Université de Paris Sud XI
E.S.E. 'Ecologie Systématique et Evolution'
F-91405 Orsay cedex
The development of a protoplast system permits regeneration of plants from a single cell. Hence protoplasts have become useful in most plant crops as a technique for their genetic improvement sensu largo, and can be considered as a technique complementary to conventional breeding Its application to somaclonal variation, genetic transformation and somatic hybridisation shows great potential. An exhaustive overview is given of the isolation and culture of protoplasts in banana.
Protoplasts are naked cells that lack cell walls (Figure 1). They are spherical with a plasmolysed cell content and are contained within a plasmalemma. In principle, each individual protoplast can reform a cell wall, and later initiate either a callus through sustained divisions, or an embryo, defined as a somatic embryo. In banana they are obtained from in vivo tissues or in vitro cultures.
Figure 1 Banana protoplasts
As early as 1902, Haberlandt  stated that individual nucleated plant cells could convert into entire plants, either directly or through a callus stage. This phenomenon is termed totipotency and denotes the recovery of a whole organism from a single cell, and is also applicable to protoplasts. In theory all cells are totipotent, but in practice it depends on the past cellular environment. Usually the morphogenetic competence is retained at the unicellular stage that corresponds to a protoplast. However since protoplasts are not exposed to the stabilizing and inductive influence from neighbouring cells, they may have lost their plant regeneration capacity. In fact, factors such as the genotype or species of the manipulated plant, and the ontogenetic state of the explant source, exert a powerful effect on the regeneration potential of protoplasts. Consequently the development of appropriate in vitro conditions for protoplast regeneration is complicated. Notwithstanding this, successful efforts have been made in isolation, cultivation and regeneration of protoplasts, since Nickell and Torrey  pointed out their merit for crop improvement.
Various tissues and an increasing number of plant species and genotypes have been successfully used in protoplast culture, but so far sufficient quantities of protoplasts for practical applications are not routinely met. While the value for agriculture of protoplasts still needs to be demonstrated, they are an invaluable tool for studies on permeability of ions and solutes [3-4], photosynthesis , phytohormones , phytochrome , and maintenance of totipotency . Moreover, protoplasts are useful for the uptake of foreign genetic material and to produce somatic hybrids through protoplast fusion. In addition, protoplasts are an excellent system for studies on cell genetics and even for plant virology.
These studies rely on isolated, clean and healthy protoplasts, which requires the appropriate choice of osmoticum, hydrolysing enzymatic solution, and pretreatment of the donor explant either in vivo or in vitro. This involves the right choice of explant age, cold treatment, phytohormone pretreatment, light intensity or photoperiod, and subculture rhythm, which affects both the internal metabolic status of the cells and cell-wall composition. Prior to cultivation isolated protoplasts need to be freed from enzymatic remains and debris, which are considered to be toxic. Culture in or on solid medium is considered to be more advantageous than in liquid medium because the development of a single protoplast into a colony can be followed up much better. For development, different factors such as medium composition are crucial, especially the nature and quantity of growth hormones. Other important factors are the physical environment, plating density, and embedding conditions or the use of feeder layers, which all contribute in one way or another to cell-wall regeneration and cell division. Once cell divisions start, the level of auxin(s), as well as the colony density, usually need to be reduced to avoid overcrowding. This is done by subculturing, which gives the additional benefit that an adequate nutrient supply is maintained. Finally, for rooting, plants are usually transferred to cytokinin-free medium, possibly containing some auxin, and exposed to high light intensities, usually in illuminated plant growth chambers.
During recent decades, plant biology workers have recognised the potential of protoplasts in many experimental systems, since an efficient enzymatic method for protoplast isolation was first established by Cocking .
2. SIGNIFICANCE AND USE OF PROTOPLASTS IN BANANA
Banana is now easily amenable to in vitro culture, and plants are regenerated from various explants through organogenesis , embryogenesis [11-12], anther culture  and even from cultured protoplasts [14-15]. This has created opportunities for other biotechnological applications. Cell suspensions, for example, opened the way for gene transfer and improved the production, the quality, and therefore the manipulation of cell suspension-derived protoplasts. Through genetic engineering it has become possible to confer new traits on banana plants, using either particle bombardment  or Agrobacterium-mediated transfer . Protoplasts facilitate the direct transformation of plant cells by DNA microinjection and electroporation . However, many characters of agricultural interest are multigenic or ill-defined, and current transformation methodologies allow the integration of only a few foreign genes . Protoplast fusion, however, allows the transfer of several useful characters, even if detailed genetic or molecular knowledge of genes encoding for these desired characters is lacking . Protoplast fusion is therefore a complementary tool to increase nuclear and cytoplasmic variability and to confer desirable agronomic traits.
Important banana cultivars are susceptible to many pests and diseases, particularly Mycosphaerella fijiensis and Fusarium oxysporum f.sp. cubense, nematodes and insects. Very interesting traits of resistance have been identified in this genus for most serious diseases. However cross-breeding is very difficult in the genus since most edible cultivars are sterile even after profuse pollination by wild pollen-fertile genotypes [21-25]. Protoplast fusion is therefore an option, as it can overcome such sexual barriers which occur frequently in banana, and which cannot be bypassed even through embryo rescue. Therefore somatic hybridisation between wild and cultivated banana is expected to produce hybrids combining agronomic traits with genetic resistance to pests and pathogens.
Protoplast fusion has the added benefit that it creates the possibility of generating asymmetric fusions , whereby selected cytoplasmic organelles, chromosomes or chromosome fragments from an irradiated protoplast donor could be combined with the genome of an acceptor protoplast. This strategy is of especial interest in banana, where the subspecies balbisiana is considered to be a source of multiple resistance and therefore could be partially utilised to improve banana cultivars, of which the most important component is the subspecies acuminata. Protoplast fusion also makes the creation of synthetic triploid banana genotypes possible by, for example, combining haploid protoplasts derived from anther culture with protoplasts from a diploid improved cultivar.
