Plant Cell, 1~ssueand Organ Culture 43: 215-222, 1995. © 1995KluwerAcademicPublishers. Printedin the Netherlands.

215

Factors affecting Agrobacterium tumefaciens-mediated transformation in several black poplar clones M. Confalonieri 1, A. Balestrazzi 2, S. Bisoffi 1 & R. C e l l a 3 l lstituto di Sperimentazione per la Pioppicoltura SAF/ENCC, Via di Frassineto 35, 1-15033 Casale Monferrato (AL), Italy; 2Dipartimento di Genetica e Microbiologia "A. Buzzati Traverso ", Via Abbiategrasso 207, 1-27100 Pavia, Italy; 3Dipartimento di Biologia, Sezione di Botanica, Universita di Ferrata, Corso Porta Mare 2-44100 Ferrara, Italy Received20 May 1994;acceptedin revisedform31 July 1995 Key words: Genetic transformation, Populus nigra L., tissue culture

Abstract Transient expression of the uidA reporter gene was used in preliminary experiments with two oncogenic and two disarmed Agrobacterium tumefaciens strains in order to test the efficiency of T-DNA transfer to N084 x Populus nigra and N107 x P. nigra clones. The oncogenic strain A281 pKIWI105 produced the highest average number of GUS spots per leaf disc. In order to optimize the production of transgenic plantlets from different P. nigra clones (San Giorgio, Jean Pourtet, N084 × P. nigra and N107 x P nigra, respectively), twoA. tumefaciens strains (GV2260 p35S GUS, A281 pKIWI 105 ) and bacterial concentrations (7 x 108; 1.2 x 109 bacteria ml- 1) were used. Following co-cultivation with A281 pKIWI105, the frequency of leaf discs producing kanamycin-resistant calli was not significantly different between the clones and bacteria concentrations used. Transformed shoots were regenerated from all clones, except for Jean Pourtet. Co-cultivation of leaf discs with GV2260 p35S GUS produced very few calli which died when transferred to selective regeneration medium. In addition, the effects of acetosyringone and leaf wounding were evaluated for the San Giorgio and Jean Pourtet clones, using the same strains. Factors which significantly affected the transformation efficiency of leaf explants were the P. nigra clone, the A. tumefaciens strain, and the presence of acetosyringone. Genetic transformation of calli and regenerated plantlets was confirmed by their ability to grow and root on Woody Plant Medium containing kanamycin, by histochemical/3-glucuronidase assays, and Southern blot hybridization analyses. Abbreviations: BA - benzyladenine, GUS - #-glucuronidase, IBA - indolebutyric acid, MS - Murashige and Skoog, NAA -/3-naphthaleneacetic acid, nptll - neomycin phosphotransferase II gene, uidA - #-glucuronidase gene, WPM - Woody Plant Medium

Introduction Black poplar (Populus nigra L.), a forest tree species native to Europe and western Asia, is used worldwide to obtain interspecific hybrids (P x euramericana by crossing with Populus deltoides Bartr.). The introduction of new desirable traits in poplar species by classical breeding is delayed because of the long generation time. Development of A. tumefaciens-based transformarion procedures for black poplar could contribute to the expansion of the gene pool of this species by

allowing the introduction of useful genes such as those for resistance to disease, insect pest and herbicides. ARhough black poplar has been shown to be susceptible to transformation by A. tumefaciens (Confalonieri et al., 1994), the procedure exhibited low recovery of transformed plants (20% efficiency). This study reports the optimization of a system for transfer and integration of foreign chimeric genes in black poplar clones using a binary oncogenic A. tumefaciens strain (A281 pKIWI 105). The systematic analysis of several factors, such as P. nigra genotype, A. tumefaciens strain, bacte-

216 rial concentration, presence of acetosyringone, and leaf wounding procedure resulted in a notable improvement of the transformation efficiency of this species.

from single colonies as described by Confalonieri et al., (1994) and resuspended in liquid MS or WPM medium to a final density of 0.35-0.6 (A550).

