WO1997042332A2 - Genetically transformed cassava cells and regeneration of transgenic cassava plants - Google Patents

Abstract

The invention relates to methods for the stable genetic transformation and regeneration of Cassava.

Description

GENETICALLY TRANSFORMED CASSAVA CELLS AND REGENERATION OF TRANSGENIC CASSAVA PLANTS

The invention relates to the obtention of genetically transformed Cassava (Manihot genus) cells from which a whole Cassava plant can be regenerated which stably incorporates a genetic change induced in said cells.

Cassava plays a significant role as carbohydrate source in many tropical countries. In 1994 the world production was estimated to be as much as 158 millions tons. Despite the importance of cassava for the livelihood of millions of people, it was until recently, considered a neglected crop.

Traditional Cassava breeding programs have had some success in introducing new cultivars. However, a high degree of heterozygosity, irregular flowering in some cultivars, low seed set and variable germination rates have impeded faster progress.

It was then recognized that genetic engineering might be useful to resolve some of the problems affecting this crop.

The limiting factor for the improvement of Cassava by genetic engineering, however, has been the lack of a reproducible transformation system. A prerequisite for the genetic transformation of plants is indeed the ability of a cell to receive DNA, to stably integrate it into its genome, and to regenerate into a fully transgenic plant.

A method for regeneration of cassava plants through somatic embryogenesis has been available since

1982 (see reference 1 given at the end of the description, together with the other bibliographic references hereinafter mentioned).

However, the use of the somatic embryos generated by this culture system as target tissues for genetic transformation via Agrobacterium, particle bombardment, and electroporation, has yielded at best only chimeric embryos. This is probably due to the fact that the embryogenic process in these structures is not initiated by single cells, but via numerous cells that simultaneously undergo coordinated cell divisions.

There is therefore a need for a complete trans formation/ regeneration system enabling the efficient production of stably transformed Cassava plants.

An alternative regeneration system was recently developed consisting in cultivating Cassava friable embryogenic callus or embryogenic suspensions (derived from friable embryogenic callus).

The inventors have found a method which applies to such embryogenic structures and which allows the stable genetic transformation of Cassava. The inventors have also found that, from genetically transformed Cassava embryogenic structures, Cassava plants having stably incorporated desired genetic changes could be regenerated at industrially-appropriate production yield and time.

The inventors have found a method for producing Cassava transgenic plants at the industrial scale.

It is then an object of the present invention to provide a method for the genetic transformation of cassava cells that results in an homogenous population of cells, having integrated the introduced DNA. 5

It is another object of the invention to provide genetically modified cassava cells from which whole plants can be regenerated.

It is still another object of the invention to provide cassava plants and their progeny, having stably incorporated (a) trait (s) introduced in a cassava cell from which they are regenerated.

It is yet another object of the invention to provide seeds such as produced by said genetically transformed plants or their progeny.

The invention also relates to tissues such as derived from said plants or their progeny.

The method according to the invention for obtaining genetically transformed cassava cells which can be regenerated into transgenic cassava plants, comprises using Cassava embryogenic tissues, such as friable embryogenic callus and/or embryogenic suspensions, as target for the genetic transformation of said cassava cells. B_/ "friable embryogenic callus" is meant clusters of embryogenic cells that can be maintained in the embryogenic state on solidified culture medium. In contrast to organized structures, such as embryo clumps, this callus is unorganized and composed of small, spherical embryogenic units.

B/ "embryogenic suspensions" are meant clusters of embryogenic cells derived from friable embryogenic callus suspended in liquid medium. The embryogenic cells can retain their embryogenic potential for more than 18 months. Embryogenesis is apparently initiated by single cells on the surface of globular embryoϊds, which are produced in very large amounts in liquid culture. It has been found, according to the invention, that systems of tissues derived from friable embryogenic callus and/or embryogenic suspensions were particularly advantageous for genetic transformations. Particularly, they allow reproducible genetic transformation of cassava cells and the regeneration into transgenic plants of cassava.

More particularly, the method of the invention comprises: - introducing into the cells of said tissues exogenous DNA comprising at least one DNA coding for the desired transforming trait (s), cultivating the resulting tissues under conditions that allow the selection of the transformed tissues through at least the expression of the desired trait (s ) .

Advantageously, said DNA(s) encoding (a) desired transforming trait (s) is (are) selected from the group comprising DNAs whose expression confers resistance to pests, resistance to diseases due to pathogens such as bacteria, protozoa, fungi, virus, particularly the African Cassava Mosaic Virus, the Cassava Common Mosaic Virus, the Cassava Common Mosaic Vi rus- Brazil, The East African Cassava Mosaic Virus, the Cassava Vein Mosaic Virus, DNAs whose expression confers resistance to stress due to environmental factors such as woundings, UV light, temperature, drought, pollution, improves starch quality and/or quantity, increases tuber protein-content, reduces the cyanogenic glucoside content, leads to plant production of foreign proteins or of secondary metabolites such as biodegradable plastics, or extends the shelf -life of the harvested tubers. Preferred DNA(s) coding for a desired transforming trait is (are) selected from the group comprising viral coat protein genes, viral replicase genes, viral trans -activator genes, viral movement protein genes, viral transmission factor genes, antisense

DNA for Cassava granule-bound starch synthase (GBSS), the linamarase- related genes and the α-hydroxyni tri le lyase- related-genes , the starch-modifying enzymes, and genes involved in the production of biodegradable plastics such as PHAs and/or PHS.

Alternatively, said DNA(s) comprises (s) a reporter gene, i.e. a gene which mimics a gene coding for a desired transforming trait.

Such DNA(s) preferably code (s ) for an activity revealed by fluorescence and/or coloration and/or luminescence. A particularly appropriate reporter gene is, as illustrated in the examples, the uidA gene encoding β-glucuronidase.

Advantageously, exogenous DNA further comprises at least one gene acting as a selectable marker for selection of transformed tissues.

B/ "selectable marker", is meant a genetic marker for use in transformation, whose product confers a phenotype which can be selected for a given effect. Said selectable marker (s) comprise (s) gene (s ) conferring resistance to antibiotics and/or herbicides, and/or genes using sugars as carbon sources and used as selectacle marker genes, and more particularly genes conferring resistance to aminoglycosic antibiotics. Examples of such selectable markers include the hph genes (hygromycin resistance), the ptt gene (phos phi not ricin gene), the glyphosate gene and the npt.II gene.

A particularly useful gene is the nptll gene. The introduction step of said exogenous DNA into the cells of said tissues is advantageously carried out by using direct techniques such as DNA condensation, electroporation, microinjection, microbombardment, whiskers, or by indirect techniques, such as via virus, bacteria, nuclear fusion, polinic tubes or organelles. According to one embodiment of the invention, the introduction of said exogenous DNA is carried out via Agrobacterium strains, particularly an A. tumefaciens strain.

Particularly satisfactory results are obtained when the Agrobacterium strain is the vector of a plasmid containing a DNA coding for the desired transforming trait (s) and/or (a) reporter gene (s ) under the control of a transcriptional promoter such as the enhanced 35 S promoter of Cauliflower Mosaic Virus or the Cassava Vein Mosaic Virus promoter, and a DNA as selectable marker (s) fused to the nos promoter from Agrobacterium.