Chimerism and somaclonal variation are factors that seriously limit rapid clonal propagation of banana. By their nature, protoplasts avoid chimerism because they originate from a single cell, and protoplasts may be use to dissociate chimeric plants. On the other hand the cellular heterogeneity of protoplast populations can be useful for isolating somaclonal variants with improved characters such as increased yield or pathogenic resistance, so that somaclonal variation may also be exploited to provide new sources of genetic variability .
Banana protoplast isolation and culturing is nowadays routine, although protoplast research on banana started 15 years later than on model plants. Now it is regrettably underused because of the strong emphasis on molecular and genomic studies in banana. Nevertheless, protoplasts have much to offer for non-conventional banana breeding, since they overcome sexual incompatibilities at interspecific and even at the intergeneric level, and allow the incorporation of multigenic traits such as yield, resistance to stress, pests and diseases. Finally protoplast techniques may improve banana when other classical methods have failed.
3. REVIEW OF WORK ON ISOLATED CELLS AND PROTOPLASTS
3.1. Early experiments
The first attempts at banana protoplast culture were made in 1984 by Bakry [27-28]. Various explants (leaf and sheath from in vitro banana plants, callus from floral explants, and bract and tepals from in vivo banana plants) of diploid and triploid banana plants belonging to the AA, BB, AAA and AAB groups were tested. Enzyme solutions were composed of a modified CPW (Cell and protoplast washing solution) salt solution  (Table 1), containing 0.7 M mannitol, and a mixture of various enzymes at 0.1 to 5% (w/v) such as:
Pectinases: Macerozyme (Yakult Pharma Industries, Japan) and pectolyase Y23 (Seishin Pharma Company, Japan);
Hemicellulases: Hemicellulase (Sigma Chemical Corporation, USA) and Rhozyme (Pollock and Pool CTD, USA)
Cellulases: cellulase R-10 'Onozuka' (Yakult Pharma Industries, Japan), Cellulysin (Calbiochem Behring, USA) and Driselase (Kyowa Hakko Kogyo Co., CTD, Japan).
Table 1 Cell and protoplast washing (CPW) solution
The enzyme solution was adjusted to pH 5.6 with KOH and filter-sterilised (0.22 mm). Small explants of about 1-5 mm3 were incubated for 9 h with shaking (80 r.p.m.) at 27°C in the dark. The mixture was sieved through an 83 mm metallic sieve, followed by dilution with KMC  salt solution. After centrifugation at 100 g for 5 min, the pellet was diluted in a KMC solution. Whatever the plant material, enzymatic solution and plant preconditioning (light/dark), only isolated plasmolysed cells were obtained.
Positive results were obtained when nodular calli were used as starting material. Floral explants cultivated on MS medium  complemented with 500 mg/l casein hydrolysate, 2 mg/l IAA, and 2 mg/l BAP lead to calli covered with nodular bodies. These calli were embryogenic. With the enzyme mixture 2.5% cellulase R10 'Onozuka', 0.2% hemicellulase, 0.3% pectolyase Y23 and 0.6% macerozyme, Bakry produced the first living banana protoplasts. After calcofluor staining, protoplasts were found to be cellulose-free, perfectly round in shape, variable in size and having a dense cytoplasmic content. Although the yield was quite low the protocol was reproducible.
This pioneering work was confirmed by Cronauer and Krikorian  in 1986 (using proliferating shoot tips of the cultivar 'Lacatan' [AAA group]) and by Da Silva Conceicao  in 1989. Later on regular progress was made by several scientists. Initially banana calli from protoplasts were developed in 1992 , followed by plant regeneration directly from protoplasts [14, 15, 34], protoplast transformation  and somatic hybridisation [36-39].
3.2. Explants used as a source of protoplasts
In banana, in vivo explants are not a good source of protoplasts. Although Matsumoto et al.  obtained protoplasts from in vivo bracts, in most cases initiation started from in vitro cultures. Leaf explants [28, 32, 41-43], slices of shoot tissue, roots  or callus [28, 32, 43, 45] were the first choice as starting material for protoplast isolation because of their convenience. In fact, protoplasts can be obtained in banana from almost any tissue, including young leaves, sheaths, bracts, roots, and callus, but yields depend on the explant source. Yields are quite low with most tissues and range from 0.1 to 2.8 × 106 protoplasts per gram, but the protoplasts are unable to divide. However, cell suspension-derived protoplasts (Figure 1) gave high yields which ranged from 4 to 5 × 105 per gram with 'Long Tavoy'; from 1.1 to 3 × 105 per gram with malaccensis ; 2 to 20 × 106 per gram with 'Maça'  to 6.6 × 107 per millilitre of packed cell volume (PCV) with 'Bluggoe' . Essential was the embryogenicity of the suspension. The first sustained cell divisions in protoplast culture leading to calluses were observed by Megia et al. in 1992  with M. acuminata cell suspension-derived protoplasts. Megia et al.  and Panis et al.  improved the protocol. This was confirmed by Matsumoto et al.  and Assani et al. [42, 43] who produced plants from such 'embryogenic protoplasts'. Consequently embryogenic cell suspensions are the starting material of choice.
3.3. Banana genotypes tested
Bananas are either triploid, diploid or tetraploid in order of importance for food production, but protoplasts have been obtained so far from triploid, diploid and haploid genotypes. Below are listed the cultivars and wild types tested; the letter A denotes the contribution of the subspecies acuminata and B the contribution of the subspecies balbisiana.
