Production of transgenic diploid <Emphasis Type="Italic">Cucumis melo</Emphasis> plants

Melon (Cucumis melo L.) is considered to be a recalcitrant species for genetic transformation. Additionally, many studies have observed that regenerated transgenic plants are frequently polyploids. Here we have studied several aspects of melon transformation with the aim of improving transformation efficiency and producing diploid transformed plants. The protocol was based on using cotyledon explants from quiescent seeds that retain meristematic cells, which facilitated the regeneration of transformed diploid melon plants. In this study we evaluated the effect of using two different explant types from the proximal portion of melon seeds on the ploidy status (evaluated by flow cytometry) of regenerated plants. We also determined the transformation efficiencies obtained with these types of explants from four different genotypes. Regeneration was obtained from all explant types. Using quiescent seeds the percentage of diploid plants produced ranged from 85.2 to 94.1%, depending on the type of explant. On the other hand, only half of the plants regenerated from older-seed cotyledons (2- or 3-day-old) were diploids. Transgenic plants were produced with variable transformation efficiencies depending on the explant and which of the four melon genotypes was used. The explants with the best behavior produced transgenic plants with the highest efficiencies ever published both, in terms of plants expressing the visual marker transgenes (ranging from 4.5 to 15.4%) and the number of rooted plants in selective medium (ranging from 1.3 to 3.8%). Although the transformation efficiencies were still relatively low, they were consistent for the four very different melon genotypes tested. Furthermore, at least 85% of plants produced were diploid.     

Agrobacterium tumefaciens mediated transformation and regeneration of muskmelon plants

Transgenic muskmelon (Cucumis melo L.) plants were produced efficiently by inoculating cotyledon explants with Agrobacterium tumefaciens strain LBA4404 bearing a Ti plasmid with the NPT II gene for kanaymcin resistance. After co-cultivation for three days, expiants were transferred to melon regeneration medium with kanamycin to select for transformed tissue. Shoot regeneration occurred within 3–5 weeks; excised shoots were rooted on medium containing kanamycin before transferring to soil. Morphologically normal plants were produced in three months. Southern blot analysis confirmed that ca. 85% of the regenerated plants contained the NPT gene. Dot blot analysis and leaf callus assay of progeny of transgenic plants verified transmission of the introduced gene(s) to the next generation. Factors affecting transformation efficiency are discussed.Abbreviations ABA abscisic acid - BAP 6-benzylaminopurine - IAA indole 3 acetic acid - 2,4-D 2,4-dichlorophenoxyacetic acid - NPT II neomycin phosphotransferase II     

Transformation of cucumber (Cucumis sativus L.) plants using Agrobacterium tumefaciens and regeneration from hypocotyl explants

Summary Transgenic cucumber (Cucumis sativus L.) plants were successfully obtained from hypocotyl explants inoculated with Agrobacterium tumefaciens, which harbored a binary vector plasmid with NOS-nptII, CaMV 35S-I-gus and CaMV 35S-hph genes. Acetosyringone enhanced the efficiency of transformation at the cut surface cells of hypocotyl explants during five days of co-cultivation. Transformed cells were more effectively selected using 20–30 mg/l hygromycin B than using 50–100 mg/l kanamycin. Shoot regeneration occurred within 4–6 wks, and 12 of 21 regenerated plantlets displayed strong GUS expression in the very young leaves. All of 8 GUS-positive R0 plants examined showed single or a few positive bands by Southern blot analysis. The expression of the CaMV 35S-I-gus gene was observed in various tissues and organs of R0 and R1 transgenic cucumber plants.     

Enhanced regeneration of tomato and pepper seedling explants for Agrobacterium-mediated transformation

Seedling explants of three tomato (Lycopersicon esculentum) and four bell pepper (Capsicum annuum) cultivars consisting of the radicle, the hypocotyl and one cotyledon were obtained after removing the primary and axillary meristems. After 14 days of incubation on solid Murashige and Skoog (MS) medium without growth regulators, explants of both species regenerated multiple shoots on the cut surface (2.9–5.3 shoots per explant for tomato and 1.2–2.2 for bell pepper cultivars). After excision, the shoots were rooted on solid MS medium and acclimated to the greenhouse. This method was highly efficient in tomato and, particularly, in bell pepper, where plant regeneration is especially difficult. We used these explants to transform tomato with Agrobacterium tumefaciens containing a 35S-GUS-intron binary vector. As shown by GUS expression, 47% of the tomato explants produced transformed meristems, which differentiated into plants that exhibited a low (3%) tetraploidy ratio. Southern blots and analysis of inheritance of the foreign genes indicated that T-DNA was stably integrated into the plant genome. The use of this technique opens new prospects for plant transformation in other dicotyledoneous plants in which genetic engineering has been limited, to date, due to the difficulties in developing an efficient in vitro regeneration system.     

