CA1327173C - Method of gene transfer into plants - Google Patents

Abstract

Abstract of the Disclosure Method for gene transfer into plants, which com-prises a) isolation, from anthers, of immature pollen grains in nutrient solution and removing the tissue in which-they are embedded, b) culturing the isolated, immature pollen grains in a nutrient solution c) transferring foreign genetic material into the pol-len grains during the in vitro culturing and saturation d) bringing about complete saturation of the transformed pollen grains in vitro e) pollinating receiver plants with the transformed pollen grains and obtaining seds from the former and transgenic seeds and propagation products therefrom.

Description

METHOD OF GENE TR1~NSFER INTO PLANTS

The invention relates to a method of gene transfer into plants by means of isolated, immature and in vitro-cultured pollen grains, seeds produced therefrom, and propagation products therefrom.
Various techniques for gene transfer into plants are already available. Each has its advantages and dis-advantages (Goodman et al, Science 236: 48-53, 1987).
Aarobacterium tumefaciens is the most common vector nowadays. However, it is restricted to a certain host range. The use of A. tumefaciens as a vector still depends on the regeneration of in vitro-cultured somatic cells to produce transgenic plants which are capable of forming transgenic progeny. Direct gene transfer by electroporation, liposomes, microinjection and other physical-chemical methods largely depends on the use of protoplasts as target cells. However, regeneration from protoplasts is difficult and even impossible in many species.
~ different target cell has been suggested as an alternative to these gene transfer methods, the pollen grain. According to D. Hess, Plenum Press, New York, 519-537 ~1975), it is possible for the mature pollen to take up foreign DNA and to transfer this foreign DNA to the egg cell, as a "supervector'l, by means of natural pollination and fertilization. Similarly, X-ray treated pollen is said to be capable of transporting fragments of the irradiated genome into the egg cell (Pandey, Nature 256: 310-313, 1975). DeWet (W0 85/01856) reports that maize pollen which has been treated with exogenous DNA takes up this DNA on germination and transports it into the egg cell after fertilization.
In spite of the potentially great importance of 1327173 ~-gene transfer into, and by, pollen, other laboratories have not succeeded so far to reproduce the results of Hess, Pandey and DeWet. Thus, for example, Engvild (Theor. Appl.
Genet 69: 457-461, 1985) was not able to reproduce Pandey's experiments. In the experiments by Hess (1975) and by DeWet, the uptake of exogenous DNA into the pollen grain and the genetic and molecular proof of the transferred DNA is the critical link in the chain of reasoning. Proof by phenotype and physiological proof were not sufficient (Hess in: Genetic manipulation in plant breeding, de Gruyter, Berlin, New York 8~3-811, 1986). The experiments by Sanford et al (Biotechnology and ecology of pollen (Mulcahy D, ed.), Springer, Heidelberg, New York, 71-75, 1968) also showed that co-cultivation of mature Nicotiana Lan~sdorfii pollen with agrobacteria did not lead to gene transfer into pollen. In a large number of experiments, Negrutiu, Heberle-Bors and Potrykus (loc. cit. 65-70, 1986) did not succeed in transferring the neomycinphosphotransferase gene, which entails resistance to kanamycin, into mature pollen ~Shillito et al, Biotechnology 3: 1099-1103, 1985). Thus, it was not possible to confirm the statements by Hess and DeWet by means of methods corresponding to the recent state of the art. The fact that the results of Hess and DeWet are not reproducible can be explained by the mature pollen grains immediately starting the formation of a pollen tube as soon as they are put into an aqueous medium, which is required for the gene transfer. Obviously they are no longer fertile at this stage, or the period of time which is available for gene transfer is too short.
In Planta 170: 141-143 (1987), Pareddy et al described the culture and maturation of immature maize tassels (male inflorescences) in vitro. Pollination was carried out using the isolated mature pollen, and seeds were obtained from the pollinated plants. No proof, by means of genetic markers, for successful fertilization could be furnished.
Surprisingly, it has been found that it is , possible to culture immature pollen grains without the natural nutrient tissue containing them, furthermore, that foreign genes can be introduced into these lsolated, immature pollen grains during various stages of maturation, and that, with mature pollen grains, plants can be pollinated and fertilized to bring about normal seed formation, germination and propagation.
The present invention thus relates to a method of -gene transfer into plants, which comprises ., a) isolating, from anthers, immature pollen grains in nutrient solution and removing the tissue in which they are embedded, b) culturing the isolated, immature pollen grains in a nutrient solution -~
c) transferring foreign genetic material into the pollen grains during the in vitro culturing and maturation d) bringing about complete maturation of the transformed pollen grains in vitro e) pollinating receiver plants with the transformed pollen grains and obtaining seeds from the former.
, A completely novel strategy of gene transfer into plants is made available by the gene transfer into isolated, immature pollen grains, in vitro maturation and the use of pollen as universal "supervectors". Compared with other methods, it is considerably simplified because the cell culture phase is reduced to a large extent and the regeneration, which is accompanied by the troublesome -phenomenon of somaclonal variation, is omitted. Using this novel method, it is also possible to transform plants which were inaccessbile to date to successful gene transfer: thus, for example, many species of cereals, leguminosae and trees cannot be regenerated from single cells or protoplasts.
The potential benefit from gene transfer technology has large economical, ecological and social value. The aim is to alter plants genetically such that ~:

