Summary

Cross-breeding, consisting of gene recombination to generate variability, is difficult due to the extremely complicated genetic system of Musa: different genomic constitutions, heterozygosity, polyploidy and the sterility of edible varieties, e.g. dessert bananas, plantains and cooking bananas. The complexity of Musa genetics emphasizes the need for an additional system to support conventional breeding programs. Spontaneous mutations leading to significant increases of Musa genetic diversity are an important source of variation to breed Musa spp. Since the only difference between spontaneous and induced mutations is in their frequency, induced mutations have a high potential for bringing further genetic improvements. Vegetatively propagated crops are usually heterozygous, and mutagenic treatments may uncover recessive alleles by mutating or deleting corresponding dominant alleles. In vegetatively propagated crops, induced mutations can improve existing cultivars faster than conventional breeding methods. Induced mutations may also be viewed as a complementary tool to conventional breeding. In Musa, for example, dwarfness is an important trait, and it would be most desirable to have many more dwarf diploids. In vivo sucker irradiation before meristem tip isolation and culture was not effective, and yielded a relatively small amount of mutagenized material. The main advantage of induced mutations in vegetatively propagated plants is the ability to change one or a few characters of an outstanding cultivar without altering the remaining genotypes. Mutation techniques might be particularly important for sterile Musa species or varieties where there is no sexual reproduction to generate genetic variation. Thus it could be a complementary method to provide genetic variation, which is the basis for selection.

In vitro shoot-tip culture has been developed and used as a system for mutation induction in banana and plantains. This method has, however, some constraints:

a) In vitro propagated plants are not necessarily true-to-type. Genetic variability has been observed in many species during tissue culture. Off-type plants might be genetically or epigenetically modified.

b) The appearance of chimeras after tissue irradiation necessitates the subculturing of regenerated plantlets to dissociate chimeras, thereby slowing down the selection of true-to-type mutants.

c) Mutation techniques require the induction and screening of large plant populations, which could be extremely costly for giant plants such as bananas and plantains. Therefore, it seems important to develop a rapid in vitro screening method for the desired trait to make mutation induction more economical. However, in vitro selection would be advantageous only when the in vitro response is correlated with the manifestation of the selected characters in the field.

Tissue culture is another way to induce genetic variability, which is termed somaclonal variation. The genetic behaviour of somaclones generally appears to be similar to that of naturally occurring mutants. The genotypic background has a tremendous impact, so that somaclonal variation may not be useful in all genotypes.

Musa ECS is important for a wide range of biotechnological applications such as mass clonal propagation, mutation induction and genetic engineering. However, their establishment is time-consuming and still far from being a routine method for most genotypes. ECS cultures have been initiated successfully from immature zygotic embryos, male inflorescences, corm slices, and in vitro proliferating meristems. Several researchers showed that ECS is an ideal system for producing plants with agronomic traits similar to plants regenerated by the conventional in vitro budding method.

ECS cultures are chosen for mutation induction since plants can be regenerated from a single cell, thus avoiding chimeric variants. As a result, mutant somatic embryos can be produced directly, and ultimately regenerated into plants. ECS are also useful for rapid clonal propagation, protoplasts and genetic transformation. As ECS have proven to be very valuable and their initiation takes several months, it is important to preserve them safely, without the risk of somaclonal variation or microbial and fungal contamination. All this can be achieved with cryopreservation.

1. SCREENING FOR BLACK SIGATOKA (MYCOSPHAERELLA FIJIENSIS)

Although the field performance of any cultivar remains the ultimate reference for evaluating host plant resistance, epidemiological characteristics of black leaf streak make field evaluation unsuitable for rapid initial screening. Artificial inoculation under controlled conditions, which mimics cultivar behaviour in the field, represents a first step towards an early selection procedure, but remains time and labour consuming.

Plant tissue-culture techniques, including in vitro selection against toxins produced by a pathogen, offer an attractive approach for screening new genetic material among regenerates from explants, cells or protoplasts. To our knowledge, no resistant banana genotype obtained through such methods has been field evaluated, because of two major limitations:

a) The lack of characterised toxins that play a role in the disease.

b) The assurance that the susceptibility and/or resistance of cultured tissues to the toxin(s) reflect those of the whole plant.

