ABSTRACT

A system for the inducible inactivation of plant tissues based on a deacetylation of the non toxic phosphinothricin derivative N-acetyl-Phosphinothricin (N-ac-Pt) has been used to induce male sterility in transgenic plants. The induced cell death was achieved by the tapetum specific expression of the argE gene product (E.coli) which was identified to remove the acetyl group from the substrate resulting in the release of the herbicide phosphinothricin (glufosinate). Deacetylation was demonstrated in vivo by the incubation of transgenic plants with radioactively labeled N-ac-Pt prior to TLC of the crude protein extracts. Transgenic plants constitutively expressing the argE gene were constructed. No effect of the enzyme on plant growth and reproductivity could be traced.

The tissue specific expression of the chimaeric gene leads to a destruction of the tapetum cells only in the presence of N-ac-Pt, the developing anthers are empty thus creating male sterile plants. The untreated plants are completely fertile. The fact, that the male sterility is inducible by the application of N-ac-Pt makes the presence of a second gene to restore fertility in the F1 generation superfluous. The system presented here is therefore easier to handle and less time consuming then those formerly described since only the female partner of the crossing has to be transgenic. It was primarily developed in Nicotiana tabacum but proved to be also functional in Brassica napus.

Introduction

One of the most prominent tools of plant breeding is the production of F1 hybrid plants. Resulting from differences in parental gametes, F1 plants show increased vigor and productivity based on heterosis (hybrid vigor). Hybrid varieties are produced by controlled crosses between two distinct inbred lines, the optimal combination of which has been ascertained by extensive crosses. Since positive effects based on heterosis are only observed in the F1 generation, hybrid seeds always have to be produced by the breeder. Farmers demand a uniform genotype of the seeds, therefore self pollination of the seed producing plant has to be excluded. In order to prevent self-pollination in some plants like corn, tomato and cabbage, emasculation of the female line by hand is required, raising costs and labor expenses of seed production (for overview see Peacock 1990). The utilization of naturally occurring cytoplasmatic male-sterility (cms) is a more efficient approach. Mostly correlated with changes in the mitochodrial or chloroplast DNA, in some lines of distinct species like maize (e.g. Laughnan and Gabay-Laughnan, 1983), oilseed rape (e.g. Jarl et al.1988), rice (e.g. Kodowaki et al., 1988) and Beta beets (e.g. Hallden et al., 1988) the anthers do not develop viable pollen, consequently the flowers only contain fertile female reproductive organs making cross-pollination obligate. However, natural cms includes the presence of the appropriate cytoplasm, and the resulting uniformity may render the crop vulnerable to certain diseases as demonstrated with the T-cytoplasm of maize in the USA 1970 (e.g. Bosemark, 1979; Laughnan and Gabay-Laughnan, 1983). Accordingly, male sterile lines are not available for all crops used in agriculture. Cms results from an interaction of organelle genes with a certain nuclear phenotype (e.g. Leaver and Gray 1982). Similarly, fertility can be restored in the appropriate nuclear background, the restorer. Hence when creating F1 hybrids, the parental lines have to be selected carefully to obtain fertile and uniform plants in the field.

Nuclear male sterile mutants can be observed in more than 175 plant species and even more have been induced using mutagens. However, due to the lack of breeding lines that are homozygous for the male-sterility conferring mutation, populations with 100% male sterile plants can not be produced. Hence, these mutants are not used for hybrid seed production (Williams, 1995). Additionally, the molecular mechanism responsible for ms is still not clear and restorer genes are also required in the male parent (Kaul 1988).