yield is increased, that the plants become resistant to diseases and pests, that they become tolerant to cold, heat, drought, salinification and lack of nutrients, that they have better nutritional qualities, that they produce novel raw materials for industry, or that they fix their own nitrogen or become independent of fertilizers in another manner.
It is essential for the invention that pollen grains are cultured without the tissue around them, i.e.
isolated, so that the genetic material, coupled to a vector or naked, to be transferred has direct contact with the pollen grain. If the complete male inflorescence is cultured, it is not possible to transfer genes into the pollen grains with the aid of customary transfer methods (A.
tumefaciens, electroporation, microinjection and the like) due to obstruction by the extensive surrounding tissue.
Another essential feature of the invention is the use of immature pollen grains. Due to this, the complete period of in vitro maturation is available for gene transfer. Transfer into the immature pollen grain can be carried out in different stages of maturity, depending on the plants features. It is important that the genetic material has to pass through as few cell walls as possible to enter the genome of the sperm nuclei and become part of the zygotic genome on fertilization, in particular when vectors are used ~or gene transfer. Transfer is preferably carried out when the microspores are in their uninucleate stage, but it can also be carried out during the first pollen mitosis, in the early binucleate stage, as long as the generative cell is still attached to the pollen wall, or alternatively, in those plants whose pollen is trinucleate in the mature stage (cereals), in the stage briefly before or during the second pollen mitosis.
Specifically, the method is carried out as described below, tobacco (Nicotiana tabacum) being used as a model system. The system can be applied to all plants which ~.

can be propagated by pollination in particular to mono- and dicotyledon crop plants, such as wheat, maize, rice leguminosae, oil crops, vegetable plants, fruit plants and forest plants.
In a preliminary experiment, the development stage of the pollen of the desired plant is determined in a customary manner by isolation of one anther, preparation of a squash and observation under the microscope after the addition of, for example, acetocarmine.
When the pollen has reached the desired stage, the pollen grains are isolated under sterile conditions by squeezing them out of the anthers in a nutrient medium, passing them through a sieve, washing them, collecting them by centrifugation and resuspending them.
Cell density should be higher for pollen grains in an earlier development stage than for pollen grains in a later development stage. For tobacco, cell density of binucleate pollen grains is about 105/ml, and about twice as much in the case of uninucleate microspores.
The nutrient mediu~ used contains all the nutrients and the growth substances essential for culturing and maintaining the ability of the pollen grains to mature. Depending on the plant, various compositions may result, so that it fulfils the function of the nutritive tissue (Tapetum). Here, the main constituents of the nutrient medium are sugars, nutrient substances, mineral salts and vitamins, the pH being about 6.5 to 7.5.
In the case of uninucleate microspores, culturing up to the binucleate stage is carried out in a medium considerably enriched in sugar, for example sucrose, and further nutrient substances. For example, the additional nutrient substances may be added in the form of coconut water. When the pollen grains have reached the binucleate stage, they can be transferred to a medium less rich in nutrients.
, .