The most abundant phytotoxic compounds in culture filtrates of M. fijiensis appeared to be 2,4,8-trihydroxytetralone, which induced necrotic lesions on black sigatoka-sensitive cultivars of bananas. Crude filtrates of M. fijiensis have been used to select resistant varieties. Resistance in highly resistant cultivars was induced by the release of glucan(s) from germinating spores of M. fijiensis, and is definitely unrelated to the actions of these toxin metabolites.

2. FAO/IAEA/BADC BANANA PROJECT

FAO/IAEA started a banana CRP entitled "Cellular biology and biotechnology including mutation techniques for creation of new useful banana genotypes" in 1994, aiming to integrate radiation mutation induction with in vitro culture, molecular genetics and conventional breeding of bananas in order to obtain desirable variations such as disease resistance, dwarfism and earliness. In addition, the development of methods for large-scale rapid multiplication of mutants/segregants through somatic embryogenesis and micropropagation was promoted. The Belgium Administration for Development Cooperation (BADC) decided to fund this CRP in 1996. Since than, three RCMs have been held, in Austria, Malaysia and Sri Lanka. The fourth and final RCM was held at K.U. Leuven, Belgium, 24-28 September 2001. Participants from Belgium, Cuba, Czech Republic, Germany, Israel, Mexico, Philippines, and Sri Lanka attended this meeting. During this RCM, it was decided to publish a book, based on the results and achievements of this CRP. FAO/IAEA was represented by the technical officer, who acted as a Scientific Secretary, and by a representative from Plant Breeding and Genetics Unit, Seiberadorf. This book contains the results and achievements of CRP participants and their recommendations. The editors included additional chapters to cover various aspects of banana improvement more comprehensively.

2.1. Overall achievements

Research tools were developed for germplasm characterization and improvement through induced mutations, cryopreservation, somatic embryogenesis, somaclonal variation and genetic engineering. Some of the existing cultivars have been improved for disease tolerance and important agronomic traits. Collaborations among participating laboratories were established, including exchange of staff, training and technology transfer.

2.2. PRACTICAL ACHIEVEMENTS

Several young students benefited from this CRP in completing their masters' and Ph.D. programs in Israel, the Czech Republic, and Belgium. Some of the participants have presented their results at international conferences. A total of 51 research papers have been published in conference proceedings and international refereed journals (see Appendix 1)

Many international trainees received training on several aspects of banana tissue culture, molecular cytogenetics and molecular markers at K.U.Leuven and FUSAG; IEB; and the University of Frankfurt, Germany. The trainees came from China, Cuba, Egypt, Mexico, and Rwanda. The outcome of this training was very successful. For example, a Cuban trainee was successful in establishing in Cuba somatic embryogenic cell suspensions from local banana plant material. In addition, he successfully irradiated plant material in Cuba. In Sri Lanka, 20 persons from the countryside were given training in tissue culture technology for mass production of bananas. Post-graduate training on indexing of banana viruses was organized.

The flow cytometry protocol was improved at IEB. Flow cytometry facilities were established at the International Institute of Tropical Agriculture (IITA, Nigeria) and the Malaysian Institute for Nuclear Technology (MINT, Malaysia). The technology transfer involved staff training in IEB.

2.3. Specific achievements

(a) Somatic embryogenic cell suspension cultures (ECS) were developed for several banana cultivars including plantains (AAB).

(b) Three cryopreservation techniques were developed for long-term conservation of meristems. An INIBAP technical guideline for cryopreservation of bananas was published.

(c) Induced mutations generated a series of improved clones that were screened for different traits such as early flowering, reduced height, large fruit size, and tolerance to Fusarium (Table 1).

(d) Both Agrobacterium-mediated transformation and particle bombardment methods were used for banana transformation, and the transformation rate was shown to be cultivar-dependent.

(e) Virus indexing procedures were transferred to Sri Lanka for indexing local banana virus strains.

(f) An early screening technique was developed for Fusarium wilt using tissue culture-derived plants in a double-tray system.

(g) A selection system was developed against black Sigatoka disease using Mycosphaerella fijiensis crude extracts, and semi-purified fractions, and one purified fraction (juglone).