Plant regeneration <Emphasis Type="Italic">in vitro</Emphasis> directly from cotyledon and hypocotyl explants of <Emphasis Type="Italic">Perilla frutescens</Emphasis> and their morphological aspects

A rapid plantlet regeneration system for Perilla frutescens was established from cotyledon and hypocotyl explants. A maximum of 91.06 % cotyledon and 76.4 % hypocotyl explants could directly produce shoots (3.09 ± 0.18 shoots per explants) on Murashige and Skoog (MS) medium. The optimum hormone combinations were 4.44 μM 6-benzylaminopurine (BA) for cotyledon and 2.22 μM BA + 2.85 μM indole-3-acetic acid (IAA) for hypocotyls. Rooting was induced on half-strength hormone-free MS medium. After transplantation to soil, approximate 80 % of the regenerated plantlets could survive, flower and fruit. Moreover, some morphological abnormalities were found among the regenerated plants.     

Analysis of mannose selection used for transformation of sugar beet

Various factors affecting mannose selection for the production of transgenic plants were studied using Agrobacterium tumefaciens-mediated transformation of sugar beet (Beta vulgaris L.) cotyledonary explants. The selection system is based on the Escherichia coli phosphomannose isomerase (PMI) gene as selectable gene and mannose as selective agent. Transformation frequencies were about 10-fold higher than for kanamycin selection but were only obtained at low selection pressures (1.0–1.5 g/l mannose) where 20–30% of the explants produced shoots. The non-transgenic shoots were eliminated during the selection procedure by a stepwise increase in the mannose concentration up to 10 g/l. Analysis of the transformed shoots showed that the PMI activity varied from 2.4 mU/mg to 350 mU/mg but the expression level was independent of the selection pressure. Complete resistance to mannose of transformed shoots was observed already at low PMI activities (7.5 mU/mg). Genomic DNA blot analysis confirmed the presence of the PMI gene in all transformants analysed. The possible mode of action of mannose selection compared to other selection methods is discussed.     

Plant selection principle based on xylose isomerase

Summary The xylose isomerase genes (xylA) from Thermoanaerobacterium thermosulfurogenes and Streptomyces rubiginosus were introduced and expressed in three plant species (potato, tobacco and tomato) and transgenic plants were selected on xylose-containing medium. The xylose isomerase genes were transferred to explants of the target plant by Agrobacterium-mediated transformation. The xylose isomerase genes were expressed under the control of the enhanced cauliflower mosaic virus 35S promoter and the Ω′ translation enhancer sequence from tobacco mosaic virus. In potato and tomato, xylose isomerase selection was more efficient than the established kanamycin selection. The level of enzyme activity in the regenerated transgenic plants selected on xylose was 5–25-fold higher than the enzyme activity in control plants selected on kanamycin. The xylose isomerase system enables transgenic cells to utilize xylose as a carbohydrate source. In contrast to antibiotic or herbicide resistance-based system where transgenic cells survive on a selective medium but nontransgenic cells are killed, the xylose system is an example of a positive selection system where transgenic cells proliferate while non-transgenic cells are starved but still survive. The results show that a new selection method, is established. The xylose system is devoid of the disadvantages of antibiotic or herbicide selection, and depends on an enzyme which is already being widely utilized in specific food processes and that is generally recognized as safe for use in the starch industry.     

Leaf disc transformation of cultivated tomato (L. esculentum) using Agrobacterium tumefaciens

The leaf disc transformation/regeneration system was modified for tomato (L. esculentum). Both leaf explants and cotyledon/hypocotyl sections can be used to regenerate transformed plants. We have obtained over 300 transgenic plants from eight tomato cultivars. We have evidence for both single and multi-copy insertions of the T-DNA, and have demonstrated inheritance of the T-DNA insert in the expected Mendelian ratios. Several heterologous promoters function in tomato. A reduced efficiency of transformation was observed with binary T-DNA vectors as compared to co-integrate T-DNA vectors. The ease of the leaf disc method makes tomato a premier experimental organism for plant biotechnology.Abbreviations BA benzyl adenine - IAA indole acetic acid - LB Luria Broth     