GUS transient expression Materials and methods

Plant material and culture conditions Four clones of P nigra were examined: San Giorgio (cv. italica), Jean Pourtet, N084xpolycross P nigra and N107xpolycross P nigra. The last two clones were obtained through in vitro culture of fertilized ovaries (unpublished data). Aseptic poplar plantlets were cultured in vitro on WPM (Lloyd & McCown, 1981) supplemented with 2% sucrose (Merck), 0.6% agar (Merck) and propagated on the same medium containing 2.46 gM IBA. A modified WPM (microsalts and MS vitamins), containing 2.2 gM BA and 0.27 glVl NAA, was used for shoot regeneration from leaf discs of San Giorgio and Jean Pourtet. Shoots of N 0 8 4 x P nigra and N107xP nigra clones were regenerated from leaf discs on MS medium supplemented with 3.3 gM BA and either 0.27 or 1.34 gM NAA, respectively. All regeneration media contained 3% sucrose (Merck), 0.4% agar (Merck), 0.2% gelrite (Schweizerhall) and the pH was adjusted to 5.7 before autoclaving. In vitro plantlets (two per vessel) were cultured in sterile GA7 boxes (Magenta Corp.) and kept in a growth chamber (25 ° C, 16 h, 27-33 gmol m-2s-1; 22 ° C, 8 h, darkness).

Agrobacterium strains and plasmids The following A. tumefaciens strains were used for the transformation experiments: two disarmed (GV2260 p35S GUS (Vancanneyt et al.,1990) and LBA4404 pKIWI105 (Janssen & Gardner, 1989)) and two oncogenic strains (C58 pKIWI105 and A281 pKIWI105 (Janssen & Gardner, 1989)). The p35S GUS and pKIWI105 binary vector plasmids contain the nptlI gene under the control of nopaline synthase promoter, and the uidA reporter gene is under the control of the Cauliflower Mosaic Virus 35S promoter. In the p35S GUS construct, the uidA gene is interrupted by a plant intron and shows activity only in transformed plant cells (Vancanneyt et al., 1990). In the pKIWI105 plasmid, the uidA gene is lacking a functional bacterial ribosome binding site and, therefore, gives a negligible level of expression in A. tumefaciens (Janssen & Gardner, 1989). Bacterial cultures were obtained

Leaf discs (5 x 10 mm) obtained from in vitro -grown plantlets o f N 0 8 4 x P nigra and N107xP nigra clones were pre-cultured, co-cultivated and transferred to selective regeneration medium using the conditions described by Confalonieri et al., (1994). Four different A. tumefaciens strains (A281 pKIWI105, C58 pKIWI105, LBA4404 pKIWI105, GV2260 p35S GUS) and two different bacterial concentrations (7x 108, 1.2x 109 bacteria m1-1) were tested for each poplar clone. Fifteen leaf discs were used for each treatment. Three days after co-cultivation, explants were assayed for the presence of GUS activity as described by Jefferson (1987). Control leaf discs were cultured using the same conditions as were used for the co-cultivated leaf discs. The number of single GUS spots per explant was determined by microscopic observations.

Recovery of stable transformants In the first experiment, leaf discs of San Giorgio, Jean Pourtet, N084×P nigra and N 107×P nigra were co-cultivated with A281 pKIWI105 or GV2260 p35S GUS. For each clone/strain combination, two different bacterial concentrations (7x 108; 1.2 × 109 bacteria m1-1) were used. Fifty leaf discs were employed for each experimental unit. The experimental design consist of a 4 × 2 x 2 factorial without replications. Due to clear failure of GV2260 p35S GUS strain to produce any transformed calli, the ANOVA was performed on a reduced set of data corresponding to a 4 × 2 factorial design. The clone x bacterial concentration interaction was used as an error term for testing the significance of differences among main effects. The second experiment was devised as an unreplicated 25 factorial design. The factors studied were clones (San Giorgio, Jean Pourtet), strains (GV2260 p35S GUS, A281 pKIWI105), bacterial concentrations (7 x 10s; 1.2 x 109 bacteria ml-l), absence or presence of 200 gM acetosyringone in the bacterial suspension and leaf wounding (leaves were either cut in small pieces or cut and punctured on the surface with a pinbrush). Also in this experiment, GV2260 p35S GUS did not produce transformed calli and the ANOVA was performed on a reduced set of data corresponding to a 24 factorial design. The error term was derived

