Plant Cell Reports

Plant Cell Reports (1993) 12:303 306

9 Springer-Verlag1993

Transformation of Liquidambar styraciflua using Agrobacterium tumefaciens Janet Sullivan and L. Mark Lagrimini Department of Horticulture, The Ohio State University, Columbus, Ohio 43210-1096, USA Received May 14, 1992/Revised version received January 28, 1993 - Commuuicated by J. J. Finer

Summary. We describe the molecular transformation of Liquidambar styraciflua using A grobacteriurn tum efaciens. A binary TI-plasmid vector containing a chimeric neomycin phosphotransferase gene which confers resistance to kanamycin and either a chimeric Bacillus thuringiensis toxin gene, a chimeric E. coli [~-glueuronidase (GUS), or a chimeric tobacco anionic peroxidase gene was introduced into sweetgum by co-cultivation with A grobaeteriurn turn efaciens. Sweetgum shoots regenerated in the presence of kanamyein were confirmed to be transformed by genomic DNA blots or the presence of GUS activity. The optimization of the transformation protocol and the incorporation of molecular transformation into a rapid germplasm improvement protocol are discussed. Introduction The introduction of new traits into tree species is frequently slowed by exceptionally long generation times which can exceed 30 years seed-to-seed. For example, sweetgum (Liquidambar styraciflua) typically takes 15 years to reach sexual maturity. This lengthy generation time often renders traditional breeding ineffective for germplasm improvement. However, because a micropropagation method for sweetgum has been developed which results in the production of large numbers of shoots from excised leaf tissues (Brand and Lineberger, 1988), it may now be possible to introduce specific genes through genetic engineering. This could ultimately result in a significant time reduction for the commercial production of genetically improved tree species. Regenerated plants can then be put into the Ohio Production System (OPS) where they can be forced to flower in three years through modified cultural practices (Struve, 1990). Potential improvements of sweetgum through genetic engineering include improved pulping quality, reduction or elimination of obnoxious fruit (gum balls), and Correspondence to: L. M. Lagrimini

resistance to insect defoliation. For example, there are two known defoliators of sweetgum; fall web worm (Pirone et al., 1960) and the gypsy moth (Gerardi and Grimm, 1967). The fall web worm is native to North America and tends to feed on ornamental and shade trees with a preferred host being sweetgum. As many as four generations of the fall webworm occur per year (Johnson and Lyon, 1976). In addition, the gypsy moth is rapidly spreading throughout the Northeastern U.S. and into the Mid-West, causing the defoliation of many thousands of acres of sweetgum trees. The increasing occurrence of the gypsy moth in the Western U.S. has raised serious concerns about present control programs which require spraying trees with chemical and B.t. toxin-based insecticides (Johnson and Lyon, 1976). As an alternative to spraying, the B.t. toxin gene can be introduced directly into plants via gene transfer (Barton et al., 1987; Fischoff et al., 1987; Vaeck et al., 1987). Transgenic plants can synthesize the B.t. toxin, thus rendering them resistant or at least tolerant to infestation by. certain insects. Additional improvements through genetic engineering might also include such things as increased drought tolerance, increased disease resistance and lower lignin levels which, in turn, might improve pulping quality. For example, the overproduction of a tobacco anionic peroxidase in transgenie tobacco plants, an enzyme involved in cell growth and cell wall development, has been shown to increase root mass (Lagrimini et al., 1990) and have some effect on insect resistance (Dowd and Lagrimini, unpublished results). These initial observations indicate that the tobacco anionic peroxidase gene could be incorporated into sweetgum improvement programs. Through the use of Agrobactevium-mediated transformation, we have adopted a protocol to introduce a chimeric B.t. toxin gene, a chimeric anionic peroxidase gene and a chimeric E. coli GUS gene into sweetgum trees. We present in this paper a protocol for the transformation and regeneration of sweetgum, and demonstrate conclusive-

304 ly the introduction of all of these genes into the genome of L. styraciflua.

kanamycin resistance was used in transformation studies as described above.