Agrobacterium tumefaciens-mediated transformation of the aromatic shrub Lavandula latifolia

Transgenic plants of the aromatic shrub Lavandula latifolia (Lamiaceae) were produced using Agrobacterium tumefaciens-mediated gene transfer. Leaf and hypocotyl explants from 35–40-day old lavender seedlings were inoculated with the EHA105 strain carrying the nptII gene, as selectable marker, and the reporter gusA gene with an intron. Some of the factors influencing T-DNA transfer to L. latifolia explants were assessed. Optimal transformation rates (6.0 ± 1.6% in three different experiments) were obtained when leaf explants precultured for 1 day on regeneration medium were subcultured on selection medium after a 24 h co-cultivation with Agrobacterium. Evidence for stable integration was obtained by GUS assay, PCR and Southern hybridisation. More than 250 transgenic plants were obtained from 37 independent transformation events. Twenty-four transgenic plants from 7 of those events were successfully established in soil. -glucuronidase activity and kanamycin resistance assays in greenhouse-grown plants from two independent transgenic lines confirmed the stable expression of both gusA and nptII genes two years after the initial transformation. Evidence from PCR data, GUS assays and regeneration in the presence of kanamycin demonstrated a 1:15 Mendelian segregation of both transgenes among seedlings of the T1 progeny of two plants from one transgenic L. latifolia line.     

Genetic transformation of <Emphasis Type="Italic">Torenia fournieri</Emphasis> using the PMI/mannose selection system

Transgenic torenia plants were obtained using the selectable marker gene phosphomannose isomerase (manA), which encodes the enzyme phosphomannose isomerase (PMI) to enable selection of transformed cells on media containing mannose. We found that shoot organogenesis in torenia leaf explants was effectively suppressed on medium supplemented with mannose, which indicated that torenia cells had little or no PMI activity and could not utilize mannose as a carbon source. Leaf pieces from in vitro-germinated plants were inoculated with Agrobacterium tumefaciens EHA105 containing the binary vector pKPJ with both hpt and ManA genes, and subsequently selected on shoot induction (SI) medium (half strength MS basal + 4.4 μM BA + 0.5 μM NAA) supplemented with 20 g l⁻¹ mannose and 5 g l⁻¹ sucrose as carbon sources. Transformed plants were confirmed by PCR and Southern blot. The transgene expression was evaluated using Northern blot and the chlorophenol red assay. The transformation efficiency ranged from 7% to 10%, which is 1–3% higher than that obtained by selection with hygromycin. This system provides an efficient manner for selecting transgenic flower plants without using antibiotics or herbicides.     

The ploidy level of transgenic plants in <Emphasis Type="Italic">Agrobacterium</Emphasis>-mediated transformation of tomato cotyledons (<Emphasis Type="Italic">Lycopersicon esculentum</Emphasis> L.Mill.) is genotype and procedure dependent

A protocol avoiding the feeder-layer cell system was optimized for Agrobacterium-mediated transformation of tomato cotyledonary explants. Over 500 transgenic plants from five tomato cultivars were regenerated in 15 independent experiments. Depending on both genotype and procedure, transformation frequencies ranged from 1.8% to 11.3%. The optimal transformation rate was obtained by inoculating explants with a bacterial suspension in exponential growth ( D(600) = 10(2)-10(3) cells/ml) and transferring cotyledon explants to fresh selective regeneration medium every 3 weeks. The ploidy level of both tomato genotypes used as explant source and primary transformants, was studied by flow cytometry. The inbred lines and cultivars were diploid but a polysomatic pattern in the cotyledon explant was confirmed. The rate of tetraploid transgenic plants ranged from 24.5% to 80% and depended on both the genotype and the transformation procedure. Surprisingly, the percentages of transformed plants with higher ploidy levels were not related to the proportion of 4C and 8C nuclei in the cotyledonary tissue. For some genotypes the optimisation of the transformation rate resulted in an increase of tetraploid transgenic plants. Results obtained in this work indicate the convenience of checking the ploidy level of the primary transformants before performing basic studies or introducing tomato transgenic material in a breeding program.     

