III. VERIFICATION AND GENETIC DIVERSITY ANALYSIS OF INDONESIAN BANANA CULTIVARS CONTAINING
B GENOME BY MICROSATELLITE
ABSTRACT
Genetic diversity of bananas in Indonesia is considered to be high. Phenotypic identification of banana accessions is sometimes incorrect due to
instability of morphological characters. The aims of this study were to verify banana accessions containing the B genome previously identified based on
morphological characters, to construct molecular determination key for the genomic groups, and to reveal genetic relationships among the accessions based
on the microsatellite markers. The DNA of 92 banana accessions containing the B genome was analyzed using 8 microsatellite primers. The results showed that 9
accessions were incorrectly grouped into banana with the B genome. Those accessions should be placed into Musa acuminata due to the absence of the
MaCIR108 alleles which longer than 270 bp. Four alleles of the MaCIR108 were considered as the diagnostic characters for the B genome, and 1 allele of the Ma-
3-90 could be used to differentiate AAB from ABB genomic groups. Alleles of the MaCIR108 and Ma-3-90 with the size of 295 bp and 152 bp were only found
in BB and ABB and never been detected in AAB. The molecular determination key divided 83 accessions containing the B genome into 3 genomic groups; which
are 8 accessions of BB, 33 accessions of AAB, and 42 accessions of ABB. Twenty four of the 83 accessions formed 8 identical genotypes, therefore this
study detected only 67 genotypes. These accessions clustered according to their genomic groups. The ABB formed fewer clusters than the AAB genomic groups
indicating that the contribution of the B genome is less diverse than that of the A genome.
Key words: allele, genomic group, genetic relationships, molecular determination key
INTRODUCTION
For over 50 years, genomic constitution in banana has traditionally been determined based on morphological characters following the scoring system as
described by Simmonds and Shepherd 1955. The modified version proposed by Silayoi and Chomchalow 1987 suggested that 15 morphological characters be
used for identifying genomic group of AA; AAA; AAB; AB or AABB; ABB or ABBB; and BB or BBB. The phenotypic identification of banana accessions are
sometimes incorrect due to instability of morphological characters Vicente et al. 2005, mainly in identifying accessions which have intermediate characteristics
between Musa acuminata and M. balbisiana. Both species are designated as AA and BB genomes Stover and Simmonds 1987. Incorrect identification tended to
increase among closely related cultivars Vicente et al. 2005. Jumari 2000 reported several banana accessions of AAB genomic group were detected as the
ABB or vice versa in the study of Pudjoarinto et al. 1994 and Hadisunarso et al. 1995.
Genomic groups have an important role in cultivated banana classification. However, genomic composition among banana accessions was hardly interpreted
due to the occurrences of hybridization and polyploidy Pillay et al. 2004. Currently, several genomic groups were questionable due to the restricted
variability on the morphological characters, so that it is necessary to verify the groups using high accurate investigations. Cultivar characteristics could be easily
detected through nucleotide distinction of the DNA. Resolution of molecular markers is commonly higher than that of morphological appearance Ford-Lloyd
et al . 1997. Vicente et al. 2005 suggested that DNA nucleotides are better than
morphological markers in identifying cultivars. Microsatellites have been employed to investigate the genetic diversity of
Musa species Creste et al. 2003, 2004. The markers were powerful for
estimating an intraspecific genetic distance and effective for analyzing polyploidy Bruvo et al. 2004 especially for allopolipoidy. Microsatellites are chosen as
marker in the genetics analysis since they produce high polymorphic alleles even in the closely related lines Semagn et al. 2006. Therefore, microsatellite markers
were employed to characterize and identify cultivars of fruit crops, such as grape Arnold et al. 2002, strawberry Ashley et al. 2003, sweet orange Novelli et al.
2005, apple Galli et al. 2005 rasberry and blackberry Stafne et al. 2005, and mango Ukoskit 2007. The first study in M. acuminata cultivars Chapter II
showed that microsatellites could be used for clarifying taxonomic status of banana containing the A genome alone. The absence of diagnostic characters of
the B genome determined taxonomic status of the accessions. That microsatellite markers were able to be used for genomic fingerprint Kaemmer et al. 1997;
Creste et al. 2003, and banana germplasms wealth in Indonesia are considered to
be high Daniells et al. 2001 support the opportunity for constructing a determination key using molecular data.
The goals of the study were to verify cultivars of banana containing the B genome previously identified based on morphological characters, to construct the
molecular determination key for banana genomic groups and to investigate the genetic relationships among the accessions using microsatellite markers.
MATERIALS AND METHODS Plant materials
This work used genomic DNA from fresh leaves of 92 banana accessions previously morphologically identified and classified into BB; BB or BBB; AAB;
and ABB genomic groups Jumari 2000; Jumari and Pudjoarinto 2000; Edison et al
. 2004; INIBAP 2002a; Siddiqah 2002. The accessions originated from several regions in Indonesia Table 3.1.