3.3.1. Triploid genotypes
3.3.2. Diploid genotypes
These haploids were derived from anther-culture calli: either the callus © or leaves from regenerating plants obtained from haploid calli was the source (l).
'Pisang Klutuk Wulung' (BB) ©; 'Pisang Klutuk' (BB) (c and l); 'Pisang Batu' (BB) © [42, 43].
3.4. Protoplast extraction and purification
Various enzyme mixtures are used depending on the author and banana explant (CS = Cell Suspension; P = Plant explant; C = callus) (see section 3.4.2.). They are commonly composed of:
Cellulases (Cellulase Onozuka R10, Cellulase Onozuka RS, Cellulysin, Macerase, Driselase)
Pectinases (Pectinase; Macerozyme R10; Pectolyase Y 23)
Hemicellulase (Rhozyme HP -150; Glusulase)
3.4.2. Protoplast isolation media
Different protoplast isolation media exist and are listed here:
0.5% Cellulysin, 0.5% Rhozyme HP-150, 0.125% pectolyase Y23, 100 mg CaCl2.2H2O, 10 mg NaH2PO4.H2O, 58.5 mg MES 2-(N-morpholino) ethanesulphonic acid, 12.6 g sugar, pH 5.5 ;
1% Cellulysin, 1% Rhozyme, 1% pectinase, 0.3 M mannitol, 0.3 M sorbitol, pH 5.8 (P) ;
1% cellulase, 1% Macerozyme R10, 1% pectinase, 0.55 M mannitol, 3 µM MES pH 5.8 20-24 h ;
1% cellulase (Onozuka R10), 1% Macerozyme R10, 1% pectinase (Serva), 7 mM CaCl2.2H2O, 0.7 mM NaH2PO4.2H2O, 3 mM MES, 10% mannitol, pH 5.8, 24 h [14, 34];
5% cellulase RS (Yakult), 2% pectolyase Y23 (Sheishin Co.), 3% KCl, 0.5% CaCl2, pH 5.6; 5 h ;
1.5% cellulase RS (Yakult), 0.2% pectolyase Y23 (Sheishin Co.), 0.2% hemicellulase Sigma Co, 3% KCl, 0.5% CaCl2, pH 5.6, 5 h ;
1.5% cellulase (Onozuka RS, Yakult Japan), 0.2% pectolyase Y23 (Seishin Co.), 0.6 M mannitol, 0.1 mM CaCl2, 0.5% PVP 40, 3.5 mM MES, 8 mg/l Bromocresol Purple, pH 5.7 15 h ;
2% cellulase RS (Yakult Honsha Co. Japan), 0.5% Macerozyme (Sigma), 0.2% hemicellulase (Sigma), 0.25% pectolyase Y23 (Seinshin Pharmaceutical Co., Tokyo, Japan), 3% KCl, 0.5% CaCl2, pH 5.6, 15-17 h ;
1.5% Cellulase RS (Yakult Honsha Co., Japan), 0.15% Pectolyase Y23 (Seishin Pharmaceutical Co.), 1% Macerozyme (Sigma), 204 mM KCl, 67 mM CaCl2, pH 5.6, 12-14 h ;
The same as above but without macerozyme ;
1.5% cellulase RS (Yakult Honsha Co., Japan), 0.15% pectolyase Y23 (Seishin Pharmaceutical Co.) 0.5% Macerozyme (Sigma), 204 mM KCl, 67 mM CaCl2, pH 5.6, 12-14 h .
It is standard practice to incubate the plant material in the enzyme solution for a digestion period of 1-24 h either with shaking (40 r.p.m.) or no shaking, generally at 27°C in the dark. The best source of protoplasts is from embryogenic cell suspension cultivated in the light (30 or 65 mE m-2 s-1) or darkness  and regularly subcultured. Initiation should start 3-5 days , 7-10 days  or 2-15 days  after the last subculture, followed by sieving through a 200-400 mm mesh just before enzyme incubation .
3.4.3. Protoplast purification
Protoplasts are purified from debris by sieving (100 mm and subsequently or directly through 56/32/25 mm sieves), which is preferred to flotation [34, 40, 42]. With flotation purification, a 21% sucrose solution and centrifugation at 120 g is used .
3.4.4. Protoplast washing
After sieving, protoplasts are generally repeatedly washed to remove enzymes. Protoplast pellets obtained by centrifugation (50/66/90 g) are washed with protoplast isolation medium without enzymes ; cell suspension culture medium without growth regulator plus 10% mannitol ; or 3% CaCl2 plus 0.5% KCl solution ; or 3% CaCl2 0.5% KCl solution and 10% mannitol ; or 0.6 M D-mannitol, 0.1 mM CaCl2, 0.5% PVP-40 and 3.5 mM MES .
3.5. Protoplast quality control
Complete cell wall degradation is confirmed both by the spherical shape of the released protoplasts and by the absence of fluorescence after staining with calcofluor white . Evans blue or FDA staining  are commonly used for viability assessment.
3.6. Protoplast culture media
For banana protoplast culture, derived MS  medium (supplemented with 2,4-D, or 2,4-D plus zeatin) or N-medium composed with N6 salt , vitamins, organic acid and sugar alcohol  are commonly used. Due to difficulties in obtaining favourable protoplast development, many culture systems have been tested. Liquid and solid media (derived from MS) were initially tried for protoplast cultivation, but it was found that liquid cultures are not suitable for Musa protoplasts. In contrast, solid media supported sustained divisions.