Biochemical and phenotypical characterization of transgenic tomato plants overexpressing a basic peroxidase

Tomato plants ( Lycopersicon esculentum Mill. cv. Pera) were transformed via Agrobacterium tumefaciens with the binary vector pKYLX71 containing a tomato basic peroxidase (EC 1.11.1.7) gene, tpx1 , under the control of the cauliflower mosaic virus (CaMV35S) promoter. Transgenic plants showed a 2–5-fold increase in the activity of the peroxidase ionically bound to the cell wall, whereas soluble peroxidase activity remained similar or even lower than wild-type plants. Isoelectric focusing showed the presence of a new isoperoxidase of pI ca 9 in the ionically bound extract. Western blot also showed the presence of a new band at 41 kDa that was absent in the wild-type extract. A 40–220% increment of lignin content of the leaf was found in transgenic plants. Shoot phenotype of transgenic plants was similar to wild type, although under stress, the plants appeared wilted and the new leaves had a reduced area and were thicker than wild-type or older transgenic leaves. The root system was underdeveloped in transgenic plants, but the rooting ability of the stem was not affected by the overexpression of peroxidase. Finally, the morphogenetic response of cotyledon and hypocotyl explants from transgenic plants was evaluated. In the case of cotyledons, the percentage of explants with shoot was not different from wild-type plants. For hypocotyl, one of the transgenic lines showed a 30% reduction in the percentage of shoot organogenesis. The results are discussed in relation to the role of tpx1 in lignin synthesis.     

Transgenic Acacia sinuata from Agrobacterium tumefaciens-mediated transformation of hypocotyls

Transgenic herbicide tolerant Acacia sinuata plants were produced by transformation with the bar gene conferring phosphinothricin resistance. Precultured hypocotyl explants were infected with Agrobacterium tumefaciens strain EHA105 in the presence of 100 μM acetosyringone and shoots regenerated on MS (Murashige and Skoog, 1962, Physiol Plant 15:473–497) medium with 13.3 μM benzylaminopurine, 2.6 μM indole-3-acetic acid, 1 g l⁻¹ activated charcoal, 1.5 mg l⁻¹ phosphinothricin, and 300 mg l⁻¹ cefotaxime. Phosphinothricin at 1.5 mg l⁻¹ was used for the selection. Shoots surviving selection on medium with phosphinothricin expressed GUS. Following Southern hybridization, eight independent shoots regenerated of 500 cocultivated explants were demonstrated to be transgenic, which represented transformation frequency of 1.6%. The transgenics carried one to four copies of the transgene. Transgenic shoots were propagated as microcuttings in MS medium with 6.6 μM 6-benzylaminopurine and 1.5 mg l⁻¹ phosphinothricin. Shoots elongated and rooted in MS medium with gibberellic acid and indole-3-butyric acid, respectively both supplemented with 1.5 mg l⁻¹ phosphinothricin. Micropropagation of transgenic plants by microcuttings proved to be a simple means to bulk up the material. Several transgenic plants were found to be resistant to leaf painting with the herbicide Basta.     

The use of phenotypic markers to identify Brassica oleracea genotypes for routine high-throughput Agrobacterium-mediated transformation

Doubled haploid (DH) genotypes from a genetic mapping population of Brassica oleracea were screened for ease of transformation. Candidate genotypes were selected based on prior knowledge of three phenotypic markers: susceptibility to Agrobacterium tumefaciens, shoot regeneration potential and mode of shoot regeneration. Mode of regeneration was found to be the most significant of the three factors. Transgenic plants were successfully obtained from genotypes that regenerated multiple shoots via a distinct swelling or callus phase. The absence of tissue culture blackening (associated with genotypes that formed callus) was found to be critical for transformation success. Transgenic shoots were obtained from genotypes that regenerated via an indirect callus mode, even when susceptibility to Agrobacterium was low. The most efficient genotype (DH AG1012) produced transgenic shoots at an average rate of 15% (percentage of inoculated explants giving rise to transgenic plants). The speed and efficiency of regeneration enabled the isolation of transgenic shoots 5–6 weeks after inoculation with A. tumefaciens. This line was also self-compatible, enabling the production of seed without the need for hand-pollination. A genetically uniform DH genotype, with an associated genetic map, make DH AG1012 highly desirable as a potential model B. oleracea genotype for studying gene function. The possibility of applying the same phenotypic tissue culture markers to other Brassica species is discussed.     