Materials and methods

Plant Transformation and Regeneration. A. tumefaciens harboring either

Plant Matcrial. Seed of L iquidambar styraciflua (American Sweetgum)

was obtained from Dr. Kim Steiner (Pennsylvania State University; #85KS003 progeny from accession #75KS314, genetic origin: New Brunswick, N J). Aseptic shoot tip cultures were initiated and maintained on woody plant medium (WPM) supplemented with 2.5 mg/i B A (benzyladenine) according to Brand and Lineberger (1988). Determination o f K anam yein Sensitivity. Aseptic sweetgum leafexplants

(10 per treatment) were placed onto shooting medium (WPM + 2.5 mg/l BA) containing either 0, 5, 10, 25, 50, or 100 ~tg/ml kanamycin. Filter sterilized kanamycin was added to the media after autoclaving. Followhag culture for four weeks on the kanamycin-containing medium, the explants were scored for shooting. Bacillus thuringiensis Toxin Gene Vector. A truncated B.t. toxin gene (subsp. kurstak 0 transcribed from the Cauliflower Mosaic Virus 35S promoter was described previously (Carozzi et al., 1992). The chimeric B.t. toxin gene was contained on the binary plant transformation vector, pCibl0 (Rothstein et al., 1987), which includes a chimeric neomycin

phosphotransferase gene to confer kanamycin resistance in plants. The resulting plasmid construct is called pCib1322 (Fig 1.) (Carozzi et al., 1992). This binary TI-plasmid was transferred by mating to Agrobacterium turn efaciens strain A 136 which harbors the disarmed helper TIplasmid pCib 542. The resultingAgrobacterium host was used in plant transformations. 0 KB

PsiI ~

~

~

n

c

~

e

~

%

BamHI

3rS~nmnt.r ~ ~,---4KB

(~

i ......

SalI/BamHI

_...,_...2

,

.........

Pst I

Fig. 1. Restriction map ofpCibl322 (Carozzi et al., 1992).

Chimeric E. coli fl-glucuronidase Gene. A TI-plasmid based vector which includes the E. cell reporter gene 13-glueuronidase (GUS) fused to the Cauliflower Mosaic Virus 35S promoter was also used for sweetgum transformation studies (Jefferson et al., 1987). In addition to the chimeric GUS gene, this vector (pBll21) also has a chimeric neomycin phosphotransferase (NPT) gene which confers kanamycin resistance. This plasmid was transferred into .4. tumefaciens for transformation studies as described above. Chimeric Tobacco Anionic Peroxidase Gene. The tobacco anionic peroxidase was previously joined to the Cauliflower Mosaic Virus 35S promoter and terminator which was contained within the binary transformation vector pCibl0 (Lagrimini et al., 1990). This plasmid (pML507) which directs high levels ofperoxidase expression and confers