217 by pooling the 3-way and 4-way interactions. As no 2-way interactions were significant (a < 0.05), they were also included in the error term used for testing the significance of the main effects. Fifty leaf discs were employed per experimental unit. The explants were pre-cultured, co-cultivated and transferred to selective regeneration medium as described by Confalonieri et al. (1994). Regeneration control plates were used for each clone by culturing 20 explants using the same conditions as were used for the co-cultivated leaf discs. Following co-cultivation, the explants were incubated in the dark for seven days, then transferred to light and subcultured every two weeks for the first forty days. Subsequently, all calli which grew from each leaf disc were excised and cultured on the same medium for one month. These calli were treated as independent transformants. Portions of kanamycin resistant calli were examined for GUS activity by histochemical assays. In order to evaluate autotrophic growth, the calli obtained by co-cultivation with A281 pKIWI105 were also placed on selective WPM medium lacking growth regulators. As a control, 20 untransformed calli per clone were cultured on the same basal medium with and without kanamycin. The regenerating calli were then cultured on modified selective MS medium (WPM macrosalts) containing 0.22 gM BA and subsequently transferred to fresh medium every month. Resulting normal and teratomic shoots were excised from the calli, placed to root on selective WPM medium and assayed for the expression of the uidA gene. All cultures were incubated under the same conditions of light and temperature as described in the plant materials section. The transformation efficiency was expressed as the percentage of leaf explants producing kanamycin resistant calli or plantlets relative to the total number of co-cultivated leaf explants. The arcsin x/x transformation was performed on percent data before ANOVA. Molecular analyses Southern blot analysis was performed on DNA sampies extracted from leaves of one N107 x P. nigra line transformed with GV2260 p35S GUS, from leaves of different P. nigra clones (San Giorgio, Jean Pourtet, N084xP. nigra and N107xP. nigra, respectively) transformed with A281 pKIW1105, from an oncogenic callus line of each of the above mentioned clones transformed with A281 pKIWI105 and from leaves of untransformed poplar clones. DNA was isolated according to Rogers & Bendich (1988), except that the

concentration of CTAB in the extraction buffer was increased from 2% to 4%. For Southern blot analysis, aliquots of poplar DNA (5-10 gg) were digested with EcoRV (Promega), separated on 0.8% agarose gels and transferred to Hybond N + membranes (Amersham) according to manufacturer's instructions. Prehybridization and hybridization were performed as described by Confalonieri et al. (1994). In order to detect the presence of the nptII gene, we used a probe consisting of the EcoRV fragment (containing the translated region of the nptlI gene) obtained from plasmid pABDI (kindly supplied by Prof. I Potrykus, ETH, Ztirich). In order to detect the presence of oncogenic T-DNA, the BamHI fragment (10 kbp) derived from the agropine type Ti plasmid (pTiBo542) was used as a probe. Plasmid pEH136, containing this fragment, was kindly supplied by Prof. E. Hood (Utah State University, Logan USA). Both the EcoRV (1.2 kbp) and the BamHI fragments were used as positive controls and run on the same agarose gel with poplar DNA (see Pig. 1 and 2). Probes were labelled with the "Ready to go" DNA labelling kit (Pharmacia) according to the manufacturer's instructions. Filters were washed twice in 2 x SSC, 0.1% SDS (10 min) at room temperature, and once in 1 x SSC, 0.1% SDS (15 min) at 65 ° C. Results were obtained following 1-4 days of exposure using HyperfilmTM-Mp (Amersham) with intensifying screens.

Results and discussion GUS transient expression N107x P. nigra resulted in a higher average number of GUS spots per explant (15.6) compared to N084 x P. nigra (9.0). Co-cultivation with A281 pKIWI105 and LBA4404 pKIWI105 showed the highest (38.5) and the lowest (0.4) average number of transformation events, respectively. These data are similar to those obtained by cocultivating two other P. nigra clones using the same strains (Confalonieri et al., 1994). GV2260 p35S GUS and C58 pKIWI105 gave a low TDNA transfer efficiency: the average number of GUS spots per explant was 7.2 and 3.2, respectively. No differences were observed between the two different bacterial concentrations tested.

