Plant Cell Reports

Plant Cell Reports (1991) 10:30-34

9 Springer-Verlag1991

A grobacterium-mediated transformation of Solanum tuberosum L. ev. 'Russet Burbank' C.A. Newell 1, R. Rozman, M.A. Hinchee, E.C. Lawson, L. Haley, P. Sanders, W. Kaniewski, N.E. Tumer, R.B. Horsch, and R.T. Fraley Monsanto Company, 700 ChesterfieldVillage Parkway, St. Louis, Missouri 63198, USA i Present address: Applied Plant TechnologyLaboratory, Agricultural Genetics Company, Babraham, Cambridge CB2 4AZ, UK Received July 30, 1990/Revisedversion receivedDecember20, 1990 - Communicatedby R.N. Beachy

Abstract. Stem sections from shoot cultures maintained in vitro were used to produce transgenic plants of the potato, Solanum tuberosum L. cv. 'Russet Burbank'. Stem internode pieces inoculated with Agrobactetium tumefaciens containing coat protein genes from potato

virus X and potato virus Y, produced shoots with a frequency of 60% in the absence of selection and 10% on medium containing 100 mg/l kanamycin monosulfate. Regenerated shoots were assayed for kanamycin resistance by placing stem segments on callus induction medium containing an increased level of kanamycin. Of a total 255 regenerated shoots, 47 (18%) were kanamycin resistant. Of the kanamycin resistant shoots, 25 (53%) expressed the PVX or PVY coat protein genes as assayed by enzyme-linked immunosorbent assay or Western immunoblot analysis.

Introduction The potato, Solanurn tuberosum L., is one of the major food crops of the world today and ranks fourth in world production after wheat, maize and rice (Ross, 1986). The process of varietal improvement via traditional breeding methods is complicated by tetrasomic segregation patterns and incomplete fertility in many tetraploid commercial cultivars. Genetic engineering techniques such as Agrobacterium-mediated DNA transfer thus have great potential to improve already acceptable cultivars by introducing genes of interest without perturbing the commercially desirable phenotype. In order for genetic modification via Agrobacterium to be successful, both a reproducible regeneration system and an effective transformation system directed towards those cells capable of regeneration, are mandatory. Successful production of transgenic potato plants is a relatively recent phenomenon. Early reports of transformation in the European cultivars Maris Bard and Desiree (Ooms et aL, 1983; 1985) described the use of armedAgrobacterium strains, with the result that regenerated plants were morphologically abnormal. Gene transfer via disarmed Agrobacterium was reported subsequently for a tetraploid line (Shahin and Simpson, 1986) and for the cultivars Offprint requests to+ C.A. Newell

Maris Bard and Desiree (Ooms et al., 1987; Twell and Ooms, 1987). In the last two years transformed plants have been produced from a number of potato cultivars using tuber discs (Hoekema et al., 1989; Sheerman and Bevan, 1988; Stiekema et al., 1988), microtuber discs (Ishida et aL, 1989), and leaf pieces (de Block, 1988; Tavazza et al., 1988; Wenzler et aL, 1989). Our goal was to develop a reproducible transformation system for the cultivar Russet Burbank, which accounts for approximately 40% of the commercial acreage in the U.S.A. In vitro regeneration of whole plants of Russet Burbank was achieved in 1977 by Shepard and Totten, using leaf mesophyll protoplasts and tuber discs as source material. Subsequent research demonstrated that tuber tissue could be used for shoot production in Russet Burbank (Jarret et al., 1980a,b; Silva, 1985). Leaf tissue was also reported to be regenerable by Chang and Loescher (1987). Despite success with regeneration, however, Russet Burbank has proven less amenable to transformation than many other cultivars, and at the time this study was initiated there were no reports in the literature of successful transformation. Recently, transgenic shoots have been reported for Russet Burbank from both leaf tissue (de Block, 1988) and microtuber discs (Ishida et al., 1989). We report here the use of stem sections from in vitro grown plantlets, to produce transgenic plants of Russet Burbank expressing coat protein genes derived from potato virus X (PVX) and potato virus Y (PVY). Those plants exhibiting detectable levels of PVX and PVY coat protein were analyzed to determine the extent of cross protection against subsequent inoculation of virus; results from growth chamber and field tests are discussed by Lawson et al. (1990) and Kaniewski et aL (1990), respectively.