Table 3.1 List of 92 accessions used in study of banana containing the B genome
Accession Subgroup
a
Genomic group
Region Source
b
Batu Klutuk BB
Medan Diperta
Batu Krawang Klutuk
BB Unknown
Diperta Klutuk Awu
- BB
Uknown RIF
Klutuk Batu -
BB Uknown
RIF Klutuk Hitam
- BB
Jasinga Bogor Bogor
Klutuk Warangan Klutuk
BB Kulon Progo
Diperta Klutuk Watu
Klutuk BB
Uknown Diperta
Klutuk Wulung Klutuk
BB Cilongok Banyumas
Diperta Boi
- BB or BBB
Genyem Jayapura RIF
Angleng - AAB
Jasinga Bogor
Bogor Austroli Plantain
AAB Dongkelan
Yogyakarta Diperta
Bangkahulu - AAB Unknown Diperta
Beleum - AAB
Jasinga Bogor
Bogor Burlangge - AAB
Sentani Jayapura
RIF Haseum -
AAB Jasinga
Bogor Bogor
Jambe Saat Triolin
AAB Cipaku Bogor
Diperta Kapas -
AAB Jasinga
Bogor Bogor
Ketan - AAB
East Java
Diperta Ketip Gunungsari
- AAB
Cipaku Bogor Diperta
Koja Susu -
AAB Unknown
Diperta Longong -
AAB Cipaku
Bogor Diperta
Nangka - AAB
Jasinga Bogor
Bogor Nangka -
AAB Unknown
RIF Pisang Seribu
Pisang Seribu AAB Unknown
Diperta Pulut Pulut
AAB Sukoharjo
Diperta Raja -
AAB Uknown
RIF
Table 3.1 List of 92 accessions used in study of banana containing the B genome continued
Accession Subgroup
a
Genomic group
Region Source
b
Raja Bagus Raja
AAB Sleman
Diperta Raja Bandung
- AAB
Bantul Diperta
Raja Bulu -
AAB Jasinga Bogor
Bogor Raja Kasman
Plantain AAB
Kulon Progo Diperta
Raja Lini Raja
AAB Guning Kidul
Diperta Raja Marto
Plantain AAB
Tembarak Temanggung Diperta Raja Nangka
Plantain AAB
Giwangan Yogyakarta Diperta
Raja Puser Plantain
AAB Gunung Kidul
Diperta Raja Sabrang
- AAB
Kulon Progo Diperta
Raja Sereh -
AAB Jasinga Bogor
Bogor Raja Sereh
- AAB
Uknown RIF
Raja Sereh -
AAB Purworejo
Diperta Raja Talun
- AAB
Kotagede Yogyakarta Diperta
Roid - AAB
Cipaku Bogor
Diperta Rojomolo Plantain
AAB Malang
Diperta Susu -
AAB Jasinga
Bogor Bogor
Tanduk - AAB
Jasinga Bogor
Bogor Tanduk ex Kedondong
- AAB
Cipaku Bogor Diperta
Tanduk Hijau Plantain
AAB Surakarta
Diperta Triolin Triolin
AAB Bantul Diperta
Abu Awak
Awak ABB Unknown Diperta
Apu - ABB
Jasinga Bogor
Bogor Awak
Awak ABB Samarinda Diperta
Awak Rawa
Awak ABB Unknown Diperta
Bawean Kepok ABB
Banyumas Diperta
Brentel Kepok ABB
Tlekung Malang
Diperta Byar -
ABB Purworejo
Diperta Comot
Awak ABB Gunung Kidul
Diperta Gablok Sobo
ABB Unknown
Diperta Gandul Sobo
ABB Giri
Banyuwangi Diperta
Gedah Awak ABB Cipaku
Bogor Diperta
Kapas - ABB
Tirto Pekalongan
Diperta Kaso
Awak ABB Cipaku Bogor
Diperta Kates Kepok-
ABB Giri
Banyuwangi Diperta
Kepok - ABB
Jasinga Bogor
Bogor Kepok Asam
- ABB
Bantul Diperta
Kepok Awu -
ABB Gunung Kidul
Diperta Kepok Brot
Kepok ABB
Purworejo Diperta
Kepok Byar Kepok
ABB Unknown
Diperta Kepok Gandul
Kepok ABB
Gunung Kidul Diperta
Kepok Gajih Kepok
ABB Unknown
Diperta Kepok Klutuk
- ABB
Bantul Diperta
Kepok Kuning Kepok
ABB Dongkelan Yogyakarta
Diperta Kepok Kuning
- ABB
Unknown RIF
Kepok Kuningan Kepok
ABB Tegalsari Girimulyo
Diperta Kepok Ladrang
Kepok ABB
Unknown Diperta
Kepok Lumut -
ABB Bantul