Conditions of transformation and regeneration of `Induka' and `Elista' strawberry plants

Efficiency of plants' transformation depends on many factors. The genotype, applied techniques and conditions of plant's modification and modified plant regeneration are the most important among them. In our studies regeneration and transformation conditions for two strawberry cultivars were determined and compared. Plants were transformed by Agrobacterium tumefaciens LBA₄₄₀₄ strain containing plasmid pBIN19 with nptII and gus-reporter genes. Experiment was carried out on more than 1300 leaf explants from each cultivar. Generally, `Induka' plants characterized with higher regeneration potential than `Elista'. The highest number of regenerated shoots was obtained on MS medium with 0.4 mg l ^–1 IBA and 1.8 mg l^–1 BA (3.5 and 1.8 shoots/explant for `Induka' and `Elista', respectively). After plant transformation number of regenerated, transgenic shoots was higher for `Elista' (on the average: 8.3 shoots/100 explants). The number of transgenic `Induka' shoots, obtained at the same conditions, was twice lower (4.2). Simultaneously `Induka' plants needed higher kanamycin concentration for transgenic explants selection than `Elista' (25 mg l^–1). Preliminary incubation of A. tumefaciens in LB or MS medium with acetosyringone and IAA resulted in increasing transgenic shoots number (per 100 explants: `Induka' 4.5, `Elista' 8.0–9.5 shoots). After using untreated bacteria for plants' transformation, number of transgenic plants varied (dependently on cultivar) from 3.8 to 7.0/100 explants. Applying LB or MS as basic medium as well as adding tobacco plant extract to these media did not significantly influence transformation efficiency.     

Plant regeneration and stable transformation in the floricultural plant Cleome spinosa, a C3 plant closely related to the C4 plant C. gynandra

Cleome spinosa is widely used as a garden ornamental in many countries. Here we determined the optimal conditions for plant regeneration from different tissue explants grown in vitro. Induction medium containing MS salts, MS vitamins, 3% sucrose, 1 mg l?1 BA, 200 mg l?1 timentin, and 0.8% agar was sufficient for shoot regeneration of all the tissue explants examined, including leaf, hypocotyl, and cotyledon. Subsequently, an Agrobacterium tumefaciens-mediated method was developed to transform the vector pCHS, which carries the transgenes Petunia chalcone synthase (chs) and selection marker neomycin phosphotransferase II (nptII), into C. spinosa. From a total of 368 cotyledon explants, 13 putative transgenic lines were regenerated from selection medium supplemented with 50 mg l?1 kanamycin and 200 mg l?1 timentin, and transferred to the greenhouse. Genomic PCR and Southern blot analyses revealed that the nptII transgene was present in all 13 transgenic plants. Similarly, when the Petunia chs transgene was used as a probe in Southern blot analysis, single or multiple hybridization bands were detected in 12 out of the 13 transgenic plants. In addition, T? progeny assay from selected transformants showed that the nptII transgene can be transmitted in a Mendelian manner from transgenic parents into their progeny. This is the first report of stable transformation of the C? dicotyledon C. spinosa, which will facilitate functional comparison of cell-type specific genes with counterpart C? dicotyledon C. gynandra using transgenic approaches.     

High frequency organogenesis in hypocotyl,cotyledon, leaf and petiole explants of broccoli (Brassica oleracea L. var. italica), an important vegetable crop

Broccoli (Brassica oleracea L. var. italica) is an important, nutritionally rich vegetable crop, but severely affected by environmental stresses, pests and diseases which cause massive yield and quality losses. Genetic manipulation is becoming an important method for broccoli improvement. In the present study, a reproducible and highly efficient protocol for obtaining organogenesis from hypocotyl, cotyledon, leaf and petiole explants of broccoli (Brassica oleracea L. var. italica cv. Solan green head) has been developed. Hypocotyl and cotyledon explants were used from 10 to 12 days old aseptically grown seedlings whereas leaf and petiole explants were excised from 18 to 20 days old green house grown seedlings and surface sterilized. These explants were cultured on shoot induction medium containing different concentration and combination of BAP and NAA. High efficiency shoot regeneration has been achieved in hypocotyl (83.33 %), cotyledon (90.11 %), leaf (62.96 %) and petiole (91.10 %) explants on MS medium supplemented with 3.5 mg/l BAP + 0.019 mg/l NAA 2.5 mg/l BAP + 0.5 mg/l NAA, 4.0 mg/l BAP + 0.5 mg/l NAA and 4.5 mg/l BAP + 0.019 mg/l NAA respectively. Petiole explants showed maximum shoot regeneration response as compared to other explants. MS medium supplemented with 0.10 mg/l NAA was found best for root regeneration (100 %) from in vitro developed shoots. The regenerated complete plantlets were transferred to the pots containing cocopeat and successfully acclimatized. This optimized regeneration protocol can be efficiently used for genetic transformation in broccoli. This is the first comparative report on multiple shoot induction using four different types of explants viz. hypocotyl, cotyledon, leaf and petiole.     