To overcome the low growth response, other media were tested, such as semi-solid media, protoplasts embedded in solid or semi-solid media, nurse cultures, and nurse cultures with a feeder layer. Cocultivation in banana protoplast cultures was considered as it was reported that coculturing protoplasts of recalcitrant species, especially monocotyledons, with a reliable feeder culture, induced cell division . The banana protoplast literature describes various combinations of media and culture systems but, whatever the protocol, a high plating density (105-106/ml) was crucial , as well as the use of feeder layer , or nurse culture  combined with a feeder layer [42,43,45]. Protoplasts are cultured at 27 ± 1°C in the dark.
3.6.1. Culture in liquid media
For culturing protoplasts in liquid media (Figure 2A), protoplasts with a density of 105-106 per millilitre are suspended in liquid media, with or without shaking. The following options are open:
N6 salt  and KM vitamins, organic acids and sugar alcohols according to Kao and Michayluk , 6.3% (w/v) glucose and 4% (w/v) sucrose as osmoticum, 260 mg/l KH2PO4 and 100 mg/l 2-(N-morpholino) ethanesulphonic acid MES ;
N6 salts , KM vitamins, organic acids and sugar alcohols , vitamins of Morel , 0.4 M glucose, 117 mM sucrose, 0.5 mM MES, 1.9 mM KH2PO4 ;
N6 salts , KM vitamins, organic acids and sugar alcohols , vitamins of Morel , 0.4 M glucose, 117 mM sucrose, 1.9 mM KH2PO4, 2.28 µM zeatin, 0.90 µM 2,4-D and 5.4 µM NAA ;
MS modified medium, 5 mM 2,4-D and 10% mannitol ;
1/2 MS, 10% mannitol ;
MS medium, 5 mM 2,4-D, 1 mM zeatin, 0.55 M mannitol ;
MS minor nutrients, 1/2 major nutrients, MS vitamins, 10 mg/l ascorbic acid, 20 g/l sucrose, 5 µM 2,4-D; 0.55 M D-mannitol ;
MS preconditioned medium, 5 mM 2,4-D, 10% filter-sterilised medium from a one-week-old cell culture .
Figure 2 The different protoplast culture techniques (modified from Matsumoto and Oka 1998).
3.6.2. Culture on solid medium
For culture on solid medium (Figure 2B) protoplasts are directly plated at high density on a gelatinous medium or on a membrane which covers the gelatinous medium. Such media are composed of:
MS minor nutrients, 1/2 major nutrients, MS vitamins, 10 mg/l ascorbic acid, 20 g/l sucrose, 5 µM 2,4-D, 0.275 M D-mannitol, 0.2% gelling gum ; or
MS 1/2 salt solution, 5% mannitol, 5 µM 2,4-D, 0.8% agarose or 0.2% Gelrite .
3.6.3. Culture in semi-solid medium
For culture on semi-solid medium (Figure 2C) protoplasts are cultivated on:
0.6% agarose, N6 salt , vitamins of Morel , vitamins of Kao and Michayluk ;
0.6% agarose, 0.35 M glucose, 40 g/l saccharose, 260 mg/l KH2PO4, 3 mM MES, and Pichloram, BAP, NAA, L-asparagine and L-proline at various concentrations ;
Protoplasts at twice final desired concentration in culture medium, are mixed with the same medium containing 0.3/0.8% agarose, 5-10% mannitol, 0-10 µM 2,4-D .
3.6.4. Culture embedded in solid medium
For embedment in solid medium (Figure 2D) protoplasts at twice the desired concentration (2 × 106) are placed in a twofold concentrated culture medium: MS minor nutrients (× 2), MS major nutrients, MS vitamins (× 2), 20 mg/l ascorbic acid, 40 g/l sucrose, 10 µM 2,4-D, but with 0.55 M D-mannitol. This is followed by mixing with an equal volume of solution containing a gelling gum at 0.6%, 0.55 M D-mannitol, previously melted and maintained at 55°C .
3.6.5. Culture in alginate or solid medium floating in liquid medium (bead type culture)
Protoplasts are cultured in a solid medium as in section 3.6.4, or in calcium alginate beads according to Schilde-Rentschler et al.  (Figure 2E). For this purpose, the alginate solution (2.8% alginic acid, 0.4 M mannitol) is first autoclaved. Protoplasts are added to the alginate solution to obtain a concentration of 1.0 × 106 protoplasts per millilitre. Later, drops of the protoplast-alginate mixture are polymerised in 0.4 M mannitol/50 mM CaCl2 solution, and transferred into protoplast liquid culture medium (see below) in small Petri dishes. Pieces of solid medium containing protoplasts (Section 3.7.4) are transferred into liquid medium:
MS medium, 10 mg/l ascorbic acid, 20 g/l sucrose, 5 µM 2,4-D, 0.275 M D-mannitol, with or without 0.3% activated charcoal ;
'A medium' consisting of N6 salts , KM vitamins, organic acids and sugar alcohols , vitamins of Morel , 117 mM sucrose, 1.9 mM KH2PO4 ;
'B medium' consisting of the 'A medium' containing 2.28 µM zeatin, 0.90 µM 2,4-D and 5.4 µM NAA .
3.6.6. Nurse cultures
188.8.131.52. Living cells as nurse in liquid medium
Protoplasts can be maintained according to one of the following procedures:
Nurse cell suspensions are maintained in N6 salt  with MS vitamins, 10 µM L-proline, 300 mg/l casein hydrolysate, 5 µM 2,4-D, and 3% (w/v) sucrose at pH 5.8;
1 ml PCV of nurse cells is freshly subcultured and mixed with 50 ml of MS minor nutrients, 1/2 major nutrients, MS vitamins, 10 mg/l ascorbic acid, 20 g/l sucrose, 5 µM 2,4-D and 0.275 M D-mannitol. Embedded protoplasts prepared as above (Sections 3.6.4 and 3.6.5.) are then immersed in the nurse solution. Cultures are placed on a rotary shaker at 120 r.p.m. .