Diperta Kepok Ungu
Sobo ABB
Pendowoharjo Bantul Diperta
Table 3.1 List of 92 accessions used in study of banana containing the B genome continued
Accession Subgroup
a
Genomic group
Region Source
b
Kepok Urang Kepok
ABB Gunung Kidul
Diperta Klutuk Susu
Awak ABB
Umbulharjo Yogyakarta Diperta Lempeneng Sobo
ABB Cilongok
Banyumas Diperta
Raja Bali Awak
ABB Bantul
Diperta Raja Entog
Awak ABB
Purworejo Diperta
Raja Kul Kepok
ABB Malang
Diperta Raja Pendopo
Sobo ABB
Purworejo Diperta
Raja Siem -
ABB Unknown
RIF Raja Utri
Awak ABB
Gunung Kidul Diperta
Raja Wesi Awak
ABB Tawangsari Sukoharjo
Diperta Selayar -
ABB Sentani
Jayapura RIF
Siem - ABB
Jasinga Bogor
Bogor Sobo Kapuk
Sobo ABB
Giri Banyuwangi Diperta
Sobo Kepok -
ABB Unknown
Diperta Sobo Kerik
- ABB
Tlekung Malang Diperta
Sobo Londoijo Sobo
ABB Unknown
Diperta Sobo Londoputih
Sobo ABB
Sukoharjo Diperta
Sobo Madura Sobo
ABB Giri Banyuwangi
Diperta
a.
following the subgroup determination key constructed by Jumari 2000 and Jumari and Pudjoarinto 2000
b1
Diperta, Dinas Pertanian dan Kehewanan, Yogyakarta
b2
RIF, Research Institute of Fruits, Solok West Sumatera
Microsatellite analysis
Eight primers producing discrete and repeatable alleles in the first study Chapter II were used in this study Table 3.2. The MaCIR108 alleles were
confirmed as fingerprint which could differentiate banana cultivar containing the B genome from banana with the A genome alone Kaemmers et al. 1997; Creste
et al. 2003. A genomic DNA was extracted following Dixit 1998 procedure
with modification. Leaves were crushed in liquid nitrogen and homogenized in the extraction buffer containing 100 mM Tris pH 8.0, 50 mM EDTA pH 8.0, 500
mM NaCl, 10 mM beta-mercaptoethanol, and 20 SDS. After purification using 10 mgml of RNAse free at 37ºC for 1 hour, precipitation of DNA was carried out
without PEG solution treatment. DNA concentration was measured by spectrophotometer and then the DNA was diluted in TE to a suitable
concentration.
Microsatellites were
amplified using PCR in a 10 µl reaction solution
containing 15-20 ng of genomic DNA, 1 μl 10 x PCR buffer with 20 mM MgCl
2
, 0.2
μl 10 mM dNTPs, 0.2 μl 10 µM of each primer, and 0.06 μl 5 u μl
-1
Taq DNA Polymerase Native GenScript Corporation®. The amplification was performed
for 35 cycles consisting of denaturation at 95ºC for 30 s, annealing at 50, 53, 55 and 56ºC depending on the primers for 30 s, and extention at 72ºC for 30 s.
These cycles were preceded by a 4 min denaturation at 95ºC, and ended by a 10 min final extention at 72ºC Kaemmer et al. 1997. PCR reactions were carried
out in a Perkin Elmer 2400 thermocycler Applied Biosystems, Foster City, CA, USA®. The PCR products were then separated on a gel containing 6 wv
polyacrylamide and 7 M Urea Sigma-Aldrich Chemie, Germany® at 45-60 W constant power for 2 to 3 hours. Then, the gel was stained using a modified silver
staining method described by Creste et al. 2001. All solutions were prepared using distilled water.