(h) Screening techniques for nematode resistance were developed in Musa under shade-house and field conditions. Aseptic cultures of Radopholus similis and Pratylenchus coffeae were established using alfalfa calli, and their pathogenicity was confirmed after greenhouse tests.

(i) DNA flow cytometry was used for detection of polyploidy, monitoring of cytochimera dissociation, and analysis of karyological stability of ECS.

(j) DNA methylation polymorphism was detected in banana micropropagated plants with AFLP.

(k) Transposon mutagenesis was explored for gene tagging, using the maize Ac element, in the banana genome. A substantial number of distinct mutants were generated and characterized.

(l) FISH protocol was developed for Musa for detailed study of karyotypes, providing distinct chromosome landmarks, gene localization, analysis of long-range chromosome structure, and linkage to physical and genetic maps.

(m) A total of 28 allele-specific SSR markers were generated for Musa, and used to detect polymorphisms between the A and B genomes, identify hybrids, and trace back the B genome in hybrids. These markers are now used within the CRP and worldwide. A total of 24 locus-specific, highly polymorphic SSR markers were also produced for Mycosphaerella fijiensis to discriminate it from other species.

2.4. Specific recommendations

The participants in the final banana RCM gave recommendations to further advance international banana improvement programs for food security, nutrition, and employment generation.

2.4.1. Characterization and evaluation of germplasm and selections

a) There is certainly a great need to increase awareness and access to the "Descriptor for banana (Musa spp.)" published by INIBAP/IPGRI/CIRAD and the INIBAP Technical Guidelines for the "Evaluation of Musa germplasm for resistance to black sigatoka and Fusarium wilt", published by INIBAP/IPGRI/CTA/PROMUSA. The "Descriptor for banana" will standardize the morphological evaluation of the selections, and technical guidelines will standardize the resistance evaluation for black sigatoka and Fusarium wilt. Furthermore, scientists and farmers should be made more aware of the existence of "Musalogue" to provide the latest information on agronomy, pathology, taxonomy, and the post-harvest. These documents would certainly benefit the researchers and growers. (www.inibap.org).

b) The participants from Cuba, Malaysia, Philippines and Sri Lanka should continue evaluation for disease resistance and agronomic performance of the field selections. Selected mutants should be further tested in the farmers' fields.

c) Interesting desirable mutants and somaclonal variants should be introduced into ITC under MTA, and made available worldwide.

d) With the availability of interesting desirable mutants and somaclonal variants from this project, additional funding would be needed to scale up to on-farm testing and worldwide distribution. It would certainly have a significant socio-economic impact on banana growers.

e) In the Philippines, Sri Lanka, and Malaysia local germplasm should be collected and characterized according to existing guidelines.

f) Flow cytometry should be carried out to analyse ploidy levels of the ITC collections, mutants and /or somaclonal variants.

g) Re-evaluation of the genomic constitution of accessions held at ITC using newly developed DNA techniques (GISH, SSR typing, AFLP fingerprinting, etc.) should be seriously considered.

h) High-throughput screens and parameters for GO have not yet been described for the banana genomes and require serious attention. These tools will be invaluable in future for Musa breeding.

Table 1 Examples of desirable variants/putative mutants identified for release or further confirmation trials

2.4.2. Training

a) Since the training component of CRP proved to be productive, earmarking of funding for individual and group training is recommended.

b) At FUSAGx and K.U.Leuven, training should continue on Mycosphaerella fijiensis toxin isolation and purification, purified toxin treatment, inoculation with pathogen, somatic embryogenic cell suspension cultures, cryopreservation, transformation, and in vitro nematode screening. PROMUSA should assist in identifying suitable trainees and support them financially or find them financial resources. However, utmost attention should be given to the establishment of regional training centres in the southern hemisphere.

c) The regional training courses should cover scalp and 'male flower' techniques for the initiation of somatic embryogenic cell suspension cultures, molecular marker analysis, and flow cytometry; trainees should be acquainted with appropriate technologies that are most applicable to their needs and technical capabilities.