A MillicellTM-HA (Millipore, 45 µm pore size) may be used as a physical barrier, to separate protoplasts from nurse cells during culture .
184.108.40.206. Feeder-layer culture of living cells
For banana protoplast culture, Oryza, Lolium multiflorum, Triticum monococcum (Poaceae) and banana cell suspensions are useful as nurse cells (Figure 2H). Feeder cultures are prepared one day before protoplast isolation by mixing 3 ml of feeder cell suspension with 100 ml of a medium containing 0.8% (w/v) melted Sea plaque agarose, MS basal salt, vitamins of Morel , 2 mg/l 2,4-D, 20 g/l sucrose, 72 g/l maltose, 250 mg/l glucose and 20 ml/l fresh coconut water . Feeder cells are then separated from overlaying protoplasts by a 10 mm nylon filter or 0.22 mm Millipore membrane.
Others options are:
2 ml of a nurse culture are poured on 20ml 1/2 MS, 5% mannitol, 5 µM 2,4-D, 0.2% Gelrite. The nurse culture is made from 1 ml of one-week-old suspension concentrated to 25% PCV mixed with 1 ml 1/2 MS supplemented with 10% mannitol, 200 mg/l myoinositol, 5 µM 2,4-D and 1.6% agarose .
Nurse cells are mixed in melted medium MS (but with major nutrients at half strength), 10 mg/l ascorbic acid, 5 µM 2,4-D, sucrose 20 g/l, 0.55 M D-mannitol, 0.8% agarose (Sea plaque) .
1 or 2 days before protoplast culture, prepare the feeder layer.
The cell suspensions are sieved through a 250 µm nylon filter to select small cell aggregates.
The PCM (Protoplast Culture Medium) liquid medium, which consists of MS salts, 9 µM 2,4-D, vitamins of Morel , 2.8 mM glucose, 278 mM maltose, 116 mM sucrose, 2.5 mM myoinositol (pH 5.7) is sterilised through 0.22 mm Millipore Millex GS filters (Millipore Corporation). Nurse cell suspensions are added to 100 ml double-concentration PCM culture medium to obtain a final cell concentration of 3-10% (v/v). 1.2 g agarose Sea plaque (Sigma) is dissolved in 100 ml water and then autoclaved (pH 5.7); when the agarose solution has cooled to 30-35°C, it is gently mixed with 100 ml PCM medium containing nurse cells, and 10-12 ml of the mixture is poured into small Petri dishes (5.5 cm diameter). After solidification, the medium is covered with a sterilised nitrocellulose filter (AA type, Millipore Corporation). 0.5 ml of the protoplast suspension in the A or B medium (section 3.6.4) is transferred onto the nitrocellulose filter. Cell-wall regeneration is observed with calcofluor white . All cultures (liquid culture, alginate culture and feeder-layer culture) are maintained at 27°C in the dark .
3.6.7. Culture on preconditioned medium
This cultivation technique (Figure 2I) resembles the feeder-layer technique, except that the preconditioned medium is the cell culture filtered through a Whatman filter, followed by mixing the filter-sterilised medium with the solidifying medium .
3.7. Plant regeneration
Sustained cell divisions are obtained with the feeder-layer technique, leading to plant development, while other techniques give only protoplast budding or few divisions. A high protoplast density (105-106 per millilitre) also appears to be crucial for plant regeneration. After 3-8 days of culture, first and second divisions are observed [8, 15]. After 4-6 weeks of cultivation, cell clumps consist of cells with embryogenic characters (i.e. small cell size, large nucleus with very stainable nucleolus, dense cytoplasm, small vacuoles). Since the clumps all have a similar size, it seems that they all arise at the same time from the protoplasts. These clumps exhibit a smooth surface, due to epidermis formation, and invaginate just like normal somatic embryos (Figure 3). These globules develop into rooted plantlets after 2-3 months.
For plant regeneration, cell clumps are transferred to a semi-solid medium without nurse cells, devoid of mannitol and growth regulators , or are transferred onto a medium containing 2.2 mM BAP and 2 mM pichloram for one week before transfer to MS medium containing 10 mM BAP to improve the rate of conversion of globules into plants . The membrane on which one-month-old clusters are growing is transferred onto solid medium with minor MS nutrients, 1/2 major MS nutrient strength, MS vitamins, 10 mg/l ascorbic acid, 20 g/l sucrose, 5 µM 2,4-D, 0.275 M D-mannitol, 0-2% gelling gum  to produce embroids. Their maturation is achieved by transfer on hormone-free medium without mannitol, and finally on MS solid medium without growth regulators. Plantlet regeneration increased in the presence of BAP and IAA at low concentrations (2 mM)  before transplanting to the soil in the greenhouse.
Figure 3 Somatic embryos derived from protoplast culture on feeder layer
Protoplasts in a medium containing N6 salts , KM vitamins, organic acids and sugar alcohols , vitamins of Morel , 117 mM sucrose, 1.9 mM KH2PO4, 2.28 µM zeatin, 0.90 µM 2,4-D and 5.4 µM NAA on banana feeder layer gives clusters in 3 weeks . Cultivation of these clusters on regeneration medium containing 2.2 mM BAP and 2.3 mM IAA results in somatic embryos with a 2% conversion rate of protoplast-derived cell suspension into embryos. These embryos are transferred onto MS medium supplemented with 1.2 mM NH4NO3 for rooting and elongation . Protoplast-derived bananas have now been field-established and are investigated for their performance.
