27. Early detection of dwarf off-types in banana (Musa spp.) Using AFLP, TE-AFLP and MSAP analysis - Engelborghs, I., L. Sági, R. Swennen

Laboratory of Tropical Crop Improvement
Catholic University Leuven
Kasteelpark Arenberg 13
B-3001 Leuven
Belgium

Abstract

AFLP and several variant techniques were performed on banana to characterize different dwarf types. The dwarf AAB plantain 'Curare enano' and its normal-sized in vitro generated off-type were analysed by AFLP, TE-AFLP, MSAP, cDNA-AFLP and cDNA-TE-AFLP. AFLP and TE-AFLP were also performed on four pairs of naturally occurring dwarf/normal banana cultivars. Differential AFLP patterns were obtained and different levels of polymorphism were observed depending on the technique, the primer combination and the cultivar. TE-AFLP analysis generated fewer and shorter fragments resulting in fewer polymorphisms between the dwarf and normal-sized cultivars. The phenotype switch that was observed when in vitro plantlets of the dwarf 'Curare enano' were transferred to the greenhouse might have been caused by methylation differences induced by in vitro conditions. MSAP analysis, based on the methylation (in)sensitivity of a pair of isoschizomeric restriction enzymes, appeared to be a valuable tool in revealing differential methylation. Cloning and sequencing of differential fragments did not result in homologous matches with sequences available in public databases. cDNA-AFLP analysis between the dwarf and normal 'Curare enano' revealed a normal-specific fragment, while cDNA-TE-AFLP revealed a dwarf-specific fragment. AFLP and the variant techniques have shown the potential to differentiate between closely related genotypes. More primer combinations and/or alterations of restriction enzymes will increase the chance of finding dwarf-related sequences.

Abbreviations: AFLP, amplified fragment length polymorphism technique; MSAP, methylation-sensitive amplified polymorphism technique; TE-AFLP, three-enzyme AFLP technique.

1. INTRODUCTION

Bananas (Musa spp.) are the largest 'non-woody' plants in the world. While often referred to as a 'tree', banana is best described as a large herbaceous plant, being by far the most important fruit crop in many developing countries. Musa plants show a great diversity in shape and size, with different flowering or bunch types, etc. An interesting form of banana is the dwarf type because it is less susceptible to toppling during regular tropical storms, and despite of its reduced height the bunch size is usually still large. Dwarf phenotypes have already proven to be important for plant breeding. In the 1960s and 1970s dwarfs have shown their merit during the 'Green Revolution'. Indeed, new dwarf wheat varieties were developed which were resistant to damage caused by wind and rain and had increased yield due to an improved straw/grain ratio [1].

1.1. Dwarfism in banana

Reduced height, thicker pseudostem and shorter but wider leaves are the major visible differences between the dwarfs 'Cachaco enano', 'Prata ana', 'Figue Rose Naine' and their normal-sized variants 'Cachaco', 'Prata' and 'Figue Rose', respectively. In the case of 'Cavendish', there exists a whole range of 'Cavendish' types gradually going from extra dwarf ('Dwarf Parfitt') to big plants, like the 'Giant Cavendish' [2]. Different naturally occurring dwarf Musa types exist, but this phenotype is often also obtained by in vitro culture [3-5].

Since most triploid bananas are sterile, no segregating populations are available for genetic analysis. Therefore, the above mentioned dwarf/normal pairs were subjected to AFLP and TE-AFLP analysis in order to find dwarf-specific patterns or sequences. cDNA-AFLP and MSAP analysis was performed on the off-type and true-to-type plants, to reveal fragments potentially related to differential gene expression, or to induced differential methylation, respectively.

1.2. Amplified fragment length polymorphism (AFLP) analysis

The original amplified fragment length polymorphism (AFLP) technique [6] involves digestion of total DNA by a pair of endonucleases, followed by ligation of double-stranded adapters and two amplification rounds, so-called pre- and selective amplification. By adding extra selective nucleotides to the primers in the latter PCR reaction, the number of amplified fragments can be tuned.

1.3. Three endonuclease (TE)-AFLP analysis

Unlike AFLP, three restriction enzymes are used in TE-AFLP analysis, thus increasing its discriminatory power. With TE-AFLP the number of amplified fragments is reduced by selective ligation and amplification. Three endonucleases together with only two adapters are added in a single reaction. In consequence, the number of potentially amplifiable fragments is limited to those ligated to both adapters. This diminishes competition during PCR, permitting stringent reaction conditions and thus eliminating the need for two-step amplification in fingerprinting complex genomes [7].