Table 3.2 Primers used for verification and genetic diversity analysis of banana containing the B genome
Primers Forward primer sequence
Reverse primer sequence, 5’-3’ Optimized annealing
temperature ºC MaCIR108
a
F: TAAAGGTGGGTTAGCATTAGG
R: TTTGATGTCACAATGGTGTTCC 55
Ma-1-132
a
F: GGAAAACGCGAATGTGTG
R: AGCCATATACCGAGCACTTG 53
MaCIR327b
a
F: AAGTTAGTCAAGATAGTGGGATTT R: CTTTTGCACCAGTTGTTAGGG
50 MaCIR332a
a
F: TCCCAACCCCTGCAACCACT
R: ATGACCTGTCGAACATCCTTT 53
Ma-1-17
a
F: AGGCGGGGAATCGGTAGA
R: GGCGGGAGACAGATGGAGT 56
Ma-1-27
a
F: TGAATCCCAAGTTTGGTCAAG
R: CAAAACACTGTCCCCATCTC 56
Ma-3-90
b
F: GCACGAAGAGGCATCAC
R: GGCCAAATTTGATGGACT 56
Ma-3-139
b
F: ACTGCTGCTCTCCACCTCAAC
R: GTCCCCCAAGAACCATATGATT 56
a
Primers designed by Kaemmer et al. 1997
b
Primers designed by Crouch et al. 1998
Data analysis
Banana accessions containing the B genome were distinguished from the A genome alone based on the fragment size of MaCIR108 amplified products. Each
variant fragment of microsatellite was assumed as an allele. For each allele was scored as present 1 or absent 0. The allele size was estimated using 100-bp
DNA ladder Invitrogen®. Designation of allele submitted to the relative size of
allele followed CYMMYT 2004. The loci are named alphabetically as a for the largest fragment to h for the smallest fragment. Any band below the fragment is
named according to the letter designation followed by number, for instance a
1
for
the largest of a allele, etc.
A similarity matrix was calculated based on Jaccard similarity index using Similarity of Qualitative Data SYMQUAL procedures. The matrix was then
used for clustering analysis to examine genetic relationships among the accessions. Clustering analysis was performed using Sequential, Agglomerative,
Hierarchical and Nested SAHN of Unweighted Pair-grouping Method with Arithmatic Average UPGMA by the Numerical Taxonomy and Multivariate
Analysis System NTSYSpc version 2.02 Rohlf 1998.
RESULTS AND DISCUSSION Verification of banana accessions containing the B genome
All primers produced well-defined alleles exhibiting a polymorphic banding pattern. Total 75 alleles were obtained in this study, with the range of allele size
from 110 to 436 bp Table 3.3. Nine of the 92 accessions examined did not have the MaCIR108 alleles with the size of more than 270 bp that were confirmed as
the diagnostic characters for banana containing the B genome Chapter II. It means that those 9 accessions had the A genome alone. The absence of the
MaCIR108 alleles which larger than 270 bp indicated that they should be
classified into M. acuminata as reported in the first study Chapter II.
Banana genomic groups were incorrectly identified perhaps due to the morphological expressions which were overlap between M. acuminata and M.
balbisiana Jumari 2000. However in this study, incorrect classification of those
9 accessions was considered mostly due to mislabeled Table 3.4.
Table 3.3 Microsatellite alleles revealed from analysis of banana containing the B genome using the 8 microsatellite primers
Primer Alleles size
range bp Number
of alleles Alleles of each primer
MaCIR327b 388-436 5 a
1
,a
2
,a
3
,a
4
,a
5
Ma-1-132 330-378 8 b
1
,b
2
,b
3
,b
4
,b
5
,b
6
,b
7
,b
8
MaCIR332a 260-296 10 c
1
,c
2
,c
3
,c
4
,c
5
,c
6
,c
7
,c
8
,c
9
,c
10
MaCIR108 220-295 13 d
1
,d
2
,d
3
,d
4
,d
5
,d
6
,d
7
,d
8
,d
9
,d
10
,d
11
,d
12
,d
13
Ma-3-139 132-177 11 e
1
,e
2
,e
3
,e
4
,e
5
,e
6
,e
7
,e
8
,e
9
,e
10
,e
11
Ma-3-90 132-172 12 f
1
,f
2
,f
3
,f
4
,f
5
,f
6
,f
7
,f
8
,f
9
,f
10
,f
11
,f
12
Ma-1-27 122-142 6 g
1
,g
2
,g
3
,g
4
,g
5
,g
6
Ma-1-17 110-154 10 h
1
,h
2
,h
3
,h
4
,h
5
,h
6,
h
7,
h
8,
h
9,
h
10
Total 75
‘Bawean’, ‘Kepok Klutuk’, ‘Kepok Ungu’, and ‘Sobo Madura’ are the specific names for banana containing the B genome. The inaccurate identification was
human errors that incorrectly labeled the accessions in the field. Mislabeled was common phenomenon in the ex situ germplasm collections because morphological
appearances were sometimes difficult to be interpreted due to the influence of environmental and cultivation factors Hernandez et al. 2001. The occurrence of
mislabeled in olive, barley and banana germplasm collections were reported by Lopes and Mendonca 2004, Hamza et al. 2004, and Karamura et al. 2000
respectively. This misidentification could have great influence on efficiency of the breeding improvement Lopes and Mendonga 2004.