Particularly satisfactory results are obtained when said transformation comprises contacting said tissue with Agrobacterium in a liquid medium and co-cultivating said tissue and Agrobacterium on an antibiotic-free solid medium.

According to another embodiment of the invention, the introduction of said exogenous DNA is carried out by DNA-microbombardment. Said microbombardment is advantageously carried out via microparticules coated with DNA such as plasmidic, cosmidic, chromosomal, ribosomal, bacteriophagic or viral DNA.

Said DNA for example results : from the fusion of the pUC19 plasmid, containing said DNA coding for (a) reporter gene (s ) and/or the desired transforming trait(s), such as the Cassava Vein Mosaic Virus- Brazil coat protein gene, linked to the enhanced 35 S promoter from Cauliflower Mosaic Virus, with the pM0N505 plasmid containing the nptll gene as selectable marker, or

- from the insertion of said DNA coding for (a) reporter gene (s ) and/or the desired transforming trait (s), such as the Cassava Vein Mosaic Vi rus- Brazil coat protein gene, at a polylinker site between the Cassava Vein Mosaic Virus promoter and the Nos promoter signal, into the pMON977 plasmid, in which the e35 S promoter from Cauliflower Mosaic Virus has been replaced by the Cassava Vein Mosaic Virus promoter.

The culture step of the method according to the invention is carried out through a selection step by exerting a selection pressure that allows only genetically transformed cells to grow, while being capable of a satisfactory expression of the desired trait (s ) . Suitable means for the selection step are agents which are able to kill the naturally occurring cells, i.e. the non- trans formed cells, and to which effects cells having incorporated said exogenous DNA are resistant. Examples of appropriate selection agents includes hygromycin, phos phi not ricin, glyphosate and paromomycin. A particularly preferred selection agent is paromomycin at a concentration ranging from 10 to 40 μM, preferably about 25 μM. In a preferred embodiment of the invention, the selection step comprises a pre-selection step.

In a most preferred embodiment of the invention, said pre-selection steps is exerted by paromomycin at a concentration ranging from 10 to 40 μM, preferably about 12.5 μM. A preferred embodiment of the invention includes a pre-culture step on an antibiotic-free medium for a period of less than 3 weeks, preferably for a period of 3 days, prior to pre-selection step.

The selected transformed cells resistant to the selection pressure conditions are useful for the regeneration of cassava plants.

Such selected cells may also be kept frozen for further analysis.

Accordingly, the invention also relates to a method for the regeneration of transformed cassava cells into cassava plants.

Said method comprises : using genetically transformed Cassava embryogenic structures, - cultivating and subcultivating, on induction media, tissues resistant to said selection conditions, until the obtention of calli and/or embryos and/or embryos with cotyledons and/or plantlets and/or plants and/or fragments thereof. Appropriate Cassava embryogenic structures can be selected from the group consisting of Cassava embryogenic suspensions and of Cassava friable embryogenic callus.

The culture and subculture steps comprise the initiation of embryo differentiation, the emergence of cotyledons, the embryo maturation, the induction of multiple shoots in the apical area of embryos, the rooting of shoots and the growth until the obtention of the plant.

According to an embodiment of the method of regeneration according to the invention, said genetically transformed Cassava embryogenic structures are such as obtained by the method for the stable genetic transformation of Cassava hereinabove-described.

Appropriate induction media for respectively carrying out said steps comprise picloram preferably at two different concentrations, alpha naphthaleneacetic acid (NAA) , benzylaminopurine ( B\P) and/or activated charcoal, the rooting and the growth of the plant being carried out on growth regulator free medium. Preferred concentration range of said substances added to the induction media are as follows : picloram : 25 to 100 μM, preferably about 50 μM, in a first step and/or 1 to 100 μM, preferably about 5 μM, in a second step ; NAA : 1 to 10 μM, preferably about 5 μM, in a third step; m.P : 1 to 100 μM, preferably about 4 to 4.5 μM, in a fourth step, activated charcoal : 0, 1 to 10% w/w, preferably about 0.5 % in a fifth step.

Another appropiate method can include desiccating of embryos that have matured on a growth regulator-free medium.

During the course of the culture, controls of the expression of the desired trait(s), for example in IO

histological assays and/or Southern blot analysis and/or PCR analysis, as disclosed in the examples hereinafter given, have shown the integration of the introduced exogenous DNA. The invention thus relates to a cassava cell comprising stably inserted exogenous DNA in its genome and capable of regenerating into a stably transformed whole Cassava plant, or a part thereof. The invention also encompasses a cell line, a cell suspension and a tissue such as obtained with said Cassava cell.

It more particularly relates to calli and/or embryos and/or embryos with cotyledons and/or plant lets and/or plants and/or fragments thereof, comprising stably inserted exogenous DNA, as above defined, in their genome.

The invention also relates to transgenic cassava plants, their progeny and seeds, comprising a plurality of genetically transformed cells as above defined. The invention also provides means for producing large numbers of siblings during the regeneration phase of cassava, this will advantageously reduce the time needed for the establishment of transgenic planting material. Thus, the invention provides means for large scale production of genetically transformed embryogenic units. Synthetic seed development and mass clonal multiplication of genetically improved cassava genotypes could thereby be achieved through bio reactor technology. The above disclosure generally describes the present invention. A more complete understanding can be obtained by reference to the following specific examples which are provided for purposes of illustration only, and are not intended to be limiting unless otherwise specified.

In said examples, it is referred to Figures 1 to 5 wherein :

- Figure 1 shows a flow diagram of the steps leading from microbombarded embryogenic suspensions of cassava plants to transgenic plants (time periods mentioned are indicative, - Figure 2 is a photographic plate representing the regeneration of transgenic plants from microbombarded embryogenic suspensions of cassava,

- Figure 3 shows a Southern analysis of total genomic DNA from putative transgenic embryogenic suspensions and plants. Arrows indicate the size of the expected hybridization product,

- Figure 4 shows a Western blot where Cassava plants engineered according to the invention have been analysed for their expression of the Cassava common mosaic virus-Brasil (CsCMV-Br) coat protein (the Cs CMV- Br coat protein gene being under the control of the Cassava Vein Mosaic Virus promoter), the + signs mark a positive PCR response of the same engineered Cassava plants,

- Figure 5 shows a Southern blot of genomic DNA (10 μg) of different lines of paromomycin- res is tant and

GUS-positive embryogenic suspensions and plants derived from Agrobacteriuin-mediated transformation of embryogenic suspensions of Cassava, and