Materials and Methods Plant material Virus-free, sterile plantlets of Solanum tuberosum L. cv. 'Russet Burbank' clone Idaho C (seed

stock of the Williams line of Russet Burbank, N.S. Wright, pers. comm.) were provided as starting material

31 by S.A. Slack, University of Wisconsin-Madison. Shoot cultures were maintained in 25 x 100 mm flat-bottomed glass vials, capped with polypropylene closures (Bellco Kaputs) and containing 10 ml of PM medium, made up of Murashige and Skoog (MS) (1962) inorganic salts, 30 g/l sucrose, 0.17 g/1 NaI-I2PO4H20, 0.4 mg/l thiamineHC1 and 10 mg/1 inositol at pH 6.0, solidified with 2.0 g/l Gelrite (Kelco Co.). The cultures were grown at 21~ C with a 16 h photoperiod of 60-70 # E m 2 s1. Shoots were subcultured monthly by cutting off approximately 1 cm of the shoot tip and transferring this to fresh PM. When the shoots were about 8 cm in length, the internode sections were used for regeneration and transformation experiments. Stem internodes were cut into sections 5-10 mm long, taking care to ensure that the nodal region with its axillary bud was removed. The explants were cultured on medium as described by Jarret et al. (1980b), consisting of MS inorganic salts, organic addenda as listed in Jarret et al. (1980b) including 100 mg/l inositol, 30 g/l sucrose, 3.0 mg/l 6-benzylaminopurine (BAP), 0.01 mg/1 naphthaleneacetic acid (NAA), pH 5.6, and 8 g/l Difco purified agar. This medium was designated P1. After 4 weeks, the explants were moved to medium consisting of the same salts and organics as P1, but with 0.3 mg/1 gibberellic acid (GA3) replacing NAA and BAP (Jarret et al., 1981); this was designated P2. Shoots developed from the cut ends of the explants, and when these reached a length of 1-2 cm they were excised and placed into vials of PM to root.

Transformation For transformation, stem internodes were cut into sections 5-10 mm in length, and the cut ends were smeared with Agrobacterium from a 3 day old plate culture. The stem sections were cocultured with the bacteria for 2 days at 23~ C on coculture medium composed of 1/10 strength P1 salts and organics with 1 g/l sucrose and no growth regulators, together with a tobacco cell feeder layer. For the feeder layer, 1.5 ml of a four day old tobacco cell suspension (Nicotiana tabacum), maintained in a liquid medium of MS inorganic salts, B5 medium vitamins (Gamborg et al., 1968), 30 g/l sucrose, and 2 mg/1 p-chlorophenoxyacetic acid (pCPA) at pH 5.7, were pipetted onto the surface of the coculture medium and covered with a sterile filter paper, that had been previously autoclaved in water, allowed to cool, and the excess water drained off. Some modifications to this procedure that were attempted, induded a eoculture period without a feeder layer, and the use of a dry sterile filter paper over the feeder layer instead of a wet autoclaved falter paper disc. Following coculture the explants were transferred to P1 medium containing 100 mg/l kanamycin monosulfate as a selective agent for 4 weeks for callus induction. This was followed by P2 medium containing 100 mg/1 kanamycin for a further 4 weeks, after which explants were subcultured onto fresh P2 at monthly intervals. Both media contained 500 mg/l carbeniciUin to prevent bacterial growth in pMON 9809 inoculations, and 500 mg/l carbenicillin + 100 mg/1 cefotaxime in pMON 9898 inoculations. Shoots produced by tissue under selection were

rooted on PM containing 500 mg/l carbenicillin, and then tested for transformation by means of a recallusing assay. Stem pieces were tested for the ability to produce callus in the presence of kanamycin, on MSP callusing medium consisting of MS inorganic salts, organic addenda as in P1, 30 g/1 sucrose, 2.25 mg/1 BAP, 0.186 mg/1 NAA, 10 mg/l GA3, 500 mg/l carbenicillin and 200 mg/l kanamycin.