Table 3.4 List of the 9 banana accessions incorrectly classified into banana containing the B genome
Accession
Genomic group
Based on morphology Based on microsatellites
Angleng AAB AAA
Bawean ABB AAA
Beleum AAB AAA
Kapas ABB AAA
Kepok Klutuk ABB
AAA Kepok Ungu
ABB AAA
Ketip Gunungsari AAB
AAA Roid AAB
AAA Sobo Madura
ABB AAA
Diagnostic characters and molecular key for determination of banana genomic groups
Four alleles of the MaCIR108 with the size of more than 270 bp d
1
to d
4
; alleles of the MaCIR108 with the size of 270 bp or less d
5
to d
13
; and 1 allele of the Ma-3-90 with the size of 152 bp f
7
were diagnostic characters for banana genomic groups Table 3.5. The nine accessions: 4 accessions of AAB and 5
accessions of ABB only had 3 alleles of the MaCIR108 with the size of 270 bp or less, thus they should be classified into the triploid AAA. The remaining 83
accessions were bananas containing the B genome because 1 or 2 alleles of the MaCIR108 with the size of more than 270 bp were detected in those accessions.
Table 3.5 Classification of banana accessions into genomic groups based on the primers MaCIR108 and Ma-3-90
Alleles
MaCIR108 Ma-3-90 d
1
d
2
d
3
d
4
d
5
-d
13
f
7
Genomic group based on
morphology
295 bp 289 bp
287 bp 275 bp
≤270 bp 152 bp
Genomic group based on
microsatellites
8 BB
+ - - - - +
8 BB 1 BB or BBB
+ - + - + +
1 ABB 37 AAB
- - - - +++ -
4 AAA
- - + - +++ -
23 AAB
- - - + +++ -
7 AAB
+ - - - + +
2 ABB
+ + - - + +
1 ABB 46 ABB
- - - - +++ -
5 AAA
- - + - ++ -
3 AAB
+ - + - + +
5 ABB
- - + - + +
4 ABB
+ - - - + +
16 ABB
+ + - - + +
13 ABB + = allele present
- = allele absent ++++++ = 1, 2, or 3 alleles
Allele d
1
, d
2
, d
3
, or d
4
of the MaCIR108 and allele f
7
of the Ma-3-90 were consistently presence or absence in certain genomic group. Thirty accessions of
AAB and 3 accessions of ABB did not have allele f
7
of the Ma-3-90. This allele was detected only in the other 50 accessions. Allele d
1
of the MaCIR108 was observed in all BB accessions. Allele d
1
, d
2
, or d
3
were detected in 41 accessions of ABB. Twenty three of 33 AAB accessions had allele d
3
; 2 accessions had allele d
1
; 1 accession had allele d
1
and d
2
; and the remaining 7 accessions had allele d
4
.
The size estimation of allele d
1
; d
2
; d
3
; d
4;
and f
7
were 295 bp; 289 bp; 287 bp; 275 bp; and 152 bp, respectively
Based on primer MaCIR108, only 1 accession of BB or BBB studied was known to have heterozygous genotype consisting of 3 different alleles. The two
alleles fell into the range of allele length of the B genome and 1 allele fell into the A genome. Therefore based on microsatellite marker, this accession should be
classified into ABB genomic group. Allele f
7
of the Ma-3-90 was detected in all BB 8 accessions, in 93 38 accessions of ABB, only in 9 3 accessions of
AAB and none was observed in the 9 accessions of AAA. Table 3.5 showed the 3 accessions of AAB were classified into ABB
genomic group due to existence of 1 allele of the Ma-3-90 with the size of 152 bp and 1 allele of the MaCIR108 with the size of 270 bp or less. Those accessions
were ‘Ketan’ and ‘Raja Bandung’ from Diperta collection, and ‘Susu’ from Bogor. One allele of the MaCIR108 with the size of more than 270 bp was also
detected in these accessions. On the other hands, the 3 accessions of ABB ‘Byar’, ‘Brentel’, and ‘Kates’ from Diperta collection were included into AAB genomic
group because allele of the Ma-3-90 152 bp was absent. In these accessions also found 2 alleles of the MaCIR108 with the size of 270 bp or less and 1 allele with
the size of more than 270 bp. This study exhibited that all alleles of the MaCIR108 could be considered as
diagnostic characters for banana genomic groups and only one allele of the Ma-3- 90 which could be used to differentiate AAB from ABB, especially for accessions
having allele of the MaCIR108 with the size of 287 bp and 1 allele of the MaCIR108 for the A genome. Four alleles of the MaCIR108 were thought as
diagnostic character for the B genome, and the remaining 9 alleles for the A genome. Based on those alleles, a molecular determination key for genomic
groups was constructed as the key created by Guzow-Krzeminska et al. 2001. Molecular key for determination of banana genomic groups
1a. The MaCIR108 alleles with the size of 270 bp or less……………..…………2
2a. The number of alleles, one to two ..…………AA or AAA genomic group 2b. The number of alleles, three..…….………………....AAA genomic group
1b. The MaCIR108 alleles with the size of more than 270 bp……………………3
3a. Allele with the size of 295 bp, only…………………...BB genomic group
3b. Alleles combination: 295 bp; 295 bp and 289 bp; 295 bp and 287 bp; and 1 allele of the MaCIR108 with the size of 270 bp or less
……...……………………………………………….ABB genomic group
3c. Alleles combination of 287 bp and alleles of the MaCIR108 with the size of 270 bp or less.………………………………………………………….4
4a. Alleles combination of 287 bp and 2 alleles of 270 bp or less
…………………….……………………………AAB genomic group
4b. Alleles combination of 287 bp and 1 allele of 270 bp or less ……….5
5a. The Ma-3-90 allele of 152 bp, present .…….ABB genomic group 5b. The Ma-3-90 allele of 152 bp, absent .……..AAB genomic group
3d. Allele combination of 275 bp and alleles of the MaCIR108 with the size of
270 bp or less ............................................................AAB genomic group
A fascinating characteristic of the B genome markers was that the MaCIR108 allele with the size of 295 bp and the Ma-3-90 allele with the size of
152 bp were always present in accessions with two B genomes BB and ABB, and never detected in banana with one B genome AAB. These results confirm
that genomic groups of banana accessions particularly derived from hybrid between M. acuminata and M. balbisiana could be easily and rapidly determined
by microsatellite marker. The result of determination of genomic group is summarized in Table 3.6.
