Plant CeU Reports

Plant Cell Reports (1993) 13:17-23

9 Springer-Verlag1993

Transformation of white spruce (Picea glauca) somatic embryos by croprojectile bombardment V.R. Bommineni, R.N. Chibbar, R. S. S. Datla, and E. W. T. Tsang Plant Biotechnology Institute, National Research Council of Canada, 110 Gymnasium Place Saskatoon, SK, Canada, S7N 0W9 Received 23 December 1992/Revised version received 23 July 1993 - Communicated by F. Constabel

Abstract Cotyledonary somatic embryos of white spruce [Picea glauca (Moench) Voss] were subjected to microprojectile bombardment with a gene construct containing a gus::nptll fusion gene. Somatic embryos were used to re-induce the embryogenic tissue after bombardments. Histochemical assay using X-gluc as a substrate showed that all the embryos (100%) were GUS positive 48 h after bombardment. However, only thirteen out of 605 embryos (2.2%) remained GUS positive after two months in culture. Three of those thirteen (23%) embryo-derived tissues consistently showed GUS activity for eight months in culture. These putatively transfomed embryogenic tissues were subjected to Southern blot analysis and the results suggested integration of the gus::nptll gene expression cassette in the white spruce genome.

Abbreviations: ABA = (+)abscisic acid, BA = benzyladenine, bp = base pair, 2,4-D = 2,4dichlorophenoxyacetic acid, kb = kilobase, gus = E. coil gene uid A for ~-glucuronidase, nptll = neomycin phosphotransferase II, X-gluc = 5-bromo-4-chloro3-indolyl-J3-D-glucuronic acid.

Introduction The long duration of growth and reproductive cycles make it difficult to manipulate and explore the genetics of conifers through conventional plant

Correspondence to: E. W. T. Tsang

breeding techniques (Tautorus et al, 1991). Tree improvement can be aided through presently available molecular and genetic engineering techniques (Schuch, 1991; Manders et al, 1992). Genetic transformation has some advantages over other plant improvement approaches because the technology allows direct introduction of genes encoding desirable traits into plant genomes and circumvents the long duration of conventional breeding. However, successful genetic transformation requires the optimization of several factors that include in vitro culture systems, method of gene transfer, and the expression of the introduced gene in the plant tissue (Christou, 1992; Klein et al, 1992). Several attempts have been made to transform white spruce by using both Agrobacteriummediated gene transfer methods (Ellis et al, 1989), and direct DNA delivery methods that include electroporation (Bekkaoui et al, 1990) and microprojectile particle bombardment (Ellis et al, 1991; Charest et al, 1993) in different types of explants. However, none of these reports have indicated the successful integration of foreign gene(s) in the white spruce genome We have been exploring different tissues of white spruce as suitable targets for genetic transformation through microprojectile bombardment. Among the tissues tested, cotyledonary somatic embryos were ideal for gene transfer as they were receptive to the introduction of foreign DNA. These

** during the revision of this manuscript for publication, transformation of white spruce has been reported by Ellis et al (1993) Bio/Technology 11: 8489.

18 cotyledonary somatic embryos are also capable of re-inducing embryogenic tissues for subsequent regeneration. In this report we present data for transient gene expression of gus, and stable integration of gus::nptll gene in the genome of white spruce embryogenic tissue derived from bombarded somatic embryos.

Materials and methods Embryogenic suspension cultures

culture on LP maturation medium and placed on agar-solidified half-strength Litvay medium containing BA (4.4 I~M), 2,4-D (9 I~M), sucrose (15 g/I), 0.8 g/I casein hydrolysate, 3 mM glutamine, and 0.5% agar (Sigma Chemical Company) to re-induce proliferation of embryogenic tissues and subsequent regeneration of plantlets. Stedle black gridded filter discs were placed on the agar-solidified half-strength Litvay medium for support. Twenty-five to thirty somatic embryos were placed in the center of the gridded filter discs. The embryos were incubated in the growth chamber in darkness for ten days before bombardment.