1.4. cDNA-AFLP analysis

In contrast to AFLP analysis, where total DNA is used as template, RNA is isolated and transcribed into cDNA which is used as a template for cDNA-AFLP analysis. When the analysis is limited to the transcribed part of the genome, polymorphisms between two phenotypes may be correlated with differentially expressed genes [8].

1.5. cDNA-TE-AFLP analysis

cDNA was digested with an additional enzyme for cDNA-TE-AFLP analysis, to increase further the discriminatory power by screening additional potentially polymorphic restriction sites.

1.6. Methylation-sensitive amplified polymorphism (MSAP) analysis

Methylation-sensitive amplified polymorphism (MSAP) analysis allows comparison of the degree of methylation between genomes by using either one of a pair of methylation (in)sensitive restriction enzymes in combination with Eco RI [9,10]. Reports are made of correlation between changes in methylation pattern and the switching on or off of genes in plants, i.e. alterations of gene transcription leading to morphological changes [11-13].

2. MATERIALS AND METHODS

2.1. Plant material

All the material was supplied by the International Musa germplasm collection of the INIBAP Transit Centre based in our laboratory. This comprised in vitro plantlets of four different cultivars ('Cavendish', 'Cachaco', 'Figue Rose' and 'Prata') in pairs of naturally occurring dwarf and normal-sized phenotypes. In addition, a pair of naturally occurring dwarf and in vitro-induced off-type of the cultivar 'Curare enano' were included. In vitro plantlets were cultured and multiplied on MS medium [14] supplemented with 10^-6 M indole-3-acetic acid (IAA) and 10^-6 M 6-benzylaminopurine (BAP) for regeneration, or with 10^-6 M IAA and 10^-5 M BAP for proliferation. AFLP analysis was done on true-to-type and off-type 'Curare enano' as well as on the above mentioned four naturally occurring dwarf types with their corresponding normal types. In addition, TE-AFLP analysis was performed on these varieties except for the 'Cavendish' types. The subject of the cDNA-(TE)-AFLP and MSAP analysis was the 'Curare enano' true-to-type dwarf and its normal-sized off-type.

2.2. Total DNA extraction

Total DNA was isolated from 300 mg of young leaf tissue of greenhouse or in vitro plantlets using the DNeasy DNA extraction kit (Qiagen) with some modifications [15]. Typical yields were between 30-45 µg per gram of leaf tissue.

2.3. AFLP analysis

For AFLP, DNA (250 ng) was digested with the restriction enzymes Eco RI (Pharmacia) and Mse I (New England Biolabs), ligated with adapters (Eurogentec) and amplified as earlier described [16]. Table 1 gives an overview of the sequence of adapters, and Table 2 of the primers used in this study, based on sequences previously described [6].

2.4. TE-AFLP protocol

For the TE-AFLP analysis, DNA (250 ng) was digested with Eco RI, Mse I and Bam HI (Pharmacia), ligated with adapters in a single reaction, and amplified using a modified protocol based on the method described by Van der Wurff et al. [7]. Amplification was done in one step with the following cycling conditions: 94°C for 2.5 min, followed by 10 cycles of 95°C for 30 s, 70°C for 30 s, 72°C for 60 s, followed by 40 cycles of 95°C for 30 s, 60°C for 30 s, 72°C for 60 s, and a final step of 10 min at 72°C. The sequences of the adapters and corresponding primers are given in Tables 1 and 2, respectively. They were designed so that the unlabelled primer had a higher melting temperature and the adapter a lower melting temperature than the labelled primer.

2.5. cDNA-AFLP protocol

Poly(A)⁺ RNA isolation (first and second strand) cDNA synthesis was done as described by Bachem et al. [17]. cDNA-AFLP analysis was first presented by Wiame et al. [18]. The cDNA digestion, adapter ligation, pre-amplification and selective amplification, as well as the detection and analysis of the cDNA profiles, was done as described for the AFLP analysis [16]. Sequences of adapters and primers used for cDNA-AFLP analysis are given in Tables 1 and 2, respectively.

2.6. cDNA-TE-AFLP protocol

cDNA (synthesized as described above) was digested with three enzymes (Eco RI, Mse I and Bam HI). The adapter ligation, pre-amplification, selective amplification and analysis of the cDNA-TE-AFLP profiles were done as described for TE-AFLP analysis. The sequences of adapters and primers are shown in Tables 1 and 2, respectively.