1327173 :

": ' When the in vitro maturation is complete, the pollen grains are harvested for pollination. For this purpose, they are centrifuged off, washed, and, for pollination, applied to flowers from which the anthers have previously been removed, either in an aqueous medium or dried. Seeds which are capable of germination are obtained in a generally customary manner from the plants pollinated with in vitro-matured pollen grains.
For gene transfer during in vitro culturing, the foreign genetic material to be transferred can be introduced into the pollen grains either using a customary vector, such as, for example, A. tumefaciens, or as naked DNA by direct transfer by means of electroporation, microinjection or other physical/chemical methods. The expression "foreign genetic material" denotes any genetic material produced outside the pollen grain to be transformed. When maturation is complete, the pollen grains are harvested in the abovementioned manner for pollination.
As proof of successful in vitro maturation, pollen grains of a tobacco plant having 2 marker genes (KANR, NOS) were matured in vitro, and normal wild type-plants were pollinated with these pollen grains. The seeds obtained were sterilized and germinated in a ~ana-mycin-containing germination medium. Mendelian segregation of the marker genes was confirmed by counting the seedlings in the selective medium. Seedlings which had been germinated on kanamycin-free medium, also showed medelian segregation of the NOS gene following nopaline detection by high-frequency paper electrophoresis.
This proves that it was the in vitro-matured pollen grains which carried out fertilization and not any contaminating pollen from other plants.
As proof for the expression of the foreign gene introduced by transformation during in vitro culturing, the chloramphenicolacetyltransferase activity (CAT activity) was detected in an enzyme assay in a homogenate ~repared from .. .

the transformed pollen grains.
In addition to the CAT gene, a cytochemically-detectable gene, the beta glucuronidase gene (GUS gene) was also used. In this experiment, the agrobacteria were co-cultured with the pollen and then removed or killed by daily washing with the antibiotic claforan. In a cytochemcial assay (staining blue of the pollen~ it was possible to detect more than 50% blue, in vitro-matured pollen. Non-infected pollen, or pollen infected with agrobacteria without or with a non-functional GUS gene did not show blue staining on the addition of the substrate. A fluorimetrical assay showed high GUS activity in GUS-transformed pollen, and also in pollen of GUS-transformed plants which served as a positive control. Pollen which had not been infected, or had been infected with agrobacteria free of the GUS gene, did not show GUS activity. Furthermore, infection with agrobacteria without virulence genes, but complete GUS-gene in the T-DNA did not lead to GUS-activity in the pollen (cyto-chemically and fluorimetrically).
This leads to the conclusion that gene transfer into the pollen was carried out by Aqrobacterium tumefaciens. Sucessful gene transfer into the pollen is furthermore suggested by the visible adhesion of agrobacteria to the pollen surface with formation of cellulose fibrils (electron microscope photos).
.,~....

-:

'.

Example 1: Determination of the stage of pollen development Tobacco flowers of various lengths were harvested, the anthers were isolated under sterile conditions, one of the five anthers was transferred to a slide together with a drop of acetocarmine (4% of carmine in 45% strength acetic acid), and a squash was prepared. The development stage of the pollen grains was determined under the microscope after half an hour. - -Example 2: Isolation of the pollen grains The tobacco pollen grains were isolated by --squashing the anthers carefully with a glass rod and a microscopy mortar in AMGLU medium (1), under sterile conditions, passing the resulting pollen suspension through a 7S um sieve and washing the pollen suspension twice with AMGLU medium. Finally, the pollen suspension was collected by centrifugation in an Eppendorf centrifuge at 6,5D0 rpm.
Example 3: Pollen culture for the maturation of early-binucleate pollen grains Early-binucleate tobacco pollen grains were isolated and the cell density in AMGLU medium was adjusted to 105/ml. one ml of the pollen suspension was cultured in 35 mm Petri dishes at 25C in the dark. Depending on the exact development stage of the pollen grains, they had matured after 2 to 5 days and could be harvested.
Example 4: Pollen culture for the maturation of uninucleate microspores ~
Uninucleate tobacco microspores were cultured in ~-MR24 medium (2) at a cell density of 2 X 105/ml. As soon as the pollen grains had reached the binucleate stage (after 2 to 5 days, depending on the exact age), they were collected by centrifugation and cultured further in MlS medium (3) until maturation was complete.
Alternatively, very good results were obtained when MR26 medium (2') was used for culturing. As soon as the pollen grains had reached the binucleate stage (after 3 . .