Transformation vectors The pMON 9809 vector contains the PVX coat protein gene inserted into a plant transformation vector (Hemenway et al., 1988). The PVX coat protein gene is driven by the enhanced CaMV 35S promoter (Kay et al., 1987) and the NPTII gene is driven by the 35S promoter in this vector. The construction of the pMON 9898 vector has been described previously (Lawson et al., 1990). This double gene construction has both the PVX and PVY coat protein genes driven by the enhanced CaMV 35S promoter and the NPTII gene by the unenhanced 35S promoter. Both vectors were mated into Agrobacterium tumefaciens by a triparental mating method (Ditta et al., 1980). Plate cultures ofA. tumefaciens grown on 1.5% Luria-Bertani (LB) agar containing 50 mg/1 each of streptomycin sulfate, spectinomycin and kanamycin monosulfate, and 25 mg/l chloramphenicol, were used in transformation experiments. Assays Details of the enzyme-linked immunosorbent assay (ELISA) used to detect the presence of PVX coat protein have been published elsewhere (Lawson et aL, 1990). Discs were cut from leaves of kanamycin resistant plants with a #10 cork borer and homogenized in 250/~1 phosphate buffered saline (PBS)/0.05% Tween-20/0.2% ovalbumin (PBS-T-O); 250/~1 of extract were loaded into anti-PVX IgG coated microtiter wells and incubated overnight at 4~ C. Standards of PVX virion or purified coat protein were iLncluded. The plates were rinsed, alkaline phosphatase conjugated anti-PVX IgG (1:5000) was added and incubated for 4 h at 37~ C, then rinsed 4 times prior to adding the substrate. The results were read with a Bio-Rad 3550 microplate reader. Transgenic plants were assayed for PVY coat protein expression by Western blot analysis; details of this procedure have been published in Lawson et al., 1990. Fresh leaf samples of 100 mg were ground in 300 #1 2X Laemmli buffer. After boiling for 10 min, 25/~1 of the extract (approximately 50/~g protein) were run on a 15% polyacrylamide gel at 40 mA. The gel was blotted onto nitrocellulose, and preincubated with PBS/4% bovine serum albumin (BSA) for 2 h, then 1:1000 antiPVY IgG was added and incubated for 2 h. The blots were rinsed several times with PBS, then I ~z~protein A in PBS/4% BSA was added and incubated 1 h. The blots were washed for 2 h with several changes of PBS/Triton X-100 and exposed to Kodak XAR X-ray film at -80~ C. Results

All of the uninoculated stem sections of Russet Burbank

32 Table 1. Regeneration from stem explants of Russet Burbank following inoculation with Agrobacterium tumefaciens conferring the constructs pMON 9809 and pMON 9898. Kanamycin resistance was determined by a recallusing assay, expression of PVX coat protein by ELISA, and PVY coat protein by Western immunoblot, in regenerated plants. 0 mg/l kanamycin Construct pMON 9809 pMON 9898

100 mg/l kanamycin

Total #

# with shoots(%)

Total #

50 10

32(64%) 4(40%)

825 560

# with Total shoots(%) shoots 70(8%) 69(12%)

94 161

Ave/ exp

Kant(%)

0.1 0.3

18(2%) 29(5%)