2.7. MSAP protocol

For the MSAP analysis, DNA (250 ng) was digested with either of the two isoschizomers Hpa II and Msp I (both from Pharmacia) in combination with Eco RI, ligated with adapters and amplified as described earlier [19], based on the method of Xiong et al. [9]. The isoschizomers Hpa II and Msp I cleave their recognition site CCGG differentially depending on whether the internal or external C is methylated [20]. Tables 1 and 2 list the sequence of adapters and primers used in this study.

Table 1 Sequences of adapters for the different AFLP techniques

2.8. Fluorescent detection and fragment analysis

The fluorescein label on the 5' end of the indicated primers was used to detect the separated DNA fragments on a 7% polyacrylamide gel (ReproGel High Resolution) of the A.L.F. automated sequencer (Pharmacia), as indicated by the manufacturer. Selective amplification reactions were done in duplicate for each pre-amplified DNA and run in adjacent lanes, as internal controls. Only bands that were different in both lanes were considered.

Table 2 Sequence of primers used for the different AFLP techniques

F: 5' fluorescein label; selective nucleotides are shown within brackets or in lower case

2.9. Cloning and sequencing

AFLP or MSAP patterns containing differential bands were separated on mini-acrylamide gel and silver-stained. Differential bands were scraped out of the silver-stained gel with a needle pre-wetted with PCR master mix. This needle was then placed in the mix for 1 min prior to the start of the PCR reaction [21]. Re-amplification was carried out with the same reaction conditions and with the same primers as used for the selective amplification. Cycling conditions were 94°C for 2 min, followed by 30 cycles of 94°C for 30 s, 56°C for 30 s, 72°C for 1 min and 72°C for 10 min. Differential bands were cloned using the manufacturer's instructions of the TOPO TA Cloning Kit (Invitrogen) in the pCR4-TOPO plasmid vector, and their nucleotide sequence was determined (MWG BIOTECH). The sequences obtained were compared with available sequences from the Genbank database using BLAST search (NCBI).

3. RESULTS

3.1. 'Curare enano' and off-type

3.1.1. AFLP

AFLP analysis with 28 primer combinations was performed on the true-to-type dwarf 'Curare enano' and its in vitro-generated normal-sized off-type, resulting in patterns containing around 100 fragments with a maximum size of 900 bp (Figure 1A, Table 3). Eleven primer combinations resulted in an identical pattern, whereas the other 17 primer combinations revealed 0.1-9.6% polymorphisms.

3.1.2. TE-AFLP analysis

Total DNA was used as template for a three-enzyme digestion and ligation with two adapters in a single reaction [7]. Two primer combinations were tested, with either the Eco RI primer labelled or the Bam HI primer labelled (Table 2). TE-AFLP analysis resulted in a maximum fragment size of approximately 500 bp and a reduced number of fragments compared to two-enzyme AFLP analysis (30-40 instead of about 100) (Table 3). When different quantities of the digestion/ligation reaction were used for the PCR reaction, only small differences could be found in the level of signal detection. Different amounts of the PCR reaction loaded on the sequencer gel gave only small intensity differences. This indicates that the technique is robust, because the results are insensitive to differences in DNA quantity in the PCR reaction or loaded on the gel. Differential patterns could be found with the primer combination G-Eco RI-(+C)/Bam HI-(CC)-F (Figure 1B).

3.1.3. cDNA-AFLP analysis

RNA was isolated and reverse transcribed into cDNA, which was used as a template for AFLP analysis. Two primer combinations were tested [pre-amplification primer +0/+0 with either of selective primer pairs Mse I(+G)/Eco RI(+GAC) or Mse I(+G)/Eco RI(+ACG)], resulting in band patterns of about 30 fragments with a maximum fragment size of 300 to 400 bp (Table 3). A differential pattern could be found with the selective primer combination Mse I(+G)/Eco RI(+GAC) (Figure 1C).

Figure 1 Part of the chromatograms generated by 4 variant AFLP techniques on true-to-type dwarf 'Curare enano' (D) and normal-sized off-type (N). Filled peaks indicate differential fragments. M: marker lane. (A) AFLP analysis with primer combination +0/+G for the pre-amplification and MseI(+G)/EcoRI(+GAC) for the selective amplification. (B) TE-AFLP analysis with primer combination G-EcoRI(+C)/BamHI-(CC)-F. (C) cDNA-AFLP analysis with primer combination MseI(+0)/EcoRI(+0) for the pre-amplification and MseI(+G)/EcoRI(+GAC) for the selective amplification. (D) cDNA-TE-AFLP analysis with primer combination EcoRI(+CG)-F/BamHI(C)