- Figure 6 shows a flow diagram of the steps loading from Agrojbacteri urn-inoculated embryogenic tissue of Cassava plants to transgenic plants (time periods mentioned are indicative). Example 1 : MICROBOMBARDMENT OF CASSAVA EMBRYOGENIC SUSPENSIONS (uidA gene)

la. Embjryogenic suspensions and culture media

Embryogenic suspensions were initiated from a six month old line of embryogenic callus of cassava cultivar TMS60444 (Bath, England). The embryogenic callus was transferred to 250 ml flasks with 50 ml liquid SH-1 medium. The flasks were kept on a gyratory shaker (150 rpm) at 25 ° C under a photoperiod of 16 h, at 20 - 25 umol. s^_1.m^"2 photos yntheti cally active radiation (PAR) provided by fluorescent lamps (Cool White). Stock cultures were maintained by replacing the culture medium every three to four days for a culture period of 12 - 14 days. At that time the tissue was transferred in aliquots of 1 ml settled cell volume (SCV) to fresh culture medium to start a new culture cycle. In suspensions that were used for microbombardment experiments, the culture medium was replaced with fresh medium every 2 days. For the measurement of SCV, suspensions were transferred to a 15 ml graduated centrifuge tubes. The tissue was allowed to settle for 30 min and the volume was measured. The following culture media were used : SH-1

( SH salts (2), MS vitamins (3), 50 μM picloram, and 60 g/l sucrose) ; SH-2, SH-1 with 7.5 g/l Difco-Bacto agar ; R-l, MS-salts 1/2 concentrated (3), MS vitamins, 20 g/l sucrose, 7.5 g/l Difco-Bacto agar, picloram 5 μM ; R-2, as R-l, but picloram replaced by 5 μM α -naphthalene acetic acid (NAA) ; MM, MS salts, no growth regulators, 5 g/l activated charcoal, 30 g/l sucrose ; SD, as R-l but

4.44 μM Ef\P instead of picloram ; SP, MS-salts and vitamins, 20 g/l sucrose, 2 g/l phytagel (Sigma), no growth regulators. All media were adjusted to pH 5.7 prior to autoclaving.

lb. Effect of antibiotics on non-transformed embryogenic suspensions . Genes that are frequently used to select transformed plant tissues include nptll, hph and bar, encoding neomycin phos phot rans f erase, hygromycin phos pho transferase, and phos phi not ricin acetyl transferase, respectively. These genes confer resistance to kanamycin and related ami noglycos ides , to hygromycin, and to phos phi not ricin. In order to determine the usefulness of each of these genes for selection of transformed embryogenic tissue of cassava, killing curves were established for embryogenic cell suspension cultures with a variety of antibiotics.

Embryogenic suspensions were sieved to isolate cell clusters 250-350 μm in diameter. The number of units collected was counted using a Sedgewick- Rafter Cell S50, and aliquots of 5,000 of units were transferred to 15 ml graduated centrifuge tubes containing 5 ml SH-1 medium with antibiotic. The medium was buffered with 0.5 g/l 2-

(N-morpholino) ethanes ulf onic acid (MES) to prevent pH changes resulting from the addition of antibiotics. Stock solutions of the antibiotics kanamycin, geneticin, paromomycin, hygromycin, and the herbicide phos phi not ricin were filter-sterilized and aliquots added to the autoclaved culture medium to obtain concentrations of 25, 50, 100, 200, and 300 μM. Controls were cultured without antibiotics. Four replicate tubes per treatment were placed in a tissue culture rotator (New Brunswick Scientific, USA) and rotated at 10 rpm. After seven days samples containing between 300 and 400 embryogenic cell clusters were removed from each tube and placed in the Rafter Cell. Their viability was assessed with the vital dye fluorescein diacetate (FDA) according to (4). Samples were stained for ten minutes in SH-1 medium with 0.05 %

(w/v) FDA. They were viewed under an Olympus IMT2 inverted microscope (10 x objective) with epi fluorescence attachment and with a FITC filter set and scored for their viability. An embryogenic cell cluster was assessed as alive if it gave a strong yellow fluorescent signal.

After one week of growth in culture medium with antibiotic, kanamycin (25 μM - 300 μM) had no effect on the viability of the embryogenic suspensions, geneticin killed all cells at 25 μM, while paromomycin reduced the viability to 4.5 % at 25 μM and to zero at 200 μM. Both hygromycin and phos phi not ricin were inefficient in the concentration range tested : 7.9 % and 7.3 % of the embryogenic cell clusters survived treatments with 300 μM hygromycin and phos phi not ricin, respectively. Therefore, it was decided to use the nptll gene in transformation studies and to use geneticin and paromomycin as selective agents.

lc. Plasmid used for rnicrobombardment

The enhanced 35 S promoter from Cauliflower Mosaic Virus (5) was linked to the uid-A gene coding for \≤ β-glucuronidase coupled to the 7S polyadenylation signal isolated from a β-conglycinin gene (6) and cloned into pUC19 (7) at the Pstl site. The resulting plasmid was cloned into the polylinker site of the binary vector pMON505 (8), which contains the nptll gene. This plasmid was designated pILTAB313.

Id. Microprojectile bombardment

Cell clusters 250 μm - 500 μm in size from 12 -

14 day old embryogenic suspensions were used for bombardments. Aliquots of 200 μl SCV (Settled Cell Volume) in a volume of 1 ml culture medium were pipetted onto polypropylene grids (Spectra/Mesh # 146428, opening 210 μm) in such a way that the liquid formed a drop kept in place by surface tension.

The grids were placed in Petri dishes on top of a dry filter paper. Gold particles of 1.0 μm diameter (ELoRad; USA) were coated with plasmid DNA according to (9). Five μl of the particle-DNA suspension were distributed onto each macrocarrier (BLoRad) and kept in a desiccator until used. EDmbardment took place with the use of the Particle Delivery System PDSIOOO/He (BLoRad) following the manufacturer's recommendations. The dishes with tissue were placed at a distance of 9 cm from the rupture disk retaining cap and the coated microparticles were accelerated using a pressure of 1,100 psi in a vacuum of 27 inches mercury (9.1 kPa abs. pressure). The liquid was absorbed from the droplet, leaving a monolayer of embryogenic cell clusters on the grid. Each sample was bombarded twice. To prevent dessication of the tissue between bombardments the mesh was placed in a Petri dish containing sufficient SH-1 liquid medium to just cover the tissue. After the second bombardment the embryogenic units were washed off the mesh in 25 ml liquid medium in 50 ml centrifuge tubes.

le. Selection of transformed tissue and establishment of paromomycin -res is tant cell lines.

The antibiotics geneticin and paromomycin were compared for their ability to select transformed tissue. Tissue from embryogenic suspensions was bombarded with pILTAB313, a plasmid that contains the nptll gene as selectable marker and the uidA gene (also referred to as gus gene) as a reporter: figure 3A is a shematic representation of pILTAB313, showing the restriction sites for Pstl and Hindlll and the relative size of the uidA probe (not drawn to scale).

The expression of uidA can be easily monitored by a histological assay that results in the formation of a blue stain if β-glucuronidase (GUS) is present. One month after bombardment (the tissue had been treated with antibiotic for three weeks and then moved to medium without antibiotic) histological GUS assays revealed single GU S-pos i ti ve , dark blue cells in all treatments, but there was a higher proportion of blue cell clusters in the suspension cultures treated with paromomycin versus those treated with geneticin (about 100 per bombarded sample of 0.2 ml settled cell volume (SCV) after selection with paromomycin versus about 10 after selection with geneticin). Furthermore, few tissue pieces per bombarded sample were recovered several weeks after selection in medium with geneticin (0 at 25 μM, 4 pieces at 50 μM) . Selection with paromomycin resulted in a six fold greater success (24 pieces at 12.5 μM, 28 pieces at 25 μM) .

