potential application of silkworm biotechnology for production of therapeutic proteins. These results appear very promising and it will be interesting to see if other proteins can be similarly produced with this system. Perhaps someday one could just go to a pharmacist and ask for a few meters of a human therapeutic protein from a spool.

Reference

Tomita M et al. 2003. Transgenic silkworms produce recombinant human type III procollagen in cocoons. Nature Biotech 21: 52-56.

Milk consists mainly of carbohydrates, fats, and proteins. The caseins, which make up about 80% of the milk proteins, consist of four proteins, two forms of alpha casein (alpha s1 and alpha s2), beta casein, and kappa casein. Most casein proteins are present in structures known as casein micelles, which are important in determining the physical properties of milk. Beta casein, which is the most abundant milk protein, is involved in binding calcium phosphate and thus controlling milk calcium levels. Kappa casein plays an important role in determining the cheese-making properties of milk. Rennet or chymosin cleavage of kappa casein results in the aggregation of micelles and the coagulation of milk proteins. Therefore, altering the casein content of milk would lead to not only a change in the protein content of milk but also changes in the physico-chemical properties of milk.

Transgenic cattle, which contain multiple copies of the beta and kappa casein genes, were generated by the techniques of gene transfer and cloning by nuclear transfer. One of the advantages of nuclear transfer is that the sex of the transgenic offspring can be predetermined. In this case, female bovine fibroblast cells were cotransfected with DNA fragments encoding the beta casein and kappa casein genes to ensure the generation of transgenic cows. Four doubly transgenic cell lines, containing both the beta and kappa casein genes, were isolated and examined for the correct chromosome number prior to use as nuclear donors. Overall, nine percent (11/126) of the transferred embryos derived from the doubly transgenic cells produced viable calves at weaning; whereas 20% (7/35) of the non-transfected bovine fibroblasts gave rise to viable calves. This 50% decrease in efficiency is presumably due to damage incurred during the transfection and selection steps.

Expression of the casein transgenes was examined in hormonally-induced milk from the cloned transgenic calves. Because beta and kappa caseins are post-translationally modified, a number of beta and kappa casein isoforms were detected by two-dimensional gel electrophoresis. In two of the highly expressing transgenic cows, total milk protein was increased (13–20%) and total milk casein was increased (17–35%) compared to non-transgenic control cows. These high-expressing cows also showed an 8–20% increase in beta casein and a two-fold increase in kappa casein levels, the latter of which resulted in a marked increase in the ratio of kappa casein to total casein. Expression levels of the other milk proteins, such as alpha caseins and beta lactoglobulin were unchanged in the transgenic cow milk.

This report demonstrates that the total protein and total casein content of milk can be altered by overexpression of specific milk protein genes. This is an important point to consider because it shows that the synthesis of mammary proteins is not at 100% capacity, otherwise the overexpression of milk proteins would come at the expense of synthesis of endogenous milk proteins. Now that the composition of the milk has been altered, it will be interesting to determine if these changes in milk protein content have a nutritional or economical benefit to the dairy industry.

Reference

Brophy et al. 2003. Cloned transgenic cattle produce milk with higher levels of beta-casein and kappa-casein. Nature Biotech 21: 157-162.

In general, plants respond to environmental stresses (drought, excessive salinity, and low temperature) through a wide variety of biochemical and physiological adaptive changes, such as the accumulation of compatible solutes (glycine betaine, proline, polyamines, and trehalose) and synthesis of many regulatory proteins. One such compound is trehalose, a non-reducing disaccharide of glucose, which plays an important role in stress protection in a large variety of organisms ranging from bacteria and fungi to invertebrate animals. For example, brine shrimp eggs, commercially marketed as fish food or as pet "sea monkeys," can remain dehydrated for years in a state of suspended animation due to their trehalose content. Trehalose also acts as a storage carbohydrate, and it possesses the unique feature of reversible water absorption capacity to protect biological structures from damage during desiccation. When water dissipates from the shell of macromolecules (such as protein) during severe dehydration, trehalose can act as a water substitute on the surface of the dried protein.¹ Thus, the native folding and biological activity of proteins are maintained, and denaturation and aggregation are prevented. These protective properties of trehalose are clearly superior to those of other sugars, such as sucrose, making trehalose an ideal stress protectant.