3.1.4. cDNA-TE-AFLP analysis

A combination of two variant AFLP techniques led to TE-AFLP analysis of cDNA. RNA was isolated from leaf tissue of the true-to-type 'Curare enano' and the normal-sized off-type, and reverse transcribed into cDNA, which was used as a template for three-enzyme AFLP analysis, based on the protocol described by Van der Wurff et al. [7]. A single PCR reaction was sufficient for the selective amplification of 30-40 fragments with a maximum fragment size of 500 bp (Table 3). Two primer combinations with either the Eco RI or the Bam HI-primer labelled (Eco RI(+CG)-F/Bam HI-(C) and G-Eco RI(+C)/Bam HI-(CC)-F) were tested (Table 2). Differential patterns were found with the Eco RI labelled primer (Figure 1D).

3.1.5. MSAP analysis

The MSAP analysis revealed differential methylation between the dwarf and normal-sized 'Curare enano' with primer combination Hpa I/Msp I(+GA)/Eco RI(+GAC) [19]. Briefly, in the 20 fingerprints, containing around 100 fragments with a maximum fragment size of 800 bp that were generated with five primer combinations, 14 fragments were found to differ between the dwarf and normal or between Hpa II and Msp I digestion (data not shown). The latter are due to the fact that Hpa II is sensitive to methylation of either cytosine residue at the recognition site and Msp I is sensitive only to methylation at the external cytosine. Methylation of the internal cytosine leads to the appearance of a fragment in the pattern generated after Msp I digestion but not after Hpa II digestion.

3.1.6. Cloning and sequencing

Interesting dwarf- and normal-specific bands resulting from AFLP and MSAP analysis were cloned and sequenced, but no significant homology was found with any known sequence. Further analysis of other differential fragments will be necessary.

3.2. Naturally occurring dwarf/normal banana pairs

3.2.1. AFLP

AFLP analysis was performed on the naturally occurring dwarf types 'Cachaco enano', 'Figue Rose Naine', 'Prata ana' and their normal-sized variants 'Cachaco', 'Figue Rose' and 'Prata', respectively, with 14, 8 and 14 primer combinations respectively [16]. Between 0% and 13%, 5% and 25% and 0% and 6% AFLPs were found for the dwarf/normal pairs 'Cachaco enano'/'Cachaco', 'Figue Rose naine'/'Figue Rose' and 'Prata ana'/'Prata', respectively. Between 0.7% and 10.4% polymorphism was found between the extra dwarf 'Cavendish' type 'Dwarf Parfitt', the normal-sized 'Cavendish' and the 'Giant Cavendish' with 6 of the 16 primer combinations tested. The level of polymorphism found was comparable to that found for the 'Curare enano'/off-type pair, except for the high level of AFLPs found for the 'Figue Rose (naine)' pair.

3.2.2. TE-AFLP

TE-AFLP analysis resulted in 30 to 40 fragments with maximum fragment size of approximately 500 bp (Table 3). Differential patterns could be found with primer combination Eco RI(+C)/Bam HI(CC)-F.

Table 3 Comparison of different variant AFLP techniques used in this study

4. DISCUSSION

4.1. AFLP, TE-AFLP, cDNA-AFLP, cDNA-TE-AFLP analysis

The AFLP technique has shown its potential and power for the differentiation of two closely related phenotypes. It has been used for many species (from bacteria to humans) [22, 23] and in different applications (from taxonomy to medical) [24-26]. The variant AFLP techniques tested in this study have proven their applicability to differentiate Musa spp. and for the characterization of dwarf banana types.

The five dwarf/normal pairs used in this study were analysed using 80 primer combinations. This resulted in 160 AFLP patterns each with around 100 fragments. Some primer combinations showed a high level of polymorphism in all cultivars, while others did not reveal any differences at all. The cultivar pair 'Figue Rose'/'Figue Rose naine' showed an extremely high maximum level of polymorphism, while for the 'Cavendish' types, 10 out of 16 primer combinations did not reveal any differences at all. The different genetic backgrounds might have caused this. All the dwarf types could be differentiated from their normal-sized variant at the in vitro stage. However, no dwarf-specific fragment of the same length was found to be common to all cultivars. This suggests that these dwarf types have a different origin, or that the differences found were not correlated with the dwarf phenotype.