Several approaches using recombinant DNA techniques to obtain ms in plants have been undertaken. Nuclear ms has been achieved in Petunia hybrida by depletion of flavonoid pigments in the anthers leading to the arrest of pollen maturation. This was due to an antisense (Van der Meer et al. 1992) or a cosuppression approach (Taylor and Jorgensen, 1992). In tobacco, ms has been induced by the expression of a Agrobacterium rhizogenes rolC-coding region fused to the CaMV35S promoter. Unfortunately, the ms phenotype was accompanied by several other phenotypic alterations in the transgenic plant (Schmülling et al., 1988). Restoration of fertility was achieved by the expression of an antisense RNA in the F1 hybrids (Schmülling et al., 1993). A promising system is described by Mariani et al. (1990). Here the barnase-gene is fused to the tapetum-specific TA29 promoter. The RNase leads to the degradation of the tapetum, which is essential for the pollen development, and consequently inhibits the maturing of pollen. In order to regain fertility in the offspring, the activity of the Barnase has to be inhibited by the gene product of the barstar-gene (Mariani et al., 1992). A major disadvantage of the systems described is the fact that both partners for crossing have to be transgenic. In addition to the amount of time needed for transformation, the genes have to be transferred to the optimal lines for agricultural use via selective breeding. Hence, a recombinant system needing only one transgenic partner would increase the breeders flexibility. Such conditional ms has already been demonstrated in two different approaches. O’Keefe et al. (1994) describe the expression of the cytochrom B450_SU1 in tapetal cells. The protein targeted to the chloroplast converts the sulfonyl urea compound R7402 into a 500 times more toxic form. However, possibly due to the rapid metabolism of R7402, male sterility is limited to the flowers in the appropriate developmental stage during application of the compound. Additionally, R7402 itself is toxic and starts to inhibit growth when applied in four times higher amounts than used for the induction of male sterility. Another system for the inducible destruction of tapetal cells was presented by Kriete et al. (1996). The tapetum specific expression of an enzyme (ArgE) that deacetylates the non toxic herbicide derivative N-acetyl-phosphinothricin led to the release of the active herbicide and thereby to a specific cell death in Nicotiana tabacum. While flowers treated with the inducer substance produced 100% hybrid seeds, untreated plants remained completely fertile. No reduction in yield or quality of the seeds could be detected (Kriete et al. 1996). The inducer N-ac-Pt has proven to be non toxic either for sterile or unsterile plants of a variety of different species. It is taken up by the plant, but is not metabolized by the plant for a period of more than 100 days. Additionally, it is transported preferentially to the top of the plant (Dröge et al. 1992, Dröge-Laser et al. 1994) As a pure substance, it can therefor be applied in large amounts during the early growing season and should be present in the tapetum by the time the first cells begin to express the deacetylase activity. Nevertheless, the system has never been tested in the field. The transfer of the system to the commercially important crop rape seed that can, in contrast to tobacco, be cultivated in the North of Germany without any problem should allow the necessary extensive field trails.

Results and discussion

In order to construct transgenic rape seed plants, the tapetum specific argE gene was introduced into the plasmid RE1 and transferred to the A. tumefaciens strain GV3101. Transformation of the summer rape seed variety Drakkar occurred according to the method of de Block et. al. (1989). 40 independent transgenic lines were analyzed. As proven by PCR, all transgenic lines selected on kanamycin containing media carried in addition to the nptII gene at least one copy of the argE gene. The transcription of the argE gene could be demonstrated using reverse transcriptase PCR in all lines analyzed.

Fig. 1 Using the deacetylase system to induce male sterility in transgenic rape seed, the breeder is free to use any pollen donor for the crossings.

The production of male sterile flowers was induced in the greenhouse. Flower buds were treated with 10 µl of a 5mg/l solution of D.L-N-ac-Pt. After 1-3 weeks the anthers were analyzed. Flowers directly adjacent to the point of application turn out to be necrotic. All flowers next to them produced male sterile anthers (see Fig. 2) while the flowers at the bottom of the inflorescence were still male fertile.

The die off observed for the buds directly treated with the inducer might be due to the presence of approximately 3% of Phosphinothricin which is sufficient to harm the tissue that does not express the deacetylase. The amount of Pt transported to the neighboring flowers is to low to affect the plant cell, only the tapetum cells expressing the deacetylase seem to be exposed to Pt concentrations high enough to kill the cells. This leads to the lack of Pollen production. The amount of N-ac Pt transported to flower further down at the bottom of the inflorescence is to low to lead to any visible effect. We therefore have to conclude, that in principle, the deacetylase system is functional in rape seed, nevertheless the Pt contamination of the inducer substance is to high to apply sufficient amounts of the inducer substance to prevent pollen formation in all flowers without affecting other tissues. In order to prove, that the contaminating substance is really Pt, the solution was applied to Pt-resistant transgenic tobacco plants (Broer et al. 1989). No bleaching of the tissue could be observed.