Despite the wide distribution of trehalose in microorganisms and invertebrates, trehalose had until recently only been found in a few plant species, notably highly desiccation-tolerant, resurrection plants [club mosses Selaginella lepidophylla and the angiosperm Myrothamnus flabellifolius], so named because of their unique ability to fully recover from a state of almost complete loss of water. These resurrection plants can accumulate trehalose at levels approaching 1% of dry weight under non-stress conditions, whereas the majority of plants do not appear to accumulate easily-detectable amounts of trehalose. However, genes that encode enzymes of trehalose synthesis, i.e., trehalose-6-phosphate synthase (TPS) and trehalose-6-phosphate phosphatase (TPP) (Figure 1), have been recently identified in a number of plants. This suggests that the ability to synthesize low amounts of endogenous trehalose may be widely distributed in the plant kingdom.

Figure 1. Trehalose synthesis and degradation pathway in bacteria and plants

Recently, several research groups have attempted to study, via genetic engineering, the role of trehalose in abiotic stress protection and carbohydrate metabolism in plants. In all previous studies, engineering constitutive overexpression of TPS- and/or TPP-encoding genes from yeast or Escherichia coli in tobacco or potato resulted in enhanced trehalose levels and drought tolerance. However, constitutive overexpression of these genes also leads to unfavorable developments, such as stunted plant growth, lancelet leaves, and altered roots, as well as changes in carbohydrate metabolism under normal growth conditions.^2,3,4,5

Recently, we reported an alternate strategy to engineer increased trehalose accumulation in rice in such a way that trehalose synthesis occurs only when there is abiotic stress. We used a stress-inducible promoter to drive the overexpression of Escherichia coli trehalose biosynthetic genes (otsA and otsB) as a fusion gene (TPSP) for providing abiotic stress tolerance in rice.⁶ The TPSP fusion gene⁷ has the dual advantages of necessitating only a single transformation event to introduce both genes simultaneously into the rice genome, while at the same time increasing the catalytic efficiency for trehalose formation by the bifunctional enzyme. We introduced these two genes, which are responsible for the synthesis of trehalose, into an important variety of rice plant (Pusa Basmati 1) by Agrobacterium-mediated gene transfer and created a large number of transgenic rice plants that are completely fertile and grow well under normal growth conditions.⁶

The genetically-engineered rice plants produced higher amounts of trehalose. Importantly, since our custom-designed inducible promoter that drives the fusion gene is expressed only under stress, the plants grow normally without any undesirable effects. This is in contrast to previous experiments in which researchers have constitutively expressed an individual TPS or TPP gene so that it is turned on all the time, which stunts the growth of plants.

Furthermore, the transgenic rice plants exhibited sustained plant growth, less photo-oxidative damage, and more favorable mineral balance under salt, drought, and low-temperature stress conditions as compared to non-transgenic plants, many of which died due to salt stress. Depending on growth conditions, the transgenic rice plants accumulate trehalose at levels 3 to 10 times that of the non-transgenic controls. The observation that peak trehalose levels remain well below 1 mg/g fresh leaf or root weight indicates that the primary effect of trehalose is not just serving as a compatible solute. Rather, increased trehalose accumulation correlates with higher soluble carbohydrate levels and an elevated capacity for photosynthesis under both stress and non-stress conditions, consistent with a suggested role in modulating sugar sensing and carbohydrate metabolism. These findings demonstrate the feasibility of engineering rice for increased tolerance of abiotic stress and enhanced productivity through stress-dependent or tissue-specific overproduction of trehalose.⁶

Our laboratory has experimented with six other genes, each of which provides some degree of stress tolerance. What is special about the two genes responsible for the overproduction of trehalose is that the degree of protection from stresses has been much higher than with the other genes reported previously.

