21. Detecting ploidy level instability of banana embryogenic cell suspension cultures by flow cytometry - Roux, N.S.^[42], H. Strosse^[43], A. Toloza^[44], B. Panis^[45], J. Dolezel^[46]

Abstract

During micropropagation of bananas and plantains, somaclonal variation can occur in regenerated plantlets. This variation may interfere with the use of these cultures for physical or chemical mutagenesis and/or genetic transformation. Although the causes of genetic instability are poorly understood, chromosome instability is believed to be one of the most common causes of tissue culture-induced variation. Using flow cytometry, variation in chromosome number could be detected in embryogenic cell suspensions and in plants regenerated from them. Results obtained by flow cytometry were verified by chromosome counting in meristem root-tip cells. Abnormalities in DNA content such as polyploidy and aneuploidy were detected at the cell suspension level. For the first time, a hypotriploid banana embryogenic cell line with 2n = 28 (i.e. with loss of five chromosomes) was reported. Factors affecting the genetic stability of embryogenic cell suspensions are discussed.

1. INTRODUCTION

The use of shoot-tip cultures in Musa has already allowed great progress in mass propagation (micropropagation), conservation (medium-term conservation and cryopreservation), elimination of virus diseases, and exchange of germplasm [1,2]. Unfortunately, the use of banana shoot tips as target tissues for genetic engineering strategies such as in vitro mutagenesis and genetic transformation can lead to chimeric plants [3]. Somatic embryogenesis is usually the preferred mode of regeneration to organogenesis because of the presumed single-cell origin of the regenerants [4,5]. The occurrence of off-types has delayed widespread industry acceptance of micropropagated bananas [6]. Before adopting somatic embryogenesis as a new method to support genetic improvement and mass propagation of bananas and plantains, further studies need to be undertaken to understand the phenomenon of somaclonal variation. The term 'somaclonal variation' was introduced to describe the genetic variation in plants regenerated from any form of cell culture. Larkin and Scowcroft [7] advocate the view that somaclonal variation represents a new source of variability, and therefore constitutes a powerful tool for the breeder. Nevertheless, only a narrow spectrum of variants is obtained through somaclonal variation as a breeding methodology, and it is becoming increasingly clear that somaclonal variation is usually undesirable [8]. Although the nature of somaclonal variation is poorly understood, chromosome instability is believed to be one of the commonest causes of tissue culture-induced variation [9]. Numerical chromosome aberrations (aneuploidy and polyploidy) are more common among polyploids (e.g. sugarcane, wheat, oats, triticale, potato and banana) than in diploids, since the latter cannot easily tolerate aneuploidy [10]. Aneuploidy and polyploidy have traditionally been detected by chromosome counting in regenerated plants [11,12], but this is a time-consuming and laborious procedure. Moreover, chromosome counting is not practical for large-scale screening [13]. In cell suspensions or calli the number of dividing cells is rather low, and chromosome observation difficult. Measurement of the nuclear DNA content by flow cytometry has been suggested as an alternative [14]. The method is now being increasingly used for large-scale ploidy screening of callus cultures [15] and cell suspension cultures [16]. Flow cytometric analysis has already been established in Musa spp. [17,18]. The present paper studies the possibility of verifying the stability of the ploidy level of embryogenic cell suspensions prior the formation of embryos or plants. The methodology and factors affecting the ploidy level stability of embryogenic cell suspensions are discussed.

2. MATERIAL AND METHODS

2.1. Plant material

Embryogenic cell suspensions (ECS) of the triploid (2n = 3x = 33) clones Grande Naine (ITC.1256), Williams (ITC.0365), Three Hand Planty (ITC. 0185) and Bluggoe (ITC.0010) were obtained from the Laboratory of Tropical Crop Improvement at the Catholic University of Leuven (K.U.Leuven), Belgium.

2.2. Embryogenic cell suspension culture

The embryogenic cell suspensions (ECS) were initiated via the scalp method, according to Dhed'a et al. [19] and Schoofs [20]. The method consists of three steps: (a) the preparation of highly proliferating material; (b) the induction of embryogenic cell aggregates; and (c) the transfer of embryogenic cell aggregates in liquid medium and their maintenance. These steps were executed at K.U.Leuven (see Chapter 4 in this book).