4. PROTOPLAST HYBRIDISATION
Few groups have attempted somatic hybridisation in banana. The first experiments were made by Chen and Ku  using PEG (polyethylene glycol) fusion, but at that time a protocol for plant regeneration from banana protoplasts was not yet available. Nowadays such research continues at EMBRAPA (Empresa Brasiliera de Pesquisa Agropecuaria) and at Paris-Sud University, France, using electrofusion approaches [37-39,42,46] (Figure 4). This resulted in plant regeneration (Figure 5), and the first putative hybrids have been selected for ploidy by flow cytometry and established for field evaluation.
5. PROTOPLAST TRANSFORMATION
Banana transformation has much to offer for banana improvement, but technologies relying on protoplasts have not been fully explored yet. The only reports come from KULeuven (Catholic University of Leuven), Belgium [35,56,57] and show that foreign genes can be inserted through electroporation. Embryogenic cell suspensions from the cultivar 'Bluggoe' were subjected to electroporation conditions consisting of 500 and 960 mF capacitance, PEG treatment, electroporation buffer, and heat shock, resulting in at most 1.8% protoplasts expressing the introduced DNA. These results show that non-chimeric transformed plants could potentially be produced from unicellular plant material.
Figure 4 Protoplast chains during electrofusion
Figure 5 Banana plant regeneration from protoplast derived embryos
6. PRACTICAL PROTOCOL FOR PROTOPLAST ISOLATION AND CULTURE IN BANANA
6.1. Isolation of protoplasts from cell suspensions
For the isolation of protoplasts from embryogenic cell suspension cultures, suspensions should be selected 3-4 days after their last subculture. They then need to be sieved through sterile 500 and 200 mm stainless mesh. The enzyme solution is made of 1.5% cellulase RS (Yakult Honsha Co., Tokyo, Japan), 0.15% pectolyase Y23 (Seishin Pharmaceutical Co., Tokyo, Japan), 204 mM KCl, 67 mM CaCl2 (pH 5.6) and centrifuged at 3000 r.p.m. for 1 h, after which the supernatant is filtered through a 0.22 mm sterile Millipore filter. Two millilitres of enzyme mixture is added to 1 ml of cell suspension, followed by overnight (12-14 h) incubation at 27°C without shaking.
6.2. Isolation of protoplasts from callus and leaves
Embryogenic calli, or young leaves from in vitro plants (4-5 weeks old) are sliced and transferred into 150 ml Erlenmeyer flasks supplemented by a side nozzle connected to a Millipore filter (0.22 mm pore size, Millex GS filters, Millipore Corporation). Then 1 g of callus or leaves is mixed with 10 ml enzyme solution. For callus digestion, the enzyme solution consists of 1.5% cellulase RS (Yakult Honsha Co.), 0.15% pectolyase Y23 (Seishin Pharmaceutical Co.), 1% macerozyme (Sigma), 204 mM KCl and 67 mM CaCl2 (pH 5.6). For leaf digestion, the enzyme solution consists of the same composition except for cellulase RS (1%), and the addition of 0.5 M mannitol. The mixture is incubated on a rotary shaker (30 r.p.m.) for 12-14 h at 27°C in the dark.
6.3. Purification of protoplasts
For purification, the digestion mixture is filtered through a combined 100/25 mm metallic mesh to remove debris and large cell colonies. Protoplasts are then centrifuged at 66 g for 5 min. The pellet is washed twice (centrifugation at 66 g for 5 min) with a mixture of 204 mM KCl and 67 mM CaCl2. Protoplast viability is then determined by fluorescein diacetate (FDA) and the yield counted on a Nageotte haemocytometer.
6.4. Culture of protoplasts
To induce cell divisions in cultured protoplasts, nurse cultures have to be prepared a day before protoplast isolation. Cell suspensions in the exponential growth phase are sieved through a 250 mm metallic mesh to select only small cell aggregates. The PCM medium is prepared by mixing MS salts, 9 mM 2,4-D, vitamins of Morel, 2.8 mM glucose, 278 mM maltose, 116 mM saccharose and 2.5 mM myoinositol (pH 5.7) and sterilised by filtration. Sieved cell suspensions are mixed with 100 ml double-concentration PCM liquid medium, to obtain a final cell concentration of 10%.
The subsequent steps are then:
Dissolve 1.2 g agarose Sea plaque (Sigma) in 100 ml double-distilled water followed by autoclaving (pH 5.7). As soon the temperature of the agarose solution has decreased to 30-35°C, 100 ml of PCM medium containing nurse cells is gently mixed into it. Then 10-12 ml of the mixture is poured into small Petri dishes (5.5 cm diameter) and covered by a sterilised nitrocellulose filter (AA type, Millipore Corporation).
Suspend 0.5 × 106 freshly isolated protoplasts in 0.5 ml liquid culture medium (1.0 × 106 protoplasts per millilitre) and transfer 300 ml aliquots onto a nitrocellulose filter.
The protoplast culture medium is composed of N6 salts , vitamins, organic acids and sugar alcohols , Morel vitamins , 117 mM sucrose, 0.4 M glucose, 0.5 mM 2-(N-morpholino) ethanesulphonic acid (MES), 1.9 mM KH2PO4, 2.3 mM zeatin, 0.9 mM 2,4-D and 4.4 mM NAA (pH 5.7). Sterilise by filtration. The cultures are maintained at 27°C in the dark. Cell wall regeneration is observed with calcofluor white  under a UV microscope.
6.5. Somatic embryogenesis
Protoplast-derived embryos are individually picked up from the feeder layer after about 4-8 weeks of culture (depending on the genotype) and gently transferred onto regeneration medium containing MS salts, Morel vitamins, 88 mM saccharose, 2.3 mM IAA, 2.2 mM BAP, and 7.5 g/l agarose sea plaque (pH 5.7). They are maintained at 27°C in the dark. Regenerated plants are transferred onto solidified growth regulator-free MS medium with 1.2 mM NH4NO3 (pH 5.7) and cultured under a 16 h photoperiod (65 mE m-2 s-1) at 27°C. After reaching a size of 10 cm, the plants are transplanted to the soil.