In conclusion, we have demonstrated that engineering trehalose overproduction in rice can be achieved by stress-inducible or tissue-specific expression of a bifunctional TPSP fusion gene without any detrimental effect on plant growth or grain yield. During abiotic stress, transgenic plants accumulated increased amounts of trehalose and showed high levels of tolerance to salt, drought, and low-temperature stresses, as compared to non-transgenic plants. These results demonstrate the potential use of our transgenic approach in developing new rice cultivars with increased abiotic stress tolerance and enhanced rice productivity. In principle, this same technique can be used to confer stress tolerance on other high-value, sensitive crops such as wheat and corn.

References
1. Crowe JH, Hoekstra FA, Crowe LM. 1992. Anhydrobiosis. Annu Rev Physiol 54:579-599.

2. Goddijn OJ et al. 1997. Inhibition of trehalase activity enhances trehalose accumulation in transgenic plants. Plant Physiol 113:181-190.

3. Holmstrom KO et al. 1996. Drought tolerance in tobacco. Nature 379:683-684.

4. Pilon-Smits EAH et al. 1998. Trehalose-producing transgenic tobacco plants show improved growth performance under drought stress. J Plant Physiol 152: 525-532.

5. Romero C et al. 1997. Expression of the yeast trehalose-6-phosphate synthase gene in transgenic tobacco plants: pleiotropic phenotypes include drought tolerance. Planta 201: 293-297.

6. Garg AK et al. 2002. Trehalose accumulation in rice plants confers high tolerance levels to different abiotic stresses. Proc Natl Acad Sci USA 99: 15898-15903.

7. Seo HS et al. 2000. Characterization of a bifunctional enzyme fusion of trehalose-6-phosphate synthetase and trehalose-6-phosphate phosphatase of Escherichia coli. Appl Environ Microbiol 66: 2484-2490.

SALT TOLERANCE TRANSFERRED FROM WILD TRITICEAE SPECIES INTO WHEAT
Zan-Min Hu

Abiotic stress is a major limiting factor in agricultural crop production in many countries. The major abiotic stresses of economic importance include drought, cold, heat, salinity, soil mineral deficiency, and toxicity. Salinity is usually exacerbated by intensive irrigation. In the USA, yield reductions because of soil salinity occur on an estimated 30% of all irrigated land. In China, 52.3 million hectares were cultivated under irrigation in 1998. Yield reduction due to salinity occurred on 40% of the irrigated areas. World wide, crop production is affected by salinity on nearly 50% of all irrigated land. Approximately 10 million hectares are not being used for agricultural production due to salinity problems.

Although some wheat cultivars and landraces are considered moderately tolerant to high salinity, they are still less salt-tolerant than many wild Triticeae species, especially those in the genus Thinopyrum. There is a long-term effort by plant scientists attempting to transfer salt tolerance from Thinopyrum species into wheat.^1,2,3 A milestone has recently been achieved in this pursuit by scientists at USDA-ARS Forage & Range Research Laboratory, Logan, Utah, assisted by several visiting scientists from China as well as scientists at both USDA-ARS George E. Brown, Jr. Salinity Laboratory, Riverside, California and the International Maize and Wheat Improvement Center (CIMMYT), Mexico.⁴

The researchers first screened and identified a wheat addition line, AJDAj5, as having the highest salt tolerance among 11 wheat addition lines developed by a French scientist. AJDAj5 carries all 21 pairs of wheat chromosomes plus a pair of wheatgrass chromosomes from the species Thinopyrum junceum, which is well known for its salt tolerance. AJDAj5 is still less tolerant to salinity than wheat-wheatgrass partial amphidiploids, which have seven or more pairs of wheatgrass chromosomes in addition to the full complement of wheat chromosomes. This is because salinity tolerance is a complex quantitative trait, which is controlled by genes on several wheatgrass chromosomes that exert additive gene effects.