2.2.1. Maintenance of ECS

The suspensions were maintained in 100 ml Erlenmeyer flasks containing 15 ml of liquid half strength MS [21] salts supplemented with 0.3 mg/l thiamine, 10 mg/l ascorbic acid, 5 µM 2,4-D, 1 µM zeatin, and 30 g/l sucrose, and the pH was adjusted to 5.8 before autoclaving. Embryogenic cultures originating from different scalps (cell lines) were not mixed. Erlenmeyer flasks were placed on an orbital shaker at 70 r.p.m.. The medium was refreshed every 2 weeks. These ECS were multiplied until a sufficient number of 100 ml flasks were obtained for further experiments.

2.2.2. Regeneration of ECS

Regeneration of ECS into pro-embryos, embryos and finally plantlets was achieved in three steps on three different media as described by Schoofs [20]: R1 (hormone free), R2 (containing BAP), and R3 (containing BAP and IAA) respectively. Regeneration was executed on semi-solid medium in Petri dishes (9 cm diameter) for R1 and R2 media and in glass tubes (2.5 × 15 cm) for R3 medium. Embryos developed in R2 were also regenerated in Temporary Immersion System (TIS) [22] containing R3 medium.

2.3. Methods for genetic instability detection in DNA content

2.3.1. Flow cytometric analysis

Flow cytometric analysis was performed with a PA-I flow cytometer (Partec, Münster, Germany). Samples were prepared according to Dolezel et al. [17,18] with some modifications. Briefly, 3-4 drops of highly concentrated cell suspension were transferred to a plastic Petri dish, and the liquid medium was removed before adding 0.5 ml of OTTO I buffer, consisting of 0.1 M citric acid and 0.5% Tween 20 [23]. The cell clumps where then chopped with a sharp razor blade together with a 20 mg of leaf sample of the diploid banana 'Pisang Mas' (2n = 2x = 22) used as internal reference standard. Alternatively, chicken red blood cell nuclei (CRBC), prepared according to Galbraith et al. [24] were added to the suspension of released nuclei as an internal reference standard. The sample was then filtered through a 50 µm nylon mesh. In some cases, instead of chopping a cell suspension, protoplasts were isolated from ECS according to Panis et al. [25]. To stain nuclear DNA, 2 ml OTTO II buffer containing 0.4 M sodium hydrogen phosphate and 5 µM DAPI (4',6-diamidino-2-phenylindole)[23] was added to the suspension of released nuclei. The gain of the instrument was adjusted so that the G_o/G₁ peak of the internal standard CRBC or 'Pisang Mas' nuclei was positioned approximately at channel 100. The relative DNA content of Musa was then determined by comparing peak positions of the sample nuclei with the internal 'Pisang Mas' standard or CRBC nuclei. In each sample, a total of 5000-10,000 nuclei were analysed.

2.3.2. Chromosome counting

Slides for chromosome counting were prepared according to Dolezel et al. [26]. Meristems of actively growing roots were treated with 0.05% 8-hydroxyquinoline for 3 h at room temperature and fixed overnight in ethanol:acetic acid (3:1). The roots were stored in 70% ethanol at 4ºC for up to several months. Meristem tips of 5-15 roots per plant were digested for 60 min at 30ºC in an enzyme mixture consisting of 1% pectinase (Sigma P-2401), 0.5% pectolyase (Sigma P-3026) and 0.5% cellulase (Serva 16419) prepared in 0.1 M citrate buffer (pH 4.7). The suspension of released protoplasts was filtered through a 150 µm nylon mesh and washed in 75 mM KCl and 7.5 mM EDTA (pH 4). Finally, protoplasts were stored in 70% ethanol at -20ºC for up to several months. To prepare a slide, 7 µl of protoplast suspension in 70% ethanol was dropped onto an ice-cold slide. The suspension was allowed to spread out and air-dried. Shortly before complete drying out, 7 µl ice-cold ethanol:acetic acid (3:1) fixative was added to the drop to induce cell bursting. Just before complete drying, the slide was briefly rinsed in 100% ethanol and air-dried at room temperature. Chromosomal DNA was stained with 100 µl DAPI (2 µg/ml) on the slides for 10 min. The stain was removed by washing the slide with 2 × SSC (0.3 M NaCl, 0.03 M Na citrate, pH 7), and the slides mounted in antifade solution (Vectashield^®, Vector Laboratories). Slides were observed under a fluorescence microscope (Nikon, Eclipse E-800), and chromosomes counted under a 100 ×, NA 1.35 oil immersion objective. Images were captured using a CCD camera and processed using image analysis software (Lucia, version 4.21). For each plant, at least ten metaphase cells showing well scattered and contracted chromosomes were counted.