218

Fig. 1. Results of Southern blot hybridization analysis performed on transformed poplar lines. The EcoRV fragment (1.2 kbp) containing the coding region of the nptll gene was used as a probe. Poplar genomic DNA was digested with EcoRV. The following poplar lines were analyzed: (A) lane 1: negative control (untransformed poplar plant); lane 2: a plant line of Nl07x P. nigra transformed with GV2260 p35S GUS; lanes 3 to 6: different P. nigra lines (San Giorgio, Jean Pourtet, N084 x P. nigra and N 107 x P. nigra) transformed with A281 pKIWI 105; lane 7: positive control (the EcoRV fragment which was purified from agarose gel, following digestion of plasmid pABDI with EcoRV). (B) lane 1: negative control (untransformed poplar callus); lanes 2 to 5: different P. nigra callus oncogenic lines (San Giorgio, Jean Pourtet, N084 x P. nigra and N107x P. nigra, respectively) transformed with A281 pKIWI105 strain; lane 6: positive control (EcoRV fragment obtained as previously described).

Recovery of stable transformants While T-DNA transfer in poplar cells was quite successful according to the transient expression data, no stable integration was recorded for most of the tested clones (data not shown). In our hands, GUS transient expression was not a reliable indicator of the ability of GV2260 p35S GUS to mediate stable transformation. However, one transformed plant of N107x P nigra was obtained. By contrast, leaf explants co-cultivated with A281 pKIW1105 produced a large number of cal-

li. Most of these calli died following their culture for one month on selective regeneration medium. It is possible that the effect of kanamycin selection on poplar tissues was not immediate, so that untransformed calli and/or calli which were transformed only with oncogenic T-DNA were able to grow on leaf discs. These calli browned and died only after their excision and culture on selective regeneration medium. Approximately 6-29% of the isolated calli were kanamycin resistant, depending on the plant clone and bacterial concentration used (Table 1). Kanamycin resistant cal-

219

Fig. 2. Results of Southern blot hybridization analysis performed on oncogenic calli and plantlets transformed with A281 pKIW1105 strain, to detect the presence of oncogenic T-DNA. The BamHl fragment (10kbp), derived from agropine Ti plasmid (pTiBo542), was used as a probe. Lane 1 : negative control (untransformed poplar DNA; lanes 2, 4, 6 and 8: DNA from a transformed plant line of different P. nigra clones (San Giorgio, Jean Pourtet, NO84 x P nigra and NI07 x P nigra, respectively). Lanes 3, 5, 7 and 9: DNA from oncogenic callus line of the above mentioned clones; lane 10: positive control (the BamHI fragment which was purified from agarose gel, following digestion of plasmid pEHI36 with BamHl).

Table 1. Effects on transformation efficiency of different P nigra clones following co-cultivation of 50 leaf discs per treatment combination with A281 pKIWI 105 Clone

Bacterial

Calli

concentration obtained (bacteria m l - t) (no.)

Calli kanamycin resistant Total no,(%)

Plantlets

obtained GUS+ GUS+/- Transformation (no.) % % frequency (%) '~

Plantlets kanamycin resistant Total no.(%)

GUS+ Transformation % frequency (%) "

N084 x P nigra

7 x 108 1.2 ×109

242 110

21(8.7) 14(12.7)

42.8 35.7

47.6 57.1

26 18

39 7

32(82.0) 2(28.6)

0,0 50.0

6 4

N107 x P. nigra

7 x l08 1.2 x 109

131 498

38(29.0) 135(27.0)

60.5 54.1

28.9 25.2

28 68

0 57

53(92.9)

50.9

10

SanGiorgio

7 x 108 1,2 x 109

235 32

64(27.2) 2(6.2)

43.7 50.0

43.7 0,0

28 4

46 0

43(93.5)

46,5

12

Jean Pourtet

7 xl08 1.2 x 109

91 220

12(13.2) 24(10.9)

66.6 45.8

33.3 25.0

8 26

2 14

0 0

a Calculated based on the number of leaf discs producing kanamycin resistant calli or plantlets as a percentage of the total of co-cultivmed leaf discs.