PVX § 5 19

PVY §

12

#: number of stem explants Ave/exp: Average number of shoots per explant, based ontotal explant number

placed on P1 medium without kanamycin enlarged and produced a limited amount of semi-friable callus. Two to four weeks after transfer to P2 medium small shoots developed from callus produced by the cut ends of the stem pieces; 40% of explants produced shoots (12/30; data not shown) over a period of two to three months. In the presence of 100 mg/1 kanamycin, uninoculated explants lost green pigment and died without callus or shoot formation. Results of inoculations with pMON 9809 and pMON 9898 are shown in Table 1. In the absence of selection, 95% of the inoculated stem pieces produced callus (data not shown); 64% and 40% of segments transformed with pMON 9809 and pMON 9898, respectively, produced shoots, indicating that Agrobacterium inoculation alone did not reduce shoot production. In the presence of 100 mg/l kanamycin, the central portion of inoculated explants bleached and turned brown, while small, hard, green knobs of callus developed at one or both cut ends of the stem segment. Shoots started to develop under selection 6-8 weeks after inoculation, but the frequency was reduced to 8% for pMON 9809 and 12% for pMON 9898. Potentially transgenic shoots continued to form over a two to three month period. Transgenic shoots that displayed resistance to kanamycin on the basis of a recallusing assay were produced at a frequency of 2% for pMON 9809 (18 shoots from 825 explants) and 5% for pMON 9898 (29 shoots from 560 explants; Table 1). Following transformation with pMON 9809, 5 of the 18 kanamycin resistant shoots accumulated detectable levels of PVX coat protein as determined by an ELISA assay. Of the 29 kanamycin resistant shoots obtained following inoculation with pMON 9898, 19 were positive for the PVX coat protein, while 12 were positive for the PVY coat protein. Of the latter, 11 plants expressed genes for both PVX and PVY coat proteins. Table 2 shows the results of analyses for kanamycin resistance and coat protein accumulation in individual regenerated shoots. Shoots that were harvested from the same callus clump were assigned a letter of the alphabet in the order in which they were harvested, so that a record Could be kept of shoots potentially of clonal origin. Shoots arising from the same location on the stem piece were of three classes, kanamycin negative, kanamycin positive and coat protein negative, or kanamycin positive and positive for

either or both coat proteins. When inoculated explants were cocultured without a feeder layer, 64% produced callus and none produced shoots, while 83% of explants cocultured with a feeder layer produced callus and 11% produced shoots. Similarly, a comparison of coculture conditions using a dry versus a wet autoclaved filter paper over the tobacco cell feeder layer, gave the following frequencies: callus and shoot production were 45% and 4%, respectively, for explarrts on dry autoclaved filter paper; and, 80% and 34%, respectively, for explants on wet autoclaved fdter paper. Kanamycin resistant shoots were obtained with a frequency of 6% on wet autoclaved f'dter paper and 1% on dry autoclaved filter paper. The same method of transformation was tested on stem sections of the cultivars Desiree and Kennebec (Table 3). Following inoculation with pMON 9809, transformation frequency was 13% for Desiree and 14% for Kennebec, compared with 2% for Russet Burbank. Transformation results were based on a recallusing assay in the presence of kanamycin. Discussion

At the time this study was initiated there were no published procedures for producing transgenic shoots of Russet Burbank, a cultivar which is relatively recalcitrant in terms of transformation and regeneration compared with other cultivars of Solanum tuberosum. By using stem sections from in vitro grown plantlets as starting material, and by following the regeneration procedure that was successful for regeneration from tuber discs, it was possible to produce transgenic plants of Russet Burbank with a low but reproducible efficiency. The two methods for transformation of Russet Burbank previously published use leaves (de Block, 1988) and microtuber discs (Ishida et al., 1989) rather than stem sections. The stem section method appears advantageous compared to the microtuber disc method as potentially transformed shoots first appear after approximately 6-8 weeks, instead of the 12-14 weeks reported by Ishida et al. (1989); the percentage of shoots that are transformed (18%) also compares favorably with the microtuber disc method, where approximately 5% of the regenerated shoots were resistant to kanamycin. Furthermore, the use of stem sections obviates the need

33 Table 2. Expression analysis of kanamycin resistance by recallusing, PVX coat protein (ng/mg protein ) by ELISA, and PVY coat protein by Western immunoblot, in regenerated plants of Russet Burbank following inoculation of stem explants with pMON 9809 and pMON 9898.

Table 3. Comparison of transformation frequencies in potato cultivars Desiree, Kennebec and Russet Burbank following inoculation of stem explants withAgrobacterium containing pMON 9809 conferring kanamycin resistance. Kan.