This work was performed within the framework of the European INCO-DC Project ERBIC18- CT97 0204 and C.M.I.F.M. AI n° 00/230/SVS.
 HABERLANDT, G., Kulturversuche mit isolierten Planzenzellen, S.B. Weisen Wien Mathnaturw. 111 (1902) 69-92.
 NICKELL, L.G., TORREY, J.G., Crop improvement through plant cell and tissue culture, Science 166 (1969) 1068.
 CORNEL, D., et al., Measurement of intracellular potassium activity in protoplasts of Acer pseudoplatanus: origin of their electropositivity, Physiol. Plant. 57 (1983) 203-209.
 RAHAT, M., REINHOLD, L., Rb+ uptake by isolated pea mesophyll protoplasts in light and darkness, Physiol. Plant. 59 (1983) 83-90.
 CHAPMAN, K.S.R., HATCH, M.D., Intracellular location of phosphoenol pyruvate carboxykinase and other C-4 photosynthetic enzymes in mesophyll and bundle sheath protoplasts of Panicum maximum, Plant Sci. Lett. 29 (1983) 145-154.
 CHANG, T.Y., et al., Effect of abscisic acid on maize root protoplasts, Z. Planzenphysiol. 110 (1983) 127-133.
 KIM, I.S., SONG, P.S., Binding of phytochrome to liposomes and protoplasts, Biochemistry 20 (1981) 5482-5489.
 ASSANI, A., et al., Plant regeneration from protoplasts of dessert banana cv. Grande Naine (Musa spp, Cavendish subgroup AAA) via somatic embryogenesis, Plant Cell Rep. 20 (2001) 482-488.
 COCKING, E.C., A method for isolation of plant protoplasts and vacuoles, Nature 187 (1960) 927-929.
 CRONAUER, S.S., KRIKORIAN, A.D., "Banana (Musa spp.)", Biotechnology in agriculture and forestry, (BAJAJ, Y.P.S., Ed.), Springer Verlag, Berlin 1 (1986) 233-252.
 NOVAK F.J., et al., Somatic embryogenesis and plant regeneration in suspension cultures of dessert (AA, AAA) and cooking (AAB) bananas, BioTechnology 46 (1989) 125-135.
 COTE, F.X., et al., Embryogenic cell suspension from the male flower of Musa AAA cv. Grande Naine (AAA), Physiol. Plant. 97 (1996) 285-290.
 KERBELEC, F., Etablissement d'une technique d'androgenèse pour l'amélioration génétique du bananier (Musa spp.) Thèse de Doctorat, Ecole Nationale Agronomique de Rennes (1996).
 PANIS, B., et al., Plant regeneration through direct somatic embryogenesis from protoplasts of banana (Musa spp.), Plant Cell Rep. 12 (1993) 403-407.
 MEGIA, R., et al., Plant regeneration from cultured protoplasts of the cooking banana cv. Matavia (Musa ssp., ABB group), Plant Cell Rep. 13 (1993) 41-44.
 SAGI, L., et al.,. Genetic transformation of banana and plantain (Musa spp.) via particle bombardment, BioTechnology 13 (1995) 481-485.
 MAY, G.D., et al., Generation of transgenic banana (Musa acuminata) plants via Agrobacterium-mediated transformation, BioTechnology 13 (1995) 486-492.
 SIHACHAKR, D., et al., "Regeneration of plant from protoplasts of eggplant (Solanum melongena L.)", Plant protoplasts and genetic engineering IV, (BAJAJ, Y.P.S., Ed.), Springer Verlag, Berlin, Biotechnology in Agriculture and Forestry 23 (1993) 108-121.
 DUCREUX, G., et al., Exploitation of genetic and physiological variability in Solanaceae: the examples of potato and eggplant, Acta Hort. 289 (1991) 65-75.
 JONES, M., Fusing plant protoplasts, Trends Biotechnol. 6 (1988) 153-158.
 SIMMONDS, N.W., The evolution of the bananas, Longman, London (1962).
 SIMMONDS, N.W., Bananas, Longman, London (1966).
 ROWE, P.R., L'hybridation pour l'amélioration des plantains et autres bananes à cuire, Fruits 38 (1983) 256-260.
 STOVER, R.H., BUDDENHAGEN, I.W., Banana breeding: polyploidy, disease resistance and productivity, Fruits 40 (1986) 175-191.
 ROWE, P.R., ROSALES, F., Genetic improvement of banana, plantains and cooking banana in FHIA, Honduras, Breeding banana and plantain for resistance to diseases and pests, (GANRY, J., Ed.), INIBAP, Montpellier, France, (1993) 243-266.
 ZUBA, M., BINDING, H., "Isolation and culture of potato protoplasts", Plant protoplasts and genetic engineering I, (BAJAJ, Y.P.S., Ed.), Springer Verlag, Berlin, Biotechnology in agriculture and forestry 8 (1989) 124-146.
 BAKRY, F., Application des techniques de culture in vitro pour l'amélioration du bananier (Musa sp.), Thesis Paris-Sud University (1984).
 BAKRY, F., Choix du matériel à utiliser pour l'isolement de protoplastes de bananier (Musa sp.), Fruits 39 (1984) 449-452.
 FREARSON, E.M., et al., The isolation, culture and regeneration of Petunia leaf protoplasts, Dev. Biol. 33 (1973) 130-137.
 HARM, C.T., POTRYKUS, J., Fractionation of plant protoplasts by iso-osmotic density gradient centrifugation, Theor. Appl. Genet. 53 (1978) 57-63.