Embryogenic liquid suspension cultures of white spruce were initiated from the cryopreserved line WS1 (Dunstan e t a l , 1991). Cultures were maintained in half-strength Litvay medium (Litvay, 1981) supplemented with BA (4.4 ~M), 2,4-D (9 t~M), 15 g/I sucrose, 0.8 g/I casein hydrolysate, 3 mM glutamine and placed on a continuous shaker (150 rpm) in the dark at 25 + 2 0C. For routine transfers, 10 ml of one-week-old suspensions were dispensed into 50 ml of fresh medium in a 250 ml flask. Embryogenic suspension cultures were collected on miracloth six days after subculture (Dunstan et al, 1991) and rinsed twice with 50 ml hormone-free half-strength Litvay medium. Two grams of collected tissue was dispensed in 50 ml of fresh half-strength Litvay medium containing 15 g/I sucrose and BA (1 mg/I) (Dunstan et al, 1993) and placed on a shaker (150 rpm) in the dark prior to ABA treatment.

cassette (approximately 3.5 kb) and is schematically represented in Fig. 1. The gene construct consists of the gus::nptll fusion gene under the control of a duplicated 35S RNA promoter from cauliflower mosaic virus (E35S, Kay et al, 1987) with a 50 bp alfalfa mosaic virus (AMV) leader sequence positioned between the promoter and 5' end of the fusion gene (Fig. 1). The fusion gene is terminated by the nopaline synthase (nos) terminator and inserted in the pUC 19 vector.

Maturation of somatic embryos

--llll2 i il II

Gene construct pBI 426 (approximately 6.2 kb) contains a

gus::nptll fusion gene (Datla et al, 1991) expression

Hind TT[

I~---" 0.6 H~.

Six days after subculture in medium containing BA, the somatic embryo suspension cultures were collected as described above rinsed twice and resuspended (20% w/v) in hormone-free half-strength Litvay medium. Aliquots of 0.75 ml were then plated on half-strength LP medium (von Arnold and Eriksson, 1981) supplemented with 48 t~M (+) ABA, 10 g/I sucrose and solidified with 0.5% agar (LP maturation medium). Sterile black gridded filter discs (0.8 p.m, 47 mm diameter, no. AABGO4750, Millipore, Bedford, MA) were placed on maturation medium for support (Dunstan et al, 1991). The cultures were incubated in diffused white fluorescent lights with 16 h photoperiod at 25 + 2 0C for a period of 50 days in a growth chamber.

Somatic embryos Cotyledonary somatic embryos (stage III, Dunstan et al, 1991) were collected 50 days after

/XbaI

;iiiii N N'0.05 kb'Nlll

2.6kb

-'::

0.25kb -N

Fig. 1: Schematic representation of gene construct (3.5 kb) in plasmid pBI 426. The fused genes are under the control of duplicated 35S promoter (E35S), AMV leader sequence and nos terminator.

Preparation of DNA for bombardment of somatic embryos The procedure of Klein et al (1987) was used to precipitate and coat the DNA onto M10 (ll~m) tungsten particles. Fifty microliters of DNA (l~g/~l) was used to coat 30 mg of tungsten particles; a final volume of 350 p.I ethanol was used to prepare for fifty bombardments. Six microliters of DNA-coated tungsten particles (0.9 p.g DNA) were placed on the flying disks (kapton membranes) dried in a

19 desiccator (Russel et al, 1992) and delivered by the PDS 1000/He gun (BioRad) under the following conditions: 1100 psi helium pressure, 26 inches Hg vacuum, and the flight distance was 9.5 cm from the target tissue. Ten days after incubation on agar-solidified half-strength Litvay medium, the somatic embryos were subjected to two consecutive bombardments with pBI 426 plasmid-coated tungsten particles (Fig. 1). The somatic embryos were transferred to fresh agar-solidified half-strength Litvay medium without gridded filter discs two days after bombardment and were maintained by transferring them to fresh medium every three to four weeks.

further subdivided into clumps in every transfer. These transfers were repeated at least five times every three to four weeks until the tissue appeared homogeneous for GUS activity observed by light microscopy.