2.4.3. Technology refinement and development

2.4.3.1. Cell and tissue culture

a) Production of somatic embryos should be initiated from shoot tip cultures, especially from virus-indexed in vitro plants, which are readily available throughout the year.

b) Different factors, such as nitrogen sources, pH, etc., affecting the growth behaviour and regeneration capacity of somatic embryogenic cell suspension cultures require more investigation. Slow-growing cell suspensions should be discarded.

c) An international cryo-storage bank should be developed at ITC as a Global Service Centre, so that highly regenerable ECS can be cryopreserved for the long term, to prevent contamination and somaclonal variation. The donors should deposit new ECS samples in the cryo-storage bank within one year after establishment.

d) Protein patterns of somatic embryogenic and non-embryogenic cell suspension cultures should be analysed to detect correlations with the different growth phases.

e) More efforts should be made to speed up the initiation of ECS, without increasing the level of somaclonal variation.

f) ECS of diploid varieties are urgently needed for fundamental studies.

g) Somatic embryogenic cell suspension cultures of important local varieties should be established.

h) IAEA should continue to provide service for irradiation of ECS and other material.

2.4.3.2. Flow cytometry

a) Flow cytometric analysis should be carried out on proliferating meristem cultures and other intermediate steps leading to ECS.

b) Ploidy of somatic embryogenic cell suspension cultures should be measured.

c) Cell-cycle analysis of somatic embryogenic cell suspension cultures needs to be addressed.

d) It is recommended that ploidy measurements should be made on local cultivars, wild types and in vitro multiplied banana plants in banana-producing countries.

e) IAEA/IEB should continue to provide ploidy measurement services to the Member States.

f) After thawing, cryopreserved meristems and ECS should be evaluated for ploidy changes by flow cytometry before plant regeneration.

2.4.3.3. Somaclonal variation

a) Somaclonal variation can become a serious problem in micropropagated banana plantlets. It is desirable to refine the tissue culture protocol to minimise somaclonal variation in regenerated plants.

b) Subculture number of differentiated shoots and somatic embryogenic cell cultures should be optimised to enhance genetic stability among regenerated plants.

c) MSAP should be used to investigate the relationship between methylation and somaclonal variation.

d) More investigation is needed to find out the relationship between retrotransposon activation and somaclonal variation.

e) Further studies on retrotransposon activation are needed to investigate the influence of other stress factors, such as irradiation, low temperature, and pathogenesis on somaclonal variation.

f) Field studies should be made frequently to evaluate the rate of somaclonal variation among ECS-derived plants.

g) Molecular markers are needed for the early detection of somaclonal variants, such as dwarf types, erect and drooping leaves, etc.

h) Aneuploids should be retained for genomic studies.

2.4.3.4. Genome analysis

a) A large number of simple sequence repeat (SSR) markers from Musa acuminata should be made readily available.

b) The development of SSR markers for Musa balbisiana is needed.

c) The detection of retrotransposon-like elements is recommended, and this should be extended to other Musa genotypes. Large insert DNA libraries should be constructed.

d) The structural analysis of retrotransposable elements is recommended.

e) The production by gamma and fast neutron irradiation of a large number of useful mutants is recommended for facilitating genomic studies and banana variety improvement.

f) There is a great immediate need for the creation of large segregating populations for studies on: (a) black sigatoka, (b) yellow sigatoka, and (c) parthenocarpy. The crosses should be made and corresponding populations grown in a banana-growing country, e.g. Cuba that is supported by IAEA.

g) Mapping should be done in at least two laboratories having expertise in molecular marker techniques. The resulting materials should be made available to all interested banana research institutions.

h) ECS of the fertile diploid Calcutta 4 should be established, to validate candidate genes involved in resistance towards black Sigatoka disease and other important agronomic traits.

i) BAC library of the B genome should be established to complement the already existing library of the A genome.

j) The number of chromosome-specific markers should be increased for physical mapping, and linked to genetic and physical maps.

k) The number of annotated functional genes from banana in the public domain should be increased.

l) Stress-induced, developmentally regulated and tissue-specific promoters should be discovered in banana.

m) Recommendations are given for screening and characterization of M. fijiensis and M. musicola genes and metabolites involved in pathogenicity.

n) An internationally accepted host differential set should be established (a group of cultivars with different resistance genes) for pathotyping M. fijiensis and M. musicola isolates.

o) Recommendations are given for the establishment of matings and populations segregating for virulence, to tag and isolate avirulence genes of M. fijiensis and M. musicola.

p) The entire genome of M. fijiensis should be sequenced as a short cut for gene discovery.