For all the following experiments a concentration of 25 μM paromomycin was chosen to select resistant cells following bombardment of embryogenic tissues with the nptll gene. The general protocol for the selection and regeneration is described in Figure 1.

Preliminary experiments indicated that a period of growth after microbombardment in absence of antibiotics (Fig. 1, step 1) increases the number of antibiotic-resistant cell clusters that can be obtained after the subsequent selection phase (Fig. 1, step 2). However, if this period is longer than three weeks, the cell clusters increase significantly in size and break into smaller pieces at the end of the selection phase. As a compromise a period of ten to fourteen days was adapted for the initial growth without antibiotic (SH-1 medium). In the experiment that resulted in the first transgenic plants the initial growth phase without antibiotics (Fig. 1, step 1) was three days. During the following four to five weeks in medium with antibiotic (SH-1 + paromomycin 25 μM) most of the tissue is killed. Under a stereomicroscope, small

(0.5 - 1.5 mm), yellowish embryogenic units can be distinguished among white, dead units : figure 2B (bar 2 mm) shows a sample of an embryogenic suspension that was cultured without antibiotic for ten days after bombardment followed by 4 weeks in medium with 25 μM 15 paromomycin. The number of yellowish embryogenic units ranges from 20 to 100 per bombarded tissue sample. About one third of these units continue to grow when cultured individually on solidified selection medium SH-2 with 25 μM paromomycin (Fig. 1, step 3) and produce friable, embryogenic callus.

If. Plant regeneration from paromomycin -res is tant embryogenic tissue.

The times for the following steps leading to embryo development, shoot induction and finally plant regeneration varied between experiments, as indicated in Figure 1. To increase the chance of plant regeneration, paromomycin- resistant embryogenic tissue was amplified either by culture in liquid SH-1 medium or on solidified SH-2 (Fig. 1, step 4). In liquid medium the volume of tissue doubles within two to four days if the culture medium is renewed every two days. On solidified medium the growth rate is much slower, i.e., the volume of tissue doubles within three to four weeks.

Using the protocol outlined in Figure 1, 18 independently transformed lines were established from embryogenic callus or suspensions, 16 of which were GUS- positive in histological assays. Attempts were made to regenerate plants from seven of the lines. From all lines embryos with green cotyledons were recovered (Fig. 1, steps 5 -7). Transfer of these embryos to medium with 4.4 μM benzylaminopurine ( EAP) induced single or multiple shoot formation : figure 2H (bar 0.5 cm) shows organogenic tissue 5 1/2 months after bombardment. Subsequent culture of 1 - 2 cm long shoots in medium without growth regulators led to rooting and plantlet development : figure 2G (bar 0.5 cm) shows a plantlet, developed from figure 2H organogenic tissue, 6 months after bombardment. Shoot formation has been induced in six lines coming from three independent bombardments.

Plantlets with a shoot length greater than 5 cm from two of these lines were grown in vitro on SP medium without growth regulators, and plants from another line were transferred to soil for growth in a greenhouse : figure

21 shows a transgenic plant of line 44.3, 14 months after bombardment.

Somatic embryos of cassava can be regenerated into plants on a variety of media containing growth regulators, particularly EA.P. Under these conditions shoot development is first induced, followed by rooting on growth regulator-free medium.

Another appropiate method can include desiccating of embryos that have matured on medium without growth regulators. These embryos can then germinate by simultaneous development of root and shoot. When embryos were placed on media with EAP, shoot formation was induced in six out of seven lines of putative transformed embryogenic tissue.

l.g Histoσhemical assay for β-glucuronidase (GUS)

A modification of the assay disclosed in (10) was used in which the assay buffer included 0.08 M sodium phosphate buffer at pH 7.0, 7.7 mM X-Gluc (5-bromo-4- chroro-3-indolyl-β-D-glucuronide cyclohexylammonium salt), 20 % (v/v) methanol and 0.16 % Triton X 100.

Potassium ferricyanide and potassium ferrocyanide were added to the buffer at 6.4 mM for embryogenic suspensions and 0.64 mM for roots, stems and leaves. Tissues were covered with assay buffer and kept for 2 h (embryogenic suspensions) or 16 h (stems, leaves, roots) at 37 °C in darkness.

Assays were stopped by washing the tissue several times with water. For clearing and long-time storage the tissue then was transferred to 70 % ethanol.

During the course of culture of tissues subsequent to bombardment with pILTAB313, the following general pattern of GUS expression was observed : three days after bombardment many single, dark blue cells were visible : figure 2A (bar 2 mm) shows embryogenic units with GUS-positive cells, 3 days after bombardment ; two to four weeks after bombardment multi cellular, dark blue units were observed : figure 2C (bar 50 urn) shows GUS- positive embryoid, 4 weeks after bombardment. When the GUS-positive units had a size larger than about 200 μm

(four to eight weeks after bombardment), the pattern of

GUS activity became less uniform. The units had a light blue center with sections of dark blue cells on their surface, as well as emerging dark blue secondary units. During the subsequent series of steps, the assay buffer needed to be adapted since the overall intensity of the GUS-staining was low in differentiating tissues using the conditions described above. By reducing the ferri- and ferrocyanide concentrations to 0.5 mM and by extending the incubation time from 2 h to overnight, good development of the blue stain was achieved. In some lines emerging cotyledons (Fig. 1, step 6) stained completely blue : figure 2F (bar 2.5 mm) shows GUS- positive embryos, 10 months after bombardment. In other lines, the stain was restricted to veins and stomata. At this stage the root pole usually stained blue. In six lines shoot formation could be induced. Four of these were GUS positive, with blue stain restricted mainly to vascular tissues and to leaves : figures 2D (bar 5 mm) and 2E (bar 2.5 ram) show GUS expression in a stem and a leaf, respectively, of plant line 44.1.9 one year after bombardment.

The shoot staining patterns were very similar in independently recovered lines and were stable over time. Blue stain was always detected in all leaves of a plantlet, as well as in vascular tissue from the shoot base up to the tip. Moreover, the stain was not exclusively restricted to vascular tissues and leaves, but was observed also in single cells or small groups of cells distributed throughout all tissues.

lh. DNA isolation and Southern analysis of cassava tissues.

The fact that paromomycin- resistant embryogenic tissue could be recovered and that most of the lines derived from resistant tissue were GUS-positive in histological assays strongly suggested the stable integration of at least the nptll gene, and in most cases of both the nptll and the uidA gene.