 MURASHIGE, T., SKOOG, F., A revised medium for rapid growth and bioassays with tobacco tissue culture, Physiol. Plant. 15 (1962) 473-497.
 DA SILVA CONCEICAO, Isolement et culture de protoplastes de bananiers (Musa sp.). Etude de divers facteurs. DEA génétique et sélection Animale et Végétale, Université Paris Sud Orsay (1989).
 MEGIA, R., et al., Callus formation from cultured protoplasts of banana (Musa sp.), Plant Sci. 85 (1992) 91-98.
 PANIS, B., et al., "Regeneration of plant from protoplasts of Musa species (Banana)", Plant protoplasts and genetic engineering V, (BAJAJ, Y.P.S., Ed.), Springer Verlag, Berlin, Biotechnology in Agriculture and Forestry 29 (1994) 102-114.
 SAGI, L., et al., Transient gene expression in electroporated banana (Musa ssp., cv. Bluggoe, ABB group) protoplasts isolated from regenerable embryogenetic cell suspensions, Plant Cell. Rep. 13 (1994) 262-266.
 HAÏCOUR, R., et al., "Use of cell and protoplast cultures for banana improvement biotechnologies", Breeding Banana and Plantain for resistance to diseases and pests (CIRAD, Ed.), (1993) 327-338.
 MATSUMOTO, K., et al., Isolamento e fusao dos protoplastos de banana, Revista Brasiliera de Fruticultura, Cruz das Almas 14 (1992) 27-33.
 MATSUMOTO, K., et al., "Somatic hybridization by electrofusion of banana protoplasts", XXV International Horticultural Congress, Brussels. (Abstracts, XXV International Horticultural Congress, Science and Horticulture Interface and Interactions), Brussels: International Society for Horticultural Science (1998) 433.
 MATSUMOTO, K., Hibridos somaticos, fusao celular para a obtencao de hibridos interespecies de banana, Biotecnologia Ciencia Desenvolvimento 20 (2001) 26-28.
 MATSUMOTO, K., et al., "Isolation and culture of bract protoplasts of banana", International Rice Research Institute (IRRI) and Academia Sinica (Eds), Genetic manipulation in crops, Cassell Tycooly, Philadelphia USA (1988) 414-415.
 CHEN, W.H., KU, Z.C., Isolation of mesophyll cells and protoplasts and protoplast fusion and culture in banana, J. Agric. Assoc. China 129 (1985) 56-67 (in Chinese).
 ASSANI, A., et al., Les progrès réalisés dans la regénération des protoplastes de bananiers (Musa ssp). Des modèles biologiques à l'amélioration des plantes, IRD Editions (2001) 193-206.
 ASSANI, A., et al., Influence of donor material and genotype on protoplast regeneration in banana and plantain cultivars (Musa spp.), Plant Sci. 162 (2002) 355-362.
 CRONAUER, S.S., In vitro growth responses of Musa, Ph.D. Thesis, State University of New York at Stony Brook, USA (1986).
 MATSUMOTO, K., OKA, S., Plant regeneration from protoplasts of a Brazilian dessert banana (Musa ssp., AAB group), Acta Hort. 490 (1998) 455-462.
 HAÏCOUR, R., et al., Maîtrise de la culture et de la regénération des protoplastes de bananier, en vue de la création de nouvelles structures génétiques. Quel avenir pour l'amélioration des plantes, (LIBBEY, J., Ed.), Actualités scientifiques, Eurotext, Paris, UREF AUPELF (1994) 211-225.
 GALBRAITH, D.W., Microfluorometric quantitation of cellulose biosynthesis by plant protoplasts using calcofluor white, Physiol. Plant. 53 (1981) 111-116.
 WIDHOLM, J.M., The use of fluorescein diacetate and phenosafranin for determining the viability of cultured cells, Stain Technol. 47 (1972) 189-194.
 CHU, C.C., et al., Establishment of efficient medium for anther culture of rice through comparative experiments on the nitrogen source, Scientia Sinica 17 (1975) 659-668.
 KAO, K.N., MICHAYLUK, M.R., Nutritional requirements for growth of Vicia hajastana cells and protoplasts at a very low population density in liquid media, Planta 126 (1975) 105-110.
 SMITH, J.A., et al., Feeder layer support of low density population of Zea mays L. suspension cells, Plant. Sci. Lett. 36 (1984) 67-72.
 MOREL, G., WETMORE, R.H., Fern tissue culture, Am. J. Bot. 38 (1951) 141-143.
 SCHILDE-RENTSCHLER, L., et al., Somatic hybridization of dihaploid potato breeding, II Progress in plant protoplast research, (KRENS, K.J., Ed.), Current Plant Science and Biotechnology, Kluwer Academic Publishers, Dordrecht (1988) 195-196.
 NAGATA, T., TAKEBE, I., Cell wall regeneration and cell division in isolated tobacco mesophyll protoplasts, Planta 92 (1970) 301-308.
 MATSUMOTO, K., OKA, S., "Studies of nurse culture methods on banana protoplast culture", VI Congresso Brasileiro De Fisiologia Vegetal, Belém. Resumos, VI Congresso Brasileiro de Fisiologia Vegetal. Belém: Sociedade Brasileiro de Fisiologia Vegetal (1997) 154-154.
 SAGI, L., et al., Transient expression in banana protoplasts, Banana Newsletter 15 (1992) 42-42.
 SAGI, L., et al., Genetic Transformation in Musa species (Banana), Plant protoplasts and genetic engineering IV, (BAJAJ, Y.P.S., Ed.), Springer Verlag, Berlin, Biotechnology in Agriculture and Forestry 34 (1995) 214-237.