The researchers then crossed AJDAj5 with a unique wheat stock carrying a gene, Ph^I, which inhibits the action of another gene, Ph1b. When Ph1b was not inhibited, wheatgrass chromosomes would not come in close contact with wheat chromosomes; thus there would not be any genetic exchanges between them. The researchers used Ph^I to enhance the occurrence of genetic recombination in the hybrid between the wheatgrass chromosome from AJDAj5 and its counterpart wheat chromosomes from the Ph inhibitor line.

Indeed, the researchers subsequently identified four putative recombinant lines from 30 hybrid-derived families after screening them in high saline solutions. One of the four was totally sterile and did not set seed. The other three were confirmed to be recombinant wheat lines having only 42 wheat chromosomes, one pair of which had a tiny wheatgrass chromosome segment inserted in the middle of an arm. This type of gene transfer makes these recombinant lines more useful, because the smaller the alien chromosome segment, the lesser the chance that they carry deleterious traits from the alien species.

Figure 1. (Top left) In contrast to the stunted and dead Chinese Spring plant, salt-tolerant recombinant wheat germplasm line W4910 survived and produced seeds after the treatment of a salt solution at EC=42 dS/m. (Top right) An interphase mitotic nucleus of the salt-tolerant recombinant wheat germplasm line W4910 showing greenish-yellow hybridization signals for the introgression of Thinopyrum junceum chromatin into wheat chromosomes. (Bottom) After a salt screening at EC=36 dS/m, all plants of 'Yecora Rojo' were dying whereas F2 plants of the cross Yecora Rojo X W4910 were segregating into salt-tolerant and salt-sensitive plants.

The salinity tolerance in these three recombinant lines was subsequently confirmed in a greenhouse test conducted at the Salinity Laboratory in Riverside, CA. It was also substantiated by a field test in Mexico conducted by collaborators at CIMMYT. Based on the results of Riverside and Logan tests, the researchers concluded that two of the three recombinant lines have received genes for salinity tolerance from both parents while the third one has a gene from AJDAj5 only. As a result, those two lines showing higher salinity tolerance than both parents were released as germplasm lines W4909 and W4910. They have been registered with Crop Science Society of America (CSSA).⁵ Because of their potential usefulness in further increasing salinity tolerance of moderately salt-tolerant wheat cultivars, these released and registered germplasm lines have been requested by and distributed to wheat breeders in United States and many foreign countries.

The researchers are planning to map, and ultimately isolate, the inserted genes from AJDAj5 and the Ph inhibitor line to determine if these genes are QTL for salt tolerance. They also plan to study the mechanisms of salt tolerance in these new germplasm lines to develop efficient selection methodologies for breeding programs using the best wheat cultivars from different countries as parents in crosses.

These germplasm lines, developed through traditional "chromosome engineering," are now available for molecular studies to create new biotechnology tools.

References

1. Dewey DR. 1960. Salt tolerance of twenty-five strains of Agropyron Agron J 52: 631-635.

2. McGuire PE and Dvorak J. 1981. High salt-tolerance potential in wheatgrasses. Crop Sci 21: 702-705.

3. Forster BP, Miller TE and Law CN. 1988. Salt tolerance of two wheat Agropyron junceum disomic addition lines. Genome 30: 559-564.

4. Wang RR-C et al. 2003. Development of salinity-tolerant wheat recombinant lines from a wheat disomic addition line carrying a Thinopyrum junceum chromosome. Int J Plant Sci 164: 25-33.

5. Wang RR-C et al. 2003. Registration of W4909 and W4910 bread wheat germplasm lines with high salinity tolerance. Crop Sci 43: (in press)

TRANSFORMING WHEAT: RIDING A SUPERBINARY VEHICLE ON THE POLYAMINE ROUTE
Harjeet K. Khanna

Wheat is one of the world's major food crops. Numerous laboratories have attempted to develop protocols for transforming wheat using the microprojectile-mediated and, more recently, the Agrobacterium tumefaciens-mediated approach. However, the recovery rate of transformants using either approach is quite low, mainly because regeneration of green plantlets from transformed wheat calli, which following selction are at least 8–10 weeks old, is inefficient. In the case of the Agrobacterium-mediated method, an additional limitation is the low frequency of T-DNA transfer from ordinary binary plasmids, even if they are present in the supervirulent strains. While working on a Grain Research Development Corporation project on developing frost-resistant wheat at the University of Southern Queensland, Australia, our group addressed both these issues and devised an improved protocol for transforming wheat.