3. RESULTS

3.1. Ploidy of ECS regenerating normal plants

The quality of embryogenic cell suspensions differed from one genotype to the other. For example, ECS from Three Hand Planty were yellower and tended to form clumps after four to five subcultures (Figure 1B), whereas ECS from Williams were finer and whiter (Figure 1A). The ECS were checked regularly for their genetic stability. Most of the ECS displayed a ploidy level similar to the accession they originated from. A 4-year-old embryogenic suspension culture of the cultivar Three Hand Planty and a one-year-old embryogenic suspension culture of Williams were analysed, and the results showed normal triploid peaks (Figure 1), and the phenotype of young plants growing in the greenhouse looked normal.

Figure 1 Triploid embryogenic cell suspensions in maintenance medium and relative DNA content obtained after simultaneous analysis of nuclei isolated from Musa embryogenic cell suspensions and chicken red blood cell nuclei (CRBC, internal standard). Flow cytometer was adjusted to have the CRBC nuclei peak at channel 100.

A 'Williams', 2n=3x (WIL3x); B 'Three Hand Planty', 2n=3x (THP3x).

Relative DNA content of cell lines was expressed as a DNA index (DI) calculated according to:

2. Polyploidy detection

A non-regenerable Grande Naine suspension, characterized by the absence of large cell clumps and a white colour, was grew twice as fast as regenerable embryogenic cell suspensions. Flow cytometric analysis revealed that this unusual suspension was in fact polyploidised to the 10 x level, compared with the normal triploid Grande Naine regenerated from shoot tips (Figures 2A, B).

Figure 2 Relative nuclear DNA contents obtained after simultaneous analysis of nuclei isolated from Musa embryogenic cell suspensions and chicken red blood cell nuclei (CRBC, internal standard). (A) 'Grande Naine', 2n = 3x (GN3x); (B) Polyploid 'Grande Naine', 2n = 10x (GN10x); (C) Polyploid 'Williams', 2n = 6x (WIL6x). The flow cytometer was adjusted and the relative DNA content of cell lines calculated as described in the legend to Figure 1

A non-regenerable Williams suspension, also characterized by its fine structure, white colour and fast growth, was found to be hexaploid by flow cytometric analysis (Figure 2C). In order to avoid misinterpretation between cells in G₂ and polyploidised cells, we measured the ploidy level 15 days after subculture, when most of the cells are presumed to be in the G₁ phase of the cell cycle due to the reduced proportion of cycling cells.

Figure 3 Effect of the number of subcultures (vegetative generations) on polyploidy induction of embryogenic cell suspensions. A 'Three Hand Planty' cell line THP5. B 'Williams' cell line WIL124T. Flow cytometric measurements were made just before subculture. The columns at each vegetative generation represent the mean of four replicates

3.3. Effect of the number of subcultures on the ploidy of ECS

Because of the above findings, we decided to evaluate the effect of the number of subcultures on the ploidy level of ECS. Since ECS are subcultured every 2 weeks, 3-4 drops of highly concentrated cell suspensions were measured by flow cytometry just before subculture. The measurement were made on two cell lines from two different genotypes: Three Hand Planty (THP 5A) and Williams (Wil 124T). Four 100 ml Erlenmeyer flacks were chosen for each cell line. The suspensions were followed during ten subcultures (i.e. 20 weeks) (Figures 3A, 3B). Figure 3A shows a subpopulation of apparently hexaploid cells. These represent most probably triploid cycling cells in the G₂ phase, since the proportion of these nuclei never exceeds 10% of the total number of nuclei measured. We can conclude that during ten subcultures the ploidy of Three Hand Planty (AAB) ECS remained stable. The ploidy of Williams (AAA) ECS, however, proved to be unstable; after seven subcultures all cells of cell line Wil124T were polyploidised to the 6x level. Occasionally cells of each genotype were transferred to a regeneration medium. The hexaploid cells died after 2 weeks and could not regenerate into plants.