pCib1322, pML507, or pBI121 was grown overnight at 28~ in AB minimal medium containing 25 ~tg/ml of kanamycin sulphate (Rothstein et al., 1987). Leaf blades were removed from aseptically propagated plants, one transverse cut was made through the mid-rib and the leaves were immediately mixed with the bacteria. After incubating for seven minutes at room temperature the leaf explants were transferred to WPM supplemented with 2.5 mg/! BA (Brand and Lineberger, 1988). The explants were co-cultivated with A. turn efaciens at 28~ for three days in the dark, after which visible bacterial growth was observed. One-half of the explants were then transferred to non-selective WPM containing 500 lag/ml eefotaxime (to prevent further growth of the bacteria), and the remaining explants were transferred to selective WPM containing 25 lag/ml kanamycin sulphate in addition to 500 lag/ml cefotaxime. Controls consisted of leaf explants that received similar treatment but they were not inoculated with A. turn efaciens. All explants were transferred to fresh medium every 12 to 14 days. Leaf tissues growing on WPM + 2.5 mg/1 BA containing only cefotaxime were allowed to generate callus before transfer to selective medium containing 25 lag/ml kanamycin. Callus formed primarily along the cut end of the petiole and the midvein. Resulting shoots were then excised and transferred to Magenta GA-7 boxes containing WPM supplemented with 25 lag/ml kanamycin, 500 ~tg/ml cefotaxime and 2.5 mg/1 BA. Control explants died shortly after transfer to selective medium. Leaf tissue from four sweetgum regenerates suspected to harbor the GUS gene was assayed for GUS activity in situ with 5-bromo-4-chloro-3-indoyl glucuronidr (X-GLUC) as described by Jefferson (1987). This tissue was also assayed for GUS activity in vitro fluorometrically with the substrate 4-methyl umbelliferyl glucuronide (MUGXMolecular Probes, Inc., Eugene, OR). The extraction of the tissue and fluorometric assays were essentially as described by Jefferson (1987). Four hundred milligrams of leaftissue was homogenized in 800 laL of cold extraction buffer (50 mM NaH2PO4 [pH 7.4], 10 mM 2-mercaptoethanol, 10 mM NaaEDTA, 0.1% sodium lauryl sarcosine,0.1% Triton X-100, 1.0% PVP-40), then centrifuged to obtain a high speed supematant. Addition of soluble polyvinylpyrrolidone (PVP-40) was required for extraction of sweetgum tissue to avoid the precipitation of proteins by phenolics. Extracts (100 lal) were incubated for 0, 10, 20, and 30 min. at 37~ in the presence of MUG. A Heeler Model TKO-100 mini-fluorimeter (Hoefer Scientific, San Francisco, CA) was used with an excitation wavelength of 365 nm and an emission wavelength of 460 nm.

Enzymatic Assay f o r GUS Activity.

DNA Extrc~tion and Oenomic Blot Analysis. High molecular weight

genomic DNA was extracted from 0.3 g of tissue from two independent four-month-old B.t. transformants as well as from one control plant. The CTAB (cetyltrimethylammonium bromide) DNA extraction procedure of Doyle and Doyle (1991) was used with the addition of multiple phenol/chloroform extractions after RNase I treatment. The CTAB extraction buffer was prepared as described with the addition of lmM EDTA, 1% PVP-40 and 0.1% sodium metabisuiflte. Genomic DNA (10 lag) from both control and transformed plants was digested to completion with 100 units of Barn HI. Additionally, 600 ng ofpCibl322 was digested with Sal I and Sst I. The digested DNA was subjected to agarose gel electrophoresis (Sambrook et al., 1990) and transferred overnight onto Hybond-N nylon membrane (Amersham, Chicago, IL). The membrane was washed in 2x SSC for 5 minutes, and the DNA was crosslinked to the membrane by a 2 min. exposure to ultraviolet light (Khandjian, 1987). Pre-hybridization and hybridization were performed according to Sambrook et aL (1990). A 8al I fragment from pCib1322 (Fig. 1) was gel purified (Qian and Wilkinson, 1991), and labelled using the random primer method (Feinberg and Vogelstein, 1983). Following hybridization, the membrane was washed twice in 2x SSC for 15 rain. at room temperature followed by a wash in 0.1X SSC, 1% SDS for 30 min.

305 at 55~ The membranewas then exposedto KodakXAR-5film at -80~ for 24 hours with an intensifying screen. Results and discussion

Sensitivity of leaf explants of aseptically-grown L. styraciflua to kanamycin was determined by placing leaf