li were also able to grow on selective WPM medium in the absence of growth regulators, suggesting that oncogenie T-DNA was transferred and expressed in these calli. Control calli (not co-cultivated) did not survive on the same medium either with or without kanamycin. Selected calli derived from co-cultivation with A281 pKIWI105 strain were also assayed for GUS activity. Most of them showed either GUS + or GUS + / (both GUS + and G U S - expression), while only a few

calli were GUS- (Table 1). The lack of GUS activity could indicate that the uidA gene was either absent or present, but not expressed. No GUS expression was detected in control (not co-cultivated) calli. The frequency of leaf discs producing kanamycin resistant calli was not significantly different between the clones and bacterial concentrations used. As shown in Table 1, N107x P nigra showed the highest transformation frequency (68%) with A281 pKIW1105 at the concert-

220 Table 2. Influence of different leaf wounding procedures, bacterial concentrations, and the presence or absence of acetosyringone on genetic transformation of two clones of P, nigra co-cultivated with Agrobacterium tumefaciens strain A281 pKIW1105. Clone

San Giorgio

Leaf discs Bacterial Presence (+)/ Calli (N)I leaf discs concentration absen~ ( - ) of obtained punctered (P) (bacteria ml -~) acetosyringone (no.)

N

P

Jean Pourtet

N

P

7 xl08

+

1.2 × 1 0 9

+

7 x108

+

1.2x109

+

7 xl0 s

+

1.2 xl09

+

7 x108

+

1.2 x109

+

Catli kanamycin resistant Total no.(%)

Plantlets Planflets kanamycin resistant obtained GUS+ GLIS+I- Transformation (no.) Total GUS+ Transtbrmation % % frequency (%)" no.(%) % frequency(%)

320 235 198 32 624 451 963 126

118(36.8) 64(27.2) 73(36.9) 2(6.2) 216(34.6) 123(27.3) 306(31.8) 33(26.2)

15.2 43.7 21.9 50.0 49.1 45.5 48.4 27.3

63.5 43.7 75.3 0.0 40.1 24.4 37.9 30.3

68 28 48 4 80 42 82 30

93 46 38 0 41 91 41 18

46 91 241 220 384 18 173 154

9(t9.6) 12(13.2) 37(15.4) 24(10.9) 66(17.2) 2(1t.1) 31(17.9) 13(8.4)

22.2 66.6 29.7 45.8 68.2 50.0 58.1 30.8

66.6 33.3 51,3 25.0 24.2 50.0 35.5 69.2

18 8 40 26 66 4 42 16

0 2 0 14 4 0 10 0

82(88.2) 43(93.5) 35(92.1)

26.8 46.5 71.4

20 t2 14

34(82.9) 78(85.7) 34(82.9) t0(55.5)

23.5 56.4 58.8 100

24 8 40 4

0 0 3(75)

I00

10(100)

80

a Calculated based on the number of leaf discs producing kanamycin resistant calli or plantlets as a percentage of the total number of co-cultivated leaf discs,

tration of 1.2 x 109 bacteria m1-1. With Jean Pourtet, the highest value (26%) was obtained using the same bacterial concentration. For San Giorgio and N084x P nigra, transformation efficiency of 28% and 26%, respectively, were observed following co-cultivation with a concentration of 7 x 108 bacteria ml-1. Regeneration of transformed shoots was obtained from San Giorgio, N107x P nigra and N084 × P nigra clones. Transgenic plantlets were analyzed for GUS activity. As shown in Table 1, 50% or more of them were GUS negative. In this experiment, the rate of leaf discs producing kanamycin resistant plantlets was 12%, at the maximum. Moreover, a few kanarnycin resistant shoots with abnormal morphology (reduced internodes, small leaves) were obtained from San Giorgio and N107x P nigra clones. They developed callus and new teratomic shoots and did not root when transferred to medium lacking growth regulators (data not shown). Similar results were reported by Brasileiro et al. (1991). Two different A. tumefaciens strains, bacterial concentrations, leaf wounding procedures and the presence of the externally added acetosyringone were tested, in order to improve the transformation efficiency of the San Giorgio and Jean Pourtet clones. In our hands, GV2260 p35S GUS strain did not produce sta-