Figure 4 Aneuploid detection in embryogenic cell suspensions and in vitro derived plants. A Two plants of Bluggoe: An off-type plant (chlorotic and dwarf) regenerated from a 9 year old cell suspension culture and a true-to-type plant regenerated from shoot tip culture; B Histograms of relative nuclear DNA content obtained after simultaneous analysis of nuclei isolated from B1 Musa embryogenic cell suspensions (BG3x-?) or from B2 protoplasts (BG3x-?), leaves of a diploid banana plant Pisang Mas (PM2x) and a triploid true-to-type Bluggoe (BG3x), which served as internal standards. Flow cytometer was adjusted so that the peak representing PM2x nuclei was localized at channel 100. DI represents the DNA index and CV represents the coefficient of variation calculated for each peak. C Mitotic metaphase plates of an off-type plant of Bluggoe (2n=33-5=28) and of a true-to-type plant of Bluggoe (2n=3x=33)

3.4. Aneuploidy detection

Cell suspension from the cultivar Bluggoe (ABB, cooking banana) regenerated normal plants for several years. After 9 years the regeneration capacity of this cell suspension declined, and the regenerants were dwarf and chlorotic when compared with the normal type (Figure 4A). Flow cytometric analysis of nuclei isolated from suspension cultures resulted in a histogram with a DNA peak positioned between the peak reference standard diploid banana Pisang Mas (2n = 22) and the peak of the triploid control banana (2n = 33) obtained from leaves of true-to-type Bluggoe plants regenerated from shoot tips (Figure 4B). The analysis of nuclei isolated from protoplasts obtained from embryogenic cell suspensions resulted in narrower DNA peaks than in the chopped cell suspensions. The fact that the peak was narrow (coefficient of variation = 1.73%) indicated that the suspension became 100% aneuploid during the extended period of subculturing (9 years). By linear extrapolation we assumed that this Bluggoe suspension consisted of cells with 28-29 chromosomes. Conclusive proof was obtained after regenerating plants from this cell suspension and counting the chromosomes in their root tip meristems; metaphase plates revealed that cells from this suspension were indeed hypotriploid (2n = 28) (Figure 4C).

4. DISCUSSION

4.1. Evaluation of the method

In this report we show that flow cytometry is a useful method to monitor the genetic stability of ECS cultures. It also proves that it is possible to release intact nuclei by mechanical chopping of fresh intact suspension cells. The method is easy to use, fast, and gives good results with all cell lines tested. Nevertheless, in some cases the resolution of DNA peaks could be improved by analysing nuclei isolated by protoplast lysis instead of chopping fresh tissue, without affecting the ploidy of Bluggoe cell suspensions (Figures 4A, B). The characteristics of the cell suspension played an important role in the quality of the DNA content histograms. ECS producing a high level of polyphenols were characterised by histograms with a high background and a higher coefficient of variation of DNA peaks (results not shown). Nevertheless, cell suspensions that do produce high amounts of polyphenols, but that consist of relatively large cell clumps, could be reliably analysed. We believe that the presence of large cell clumps facilitates isolation of nuclei by chopping. If the cell suspension was too fine (containing no cell clumps at all), DNA peaks could not be distinguished from the debris background even when using the protoplast isolation technique. This could be due to the low yield of protoplasts and/or nuclear damage, because of extended treatment with cell-wall degrading enzymes. In order to use protoplasts isolated from fine ECS for flow cytometric analysis, the protoplast isolation protocol probably needs to be further refined.

For polyploidy detection, the inclusion of an internal reference standard is not always essential, but is nevertheless recommended for the estimation of a precise ploidy level. For aneuploidy detection an internal reference standard is always necessary. Most of the figures in this paper show the diploid Musa Pisang Mas used as internal reference standard. However, we now prefer to use Chicken Red Blood Cell (CRBC) nuclei, since their DNA content is closer to triploid Musa species. The use of CRBC nuclei is more convenient, the procedure is more reproducible, and leads to DNA histograms with narrower peaks.

4.2. Type of variation detected

Polyploid ECS with a high proliferation rate have a tendency to produce very white cells. Nevertheless it is difficult to find phenotypic markers that can help in identifying cell suspensions that are genetically unstable, since the colour of the cell suspension is genotype-dependent, and some euploid genotypes are also characterized by a whitish colour.