tissue on regeneration medium containing 0, 5, 10, 25, 50, or 100 ~tg/ml kanamycin. No shoots were detected on explants plated on medium containing as little as 25 ~tg/ml kanamycin; therefore, a concentration of 25 ~tg/ml kanamyein was chosen for transformation studies. A total of sixteen explants were inoculated with A. tumefaciens containing the B.t. plasmid pCib1322. The eight explants that were placed directly on selective WPM following co-cultivation with A. turnefaciens harboring the B.t. plasmid pCib 1322 showed no shoot proliferation after eight weeks in culture. However, each of the eight explants which were placed on non-selective medium showed callus proliferation at the petiole and at the cut midvein after three weeks in culture. At this point, the eight explants showing callus proliferation were transferred to selective medium containing kanamycin, where they were allowed to grow for an additional five weeks. After five weeks, four of these explants produced clusters of three to four healthy shoots at the petiole and the midvein while the remaining four became necrotic and died. Individual shoot clusters were excised away from callus tissue and transferred to Magenta GA-7 boxes containing selective WPM where they continued to grow (Fig. 2), In total, 50% of the leaf explants which were allowed to grow briefly in the absence of selection produced shoots which were resistant to kanamycin. Control explants died shortly after transfer to selective

Fig. 2. L. styraciflua Uansfonned shoot harboring the chimeric B.t. toxin gene.

medium. Southern hybridization analysis revealed that putative B.t. transformants possessed a 4kb restriction fragment that hybridized to the B.t. gene probe (Fig. 3). This size Barn HI fragment was predicted for an intact B.t. gene. The difference in band intensity between the two transformants presumably reflects variability in copy number of the integrated B.t. gene. Untransformed sweetgum DNA did not show any DNA fragment hybridizing to the B.t. gene. Genomie DNA was also digested with Pst I which cuts once within the T-DNA borders, Unique sized fragments in putative transformants confirmed integration of the B.t. gone (data not shown).

Hg. 3. Southern blot of wild-type (lane 1) and B.t. hansformed (lanes 2 and 3) shoots.

Similar transformation efficieneies were obtained with leaf explants co-cultivated with Agrobaeteriurn harboring the GUS plasmid PB1121. Of eight explants that were infected with A grobacterium harboring pBI122, six produced healthy shoots in the presence of kanamycin. Intact leaves from four of these six putative transformants were developed for GUS activity in situ with X-GLUC. An intense blue staining was seen in the vascular tissue in each of the transformants tested, but no staining was detected in untransfonned tissue (data not shown). Additionally, cell free homogenates of leaf tissue from one control and two putative transformants were assayed for GUS enzymatic activity by fluorescence spectroscopy (Table 1). Sweetgum tissue transformed with pBI 121 expressed GUS activity 3.5 - 13 fold higher than untransformed tissue, thus eonfirming transformation. Sweetgum tissue transformed with the chimeric anionic peroxidase gone (pMLS07) produced similar numbers of shoots as with the B.t. and GUS constructs. Southern hybridization analysis was carried out on one putative peroxidase overproducer and revealed the presence of a 1.2 kb restriction fragment that hybridized to the tobacco peroxidase gene probe (data not shown). This size

306 Table 1. GUS activityin transformedsweetgumplants. Solubleextracts were prepared from leaf tissue of tobacco, untransfonnedsweetgum,and two regenerates transformed with the 35S/GUS construct. GUS activity was measuredwith a fluorometerusing MUG as substrate. GUS activity is expressed as pmol MU/min/mgfresh weight. The results are presented as the mean and standard deviation from three experiments.

PLANT untransformedsweetgum 35S/GUS sweetgum (A) 35S/GUS sweetgum (B)

GUS ACTIVITY 7.6 + 0.50 101.5 9 3.30 26.7 • 4.50

fragment was predicted for an intact tobacco peroxidase gene. As with the B.t. transformants, genomic D N A from the peroxidase transformant was also digested with Pst I. Again, unique sized fragments confirmed integration of the tobacco peroxidase gene. Rooting of pML507 shoots was obtained in three to four weeks after placing shoots on growth regulator-free W P M and rooted plants were transferred to potting medium and placed in the greenhouse (Fig. 4). Future experiments w i l l b e directed at investigating peroxidase activity in pML507 sweetgum transformants.

Fig. 4. L. styraciflua transformed shoot harboring the chimeric E. call GUS gene.