As expected, the AFLP patterns contained more fragments than the TE-AFLP and cDNA-AFLP patterns (Table 3). Both the number and the intensity of the polymorphisms found with AFLP analysis was higher than with the TE-AFLP and cDNA-(TE)-AFLP techniques. The AFLP technique is not only the most time-consuming technique, but it also requires expensive equipment (automated or manual sequencer). However, the latter is faster and easier in separation and detection, i.e. mini-acrylamide gel electrophoresis and silver staining.

TE-AFLP allows a significant reduction in preparatory steps and thus in time involved, i.e. restriction and ligation are done in a single reaction, followed by only one amplification reaction instead of the traditional pre- and selective amplification. The use of a second rare cutter in TE-AFLP reduces the number of potentially amplifiable fragments. In this study a twofold reduction was observed (up to 40 fragments with a maximum size of 500 bp). The cDNA-AFLP technique has already proven to be a useful tool for gene discovery [27]. A few differential fragments were found in this study, which can be cloned and sequenced in the future. More cDNA-AFLP analysis should be performed to produce more differential fragments to increase the chance of finding a dwarf related sequence. The combination of cDNA-AFLP and TE-AFLP has not been described yet, but it may serve as a fast screening technique prior to a more detailed AFLP analysis of the cDNA. RNA isolation followed by cDNA synthesis renders the cDNA-AFLP technique even more laborious than AFLP, whereas cDNA-TE-AFLP reduces time and effort by eliminating separate digestion and ligation and using only one PCR reaction. AFLP, cDNA-AFLP and MSAP analysis (discussed below) each target different sequences, which makes them hard to compare. The choice of which technique to use will depend on the complexity of the organism, its genome, available skills and equipment, and the purpose of the research.

4.2. MSAP analysis

The MSAP technique described by Xiong et al. [9] was adapted in this study for the detection of differential cytosine methylation in the dwarf banana 'Curare enano' compared with its normal-sized in vitro-generated off-type. Results show typical patterns containing about 100 nicely separated fragments. This is in contrast with the weak rice MSAP patterns generated by Xiong et al. [9] and the low resolution of the patterns generated by Reyna-Lopez et al. [28]. Different states of methylation that might be correlated with the phenotypic switch were revealed between the dwarf and the off-type. However, detection is restricted to the cytosine residues in the restriction sites of the isoschizomers used, while other potentially methylated CC or CG dinucleotides are not analysed. Since no sequence homology with any known sequence in the NCBI Genbank was found, more differential fragments will have to be analysed. Sequence-specific primers could then be designed to generate a sequence-characterized amplified region (SCAR)-based marker for use in a PCR screening test [29].

5. CONCLUSIONS

AFLP, together with the variant techniques MSAP, TE-AFLP and cDNA-AFLP, proved to have the power to differentiate between two closely related phenotypes, i.e. dwarf and normal-sized banana types, and this could be done as early as the in vitro stage. Different levels of polymorphisms were observed, depending on the technique, the cultivar, and the primer combination. No fragment was found to give the same differential pattern in all dwarf/normal pairs. Further analysis will be necessary to identify sequences correlated with the dwarf phenotype.

ACKNOWLEDGEMENTS

This study was undertaken as part of the Global Program for Musa Improvement (PROMUSA) and was supported by the Catholic University of Leuven (K.U.Leuven). The authors thank the International Musa germplasm collection at the INIBAP Transit Centre (K.U.Leuven) for the supply of in vitro plantlets, and the Centre for Microbial and Plant Genetics (K.U.Leuven) for use of the A.L.F. automated sequencer.

REFERENCES

[1] PENG, J., et al., 'Green revolution' genes encode mutant gibberellin response modulators, Nature 400 (1999) 256-261.

[2] DANIELLS, J., et al., "Musalogue: a catalogue of Musa germplasm", Diversity in the Genus Musa (ARNAUD, E., SHARROCK, S., Eds), INIBAP, Montpellier, France (2001).

[3] VUYLSTEKE, D., et al., Somaclonal variation in plantains (Musa spp. AAB group) derived from shoot-tip culture, Fruits 46 (1991) 429-439.

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

[5] DAMASCO, O.P., et al., Use of a SCAR-based marker for the early detection of dwarf off-types in micropropagated Cavendish bananas, Acta Hort. 461 (1998) 157-164.

[6] VOS, P., et al., AFLP: a new technique for DNA fingerprinting, Nucl. Acids Res. 23 (1995) 4407-4414.

[7] VAN DER WURFF, A., et al., TE-AFLP: combining rapidity and robustness in DNA fingerprinting, Nucl. Acids Res. 28 (2000) 105e.