All the ECS under investigation that showed genetic instability partially or completely lost their regeneration capacity. This agrees with the observations of Kubalakova et al. [15], who showed that polyploidisation in embryogenic cell suspensions of cucumber was accompanied by a gradual loss of regeneration ability. Regeneration of embryos may act as a selection marker for cytologically normal cells [27,28]. In the case of a 9-year-old cell suspension of the cultivar Bluggoe, regeneration decreased considerably over the years. All the regenerates became abnormal. Since we found 100% aneuploidy, this means that during the extended period of subculture there was a selection towards abnormal cells. Our results showed that even though regeneration decreases, aberrant plants may still be regenerated. Screening for off-types is therefore necessary at an early stage of the somatic embryogenesis process.

4.3. Possible causes of variation

Very little is known about the causes of somaclonal variation in bananas [29]. Novak [30] explained the possible origin of somaclonal variation in Musa tissue cultures by three mechanisms: (a) variation is already present in the original explant; (b) variation is a result of stress induced by the conditions of the tissue culture environment; and (c) variation is induced by the specific mutagenic action of the tissue culture media. The results of Reuveni et al. [31] suggest that the high incidence and rate of occurrence of dwarf somaclonal variants is caused by the chimeric heterogeneity of the primary explants. Since in vitro multiplication proceeds at a higher rate than in in vivo conditions, chimeric plants also dissociated more easily in vitro than under field conditions, resulting in a higher rate of 'somaclonal variation' [32]. Previous results (see Chapter 4 in this book) show that somatic embryogenesis results in a high rate of chimerism dissociation.

We assume that somaclonal variation from cell suspensions can have two causes. Firstly, some non-embryogenic cells (with possible abnormal chromosome numbers) could have been cotransferred with embryogenic cell cultures from embryogenic calli into liquid medium. Initially, the number of these 'abnormal' cells is limited, but they may overgrow embryogenic cells if they display a faster growth rate. Secondly, a cell suspension which is embryogenic could produce a few abnormal cells due to tissue culture effects. These 'abnormal' cells could than have a comparative growth advantage with the same consequences mentioned above. In this case the 'abnormal' cells would be derived from embryogenic cells.

5. CONCLUSION

Our results prove that ploidy levels of Musa embryogenic cell suspension cultures can easily be determined by flow cytometry. In some cases, however, it may be necessary to improve the quality of DNA content histograms by measuring nuclei isolated from protoplasts rather than from chopped tissues. The results of this study also indicate that polyploidy could be induced during in vitro culture in a relatively short period, and that abnormal ploidy levels coincide with poor regeneration ability. To develop protocols that avoid variation at least in DNA content, we have now a simple and high-throughput assay at hand that can be applied immediately after tissue culture initiation, is objective, and is independent of environmental effects.

REFERENCES

[1] VAN DEN HOUWE, I., et al., Variability in storage potential of banana shoot cultures under medium term storage conditions, Plant Cell, Tissue Organ Culture 42 (1995) 269-274.

[2] PANIS, B., et al., Cryopreservation of banana (Musa spp.) meristems cultures after preculture on sucrose, Plant Sci. 21 (1996) 95-106.

[3] ROUX, N., et al., Effectiveness of three micropropagation techniques to dissociate cytochimeras in Musa spp., Plant Cell, Tissue Organ Culture 66 (2001) 189-197.

[4] THORPE, T.A., "In vitro somatic embryogenesis", Atlas of Science: Animal and Plant Science, ISI, Philadelphia (1988) 81-88.

[5] PESCHKE, V.M., PHILLIPS, R.L., Genetic implication of somaclonal variation in plants, Adv. Genet. 30 (1992) 41-75.

[6] SMITH, M., HAMILL, S., Early detection of dwarf off-types from micropropagated Cavendish bananas, Aust. J. Exp. Agr. 33 (1993) 639-644.

[7] LARKIN, P.J., SCOWCROFT, W.R., Somaclonal variation - a novel source of variability from cell culture for plant improvement, Theor. Appl. Genet. 60 (1981) 197-214.

[8] VUYLSTEKE, D.R., et al., Field performance of somaclonal variants of plantain (Musa spp., AAB group), J. Am. Soc. Hort. Sci. 121 (1996) 42-46.

[9] PHILLIPS, R.L., et al., Genetic instability of plant tissue cultures: break down of normal controls, Proc. Natl. Acad. Sci. USA 91 (1994) 5222-5226.

[10] GUPTA, P.K., "Chromosomal basis of somaclonal variation in plants", Somaclonal Variation and Induced Mutations in Crop Improvement (JAIN, S.M, et al., Eds), Kluwer Academic Publishers, Dordrecht (1998) 149-168.