Cultivar Construct

pMON 9809

pMON 9898

Shoot #

Kanr

21a* b c d e 62a b 66a b c lla b c d 15a b c d 23a b 24a b

+

PVY

2.4

(ha) (na) (na) (na) (na) (na) (ha) (na) (na) (na)

+ + + + + + +

3.0 2.0 4.0

+

+ (nd)

+ + + + + + + +

C

d e 32a b

PVX ng/mg

4.0

+

1.0 3.0 3.0 2.0 0.5 0.2

+ + + + +

-

-

-

-

+ +

+ (nd) 1.4

+ + + + +

0.5 1.0 0.5 + (nd) + (nd)

+ + + + + +

3.0 3.0 3.0 1.0 + (nd)

+

c

d e 39a b c d 44a b c 45a 58a 61a b C

63a

+ + + +

+

+

-

+

* Shoots a,b,c etc. originated from the same explant and are potentially of clonal origin. (na): not applicable since pMON 9809 does not contain the PVY coat protein gene. (nd): protein concentration not determined.

sel. # of # of Kan r Transf. (mg/1) exp. shoots shoots freq.

Desiree 100 Kennebec 200 Russet Burbank 100

108 91 825

41 24 94

14 13 18

13 % 14 % 2%

to produce and maintain microtubers in vitro. The stem section method does not require the addition of silver nitrate to the medium as is the case for leaf regeneration in Russet Burbank (de Block, 1988); however, the possible enhancement of regeneration from stem sections by addition of silver nitrate was not investigated. An additional advantage of the stem section method over leaf transformation is that stem pieces are relatively robust and can be handled easily in large numbers; wounding during handling of leaf pieces, reportedly lowers the frequency of transformation and regeneration (de Block, 1988). The successful production of transgenic plants of Russet Burbank from stem sections, leaves or microtuber discs emphasizes the importance of using in vitro grown material. Our initial attempts to produce transgenic plants from tuber discs of Russet Burbank following the procedure of Sheerman and Bevan (1988) were not successful. With this method, regeneration frequency was variable depending on the source of tubers, and even low levels of kanamycin were deleterious to tuber tissue following inoculation. In vitro grown plants provide uniform material for transformation and alleviate the necessity of having tubers of suitable age and quality available. Attempts to increase the transformation efficiency and reduce the number of untransformed plantlets that were produced, had little effect on the overall process. The presence of a feeder cell layer during coculture and the use of a wet autoclaved filter paper layer over the feeder cells, seemed to enhance transformation efficiency primarily by increasing the number of shoots that developed under selection, and consequently was adopted as part of the routine procedure. The advantage of a wet versus dry autoclaved filter paper may have resulted either by removing compounds deleterious to regeneration from the filter paper during autoclaving, or simply by providing a moister environment for the explants over the 2-day coculture period. Shoots harvested from stem explants that were potentially of clonal origin, showed differences in kanamycin resistance and coat protein levels. In the early stages of callus development, several small calli develop independently on one or both cut ends of the stem pieces, but are too small to be removed and cultured alone. By the time the calli are large enough to