We have been able to generate transformed sweetgum which contain a chimeric B. thuringiensis toxin gene, a chimeric E. coli ~-glucuronidase gene or a chimeric tobacco peroxidase gene. Although we have demonstrated function of the introduced GUS gene in

transformed tissue, it remains to be seen whether the B.t. toxin gene or the tobacco peroxidase gene will result in a desirable phenotype. However, the protocol we have presented here for the generation of transgenic sweetgum trees represents an important addition to the slow growing list of transformation and regeneration protocols that have been reported for such species as Populus (McCown et al., 1991 ; Fillatti et al., 1987), sugar pine (Loopstra et al., 1990), and Yellow-Poplar (Wilde et aL, 1992).

Acknowledgements. We would like to thank Dr. Vicki Gingas for technical assistance and critical review of the manuscript. This research was supported in part by a grant from the U.S. Departmentof Energy (DE-FG02-89ER14004)and by Stateand Federalfundsappropriatedto the Ohio Agricultural Research and Development Center, The Ohio State University. Manuscriptnumber 138-92.

References Barton KA, Whitely HR, Yang NS (1987) Plant Physiol 85:1103-1109 Brand MH, LinebergerRD (1988) Plant Science 57:173-179 Carozzi NB, Warren GW, Desai N, Jayne SM, Lotstein R, Rice DA, Evola S, Koziel MG (1992) Plant Mol Biol, in press Doyle JJ, Doyle JL (1990) Focus 12:13-15 Feinberg A, VogelsteinB (1983) Anal Biochem 132:6-13 Fillatti JJ, Sellmer J, MeCown B, Haissig B, Comai L. (1987) Mol Gen Genet 206:192-199 Gerardi MH, Grimm JK (1967) The History, Biology, Damage and Control of the Gypsy Moth. Associated UniversityPresses, Inc. Jefferson RA (1987) Plant Mol. Biol. Reporter 5:387-405 Jefferson RA, Kavanagh TA, Bevan MW (1987) EMBO J. 6:39013907 Johnson WT, Lyon HH (1976) Insects That Feed on Trees and Shrubs. Cornell UniversityPress, Ithaca, N.Y. Khandjian EW (1987) Bioteehnology5:165-167 Lagrimini LM, Bradford SA, Rothstein, S (1990) Plant Cell 2:7-18 Lagrimini LM (1991A) Altered phenotypes in plants transformed with chimerictobacco peroxidase genes. In: Lobarzewski J, Greppin H, Panel C, Gaspar T (eds) Biochemical,molecular,and physiological aspects of plant peroxidases. Universityof Geneva Press, Geneva, pp 60-67 Lagrimini LM (1991B) Plant Physiol. 96:577-583 Loopstra CA, Stomp AM, Sederoff RR (1990) Plant Molecular Biology 15:1-9 McC0wn BH, McCabe DE, Russell DR, Robison DJ, Barton KA, Raffa KF (1991) Plant Cell Reports 9:590-594 Pirone PP, Dodge BO, Riekett HW (1960) Diseases and Pests of Ornamental Plants. The Ronald Press Company Quian L, WilkinsonM (1991) Biotechniques 10:736-738 Rothstein SJ, Lahners KN, Lotstein ILl, Carozzi NB, Jayne SM, Rice DA (1987) Gene 53:153-161 Sanlbrook J, Fritsch EF, Maniatis T (1989) MolecularCloning: A Laboratory Manual. Second Edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Struve DK (1990) Int'n Plant Prop Soe 4:608-612 Vaeck M, ReynaertsA, Hotle H, Jansens S, De BeuekeleerM, Dean C, Zabeau M, Van Montagu M, Leemans J (1987) Nature 328:33-37 Wilde HD, Meagher RB, Merkle SA (1992) Plant Physiol 98:114-120

Transformation of Liquidambar styraciflua using Agrobacterium tumefaciens.

We describe the molecular transformation of Liquidambar styraciflua using Agrobacterium tumefaciens. A binary TI-plasmid vector containing a chimeric ...
801KB Sizes 0 Downloads 0 Views