[11] SHEPHERD, K., Mitotic instability in banana varieties. Aberrations in conventional triploid plants, Fruits 51 (1996) 99-103.

[12] SANDOVAL, J.A., et al., Chromosome number variations in micropropagated true-to-type and off-type banana plants (Musa AAA Grande Naine Cv.), In Vitro Cell. Dev Biol. Plant 32 (1996) 14-17.

[13] FAHLESON, J., et al., Correlation between flow cytometric determination of nuclear DNA content and chromosome number in somatic hybrids within Brassicaceae, Plant Cell Rep. 7 (1988) 74-77.

[14] DOLEZEL, J., "Flow cytometry, its application and potential for plant breeding", Current Topics in Plant Cytogenetics Related to Plant Improvement, (LELLEY, T., Ed.), Universitätsverlag, Vienna, (1998) 80-90.

[15] KUBALAKOVA, M. et al., Ploidy instability of embryogenic cucumber (Cucumis sativus L.) callus culture, Biol. Plant. 38 (1996) 475-480.

[16] ROTH, R., et al., Trisomy associated with loss of maturation capacity in a long-term embryogenic cultures of Abies alba, Theor. Appl. Genet. 95 (1997) 353-358.

[17] DOLEZEL, J., et al., Nuclear DNA amount in diploid bananas (Musa acuminata and M. balbisiana), Biol. Plant. 36 (1994) 351-357.

[18] DOLEZEL, J., et al., Use of flow cytometry for rapid ploidy determination in Musa species, InfoMusa 6(1) (1997) 6-9.

[19] DHED'A, D., et al., Plant regeneration in cell suspension cultures of the cooking banana cv. 'Bluggoe' (Musa spp. ABB group), Fruits 46 (1991) 125-135.

[20] SCHOOFS, H., The origin of embryogenic cells in Musa, Ph.D. Thesis, K.U.Leuven, Belgium, (1997).

[21] MURASHIGE, T., SKOOG, F., A revised medium for rapid growth and bioassays with tobacco tissue cultures, Physiol. Plant. 15 (1962) 473-497.

[22] ALVARD, D., et al., Comparison of methods of liquid-medium culture for banana micropropagation - effects of temporary immersion of explants, Plant Cell, Tissue Organ Culture 32 (1993) 55-60.

[23] DOLEZEL, J., GOHDE, W., Sex determination in dioecious plants Melandrium album and M. rubrum using high-resolution flow cytometry, Cytometry 19 (1995) 103-106.

[24] GALBRAITH, D.W., et al., "Nuclear DNA content and ploidy analyses in higher plants", Current Protocols in Cytometry, (ROBINSON, J.P., et al., Eds), Wiley, New York (1998) 7.6.1 - 7.6.22.

[25] PANIS, B., et al., Plant regeneration through direct somatic embryogenesis from protoplasts of banana (Musa spp.), Plant Cell Rep. 12 (1993) 403-407.

[26] DOLEZEL J., et al., A novel method to prepare slides for high resolution chromosome studies in Musa spp., InfoMusa 7(1) (1998) 3-4.

[27] BROWERS, M.A., ORTON, T.J., Transmission of gross chromosomal variability from suspension cultures into regenerated celery plants, J. Hered. 73 (1982) 159-162.

[28] NATALI, L., CAVALLINI, A., Nuclear cytology of callus and plantlets regenerated from pea (Pisum sativum L.) meristems, Protoplasma 141 (1987) 121-125.

[29] REUVENI, O., ISRAELI, Y., Measures to reduce somaclonal variation in in vitro propagated bananas, Acta Hort. 275 (1990) 307-313.

[30] NOVAK, F.J., "In vitro mutation system for crop improvement", Plant Mutation Breeding for Crop Improvement, (Proceedings of an International symposium on the contribution of plant mutation breeding to crop improvement jointly organized by the IAEA and the FAO and held in Vienna, 18-22 June 1990), Vol. 2 (1991) 327-342.

[31] REUVENI, O., et al., Factors influencing the occurrence of somaclonal variation in micropropagated bananas, Acta Hort. 336 (1993) 357-384.

[32] SWARTZ, H.J., "Post culture behavior: genetic and epigenetic effects and related problems", Micropropagation: Technology and Application, (DEBERGH, P.L., ZIMERMAN, R.H., Eds), Kluwer Academic Publishers, Dordrecht, (1991) 95-112.