DNA analysis by Southern blot Genomic DNA from white spruce embryogenic tissue was isolated according to Doyle and Doyle (1990). For Southern blot analysis, the genomic DNA was restricted with Hind III, or Xba I and Eco RI together and electrophoresed on 1% agarose gels. The Sambrook et al (1989) procedure was used for Southern blot. DNA was transferred to Hybond Nylon membrane (Amersham) and hybridized with 32p labelled nptll probe prepared by random prime labelling (Promega). After hybridization, the blots were washed 2 to 3 times with 0.1% SDS and 0.1% SSC and exposed to X-ray film for autoradiography.

Histochemical assay for gus expression The method of McCabe et al (1988) was used to assess transient GUS expression 48 h after bombardment. The somatic embryos were incubated in X-gluc solution (0.05%) overnight and the number of blue spots (GUS expression foci) per embryo were counted as an indication of GUS activity. The experiments were repeated six times, each time with up to 25 somatic embryos (Table 1). This assay for GUS activity was used to monitor the occurrence of transformed embryos in rapidly dividing embryogenic tissue two to three months after bombardment. Initially GUS positive embryogenic tissues were divided into small tissue clumps and transferred to fresh medium for propagation. Each GUS positive tissue clump was

Results and Discussion Transient expression of gus in bombarded somatic embryos Cotyledonary somatic embryos generated on LP maturation medium fifty days after plating are shown in Fig. 2a. These somatic embryos were

Table 1: Transient expression of gus 48 h after bombardment in white spruce somatic embryos. Treatment

Number of embryos

Average number of blue

sampled*

spots per embryo + S.E.**

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

Bombarded with pBI 426

118

28 + 1.5

Bombarded with no DNA

44

Nonbombarded

79

0

* embryos were randomly collected from six experiments. ** S. E. = standard error of the mean values

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

20

Fig. 2: Transient and stable expression ofgus in bombarded somatic embryos, a) developed somatic embryos on LP maturation medium 50 days after plating of embryogenic suspensions (Bar = 625 t~m), b) transient expression of gus 48 h after bombardment (Bar = 590 ~m), c) expression of gus in reinduced embryogenic tissue (line 8T) six months after bombardment (Bar = 500 ~m), d) transformed immature somatic embryos (line 8T) showing gus expression eight months after bombardment (Bar = 55 I~m).

21 isolated and transferred to embryogenic tissue induction medium ten days before the bombardments. Figure 2b illustrates the activity of GUS in white spruce somatic embryos by histochemical assay 48 h after bombardment. The embryos in Fig. 2b show a high activity of GUS as illustrated by the intensity of blue color when incubated with X-gluc solution. The areas of intense blue color were indicative of the cells expressing the gus gene introduced by microprojectile bombardment. Transient expression of gus in white spruce somatic embryos is summarized in Table 1. The embryos were collected 48 h after bombardment from six experiments to test the transient gene expression of gus. All the embryos (100%) showed GUS activity with an average of 28 blue spots per embryo (Table 1). The number of blue spot counts ranged from three to ninety spots per embryo. However, nonbombarded embryos and embryos bombarded with tungsten particles without DNA did not show any GUS activity (Table 1). Comparable transient expression data were reported previously in genetic transformation of Norway spruce using somatic embryos (Robertson et al, 1992), and white spruce using zygotic embryos (Ellis et al, 1991).