2.4.3.5. Mutagenesis

a) The protocol developed for irradiation of ECS should be applied to a wide range of genotypes to prevent chimerism, and to improve the effectiveness of mutation induction.

b) Fungal- and nematode-resistant bananas should be irradiated to knock out resistance, and identify resistance genes.

2.4.3.6. Nematodes

a) Biological, biochemical and molecular analyses should be carried out to study Arabidopsis/Radopholus similis and Pratylenchus coffeae interactions.

b) The potential of lectins and lectin-related proteins for banana nematode control should be further investigated, both in transgenic model systems and in banana plants.

c) Root-specific and wound-inducible promoters should be identified and tested for nematode control.

2.4.3.7. Genetic transformation

a) Non-antibiotic selectable markers should be tested (e.g. transposon Ac element in the isopentyltransferase (ipt) gene, or markers based on selection with sugar and amino acid analogues).

b) Resistance genes (R-genes) should be identified in banana for controlling nematodes, fungi, etc.

c) Pathogen-induced and tissue-specific promoters should be identified.

d) Bananas should be transformed with genes and regulatory elements originating from the Musa genome.

e) Countries should be urged to ratify the Cartagena protocol and set up National Biosafety Guidelines so that the field tests of already existing and promising transgenic bananas and plantains can finally start.

f) It is proposed that K.U.Leuven would provide banana transformation services to researchers engaged in bananas. When requested, selected cell-suspension cultures will be transformed with desired genes, and putative genetic transformants will be supplied to the requesting institute for biochemical, molecular and agronomical analyses.

g) Collaborators should be identified for the evaluation of food safety for transgenic bananas.

2.4.3.8. Viruses

Further studies should be carried out on the development of technologies related to molecular detection of banana viruses.

2.4.4. Screening techniques

2.4.4.1. Sigatoka

a) Greenhouse and field evaluation of the selected mutants should be carried out.

b) Detailed studies are needed to understand better the mechanisms of action of juglone (a metabolite of M. fijiensis) on banana tissues.

c) Selection against juglone toxin should be done on selections obtained through mutation induction and/or somaclonal variation.

d) More genotypes should be evaluated against juglone for the selection of disease-resistant lines.

e) Putative disease-resistant transgenic plants and mutants need to be screened by leaf-disc assays, and by glasshouse and field inoculation tests.

2.4.4.2. Fusarium

The early screening method for Fusarium tolerance should be applied to different Fusarium isolates and to the standard set of banana genotypes from the IMTP. Consequently, it should be used for evaluating transgenic banana plants and putative mutants.

2.4.4.3. Nematodes

a) Further confirmation of in vitro screening for resistance is required by comparative studies with plant responses under pot and field conditions.

b) Systematic evaluation of cultivars, wild types, breeding lines and selections should be carried out.

c) Studies with Pratylenchus spp. should be further validated.

d) A collection of nematodes, differing in pathogenicity and origin, should be established and made available to the banana scientific community.

e) The effect of the edaphic conditions on the genotype-nematode interaction should be studied.

2.4.5. Technology transfer

Existing collaboration and linkage with PROMUSA should be continued and further strengthened following this CRP.

2.4.6. Linkage to PROMUSA

a) INIBAP will continue to supply material required by researchers, provided it is virus- and bacteria-indexed and thus 'available'.

b) INIBAP will develop segregating populations for in-depth genetic studies and make leaf samples/DNA samples available to the genetic improvement-working group of PROMUSA. Meanwhile, the PROMUSA secretary will explore accessibility to existing segregating populations.

c) Although this CRP is finished, linkage among scientists should continue through FAO/IAEA and PROMUSA to maintain and increase information exchange and stimulate collaboration.

d) PROMUSA will facilitate the publication of any new technologies developed by this concluded CRP.

e) PROMUSA should facilitate the field-testing of transgenic plants.

f) Selections obtained through this CRP could be made available to IMTP through PROMUSA.