To verify this conclusion, Southern blot hybridization analyses were performed both with putative transgenic embryogenic suspensions and with plants regenerated from such suspensions. Genomic DNA was isolated according to (11). DNAs were digested either with Pstl or Hindlll restriction endonucleases (Gibco-BRL, Gaithersburg, MD) . Undigested and digested DNAs (5 μg per well) were loaded on a 0.8 % agarose gel and electrophoresis was carried out at 5 volt/cm for 6 hours. DNAs were transferred onto a Hybond-N+ nylon membrane (Amersham, Arlington Heights, IL) according to the instructions of the manufacturer. The uidA probe was a purified PCR product (size approx. 1.8 kb) that was synthesized with primers specific for sequences in the uidA coding region and with pILTAB313 as template. Labelling was done by random 9-mer priming with Exo-klenow (Stratagene, La Jolla, CA) and (α-32P)dCTP. Hybridization was carried out at 65 ° C in hybridization buffer (3 x SSC, 2 x Denhardt solution, 0.1 % SDS, 6 %

PEG) overnight. The final wash was done at 65 ° C with (0.1 x SSC ; 0.1 % SDS). The membrane was then exposed to X- ray film overnight.

Figure 3B shows the result of a hybridization reaction of genomic DNA (5 μg per lane) from four GUS- positive lines of embryogenic suspension cultures derived from three different bombardment experiments : lanes 1,2: DNA from non-bombarded control suspension ; lanes 3,4 and 9,10 : DNA of independently established suspensions derived from the same bombardment (lines 44.2.1 and

44.1.9) ; lanes 5-8 : suspensions derived from different bombardments (lines 55.1 and 62.3) ; lanes with odd numbers : undigested DNA ; lanes with even numbers : DNA digested with Pstl. In all lines the radiolabelled uidA probe hybridized with undigested high molecular weight

DNA (Fig. 3B lanes 3, 5, 7, 9). The undigested DNA of the non-trans genie control showed no hybridization with the probe. These results confirm the integration of the transforming DNA into the chromosomal DNA.

The digestion of genomic DNA with Pstl (to release the complete 35S-uidA-7S cassette) produced different banding patterns for each of the suspension cultures, as expected if integration of the introduced DNA occured at random sites (Fig. 3B lanes 4, 6, 8, 10). Furthermore, the digested DNA from all suspensions contained a fragment corresponding to the expected size

(3 kb) of the intact uidA gene cassette. It is presumed that other bands of hybridization represent integration of fragments of the uidA gene.

The banding patterns of DNA coming from two randomly selected embryogenic suspension cultures that are derived from the same bombardment experiment, but from different selected tissue pieces (Fig. 3 B- lanes 3, 4 and 9, 10) show that the individual tissue pieces collected at the end of the initial selection phase (see Fig. 1, step 2) are the result of different transformation events. This is supported by a Southern blot analysis of DNA derived from five independently established embryogenic suspension cultures from another bombardment experiment, which resulted in five different DNA hybridization patterns. Therefore, the probability of obtaining siblings from different antibiotic-selected tissue pieces coming from the same suspension culture is negligible.

In one experiment more than thirty plantlets were regenerated from a GUS-negative embryogenic callus derived from a single selected tissue piece (line 44.3). Southern blot analysis was performed using leaf DNA of Ztr four of these plantlets : figure 3C shows DNA from GUS- negative plants derived from a single selected putative transformed piece of tissue (line 44.3) ; lane 1: pILTAB313 digested with Pstl ; lanes 2-4: DNA of an untransformed control plant; lanes 5 - 16: DNA of four transformed plants, 3 lanes per plant. Lanes 2, 5, 8, 11, 14 correspond to undigested DNA; lanes 3, 6, 9, 12, 15 to DNA digested with Pstl; lanes 4, 7, 10, 13, 13, 16 to DNA digested with Hindlll. All DNAs showed an identical hybridization pattern when digested with Pstl or Hindlll, proving that the plantlets were siblings. As expected, DNA taken from GUS-negative plantlets did not contain the intact uidA gene cassette. Instead, the uidA probe bound to a fragment of about 1.4 kb in size (Fig. 3C, lanes 6, 9, 12, 15), indicating that a portion of the uidA gene was deleted during the transformation event or during regeneration.

Southern blot analysis of leaf DNA from a plantlet derived from a GUS-positive embryogenic suspension culture revealed integration of the intact uidA gene cassette : figure 3D shows DNA from a GUS-positive plantlet (line 44.1.9) ; lane 1: pILTAB313 digested with Pstl; Lane 2, undigested; lane 3: DNA digested with Pstl; lane 4: DNA digested with Hindlll.

The identity of the hybridization patterns of DNA from this plantlet and of the embryogenic suspension cultures from which it had been regenerated (Fig. 3 B, lane 10) demonstrates the stability of the integrated gene during the course of development from embryogenic cell to plantlet. During the different stages of regeneration, most of the tissues tested were GUS-positive and/or showed the presence of uidA in Southern blot analysis. In one case, GUS negative plants derived from GUS negative embryos (line 44.3) contained fragments of the uidA gene

(Fig. 3C).

Example 2 : MICROBOMBARDMENT OF CASSAVA EMBRYOGENIC SUSPENSIONS (Cassava Vein Mosaic Virus -CsCMV- coat protein gene)

A variant of the transformation procedure described in the above example 1 has been applied to a gene of interest (Cs CMV- Brazil coat protein gene) instead of a reporter gene for transformation. This gene of interest has been shown to provide resistance to CsCMV in transgenic tobacco (17).

The major variation lies in the reduction of the concentration of the antibiotic used for the selection of transformed tissue, paromomycin (from 25 uM

- see example 1 - to 12.5 μM in the present method), which led to an increase in the formation of embryos capable of germination in liquid culture medium without growth regulators. Embryogenic suspensions were established and bombarded with DNA-coated particles as described in the above example 1. The plasmid used for bombardment

(PILTAE353) was pMON977, in which the e35 S promoter from

Cauliflower Mosaic Virus (CaMV) was replaced by the promoter from Cassava Vein Mosaic Virus (CsVMV; (16). The coat protein gene from Cassava Common Mosaic Virus (CsCMV) was cloned into this plasmid at a polylinker site 2⁽o between the CsVMV promoter and the Nos termination signal.

DNA-bombardment was followed by a pre-culture on an antibiotic-free medium for 3 days (see Figure 1, step 1).

Tissue from each bombardment was then transferred to a liquid culture medium with 12.5 μM paromomycin (see Figure 1, step 2). The culture medium

(SH-1, containing 50 μM picloram) was replaced every two to three days by fresh medium. After three weeks in this medium, yellowish, globular structures had developed among whitish, dead tissue. They were picked with a forceps and transferred to solidified SH-2 medium with

25 μM paromomycin (see Figure 1, step 3). From here on, the tissue coming from visually selected single pieces of tissue was maintained as a separate line.

On solid SH-2 medium with 25 μM paromomycin about 70% of the visually selected antibiotic- resistant tissue pieces continued to grow and formed friable embryogenic callus.

Compared to the results obtained through selection as described in example 1 (see lg, If above), the present selection procedure leads to a considerable increase in the formation of embryos capable of germination in liquid culture medium without growth regulators.

Figure 1 step 4 has then not to be necessarily performed in this transformation procedure.