34 produce shoots, they may have grown into an inseparable mass on each end of the stem piece; thus several shoots taken from one end of a stem explant may or may not be derived from independent transformation events. To ensure the independent origin of transformation events, one shoot can be taken from each regenerating stem section end; however, this reduces the overall number of shoots that can be obtained to two per explant, and limits the variation that could otherwise be potentially obtained from one end of a regenerating stem explant. The shoots of possible clonal origin listed in Table 2 are alphabetized in the order in which they were removed from the regenerating explants. The first shoots to be harvested may not be transformed, but this does not preclude the likelihood that later shoots may be transgenic; thus, we routinely harvested up to five shoots from each regenerating callus clump in order to maximize the number of potential transgenic plants that expressed coat protein. None of the shoots that were analyzed appeared to be chimaerie in nature. Initially, regenerated shoots were screened by ELISA for the presence of PVX coat protein concomitant with the recallusing of stem pieces in the presence of kanamycin. Since no shoots were found that expressed the coat protein gene without also expressing the kanamycin resistance gene in the sample studied, the recallusing assay was undertaken as a primary screen to eliminate those escape shoots which were not transformed prior to the more labor intensive ELISA and Western immunoblot analyses. The recallusing assay selects shoots that are kanamycin resistant, and it was considered unlikely that kanamycin resistant shoots which did not express the coat protein gene, were escapes. However, the presence or absence of the NPTII gene in these plants was not investigated further. In summary, a reproducible system for producing transgenic plants of the potato cultivar Russet Burbank was developed using stem explants from in vitro grown plantlets. The same method was successfully applied to the potato cultivars Kennebec and Desiree, both of which gave a much improved transformation efficiency compared with Russet Burbank. This indicates that the procedure is not cultivar dependent and may serve as a useful starting point for maximizing transformation and regeneration in other recalcitrant cultivars of Solanum tuberosum.

References

Chang M-M, Loescher W (1987) In Vitro 23:68A De Block M (1988) Theor. Appl. Genet. 76:767-774 Ditta G, Stanfield S, Corbin D, Helinski DR (1980) Proc. Natl. Acad. Sci. USA 77:7347-7351 Gamborg OL, Miller RA, Ojima K (1968) Exp. Cell Res. 50:151-158 Hemenway C, Fang R-X, Kaniewski WK, Chua N-H, Tumer NE (1988) EMBO J. 7:1273-1280 Hoekema A, Huisman MJ, Molendijk L, van den Elzen PJM, Cornelisson BJC (1989) Bio/Technology 7:273278 Ishida BK, Snyder GW, Belknap WR (1989) Plant Cell Rep. 8:325-328 Jarret RL, Hasegawa PM, Erickson HT (1980a) Physiol. Plant. 49:177-184 Jarret RL, Hasegawa PM, Erickson HT (1980b) J. Amer. Soc. Hort. Sci. 105:238-242 Jarret RL, Hasegawa PM, Bressan RA (1981) In Vitro 17:825-830 Kaniewski W, Lawson C, Sammons B, Haley L, Hart J, Delannay X, Turner N (1990) Bio/Technology 8:750-754 Kay R, Chan A, Daly M, McPherson J (1987) Science 236:1299-1302 Lawson C, Kaniewski W, Haley L, Rozman R, Newell C, Sanders P, Turner NE (1990) Bio/Technology 8:127-134 Murashige T, Skoog F (1962) Physiol. Plant. 15:473-497 Ooms G, Burrell MM, Karp A, Bevan M, Hille J (1987) Theor. Appl. Genet. 73:744-750 Ooms G, Karp A, Burrell MM, TweU D, Roberts J (1985) Theor. Appl. Genet. 70:440-446 Ooms G, Karp A, Roberts J (1983) Theor. Appl. Genet. 66:169-172 Ross H (1986) Adv. Plant Breeding 13:1-132 Shahin EA, Simpson RB (1986) HortSci. 21:1199-1201 Sheerman S, Bevan MW (1988) Plant Cell Rep. 7:13-16 Shepar6 JF, Totten RE (1977) Plant Physiol. 60:313-316 Silva GH (1985) HortSci. 20:139-140 Stiekema WJ, Heidekamp F, Louwerse JD, Verhoeven HA, Dijkhuis P (1988) Plant Cell Rep. 7:47-50 Tavazza R, Tavazza M, Ordas R J, Ancora G, Benvenuto E (1988) Plant Science 59:175-181 Twell D, Ooms G (1987) Plant Mol. Biol. 9:365-375 Wenzler H, Mignery G, May G, Park W (1989) Plant Science 63:79-85

Agrobacterium-mediated transformation of Solanum tuberosum L. cv. 'Russet Burbank'.

Stem sections from shoot cultures maintained in vitro were used to produce transgenic plants of the potato, Solanum tuberosum L. cv. 'Russet Burbank'...
552KB Sizes 0 Downloads 0 Views