Expression of gus and nptll in embryogenic tissue derived from somatic embryos

in

Figures 2c and 2d show the activity of GUS embryogenic tissue six and eight months

respectively, after bombardment. Blue color due to GUS activity occurs throughout the somatic embryos, following overnight incubation in X-gluc solution. The activity of GUS is observed predominantly in the embryonic head (small arrow heads, Fig. 2d) as well as the suspensor cells (large arrow heads). This suggests that the transformed embryos are likely to originate from one or more transformed cells. The occurence of GUS activity in bombarded somatic embryos two months after bombardment is summarized in Table 2. A total of 605 embryos were bombarded with tungsten particles coated with pBI 426 gene construct. Of the 605 embryos bombarded thirteen embryos (2.2%) showed sectors positive for GUS activity in the developing embryogenic tissue. Three putatively transformed embryogenic lines (3T, 8T and 11"1") which consistently showed GUS activity at each subculture were further analysed by Southern blot analysis. This confirmed the integration of gus::nptll fusion gene in the white spruce genome (Fig. 3). The DNA fragments hybridizing with the radiolabelled nptll probe were observed only in the transformed tissues and not in the untransformed controls (Fig. 3). The following suggests the integration of introduced DNA in the white spruce genome: 1) hybridization of probe to a fragment, approximately 2.8 kb in size, observed in "transformed" DNA samples digested with Xba I and Eco RI represents the intact gus::nptll fusion gene (Fig. 3a). Some hybridized fragments with molecular

Table 2: Transformation frequency in embryogenic tissues derived from bombarded somatic embryos of white spruce. Number of embryos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Treatment

cultured .

.

.

.

.

.

.

.

positive for GUS* .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

Percent of embryos showing GUS activity

.

Bombarded with pBI 426

605

13

2.2

Bombarded with no DNA

221

0

0

Nonbombarded

199

* data from seven experiments

22

Fig.3: Southern blot analysis of white spruce embryogenic tissue. One untransformed (C), and three transformed (3T, 8T, and 11T) samples were used in the analysis. Filleen t~g of untransformed and transformed DNA, and 10 pg of plasmid DNA (P) was loaded on the 1% agarose gel. a) Xba I and Eco RI digested DNA, b) undigested DNA, and c) Hind III digested DNA. The hybridized DNA was probed with 32p labelled nptll probe.

weights higher than 2.8 kb in transformed line 3T, possibly, indicate rearrangements of the incorporated gene. 2) the hybridization of nptll probe in the high molecular weight undigested "transformed" DNA samples represents the integration of introduced DNA in the white spruce genome (Fig. 3b). 3) finally, the unique Hind III site at the beginning of the promoter in the pBI 426 gene construct gives a linearized length to the end of the nos terminator of approximately 3.5 kb (Fig. 1). To demonstrate integration events of the intact expression cassette in the white spruce genome the expected hybridized fragment size should be more than 3.5 kb. The presence of several bands (2 to 3) in Hind III

digested "transformed" DNA samples, with molecular weights higher than 3.5 kb, indicates multiple integrations of introduced DNA into the white spruce genome (Fig. 3c). In addition, some hybridizations occured with fragments of molecular weight smaller than 3.5 kb, which could represent integration of incomplete expression cassettes or rearrangements of gus and nptll genes in the white spruce genome. The presence of NPTII protein in three transformed lines was also confirmed by enzymatic assay with NPTII ELISA kit (5 prime - 3 prime, Inc.) (data not shown). The occurence of putative transformants ranged from 0.7% to 10% of the total of bombarded