Growing callus lines were transferred after 3-5 weeks to R-l medium (5 μM picloram 0.5 mM arginine, no paromomycin) (see Figure 1, step 5). The embryogenic 11 callus developed a granular structure with small globular uni ts .

After transfer to R-2 medium (5 μM) (Figure 1, step 6) these developed into elongated embryos with small cotyledons, which upon subculture on MM medium (0.5 % activated charcoal, no growth regulators) developed into embryos with green cotyledons (see Figure 1, step 7).

In some instances these embryos germinated on

MM medium, which was extremely rare when the selection and regeneration steps were performed such as described in example 1 (see IF). Figure 1 step 8 may then not be performed in this transformation procedure.

Well-developed embryos where then transferred to liquid SP medium (no growth regulators) where they formed roots and shoots (see Figure 1, step 9). These shoots then were treated as non- trans genie shoot cultures, i.e. , they were propagated by cuttings on solidified SP medium.

The engineered Cassava lines were analysed by PCR and by Western blot for their transgenic expression of the coat protein gene of Cassava Common Mosaic Virus - Brazil (CsCMV-Br).

Results are illustrated by Figure 4 which shows a Western blot analysis of Cs CMV- Br coat protein expression for eight Cassava lines engineered as described in example 2 (lanes 1-8) compared to non- trans genie plant control (C lane) and to Cs CMV- Br sample (V lane, 8 μg) , the M being the marker lane. The + signs mark a positive response for Cs CMV- Br coat protein expression by PCR analysis.

Lanes 1,2,3 and 4 in Figure 4 Western blot show high levels of Cs CMV- Br coat protein expression, while 7% lanes 6 and 7 correspond to engineered Cassava lines with low expression levels.

Lanes 5 and 8 correspond to engineered Cassava lines that have a too low a Cs CMV- Br coat protein expression for being detected by Western analysis.

Compared to example 1 procedure, the present example 2 procedure allows that more of the transformed cells are capable of regenerating into whole plants and that transformed plants are obtained faster. Through this procedure, Cassava transgenic plants expressing the Cs CMV- Br coat protein gene have therefore being obtained. Per sample of bombarded embryogenic tissue between 0.5 and 0.75 transgenic Cassava plants can be produced. That means that from one bombardment session (usually about 30 samples are bombarded on one day), 15 to 20 transgenic plants can be expected.

The average time period required for producing Cassava transgenic plant through the method of the present invention is of about 4-4.5 months.

The present invention therefore allows the production of transgenic Cassava plants at the industrial scale.

Example 3 : TRANSFORMATION OF CASSAVA EMBRYOGENIC SUSPENSIONS VIA Agrobacterium tumefaciens

. Growth of Agrobacterium

The plasmid pMON977 (12) containing the GUS gene driven by the enhanced 35 S promoter and the nptll gene fused to the nos promoter from Agrobacteri um was introduced via triparental mating into the Agrobacterium tumefaciens strain AHE (13). Bacterial cultures were initiated by plating material from a frozen glycerol stock (maintained at -80°C) on agar-solidified LB medium (14) with the antibiotics spectinomycin (100 mg/1) , kanamycin (50 mg/1) and chloramphenicol (25 mg/1). After two days growth at 30°C, 2 - 3 single colonies were inoculated into 2 ml liquid LB medium containing antibiotics and were grown for 24 h. Of this suspension, 200 μl samples were taken and transferred to tubes with 1.8 ml LB medium each, this time without antibiotics and with additional 200 μM acetosyringone. After 3 h growth, the bacterial density was measured with a spectrophotometer at 660 nm and adjusted with SH-1 medium to 5 xlO⁸ cells/ml.

. Transformation of embryogenic suspensions and selection of transgenic tissue

The general protocole for the transformation, selection and regeneration is shown on Figure 6.

An embryogenic suspension of cassava was cultured as described in Example 1. It was sieved, and the fraction of tissue units with a size ranging from 100 μm to 250 μm was used for the experiment. 1.5 ml SCV were transferred to a 9 cm Petri dish and covered with bacterial suspension. After one hour, the bacterial suspension was removed with a pipette. The inoculated embryogenic tissue was transferred to Petri dishes with a culture medium 3o composed of SH medium (mineral salts and vitamins) supplemented with 0.1 mg/1 kinetin, 0.2 mg/1 2,4-D (2, 4- dichlorophenoxyacetic acid) and 10 g/l Difco Bacto agar, pH adjusted to 5.7 (see Figure 6, step 1). Stock solutions of acetosyringone and galacturonie acid were filter-sterilized and added after autoclaving to give final concentrations of 200 μM and 250 mM, respectively.

After 24 h on this medium, the tissue was transferred to liquid SH-1 medium with 500 mg/1 carbenicillin to suppress the growth of Agrobacterium, while permitting the proliferation of embryogenic tissue

(see Figure 6, step 2). 1 ml SCV of tissue was cultured in a volume of 50 ml medium in 250 ml flasks. The suspensions were kept on a shaker, as described above, for non-trans formed embryogenic suspensions. The culture medium was replaced every three to four days by fresh medium.

Two weeks after the inoculation with A. tumefaciens, the suspensions were subcultured for one month in the same medium, but with additional 25 μM paromomycin (see Figure 6, step 3).

The tissue was subsequently subcultured in SH-1 with 500 mg/1 carbenicillin, 100 mg/1 cefotaxim, and 50 mg/1 paromomycin (see Figure 6, step 4). As above described, the culture medium was replaced with fresh medium every three to four days. After two weeks in this medium, two yellowish structures about 1 mm in size were subjected to the histological GUS assay (see example 1). Bath tissue pieces started turning blue already after five minutes in the assay buffer, indicating a strong expression of the GUS gene. . Regeneration of Cassava plants Regeneration can be conducted according to the method described in Example 1 or 2 (see Figure 6, steps 5 to 8).

Example 4: an alternative procedure for Agrobacterium- MEDIATED Cassava TRANSFORMATION

The culture of Agrobacteri um and the plasmid used for transformation (containing the uidA gene as visible and the nptll gene as selectable marker) are described in the above examples. A general protocol for transformation, selection and regeneration is shown on Figure 6.

Embryogenic suspensions of cassava were sieved, and the fraction of issued units with a size ranging from 250 μm to 500 μm was used for the experiment. Volumes of 0.5 ml SCV (settled cell volume) of cassava tissue were transferred to 250 ml flasks and covered with Agrobacterium suspension in SH-1 medium. After one hour the bacterial suspension was removed with a pipette, and the tissue was transferred to solidified SH-2 medium without antibiotics (see Figure 6, step 1).

After two days, the tissue was transferred to 250 ml flasks with 50 ml liquid SH-1 medium containing 500 mg/1 carbenicillin and 100 mg/1 cefotaxim to eliminate Agrobacteri um (see Figure 6, step 2). The flasks were kept on a shaker as above described for non- transformed embryogenic suspensions (cf. example la). The culture medium was replaced every two to three days by fresh medium. After 10 days 25 μM paromomycin was added to the fresh medium, and after another four weeks cefotaxim was omitted (see Figure 6, step 3). About 60 days after the inoculation yellow units of growing tissue had formed in the suspensions. They were picked with a forceps and transferred individually to solidified SH-2 medium with 500 mg/1 carbenicillin and 25 μM paromomycin (see Figure 6, step 4). After one month, some of the embryogenic callus lines were transferred again to liquid SH-1 medium, still with 500 mg/1 carbenicillin, but without paromomycin.