1327~3 g days), the same volume of M2S medium was added. After a further day, the pollen grains were collected by centrifugation and cultured further in M2S medium (3') at a cell density of 105/ml until maturation was complete (one day).
Example 5: Pollination and proof of fertilization The tobacco pollen grains were collected by centrifugation and washed in BK medium (Brewbaker and Kwack 1963), and the cell density was adjusted to 1.25 x 10~/ml.
The still closed anthers were removed from flowers which were just about to open ~red flower tip). As soon as the flower had opened, a 4 pl-drop of the pollen suspension was applied, with the aid of a 20 ,ul-pipette, to the stigma of the flower such that the stigma was covered completely in pollen suspension. These operations were carried out under conditions without movement of air and remote from other plants of the same species. As soon as the drop had dried on the stigma, stigma and pistil were covered with a 4 cm ;
piece of straw to prevent cross-pollination. The plants ~
were then returned to the greenhouse. ~ -Example 6: Harvest of the seeds, germination of the seeds and genetic assay As proof that it was the in vitro-matured pollen grains which had carried out the pollination and fertilization, pollen grains of a transgenic plant were matured in vitro and normal wild-type plants were pollinated ~-with these pollen grains. The transgenic plant contained the gene for neomycin-phospho-transferase (resistance to kanamycin) and the noplinsynthase gene as a marker gene.
Self-fertilizations and the reciprocal cross with in vitro-matured wild-type pollen were also carried out.
The mature seed capsules (brown and dry) obtained - -by Example 5 were harvested and the seeds were isolated by cutting off the tip of the capsule and transferring the seed grains directly into Eppendorf tube. The seeds were surface-sterilized for 5 min in a NaOCl solution (3% free ~
. :.
~ :, - 10 - .

chlorine), washed twice with sterile water and placed on a seed germination medium containing kanamycin (4). After four weeks, the number of kanR and kanS seedlings was counted. In the above crossing experiment, the two reciprocal crosses resulted in segregation of kanR:kan 1:1. Seedlings grown without kanamycin showed a segregation of Nos+:Nos- = 1:1 following a test for nopalin by high-frequency paper electrophoresis. As expected, a 0:1 segregation for both marker genes was observed for self-fertilization with in vitro-matured wild types. Self-fertilization with in vitro-matured pollen of the transgenic plant resulted in a 3:1 segregation.
Example 7: Transient expression of the CAT gene Agrobacteria (A tumefaciens without tumor genes, containing the CAT gene coupled to a 35S promoter) were pre-incubated for one day in Luria broth. Pollen grains in the early binucleate stage were isolated and cultured in AMGLU
medium. The bacterial suspension was adjusted to an OD580 of 0.2 using AMGLU medium. After a further dilution with AMGLU medium of 1:10, 20 ~l of the bacterial suspension were added to 1 ml of the pollen suspension. After 24 hours of co-culturing l ,ul of claforan (lg/2ml) per ml was added to destroy the agrobacteria. After a further two days, the pollen grains were harvested and an extract was prepared.
For this, 1.5 ml of calcium washing solution (5), pH 5.6, were added per ml of pollen suspension, the mixture was centrifuged for 5 min at 4,000 rpm and washed with a Tris buffer (2.25 M, pH 7.8). After centrifugation, 200 ul of Tris buffer were added to the pollen pellet, and -homogenization by ultrasound (3 x 15 sec) was carried out on ice. After 10 min on ice, the mixture was centrifuged and the supernatant was kept at -20C.
The CAT assay was carried out as described by Sleigh (Anal. Biochem., 56:251-256, 1986). 30 ~l of the e~tract were mixed with 20 ~l of chloramphenicol (8mM), 30 ,ul of Tris buffer and 20 ~ul of 14C-marked acetyl CoA (5 uCi/ml in 0.5 mM of cold acetyl CoA). Following incubation , , . , . . . . , . , . . . ~ , . . . . . . . . . ` . . . . . .

- .