ble transformants of two P, nigra clones. Following cocultivation of P. nigra explants with A281 pKIWI105 strain, however, a large number of calli were obtained and some transgenic plantlets were regenerated (Table 2). The frequency of leaf discs producing kanamycin resistant calli was significantly (a < 0.05) dependent on the clone. As shown in Table 3, an average transformation frequency of about 48% was observed for San Giorgio after co-cultivation with A281 pKIWI105, while for Jean Pourtet, the frequency of transformation was lower (27.5%). In nearly all comparable treatments, San Giorgio was the most responsive clone (Table 2). De Jong et aL (1993) reported that co-cultivation of punctured chrysanthemum leaf explants with A. tumefaciens reduced regeneration of adventitious shoots. In co-cultivation experiments of punctured leaf discs of San Giorgio and Jean Pourtet with A281 pKIWI105, however, the average frequency of explants producing kanamycin resistant calli and plantlets was not significantly different between punctured and unpunctured leaf discs (Table 3). In addition, no differences were observed between punctured and unpunctured leaf discs of San Giorgio and Jean Pourtet when they were co-cultivated with GV2260 p35S GUS (data not shown). With regard to the bacterial concentration of A281 pKIWI105, no significant differences

221 Table 3. Effects of several factors on the transformation efficiency of leaf discs of P nigra (San Giorgio, Jean Partout) clones co-cultivated with Agrobacterium tumefilciens strain A281 pKIWI105. Average transformation frequency and significance level of differences for each of the factors considered in the ANOVA are presented Factors/Levels

Frequency of leaf discs

Frequency of leaf discs

producing kanamycin-

producing kanamycin-

resistant calli %

o~

resistant plantlets %

o~

P. nigra clone

San Giorgio

47.8

Jean Pourtet

27.5

<_0.023

15.3 1.0

<0.0002

Leaf wounding procedure normal leaf discs

30.0

punctered leaf discs

45.3

5.8


10.5

<0.778 -

7.8

<0.078 -

Bacterial concentration 7 x l08 bacteria/ml

39.3

1.2 x lO9 bacteria/ml

36.0

8.5

<0.464

Acetosyringone level 0

19.8

200 ~tM

55.5

were found (a _< 0.05). Following co-cultivation with two bacterial concentrations (7 x 108 and 1.2 x 109 bacteria ml- l), 39.3% and 36.0% of the leaf discs, respectively, produced kanamycin resistant calli (Table 3). Acetosyringone has been reported to increase transformation efficiency in Atropa belladona (Mathews et al., 1990) and Cyphomandra betacea (Atkinson & Gardner, 1993). In our experiments, the addition of acetosyringone (200 ~M) to the bacterial suspension of A281 pKIWI105 significantly enhanced (a < 0.01) the average frequency of leaf discs producing kanamycin resistant calli and plantlets (55.5% and 13.3%, respectively) when compared to the control (19.8% and 3.0%, respectively) (Table 3). By contrast, no positive effects were scored when acetosyringone was used with the bacterial suspension of GV2260 p35S GUS strain (data not shown).

<0.OOO6 -

3.0 13.3

<0.005 -

Molecular analyses Southern blot hybridization analysis was performed on DNA samples extracted from different poplar lines transformed with A281 pKIWI 105 and GV2260 p35S GUS. Results of analyses using the 1.2 kbp fragment containing the coding region of the nptll gene as a probe are shown in Fig. 1. Fragments hybridizing to the nptll probe were found in all kanamycin resistant calli and plants tested, thus indicating the presence of nptlI gene in poplar genome. The hybridization bands appeared to be higher in size than 1.2 kbp. McClelland & Nelson (1988) reported that EcoRV activity is sensitive to modification of adenine to N6-methyladenine occuring at the recognition sequence. This modification has been already detected in cereals (Rogers & Rogers, 1995) and it could represent the possible explanation for our results. A Southern blot hybridization analysis was also performed on poplar genomic DNA in order to detect the presence and integration of pTiBo542 T-DNA (derived from A281 pKIWI105