Fig.2. Flowers treated with N-ac-Pt do not produce visible pollen. C1-2: untreated anthers, M1-4: anthers of one flower treated with N-ac-Pt + Tween 80, U1-2: anthers treated with H2O + Tween 80

References

1. Bosemark NO (1979) Genetic poverty of the sugar beet in Europe. In: Zeven AC, Harten AM van (eds) Proc Conf Broadening Genet Base Crops. Pudoc, Wageningen pp29-35

2. Broer I, Arnold W, Wohlleben W, Pühler A (1989) The phosphinothricin-N-acetyltransferase gene as a selection marker for plant genetic engeneering. In Proceedings of the Braunschweig Symposium on Applied Plant Molecular Biology. Editor G. Galling Zentrale Stelle für Weiterbildung TU Braunschweig :240-246

3. De Block M, Bottermann J, Vandewiele M, Dockx J, Thoen C, Gossele V, Movva NR, Thompson C, Van Montagu M, Leemans J (1987) Engineering herbicide resistance in plants by expression of a detoxifying enzyme. EMBO J 6:2513-2518

4. Dröge W, Broer I, Pühler A (1992) Transgenic plants containing the phosphinothricin- N-acetyltransferase gene metabolize the herbicide L-phosphinothricin (glufosinate) differently from untransformed plants. PLANTA 187:142-151

5. Dröge-Laser W, Pühler A, Broer I; The metabolites of the herbicide L-phosphinothricin (glufosinate): identification, stability and mobility in transgenic, herbicide resistant ans untransformed plants. Plant Physiology 105:pp 159-166, (1994)

6. Hallden C, BryngelssonT, Bosemark NO (1988). Two new types of cytoplasmatic male sterility found in wild Beta beets.Theor Appl Genet 75:561-568

7. Jarl CI, Ljungberg UK, Bornmann CH (1988). Correction of chlorophyll-defective male-sterile winter oilseed rape (Brassica napus) through organell exchange:Characterisation of the chlorophyll deficiencys. Pysiologica Plantarum 72:505-510.

8. Kaul MLH (1988) Male sterility in higher plants. (Monographs on theoretical and applied genetics,vol 10) Springer Verlag Berlin.

9. Kodowaki K, Osumi T, Nemoto H, Harad K,shinjyo C.(1988) Mitochondrial DNA polymorphism in male sterile cytoplasm of rice. Theor Appl Genet 75:234-236.

10. Kriete G, Niehaus K, Perlick AM, Pühler A, Broer, I (1996) Male sterility in transgenic tobacco plants induced by tapetum-specific deacetylation of the externally applied non-toxic compound N-acetyl-L -Phosphinothricin. The Plant J. 9(6): 809-818

11. Laughnan JR, Gabay-Laughnan S (1983) Cytoplasmatic male sterility in maize. Ann Rev Genet 17:27-48

12. Lever CJ, Gray MW (1982) Mitochondrial geneome organization and expression in higher plants. Ann. Rev. Plant Physiol 33: 373-402

13. Mariani C, De Beuckeleer M, Truettner J, Leemans J, Goldberg RB (1990) Induction of male sterility in plants by a chimaric ribonuclease gene. Nature 347:737-741

14. Mariani C, Gossele V, De Beuckeleer M, DeBlock M, Goldberg RB, deGreef W, Leemans J, (1992) A chimaeric ribonuclease- inhibitor gene restores fertility to male sterile plants. Nature 357, 384-387

15. O’Keefe DP, Tepperman JM, Dean C, Leto KJ, Erbes DL, Odell JT (1994) Plant expression of a bacterial cytochrom P450 that catalyzes activation of a sulfonylurea pro-herbicide. Plant Physiol. 105:473-482

16. Peacock J (1990) Ways to pollen sterility. Nature 374:714-715

17. Schmülling T, Röhrig H, Pilz S, Walden R, Schell J (1993) Restoration of fertility by antisense RNA in genetically engeneered male sterile Tobacco Plants. Mol Gen Genet237:385-394

18. Schmülling T, Schell J, Spena A (1988) Single genes from Agrobacterium rhizogenes influence plant developement. EMBO J 7:2621-2629.

19. Taylor LP, Jorgensen R (1992) Conditional male fertility in chalcone synthetase-dficient petunia J. Hered. 83, 11-17

20. Van de Meer IM, Stam ME, van Tunen AJ, Mol JNM, Stuitje AR (1992) Antisense inhibition of flavonoid biosynthsis in petunia anthers results in male sterility. Plant Cell 4: 252-263