The keys demonstrated that the genomic group determination based on morphological characters is always not accurate. Genomic group of several
hybrids and accession “BB or BBB” studied have been incorrectly identified. Actually, the existence of pure natural BBB accession remains questionable
Pillay et al. 2004. Based on microsatellite markers, all accessions of BB were correctly classified into its genomic group, but only 81 accessions of AAB and
83 accessions of ABB which have correctly been classified into their genomic groups.
Table 3.6 Number of accessions classified into the genomic groups based on the primers MaCIR108 and Ma-3-90
Genomic group based on microsatellites Genomic group
based on morphology
Number of accessions
AAA BB BBB
AAB ABB
BB 8 8
BB or BBB 1
1 AAB 37
4 0 30
3 ABB 46
5 0 3
38 Total 92
9 8 33
42
The key in the present study could be applied only for banana cultivars derived from section Eumusa Stover and Simmonds 1987. This section is the
most geographically widespread, with species ranging from India to Pacific INIBAP 2004. It contains at least 11 species The Australian Government Office
of the Gene Technology Regulator 2008, but only 2 wild species, M. acuminata and M. balbisiana, involving in development of most edible bananas Cheesman
1948; Simmonds and Shepherd 1955; Sharrock 1998. A number of distinct
cultivars of major group are derived from M. acuminata and M. balbisiana either alone or in various hybrid combinations.
Molecular as well as morphological determination key also may have limited ability in their application. The key in the study could not be used for
identifying bananas cultivars derived from the other section, such as sect. Callimusa
. According to Wong et al. 2002, Callimusa and Eumusa were genetically distinct because the basic number of chromosome of n = x = 10- and n
= x = =11- grouping based on AFLP analysis are robust and justified to separate Musa
species into different section. The similarity between conserved regions of the 2 organisms from different taxonomical groups would be less significant
Guzow-Krzeminska et al. 2001. Therefore, different primers should be designed for constructing determination keys of different taxonomical groups.
The MaCIR108 alleles were a fingerprinting for distinguishing banana cultivars containing the B genome from those containing the A genome alone.
The sequence of MaCIR108 flanking region in both genomes has been known precisely Kaemmer et al. 1997. The Ma-3-90 microsatellite allele could be only
used to differentiate AAB from ABB genomic groups. Besides the primers MaCIR108 and
Ma-3-90
, Creste et al. 2005 also had successfully designed the
other of microsatellite loci which were also confirmed as fingerprint for genomic groups, such as MaOCEN03, MaOCE12, MaOCE13 and MaOCE14.
In order to avoid incorrect identification, it is possible for identifying banana cultivars using molecular marker before they were commercially
cultivated, particularly for the cultivars which were not easily classified using morphological characters. Molecular determination key has the potentially applied
in identifying a small fragment of organisms or organisms which is difficult to be identified using traditionally keys Guzow-Krzeminska et al. 2001. The
molecular determination key also provides suitable tools for researchers who are familiar with molecular techniques.
Analysis of relationships among banana accessions containing the B genome
Microsatellite has provided important information concerning the genetic relationships among accessions of banana containing the B genome. Clustering
analysis based on the 8 primers Figure 3.1 detected only 67 genotypes. It is because 24 accessions clustered within 8 identical genotypes.