13271 73 : :

for 1 hour at 37C, the acetylated chloramphenicol was extracted by shaking with ethyl acetate (2 x 100 ~1), and the radioactivity was measured in a scintillation counter.
Radioactivity (cpm) in the CAT assay of extracts from tobacco pollen following co-culturing with A. tumefaciens cpm Pollen in AMGLU medium 400 Pollen together with agrobacteria in AMGLU medium 6,000 Pollen together with acetosyringon-activated agrobacteria in AMGLU medium6,000 ~ -Only agrobacteria in AMGLU medium500 - -Only acetosyringon-activated agrobacteria in AMGLU medium 500 ':
The strong radiactivity signal shows that the agrobacteria has successfully infected the pollen grains and that the T-DNA must have entered the nucleus of the growing cell for expression (transcription).
As a further control to show that the agrobacteria themselves did not have any CAT activity, no CAT activity was detected in an agrobacteria culture in Luria broth over a complete growth cycle. ~ -~xample 8: Expression of the GUS gene in tobacco pollen ~-Agrobacteria of the strain LBA 4404 (A.
tumefaciens, disarmed, having the beta-glucuronidase (GUS) gene, coupled with 35S promoter and terminator, (Matzke and Matzke in Plant Molecular Biology 7 : 357-365 (19863) were pre-incubated for one day in Luria broth. Pollen grains in the late uninucleate stage were isolated and an OD580 of 0.2 ~-was adjusted in MR26 medium. Following a further dilution with MR26 medium of 1:10, 20 ~1 of the bacterial suspension were added to 1 ml of the pollen suspension. After 14-20 hours of co-culturing, the pollen was collected by centrifugation, washed three times with MR26 medium containing claforan (lg/2ml) and cultured further. The ;-medium containing claforan was changed every day until maturation of the pollen. 40 ~1 of the last medium (M2S, containing claforan) were added to a Luria broth. No bacterial growth was observed.
The mature pollen was collected by centrifugation, and some was taken up in MR26 medium. Following the addition of X-Glu (5-bromo-4-chloro-3-indolyl glucuronide) to a final concentration of 1 mM and incubation for 4-12 hours at 37C (Jefferson in Plant Molecular Biology Reporter, Vol. S, No. 4, 387-405, (1987)), the formation of indigo (product of the beta-glucuronidase from X-Glu) could be detected with the aid of the blue staining of the pollen under a light microscope. More than 50% of the live mature pollen showed blue staining.
A second portion of the co-cultured and in vitro-matured pollen was germinated in GK medium (like BK medium, but twice the concentration of boric acid). GUS activity could also be detected in the pollen tubes of the germinated pollen.
In a control experiment, pollen was infected with a strain of agrobacteria of LBA4404, which, in its T DNA, contained the GUS gene without the promoter (supplied by Drs. Matzke, Salzburg). This pollen did not stain blue after in vitro maturation and the addition of X-Glu, nor did agrobacterial containing no GUS gene, but the KAN35S gene after co-culturing with pollen and the addition of X-Glu.
Pollen which was not infected with agrobacteria did not stain blue either after the addition of X-Glu. Pollen of a GUS35S transformed plant ~obtained by leaf-disk transformation), that served as a positive control did stain blue. Another strain of agrobacteria, that contained a functional GUS35S gene in the T-DNA, but no Ti-plasmid (binary vector) did not show GUS-activity in the pollen.
An extract was prepared from a third portion of the pollen. For this purpose, portions of 4 x 105 pollen were collected by centrifugation and taken up in extraction . , . . , : .'! ' ~ ' .: ~ , . ' , . .; .