Development of a phosphomannose isomerase-based <Emphasis Type="Italic">Agrobacterium</Emphasis>-mediated transformation system for chickpea (<Emphasis Type="Italic">Cicer arietinum</Emphasis> L.)

To develop an alternative genetic transformation system that is not dependent on an antibiotic selection strategy, the phosphomannose isomerase gene (pmi) system was evaluated for producing transgenic plants of chickpea (Cicer arietinum L.). A shoot morphogenesis protocol based on the thidiazuron (TDZ)-induced shoot morphogenesis system was combined with Agrobacterium-mediated transformation of the pmi gene and selection of transgenic plants on mannose. Embryo axis explants of chickpea cv. C-235 were grown on a TDZ-supplemented medium for shoot proliferation. Embryo axis explants from which the first and second flush of shoots were removed were transformed using Agrobacterium carrying the pmi gene, and emerging shoots were allowed to regenerate on a zeatin-supplemented medium with an initial selection pressure of 20 g l⁻¹ mannose. Rooting was induced in the selected shoots on an indole-3-butyric acid (IBA)-supplemented medium with a selection pressure of 15 g l⁻¹ mannose. PCR with marker gene-specific primers and chlorophenol red (CPR) assay of the shoots indicated that shoots had been transformed. RT-PCR and Southern analysis of selected regenerated plants further confirmed integration of the transgene into the chickpea genome. These positive results suggest that the pmi/mannose selection system can be used to produce transgenic plants of chickpea that are free from antibiotic resistance marker genes.     

MicroTom—a high-throughput model transformation system for functional genomics

We have developed a high-throughput Agrobacterium-mediated transformation model system using both nptII and the 5-enolpyruvylshikimate-3-phosphate synthase gene from Agrobacterium tumefaciens strain CP4 (cp4) based selections in MicroTom, a miniature rapid-cycling cherry tomato variety. With the NPTII selection system, transformation frequency calculated as independent transgenic events per inoculated explant ranged from 24 to 80% with an average of 56%, in industrial production scale transformation experiments. For CP4, with glyphosate selection, the average transformation frequency was 57%. Stable transformation frequency was positively correlated with transient expression (R=0.85), and variable with the genes of interest. DNA integration and germline transformation were confirmed by biological assay, Southern Blot analysis, and R₁ phenotype segregation. Transgene expression was observed in leaf, root, stem, flower, and fruit tissues of the transgenic plants. Ninety-five percent of transgenic events coexpressed two introduced genes based on β-glucuronidase (GUS) and neonmycin phosphotransferase II (NPTII) expression. Seventy-five percent of transgenic events contained one to two copies of the introduced uidA (GUS) gene based on Southern analysis. Transgenic plants from the cotyledon explants to the transgenic plants transferred to soil were produced within about 2–3 months depending on the genes of interest. The utility of this MicroTom model transformation system for functional genomic studies, such as identification of genes related to important agricultural traits and gene function, is discussed.     

Somaclonal variation in tomato: effect of explant source and a comparison with chemical mutagenesis

Summary Plants were regenerated from leaf, cotyledon, and hypocotyl explants of tomato cv Moneymaker. Various phenotypic alterations were observed among regenerated plants (R₁), but were not transmitted to the progenies, except for ploidy variation. Variation in ploidy level, mainly tetraploidy, occurred in R₁ plants and their R₂ progenies, and the frequency of polyploid plants depended on the explant source. More than 50% of the regenerants derived from hypocotyl explants were found to be polyploid. A correlation was observed between the percentage of polyploid cells present in the explant material in vivo and the frequency of polyploid plants. Several monogenic mutations were recovered in the R₂, four of which were shown to be allelic to known, recessive, single-gene mutants. No significant effect of explant source or duration of tissue culture period on mutant frequency or spectrum was found. For several mutant types that could be scored unambiguously, somaclonal variation was compared to variation induced by treatment of seeds with ethyl methane sulphonate (EMS). The results showed that the mutant frequencies were higher after EMS treatment than those generated through tissue culture. With respect to the mutant spectrum, no clear differences were observed between the spectra obtained after EMS treatment and those after tissue culture. However, tissue culture gave rise to polyploid plants, whereas no ploidy variants occurred after EMS treatment.