At the coefficient similarity of 0.22, the dendrogram showed 2 main
clusters. In the first main cluster I, the subcluster Ia, 8 accessions of BB were
grouped together with 42 accessions of ABB and separated from 15 accessions of
AAB. In the subcluster Ib it was observed 11 accessions of AAB. The second main cluster II consisted of 7 accessions of AAB. The accessions of AAB in this
cluster separated from those in the first main cluster due to the presence of allele MaCIR108 with the size of 275 bp. The results suggested that the AAB genomic
group in the main cluster II was showed to be the most distance from the other
genomic groups. According to Jumari and Pudjoarinto 2000, based on morphological
characters, 2 of the 7 accessions in the main cluster II namely ‘Triolin’ and
‘Jambe Saat’ belong to the AB genomic group. The accession characteristics tended to indicate the transition between cultivar groups of AA, AAA, and AAB;
and groups of BB, ABB, and ABBB. Naturally, the banana accessions of the AB genomic groups are rare Valmayor et al. 2000 so that the existence of hybrid AB
especially in Indonesia still debatable.
Figure 3.1 Dendrogram from analysis of the 83 banana containing the B genome based on the 8 microsatellite primers.
● genomic group based on morphological characters,
●● genomic group based on the primers MaCIR108 and Ma-3-90.
1.00
● ●● Accession
Coefficient
0.22 0.41
0.61 0.80
1.00
5 s51
17w 98
2w s52
32w 160
9 10
99 180
s55 s104
45w 61
45 47w
100 41w
1w 4w
51 7w
81 55
88 30
153 64
79 82
25w 84
96 6
31 41
52 56
27 101
43w 161
31w 158
s60 s89
46 87
7 s106
147 49
23w 24w
21w 154
103 39w
95 155
s68 26
69 22
97 149
78 s33
25 74
24 23
29w 33
8 157
11 77
152 s62
83
ABB ABB Kepok Kuning
ABB ABB Kepok Kuning
ABB ABB Kepok Urang
ABB ABB Kepok Lumut
ABB ABB Raja Wesi
BBB ABB Boi ABB ABB Raja
Kul ABB ABB Kepok
BB BB Klutuk
Watu BB BB Klutuk
Wulung BB BB Batu
BB BB Klutuk Hitam
BB BB Klutuk Batu
BB BB
Klutuk Awu BB
BB Batu Krawang
BB BB Klutuk Warangan
ABB ABB Kepok Asam
ABB ABB Gablok ABB ABB Kepok
Awu ABB ABB Gandul
ABB ABB Lempeneng ABB ABB Sobo
Kepok ABB ABB Sobo
Kerik ABB ABB Kepok
Ladrang AAB ABB Ketan
ABB ABB Kepok Gandul
ABB ABB Sobo Londoijo
ABB ABB Raja Pendopo
AAB ABB Susu ABB ABB Kepok
Gajih ABB ABB Sobo
Kapuk ABB ABB Kepok
Brot ABB ABB Kepok
Byar ABB ABB Sobo
Londoputih ABB ABB Kepok
Kuningan ABB
ABB Awak Rawa
ABB ABB Awak ABB ABB Raja
Entog ABB ABB Gedah
ABB ABB Abu Awak
ABB ABB Klutuk Susu
AAB ABB Raja Bandung
ABB ABB
Raja Utri ABB ABB Apu
ABB ABB Kaso ABB ABB Siem
ABB ABB Raja
Siem ABB ABB Selayar
ABB ABB Raja Bali
ABB ABB Comot AAB AAB Raja
Nangka AAB AAB Nangka
AAB AAB Nangka AAB AAB Raja
Puser AAB AAB Raja
Marto AAB AAB Raja
Kasman AAB AAB Austoli
AAB AAB Kapas
AAB AAB Tanduk Hijau
ABB AAB Byar AAB AAB Tanduk
Ex Kedondong
AAB AAB Tanduk AAB AAB Burlangge
ABB AAB Brentel ABB AAB Kates
AAB AAB Rojomolo AAB AAB Raja
Bagus AAB AAB Raja
Bulu AAB AAB Raja
Sereh AAB AAB Raja
AAB AAB Raja
Sabrang AAB AAB Raja
Talun AAB AAB Raja
Lini AAB AAB Longong
AAB AAB Pulut AAB AAB Pisang
Seribu AAB AAB Jambe
Saat AAB AAB Haseum
AAB AAB Triolin AAB AAB Koja
Susu AAB AAB Raja
Sereh AAB AAB Raja
Sereh AAB AAB Bangkahulu
Ib Ia
I
II
The microsatellite study demonstrated that 1 allele of the B genome and 2 alleles of the A genome were detected in the accessions ‘Triolin’ and ‘Jambe Saat’.
Therefore, based on microsatellite marker they should be placed into the AAB genomic group. In this case, variations of morphological characters were not
sufficient to precisely determine genomic group.