1~27173 buffer (6). The pollen was crushed open using glass beads and simultaneous ultrasound treatment. The fluorimetric GUS
assay was carried out as described by Jefferson. 50 ~l of extract were made up to 1 ml with extraction buffer, and MUG(4-methylumbelliferyl glucuronide) was added to a final . .
concentration of 1 mM. Aliquots of 200 ul were withdrawn every 10 or 30 sec, and the enzyme reaction was stopped using 800 ~l of Na2C03 (0.2 M). In a fluorimeter, the absorbence at 455 nm following excitation at 365 nm was measured, and the values were converted to concentrations in ~M/ml with the aid of MUG standard solutions. -~
: . ..
Enzyme activity of beta-lucuronidase from tobacco -pollen extracts after co-culturing with A. tumefaciens, measured by fluorimetry after 10, 20 and 30 minutes ~;
: .:
(in uM/ml): 10 min. 20 min 30 min ~ -"
Standard l 1,0 1,0 1,0 Standard 2 0,1 0,1 0,1 ;~
Pollen without agrobacteria 0,016 0,020 0,019 Pollen, co-cultured with ~ `
GUS35S agrobacteria 0,057 0,078 0,109 ~
Pollen, co-cultured with -KAN35S agrobacteria 0,023 0,019 0,021 Pollen, co-cultured with binary vector without ~-Ti-plasmid, but with GUS35S0,022 0,023 0,018 Pollen from a transgenic GUS+ plant 0,051 0,085 0,121 :. --, . .
Culture media: -(1) AMGLU medium Miller's macrosalts MS microsalts -~ .
Sucrose (0.25 N~
Glutamine (440 mg/l) `~
pH 7 --: "
., ,',;' 1327173 -:
- 13a -(2) MR24 medium MS macrosalts .
MS microsalts -;
Sucrose (0.5 M) ~ -Glutamine (440 mg/l) -Coconut water (2% by volume) `.
Lactalbumin hydrolysate (200 mg/l) Inositol (100 mg/l) -pH 7 ;

,, ' ' .

, .... ~ .

:X~ : :' `` - 14 - ~ ~
; ~2') MR26 ~ediu~ ~ ~-Like MR2b ne~ium, but lactalbumi~ hydrolys~te ~ 1 g / l ) ~ .

(3) M15 medium Mi~ler's macrosalts MS microsalts FeEDTA (10 4 M) Sucrose (0.25 M) pH 7 (3') M2S medium Kyo and Harada's salts (in Planta 186: 427-. ~ . . .
432 (1986)) Sucrose (0.25 M) --;
pH 7 ;~

(4) Seed germination medium ~ ~
MS macrosalts ; :-MS microsa~ts feEDTA (10 4 M) `; ~-: - . . .
Sucrose (1X by ~eight) `~ ~
Agar (0.8Z by ~eight) - ` ^
Kanamycin S04 (50 mgll) ---- -pH 5.5 ~ -~
S5) Calcium ~ashing solution ;
CaClz x Z~2 (0.16 M) MES buffer (O.SX by ~eight) pH 5.6 (6) Extraction buffer ;~
NaP04 S50 mM, pH 7) 2-mercaptoethanol (10 ~M) Na2EDTA (10 mM) Sodium lauroyL sarcosinate (0.1X) Triton X-100 (0.1X) * Trade Mark ~:
: .; , .
, ~ ,', :.

,~ ~

Claims (8)

1. Method for gene transfer into plants, which comprises (a) isolation, from anthers, of immature pollen grains in nutrient solution and removing the tissue in which they are embedded;
(b) culturing the isolated, immature pollen grains in d nutrient solution;
(c) transferring foreign genetic material into the pollen grains during the in vitro culturing and maturation;
(d) bringing about complete maturation of the transformed pollen grains in vitro;
(e) pollinating receiver plants with the transformed pollen grains and obtaining seeds from the former.

2. Method as claimed in claim 1, wherein the transfer of foreign genetic material is carried out in the stage of uninucleate microspores.

3. Method as claimed in claim 1, wherein the transfer of foreign genetic material is carried out during the first pollen mitosis.

4. Method as claimed in claim 1, wherein the transfer is carried out in the early binucleate stage, as long as the generative cell is still attached to the pollen wall.

5. Method as claimed in claim 1, wherein the transfer of the foreign genetic material is carried out briefly before or during the second pollen mitosis.

6. Method as claimed in claim 1, wherein the nutrient solution contains all the nutrient and growth substances essential for culturing and maintaining the ability of the pollen grains to mature.

7. Method as claimed in claim 2, wherein the gene transfer in the stage of uninucleate microspores is carried out in a medium enriched with sugar and further nutrient substances.

8. Method as claimed in claim 1, wherein the transfer of foreign genetic material is carried out by co-culturing with Agrobacterium tumefaciens.