[8] DELLAGI, A., et al., cDNA-AFLP analysis of differential gene expression in the prokaryotic plant pathogen Erwinia carotovora, Microbiology 146 (2000) 165-171.

[9] XIONG, L.Z., et al., Patterns of cytosine methylation in an elite rice hybrid and its parental lines, detected by a methylation-sensitive amplification polymorphism technique, Mol. Gen. Genet. 261 (1999) 439-446.

[10] XU, M., et al., AFLP-based detection of DNA methylation, Plant Mol. Biol. Rep. 18 (2000) 361-368.

[11] MEYER, P., et al., Evidence for cytosine methylation of non-symmetrical sequences in transgenic Petunia hybrida, EMBO J. 13 (1994) 2084-2088.

[12] ULIAN, E., et al., DNA methylation and expression of NPT II in transgenic petunias and progeny, Theor. Appl. Genet. 92 (1996) 976-981.

[13] ROSSI, V., et al., Analysis of methylation pattern of the maize Opaque-2 (O2) promoter and in vitro binding studies indicate that the O2 B-Zip protein and other endosperm factors can bind to methylated target sequences, J. Biol. Chem. 272 (1997) 13758-65.

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

[15] REMY, S., et al., Genetic transformation of banana (Musa spp.) for disease resistance by expression of antimicrobial proteins, PhD thesis, K.U.Leuven (2000).

[16] ENGELBORGHS, I., SWENNEN, R., Fluorescent AFLP detection of dwarf types in banana (Musa spp.) at early growth stage, Med. Fac. Landbouw Univ. Gent 64/5b, (1999) 427-430.

[17] BACHEM, C.W., et al., Visualization of differential gene expression using a novel method of RNA fingerprinting based on AFLP: analysis of gene expression during potato tuber development, Plant J. 9 (1996) 745-753.

[18] WIAME, I., et al., Characterization of a resistance gene analog and adaptation of cDNA-AFLP in banana, (International Symposium on the Molecular and Cellular Biology of Banana. Ithaca, USA, 22-25 March 1999), InfoMusa 8 (1999) PROMUSA XIV-XV, Abstract.

[19] ENGELBORGHS, I., et al., Fluorescent AFLP analysis on azacytidine and gibberellin treated banana (Musa spp.) plants to assess differences in cytosine methylation and the mechanism of dwarfism, Med. Fac. Landbouww. Univ. Gent 65/3b (2000) 387-396.

[20] McCLELLAND, M., et al., Effect of site-specific modification restriction endonucleases and DNA modification methyltransferases, Nucl. Acids Res. 22 (1994) 3640-3659.

[21] STUM, G., et al., A simple method for isolation of PCR fragments from silver-stained polyacrylamide gels by scratching with a fine needle, Elsevier Trends Journals Technical Tips Online No. 01115, (1997) http://www.elsevier.nl/locate/tto/.

[22] DESAI, M., et al., Fluorescent amplified fragment length polymorphism subtyping of the Salmonella enterica Serovar Enteritidis Phage Type 4 Clone complex, J. Clin. Microbiol. 39 (2001) 201-206.

[23] PROCHAZKA, M., et al., AFLP fingerprinting of the human genome, Hum. Genet. 108 (2001) 59-65.

[24] JIANG, S.C., et al., Genetic diversity of clinical and environmental isolates of Vibrio cholerae determined by AFLP fingerprinting, Appl. Environ. Microbiol. 66 (2000) 148-153.

[25] GIANNASI, N., et al., The use of amplified fragment length polymorphism in determining species trees at fine taxonomic levels: analysis of a medically important snake, Trimeresurus albolabris, Mol. Ecol. 10 (2001) 419-426.

[26] GURTLER, V., MAYALL, B.C., Genomic approaches to typing, taxonomy and evolution of bacterial isolates, Int. J. Syst. Evol. Microbiol. 51 (2001) 3-16.

[27] CAMPALANS, A., PAGES, M., Identification of differentially expressed genes by the cDNA-AFLP technique during dehydration of almond (Prunus amygdalus), Tree Physiol. 21 (2001) 633-643.

[28] REYNA-LOPEZ, G.E., et al., Differences in DNA methylation patterns are detectable during the dimorphic transition of fungi by amplification of restriction polymorphisms, Mol. Gen. Genet. 253 (1997) 703-710.

[29] XU, M., et al., Development of sequence-characterized amplified regions (SCARs) from amplified fragment length polymorphism (AFLP) markers tightly linked to the Vf gene in apple, Genome 44 (2001) 63-70.