The AAB accessions in the main cluster I were separated from BB and
ABB. The BB accessions were genetically closely related to ABB due to the existence of the MaCIR108 allele with the size of 295 bp and allele of the Ma-3-
90 with the size of 152 bp in both genomic groups. It supported the results of previous study using the AFLP technique Ude et al. 2002. Most AAB accessions
were grouped in the main cluster I due to the existence of the MaCIR108 allele
with the size of 287 bp. The differences among these AAB accessions were determined by alleles of the other primers.
The accessions of ABB, ‘Kates’ and ‘Brentel’ in the subcluster Ia based on
the primers MaCIR108 and Ma-3-90 were classified into AAB genomic group. They were genetically distinct from the others AAB. Both ‘Kates’ and ‘Brentel’
had 1 allele unique in each locus MaCIR332a and Ma-3-90 which were not be found in the others AAB. This result supported Valmayor et al. 2000 which
reported that Kates’ was recognized as a unique cultivar originating from Indonesia. Morphologically, ‘Kates’ produced large - solitary fruits per hand and
the fingers that were similar to small papaya. Based on the 8 markers, 7 of the 8 accessions of BB were known to be
identical. Seedless and parthenocarpy were considered as characteristics of the cultivated bananas Stover and Simmods 1987; Krishnamoorthy et al. 2004.
Within the diploid of BB, the occurrence of parthenocarpy was not detected Heslop-Harisson and Schwarzacher 2007, therefore it has allowed the
production of pulp with full seeds. Mutations within the cultivated bananas correspond with human intervention Carreel et al. 2002. Besides the BB
genomic group which is less variable than diploid the A genome Karamura 1998, the selection pressure within BB genomic group was also smaller than that
the other genomic groups due to the presence of seeds in their fruits. There were no different genotype of microsatellites among BB genomic groups; they were
genetically similar. However, ‘Klutuk Warangan’ was the exception. Based on their phenotypic features, the cultivars of BB were usually not morphologically
distinct to the wild type of M. balbisiana. Genetic relationship analysis among the 83 banana accessions containing
the B genome showed that the accessions clustered according to their genomic groups. Similarity coefficient at 0.61 producing 11 subclusters of AAB, 4
subclusters of ABB, and 1 subcluster of BB indicated that the diversity contribution of the B genome is less than that of the A genome. This result
supported cytometry analysis which revealed that the B genome was less diverse than the A genome Lysak et al. 1999; Kamate et al. 2001; Dolezel et al. 2004.
The ABB accessions appear closer to the BB than AAB genomic groups. It suggested that AAB genomic group was more divergence from the original
ancestors than ABB genomic group Pillay et al. 2004. Molecular markers were needed to identify the presence of multiple
genotypes with a single name or a single genotype with multiple names Vicente et al
. 2005; Semagn et al. 2006. This information is important to provide accurate identification for conservation activities. In the clonally propagated crops such as
banana, the conservation activities demand a complex and expensive procedures. Molecular data could be used for verifying the collections efficiently and ensured
the minimum repetition. The number of accessions could be reduced significantly without reducing genetic variations within germplasm collections Vicente et al.
2005. In general, allelic composition of microsatellites was relatively stable
Kaemmer et al. 1997. It is because the genotypes arising from cross breeding that occurred thousands years ago were maintained through vegetative
propagation Carreel et al. 2002. Actually, cultivated bananas were rather complex hybrids because they have been modified by various mutations and
exploited by human for food for a long time Stover and Simmonds 1987.
CONCLUSION
The MaCIR108 alleles are convincingly confirmed as diagnostic characters for distinguishing banana containing the B genome from banana containing the A
genome alone. Nine accessions were incorrectly classified into banana with the B genome. Those accessions should be placed into M. acuminata due to the absence
of the MaCIR108 alleles which longer than 270 bp. Four of 13 alleles of the MaCIR108 with the size of more than 270 bp as the diagnostic characters for the
B genome, whereas the remaining 9 alleles of the MaCIR108 with the size of 270 bp or less are considered as the diagnostic characters of the A genome. One allele
of the Ma-3-90 with the size of 152 bp can be used for separating AAB from ABB genomic groups. Based on molecular determination key, the 83 accessions
containing the B genome were divided into 3 genomic groups, which are 8 accessions of BB, 33 accessions of AAB, and 42 accessions of ABB. Analysis of
these accessions using the 8 primers detected only 67 genotypes due to 24 accessions formed 8 identical genotypes. The genetic relationship analysis showed
that banana accessions clustered according to their genomic groups. The ABB formed fewer clusters than the AAB genomic groups indicating that the
contribution of the B genome is less diverse than that of the A genome. Similarity coefficient at 0.61 produced 11 subclusters of AAB, 4 subclusters of ABB, and 1
subcluster of BB.
IV. MICROSATELLITE MARKERS FOR CLASSIFIYING AND ANALYZING GENETIC RELATIONSHIP OF