Improving regeneration from aging wheat calli
Selection of the transformed tissue following initial DNA transfer demands that the cultures be maintained in a callus phase for at least 8–10 weeks; however, the loss of morphogenetic potential with age prevents high frequency recovery of transformants from the transformed calli. Although immature embryos and primary calli derived from immature embryos still remain the best explant source for transformation of wheat thus far, the loss of regeneration potential has been a limitation in most wheat cultivars over a period of callus initiation. Therefore, developing an efficient green plant regeneration protocol from secondary and tertiary level calli is an important issue for developing an optimal transformation system for wheat.

With this objective in mind, we developed a high-efficiency, plant regeneration protocol for some elite Australian wheat cultivars. The effect of water stress as well as cellular polyamines, especially the ratio of putrescine to spermidine for wheat regeneration, was analyzed. The importance of cellular polyamines and ratio of putrescine to spermidine for plant regeneration ability in several commercially grown rice varieties had been reported previously.^1,4 We decided to study wheat regeneration in light of this information.

We examined the effects of dehydration stress as well as exogenous application of spermidine (over a broad range of concentrations) in isolation and in combination with dehydration stress on regeneration from 4–week old scutellar calli induced from immature embryos. At the four-week stage, the wheat calli are white to off-white, compact, and morphogenic with well-formed somatic embryos that differentiate into green shoots. With age, the calli turn yellowish and soften, and the organized structures dedifferentiate into a loose mass. By the time most calli reach the 20-week age, they turn yellowish brown and non-morphogenic, and some even start rooting. This study helped us develop an improved regeneration protocol for these aging wheat calli. We found that regeneration from older wheat calli could be significantly improved by subjecting the cultures to water stress before transferring them to regeneration medium and supplementing the regeneration medium with optimal levels of spermidine. The optimum levels of spermidine varied with cultivars but were generally in the range of 0.1 to1 mM. While spermidine at all tested concentrations had a negative effect on regeneration from four week old calli as compared to the untreated controls, there was a 3% to 50% improvement in the regeneration ability of the older calli (8–20 week old) of all the cultivars. Finally, application of exogenous spermidine at 1 mM concentration to 8–16 week old cultures in combination with 16 hour dehydration stress significantly improved plant regeneration from 10% to 65% in all the cultivars tested.³

Enhancing the transformation frequency
With a good regeneration system in hand, we moved on to the transformation process itself. We already had a microprojectile-mediated protocol for transforming wheat, which could produce transformants at the frequency of 0.01–0.1%. However, this was not good enough for producing large number of transgenics, considering the short season during which field-grown immature embryos are generally available. Previous studies had shown that field-grown donor plants provided the best transformable embryos. That necessitated development of an efficient, high frequency transformation protocol. For this, we turned our attention to Agrobacterium tumefaciens. The objective was to develop an efficient protocol for Agro-bacterium-mediated transformation of wheat cultivars that are difficult to transform using ordinary binary vectors.

Although a few wheat cultivars respond to agro-infection with ordinary binary vectors, most wheat cultivars still do not show a high degree of susceptibility to them. We thought of trying the superbinary vector approach. Superbinary vectors² have played an important role in the development of high efficiency transformation protocols for monocotyledons like rice, maize, and sorghum. For wheat cultivars that are difficult to infect with Agrobacterium strains carrying ordinary binary vectors, a superbinary vector offering an extra set of virulence genes from Agrobacterium strain A281 was expected to work successfully.

Since there were two limiting steps, agroinfection and regeneration from the transformed calli, two approaches were used to develop the transformation protocol: (i) using the superbinary vectors for superior agroinfection; and (ii) supplementing the regeneration media with polyamine spermidine to improve recovery of regenerants from transformed calli.