One month later samples from eight lines were frozen and stored at -80°C for later analysis. Another portion of the lines of embryogenic callus that had developed on solidified SH-2 medium with paromomycin was subjected to a procedure similar to the ones above- described for the regeneration of plants from tissue transformed through microbombardment (see example 1 or 2). Tissue was transferred subsequently to R-l, R-2, and

SP medium (see Figure 6, steps 5 to 8). From two lines eventually plantlets developed.

The collected samples were submitted to Southern blot analysis with the uidA coding sequence as hybridization probe.

Results are illustrated by Figure 5 showing a Southern blot of genomic DNA (10 μg) of different lines of paromomycin- res is tant and GUS-positive embryogenic suspensions and plants derived from Agrobacterium- mediated transformation of Cassava embryogenic suspensions. The DNA was either not digested (A), or digested with Hindlll ( B) and Xnol ( C) and hybridized to an uidA specific probe. Lanes 1, 2, non-trans formed control suspension; lanes 3-26; transformed suspensions; lanes 27-29, non- trans formed control plant; lanes 30-34, transformed plants. The numbers in the first row indicate the different lines of embryogenic suspensions and plants.

The results show that the introduced uidA DNA sequence was present in all of the tested suspensions and plant lines except the controls. In all transformed lines the undigested DNA hybridized to the probe, while in the control DNA no hybridization was detected (for plant P2 the amount of available DNA was limited, therefore the hybridization of undigested DNA was omitted). DNA of all lines digested with Hindlll produced a band at 3 kbp, the size expected for the intact uidA cassette. The additional bands are probably due integration of fragments of the uidA gene into the plant genome.

The fact that the selection of transformed cells in this system takes place in liquid medium raises the concern that after a certain period of growth tissue pieces might break apart and thus produce multiples coming from a single transformation event. Visual selection of yellow tissue pieces after the selection in liquid medium then could lead to the es tablishement of sibling-lines.

The eight lines of transformed embryogenic suspensions and the two plant lines are derived from tissue that was selected in two flasks with paromomycin- containing medium. Lines 2, 3, and 16 are derived from one flask, the five other lines and the two plant lines from another flask. Xhol has only one restriction site in the introduced DNA, i.e. , DNA fragments obtained by digestion with this enzyme and hybridizing with the uidA probe contain both introduced DNA and plant DNA. The fact that the banding pattern produced by Xhol digestion is different in all the lines derived from transformation shows that no siblings were produced.

Example 5 : TRANSFORMATION OF CASSAVA PROTOPLAST SUSPENSIONS VIA ELECTROPORATION

Cassava protoplasts were prepared from Manihot esculenta L. cv TMS60444 embryogenic cell suspension cultures (see example 1 for embryogenic cell suspension culture method). Fifty ml of a 10 day old culture (the medium was renewed every 2 days) was collected for protoplast isolation. Prior to enzymatic digestion, the cells were resuspended in 30 ml of medium containing 0.55 M mannitol, 3.2 g. I^"1 SH salts (2) MS vitamins (3), 20 mM CaCl₂, pH 5.8 (medium A). The cells were allowed to settle and medium A was replaced by enzymatic solution consisting of medium A supplemented with 2 % cellulase Onozuka RS and 0.1 % Pectolyase Y 23. Digestion was performed in the dark for 3.5 h at 27°C. Cells were gently agitated during the first hour of treatment. The incubation mixture was filtered sequentially through sieves of 100 μm and 70 μm. Protoplasts were washed three times by centrifugation at 100 g for 10 min in medium A. The number of protoplats was estimated using an hemocytometer. The purified protoplasts were resuspended to final density of IO⁶ cells ml^-1 in electroporation buffer containing 5 mM MES, 130 mM NaCl, 10 mM CaCl₂, 0.45 M mannitol, pH 5.8. Two hundred μl of electroporation buffer containing 30 μg of a plasmid according to example 1 or 2 was added to 800 μl of protoplast suspension in a 0.4 cm path-length cuvette. DNA uptake was carried out using a Gene Pulser instrument ( ELorad) delivering a 300

V pulse at a capacitance of 500 μF. After electroporation the protoplasts were incubated on ice for 30 min, after which they were resuspended at a density of IO⁵ ml^-1 in culture medium A supplemented with 2 % sucrose and 5xl0^~5 M picloram. After 24 hours of incubation in the dark at

27 °C, the protoplasts were collected by centrifugation (10 min at 100 x g) and resuspended in GUS extraction buffer (15) , pH 7.7.

After the regeneration of cell walls and the establishment of embryogenic callus or suspensions from transformed protoplasts, regeneration of plants can be carried out using the same procedures applied to transformed embryogenic callus derived from microbombarded or Agrobacteri um-treated tissues, as described in example 1, 2 or 3.

References

1. Stamp J.A. et al . , 1982, Pf lanzenphysiol. 105: 183-187.

2. Schenck RU. et al., 1972, Can. J. Bot. 50: 199-204.

3. Murashige T. et al., 1962, Physiol. Plant 15: 473-497. 4. Widholm J. M. , 1972, Stain Technol. 47: 189-

194.

5. Kay R et al., 1987, Science 236: 1299-1302. 6. Chen. Z. L. et al . , 1988, EMBO J.7: 297-302

7. Messing J. , 1983, Methods Enzymol. 101: 20- 79.

8. Horsch R B et ai., 1986, Proc. Natl. Acad. Sci. USA 83: 4428-4432.

9. Sivamani E. et al . , 1996, Plant Cell Rep. 15: 322-327.

10. Jefferson RA. , 1987, Plant Mol. Rep. 5: 387-405. 11. Dellaporta S. L. et al., 1983, Plant Mol.

BLol. Rep. 1: 19-21.

12. Rogers S. G. , 1986, Meth. Enzymol. 118, 627- 640.

13. Koncz C. et al . 1986, Mol. Gen. Genet. 204: 383-396.

14. Sambrook J. et al., 1989, Molecular Clonin. A laboratory manual, Vol. 3, Cold Spring Harbor Laboratory, p A.1.

15. Jefferson R A. et al., 1987, EMBO J. 6: 3901-3907.

16. Verdaguer, B , De Kochko, A. , Beachy, R N. , Fauquet, C. (1996) Isolation and expression in transgenic tobacco and rice plants, of the cassava vein mosaic virus (CVMV) promoter. Plant molecular ELology 31: 1129-1139. 17. Fauquet, C. M. , Bogusz, D. , Franche, C. ,

Schopke, C. , Chavarriaga, P. , & Beachy, R N. (1992). Cassava Viruses and Genetic Engineering. Biotechnology: Enhancing Research on Tropical Crops in Africa, Thottapilly, G. Monti, L.M. Mohan Raj, D. R Moore, A.W.Ed. , pp.287-296. Ibadan, Nigeria: IITA/CTA.