23 embryos in seven experiments based on histochemical GUS assay two months a1~er bombardment. A total of thirteen embryos positive for GUS activity were isolated from the 605 bombarded embryos based on histochemical GUS assay (Table 2). Embryogenic tissue from three of the thirteen putatively transformed embryos (23%) consistently showed GUS activity through each subculture by histochemical assay and the remaining ten have not yet proven consistent for GUS activity. This may be because the transformed tissue sector size is too small to recover in the remaining tissue. The toxicity of kanamycin for selection of zygotic embryos of white spruce has been reported by Tsang et al (1989). With somatic embryos our preliminary results indicated that selections with kanamycin at early stages of embryo culture caused inhibition in their development. Consequently we adopted the histochemical X-gluc assay to monitor number of transformants in the bombarded somatic embryos. In rice, isolation of transformants was achieved without applying selection (Chdstou et al, 1992). In addition, as indicated in Norway spruce transformation (Robertson et al, 1992), prolonged exposure to antibiotics may be a contributing factor to the formation of non-embryogenic tissue. The results show that transformation of white spruce can be achieved through microprojectile particle bombardment. With white spruce somatic embryos the experiments were highly reproducible with a 2.2% transformation frequency based on histochemical GUS assay two months after bombardment. The high frequency of transient gene expression in bombarded cotyledonary somatic embryos and recovery of transformed embryogenic tissue is an indication that such technology can be used for the introduction of useful traits into white spruce. Further experiments are in progress to initiate embryogenic suspension cultures from transformed tissue for use in regeneration, and subsequent characterization of transformed white spruce plants.

Acknowledgements We would like to acknowledge Dr. David Dunstan and Mr. Terry Bethune for the provision of white spruce embryogenic suspensions, Mr. Lee Steinhauer, and Ms. Shufen Hu for their excellent technical help.

References Bekkaoui F, Datla RSS, Pilon M, Tautorus TB, Crosby WL, Dunstan DI (1990) Theor. Appl. Genet. 79:353-359 Charest PJ,Calero N, Lachance D, Datla RSS, Duchesne LC, Tsang EWT (1993) Plant Cell Rep. 12:189-193 Christou P (1992) The Plant Journal 2(3): 275-281 Christou P, Ford TL, Kofron M (1992) Trends in Biotechnology 10(7): 239-246 Datla RSS, Hammerlindl JK, Pelcher LE, Crosby WL, Selvaraj G (1991) Gene 101:239-246 Doyle JJ, Doyle JL (1990) Focus 12:13-15 Dunstan DI, Bethune TD, Bock CA (1993) In Vitro Cell. Dev. Biol. (in press) Dunstan DI, Bethune TD, Abrams SR (1991) Plant Sci. 76:219-228 Ellis DD, Roberts D, Sutton B, Lazaroff W, Webb D, Flinn B (1989) Plant Cell Rep. 8:16-20 Ellis DD, McCabe D, Russel D, Martinell B, McCown BH (1991) Plant Mol. Biol. 17:19-27 Kay R, Chan A, Daly M, McPherson J (1987) Science 236:1299-1302 Klein TM, Wolf ED, Wu R, Sanford JC (1987) Nature 327:70-73 Klein TM, Arentzen R, Lewis PA, FitzpatrickMcEIligott S (1992) Bio/Technology 10: 286291 Litvay JD, Johnson MA, Verma D, Einspahr D, Weyrauch K (1981) Institute Paper Chemistry Technical Paper Series No. 115, Appleton, Wl Manders G, Davey MR, Power JB (1992) J. Expt. Bot. 43 (254):1181-1190 McCabe DE, Swain WF, Martinell BJ, Christou P (1988) BiofTechnology 6:923-926 Robertson D, Weissinger AK, Ackley R, Glover S, Sederoff RR (1992) Plant Mol. Biol. 19: 925935 Russell JA, Roy MK, Sanford JC (1992) Plant Physiol. 98:1050-1056 Sambrook J, Fdtsch EF, Maniatis T (1989) Molecular cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, C o l d Spring Harbor, NY Schuch W (1991) In Vitro Cell. Dev. Biol. 27P:99103 Tautorus TE, Fowke LC, Dunstan DI (1991) Can. J. Bot. 69:1873-1899 Tsang EW'I', David H, David A, Dunstan DI (1989) Plant Cell Rep. 8:214-216 von Arnold S, Eriksson, T (1981) Can. J. Bot. 59: 870-874