A binary vector and a superbinary vector, each carrying the visual marker gene GUS and the selectable marker gene bar, conferring resistance to herbicide Basta^®, were used to infect immature embryo-derived calli of wheat cultivar Veery5. Cultivar Veery5 is capable of producing good quality calli from immature embryos within two weeks, and this rapidly growing tissue was highly susceptible to agroinfection by Agrobacterium strain LBA4404 carrying either the superbinary vector or the binary vector. However, despite a comparable level of transient GUS expression following infection with both the binary and the superbinary vectors, only the superbinary-infected calli were found stably transformed with the GUS gene after a seven-week selection period. In the case of the super-binary-infected calli, 13% of the initial infected calli (showing 35–62% transient GUS expression) showed stable GUS expression following selection. Since we had started with 2-week-old calli, by this time the transformed calli were 9–10 weeks old. Glufosinate-ammonium was used for selecting the transformed calli. To compare the effects of polyamine supplementation of media on regeneration of selected calli, half the selected calli from each set were transferred to spermidine-supplemented regeneration media and the other half to regeneration media without spermidine.

Final analysis of the results revealed that while no transformants could be recovered from 587 immature embryo-derived calli of cultivar Veery5 infected by Agrobacterium strain LBA4404 carrying an ordinary binary vector, 17 stably transformed plants could be generated from 658 immature embryo-derived calli of the same cultivar following infection with the same Agrobacterium strain LBA4404 carrying the superbinary vector pHK21. Of these, 13 were regenerated from 329 calli on polyamine-supplemented regeneration media and four from the remaining 329 on the non-supplemented media. So while the superbinary approach alone could yield transformants at the frequency of 1.2%, modification of the regeneration media with polyamine spermidine could raise the transformation frequency to 3.9%.

One advantage of using the Agrobacterium-mediated approach, apart from better transformation frequency, is that unlike the case of microprojectile-mediated DNA delivery, the vector backbone carrying bacterial antibiotic markers and other unnecessary vector sequences are not present in most of the transformants, making them more environmentally friendly and acceptable. An analysis of the transformants generated in this study indicated absence of any backbone vector sequences and confirmed integration of the T–DNA into the wheat genome.

The transformed plants were grown to maturity in the glasshouse and were fertile and morphologically normal. Analysis of the next generation revealed that the transgenes were inherited as dominant loci and in most cases they segregated in Mendelian fashion. In 35% of the plant lines, segregation of 3:1 was observed for the marker gene, indicating the presence of a single locus–another advantage of the Agrobacterium-based DNA delivery system.

For testing the expression of the selectable marker gene, the plants were treated with 0.1% Basta^® (Hoechst Australia Ltd) and herbicide damage was observed ten days after application. Plants transformed with bar gene showed total resistance to Basta, whereas the non-transformed plants collapsed following application. This indicated that the genes were transferred intact and showed no silencing. T0 and T1 generation plants were also tested for reporter GUS gene expression. Fifteen of the plants had visible GUS gene activity in different tissues, though the young floral tissues and seeds showed relatively more intense expression. Absence of GUS expression in the remaining two plants could be due to silencing of the GUS gene.

This protocol has since been used for transferring many economically important genes into wheat and a good number of transformants have been produced using this method. Since most cereals show comparable responses to tissue culture and Agrobacterium-mediated transformation, this protocol can be extended to the transformation of other cereal cultivars with Agrobacterium, and the polyamine supplemented media can be used for recovering a higher number of transformants from transformed calli, in the case of species that show loss of regeneration with age.

References

1. Bajaj S, Rajam MV. 1995. Plant Cell Reports 14: 717-720.

2. Hiei Y., Ohta S., Komari T., Kumashiro T. 1994. The Plant Journal 6: 271-282.

3. Khanna HK and Daggard GE. 2001. Australian Journal of Plant Physiology 28: 1243-1247.

4. Shoeb F, Yadav JS, Bajaj S, Rajam MV. 2001. Plant Science 160: 1229-1235.