Claims

1/ A method for the stable genetic transformation of cassava comprising using tissues of friable embryogenic callus and/or embryogenic suspensions as target for genetic transformation.

2/ A method for the stable genetic transformation of cassava according to claim 1, comprising: - introducing into the cells of said tissues exogenous DNA comprising at least one DNA coding for the desired transforming trait (s), cultivating the resulting tissues under conditions that allow the selection of transformed tissues through at least the expression of the desired trait (s ) .

3/ The method according to claim 2, wherein said DNA encoding (a) desired transforming trait (s) comprises DNAs whose expression: -confers resistance to pests,

-confers resistance to diseases due to pathogens such as bacteria, protozoa, fungi, virus particularly the African Cassava Mosaic Virus, the Cassava Common Mosaic Virus, the Cassava Common Mosaic Virus- Brazil, the East African Cassava Mosaic Virus, the Cass ava Vei n Mos ai c Vi rus ,

-confers resistance to stress due to environmental factors such as woundings, UV light, temperature, drought, pollution, -improves starch quality and/or quantity,

-increases tuber protein-content, 3^

-leads to plant production of foreign proteins or secondary metabolites such as biodegradable plastics, -reduces the cyanogenic glucoside contents, or -extends the shelf-life of the harvested tubers.

4/ The method according to claim 3, wherein said DNA encoding said desired transforming trait (s) is selected from the group comprising viral coat protein genes, viral replicase genes, viral trans -activator genes, viral movement protein genes, viral transmission factor genes, antisense DNA for Cassava GBSS (granule- bound starch synthase), the linamarase-related genes, the α-hydroxynitrile lyase- related-genes the starch-modifying enzymes, and genes involved in the production of biodegradable plastics such as PHAs and/or PHBs.

5/ The method according to anyone of claims 2 to 4, wherein said exogenous DNA comprises a reporter gene.

6/ The method according to claim 5, wherein said reporter gene (s ) code (s ) for (an) activity (ies ) revealed by fluorescence and/or coloration and/or luminescence.

7/ The method according to claim 6, wherein said reporter gene (s ) comprise (s) a gene encoding β- glucuronidase.

8/ The method according to anyone of claims 2 to 7, wherein said exogenous DNA further comprises at least one gene acting as a selectable marker for selection of transformed tissues.

9/ The method according to claim 8, wherein said selectable marker (s) comprise (s) gene (s ) conferring resistance to antibiotics and/or herbicides, and/or genes 3q using sugars as carbon sources, particularly genes conferring resistance to aminoglycosic antibiotics.

10/ The method according to claim 9, wherein said selectable marker (s) is the nptll gene.

11/ The method according to anyone of claims 1 to 10, wherein said genetic transformation is carried out by direct techniques such as DNA condensation, electroporation, microinjection, microbombardment, whiskers or by indirect techniques such as via virus, bacteria, nuclear fusion, polinic tubes or organelles.

12/ The method according to claim 11, wherein said transformation is carried out via Agrobacterium strains, particularly an A. tumefaciens strain.

13/ The method according to claim 12, wherein said Agrobacterium strain is the vector of a plasmid containing a DNA encoding the desired transforming trait (s) and/or (a) reporter gene (s ) under the control of as transcriptional promoter such as the enhanced 35 S promoter of Cauliflower Mosaic Virus or the Cassava Vein Mosaic Virus promoter, and a DNA as selectable marker (s) fused to the nos promoter from Agrobacterium.

14/ The method according to claim 12 or 13, wherein said transformation comprises contacting said tissue with Agrobacterium in a liquid medium and co- cultivating said tissue and Agrobacterium on an antibiotic-free solid medium.

15/ The method according to claim 11, wherein said transformation is carried out by DNA- microbombardment. 16/ The method according to claim 15, wherein said DNA results: from the fusion of the pUC19 plasmid, containing said DNA coding for (a) reporter gene (s ) and/or the desired transforming trait (s), such as the Cassava Vein Mosaic Virus- Brazil coat protein gene, linked to the enhanced 35 S promoter from Cauliflower

Mosaic Virus, with the pMON505 plasmid containing the nptll gene as selectable marker, or

- from the insertion of said DNA coding for (a) reporter gene (s ) and/or the desired transforming trait(s), such as the Cassava Vein Mosaic Virus- Brazil coat protein gene, at a polylinker site between the Cassava Vein Mosaic Virus promoter and the Nos promoter signal, into the pMON977 plasmid, in which the e35 S promoter from Cauliflower Mosaic Virus has been replaced by the Cassava Vein Mosaic Virus promoter.

17/ The method according to anyone of claims 2 to 15, wherein said selection comprises a pre-selection step.

18/ The method according to claim 17, wherein said pre-selection is exerted by paromomycin at a concentration ranging from 10 to 40 μM, preferably about

12.5 μM.

19/ The method according to anyone of claims 2 to 18, wherein said selection comprises the use of paromomycin.

20/ The method according to anyone of claims 2 to 19, wherein said selection is performed after a pre- culture of said tissues on an antibiotic-free medium for a period of less than 3 weeks, preferably of 3 days. 21/ A method for the regeneration of cassava plants comprising using genetically transformed Cassava embryogenic structures;

- cultivating and subcultivating said tissues on induction media until the obtention of calli and/or embryos and/or embryos with cotyledons and/or plantlets and/or plants and/or fragments thereof.

22/ The method according to claim 21, wherein said Cassava embryogenic structures are selected from the group consisting of Cassava embryogenic suspensions and of Cassava friable embryogenic callus.

23/ The method according to claim 21 or 22, wherein the genetically transformed Cassava embryogenic structures are such as obtained by the method according to anyone of claims 1 to 20. 24/ The method according to anyone of claims 21 to 23, wherein said induction media comprise picloram at a concentration ranging from 25 to 100 μM, preferably about 50 μM in a first step and/or 1 to 100 μM, preferably about 5 μM in a second step, and/or α- naphthaleneacetic acid at a concentration ranging from 1 to 10 μM, preferably about 5 μM, in a third step, and/or benzylaminopurine at a concentration ranging from 1 to 100 μM, preferably about 4 to 4.5 μM, in a fourth step, and/or activated charcoal at a concentration ranging from 0.1 to 10 % w/w, preferably about 0.5 %, in a fifth step.

25/ A genetically engineered cassava cell such as obtained by the method according to anyone of claims 1 to 20, capable of regenerating into a stably transformed whole cassava plant or a part thereof. 26/ A cell line derived from the cell according to claim 25. 27/ A suspension of genetically engineered cassava cells characterized in that it is established from cells according to claim 25 to 26.

28/ A tissue of Cassava, characterized in that it is such as obtained by cultivating cassava cells according to anyone of claims 25 to 27.

29/ Calli and/or embryo and/or embryos with cotyledons and/or plantlets and/or plants an/or fragments thereof comprising stably inserted exogenous DNA in their genome.

30/ Transgenic cassava plants, their progeny and seeds, comprising a plurality of genetically transformed ells according to anyone of claims 25 to 27.