222 strain) in oncogenic callus lines of San Giorgio, Jean Pourtet, N 0 8 4 x P. nigra and N 1 0 7 x P. nigra clones, and in transformed plants regenerated from the same lines. The BamHI fragment 5 of pTiBo542 (10 kbp) was detected in the genome of oncogenic callus of all the P nigra lines we have tested, but it was not found in the genome of transformed plantlets and the untransformed control (Fig. 2). The in vitro growth of oncogenic poplar calli on selective medium lacking growth regulators and the molecular characterization of these calli demonstrated that they originated from cells transformed with sequences from both the pKIWI105 binary vector and pTiBo542 T-DNA. Moreover, the molecular analysis of transgenic poplar plantlets indicated that they originated either from cells transformed solely with the pKIWI105 binary vector or, possibly, from cells which were initially transformed with both the binary vector and oncogenic T-DNA, but which had lost the latter during shoot organogenesis (Brasileiro et al., 1991). This work showed that some factors such as P. nigra clone, strain of A. tumefaciens and addition of acetosyringone affect transformation efficiency of P. nigra by A. tumefaciens. Other factors examined were leaf wounding and bacterial concentration but they did not influence the transformation frequency. Although an oncogenic strain and chimeric genes were used in this study, the methodology should be suitable also for genes that could improve agronomic traits.

Acknowledgements We thank M. Rondanin and R. Bruschini for their technical contribution. This research was supported by the Italian Ministry of Agriculture in the framework of the project "Resistenze genetiche delle piante agrarie agli stress biotici e abiotici" and by a financial contribution of the European Commission within the framework of the European Community RTD Programme F O R E S T (Contract N. MA2B-CT91--0029).

References Atkinson RG & Gardner RC (1993) Regeneration of transgenic tamarillo plants. Plant Cell Rep. 12:347-351 Brasileiro ACM, Lepl6 J-C, Muzzin J, Ounnoughi D, Michel M-F & Jouanin L (1991) An alternative approach for gene transfer in trees using wild-typeAgrobacterium strains. Plant Mol. Biol. 17:441--452 Confalonieri M, Balestrazzi A & Bisoffi S (1994) Genetic transformation of Popuhts nigra by Agrobacteriztm tttmefilciens. Plant Cell Rep. 13:256-261 De Jong J, Rademaker W & van Wordragen MF (1993) Restoring adventitious shoot formation on chrysanthemum leaf explants followingco-cultivationwith Agrobacterium tumefiwiens. Plant Cell Tiss. Org. Cult. 32:263-270 Janssen BJ & Gardner RC (1989) Localized transient expression of GUS in leaf discs following cocultivation with Agrobacteri,m. Plant Mol. Biol. 14:61-72 Jefferson RA (1987) Assaying chimeric genes in plants: the GUS gene fusion system. Plant Mol. Biol. Rep. 5:387-405 Lloyd G & McCown BH (1981) Commerciallyfeasible micropropagation of mountain laurel, Kalmia lattfolia, by use of shoot-tip culture. Proc. Int. Plant Prop. 30:421-427 Mathews H, Bharathan N, Litz RE, NarayananKR, Rao PS & Bathia CR (1990) The promotion of Agrobacterium mediated transformation in Atropa belladona L. by acetosyringone. J. Plant Physiol. 136:404-409 McClelland M & Nelson M (1988) The effect of site-specific DNA methylationon restrictionendonucleasesand DNA modification methyltransferases - a review. Gene 74 : 291-304 Murashige T & Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15: 473-497 Rogers JC & Rogers SW (1995) Comparison of the effects of N 6methyldeoxyadenosine and NS-methyldeoxycytosineon transcription from nuclear gene promoters in barley. Plant Journal 7:221-223 Rogers SO & Bendich AJ (1988) Extraction of DNA from plant tissues. Plant Molecular Biology Manual A6:1-10 Vancanneyt G, Schmidt R, O Connor-Sanchez A, Willmitzer L & Rocha-Sosa M (1990) Construction of an intron - containing marker gene: Splicing of the intron in transgenic plants and its use in monitoring early events in Agrobacterium - mediated plant transformation.Mol. Gen. Genet. 220:245-250