792
Phylogenetic relationships among Secale species
revealed by amplified fragment length
polymorphisms
T. Chikmawati, B. Skovmand, and J.P. Gustafson
Abstract: Amplified fragment length polymorphism (AFLP) data were utilized to analyze the phylogenetic relationships among 29 accessions representing 14 of the most commonly recognized ranked species or subspecies in the genus Secale. We observed 789 AFLP markers of 1130 fragments utilizing 18 P-/M- and E-/M- primer combinations. All
polymorphic fragments were used to construct phenetic and phylogenetic trees. The resulting phenogram and cladogram
had similar tree topologies. Cluster analysis showed that Secale sylvestre was the most distantly related to all other
ryes. Annual forms were grouped together, and the perennial forms appeared more closely related to each other. This
suggested that life cycle could have played an important role in determining the relationships among Secale species.
Secale sylvestre was considered to be the most ancient species, whereas Secale cereale was the most recently evolved
species. Amplified fragment length polymorphism analysis clearly separated all Secale species into only 3 major species groups, within the genus Secale: S. sylvestre, Secale montanum (syn. Secale strictum) for perennial forms, and
S. cereale for annual forms. This study demonstrated that the AFLP approach is a useful tool for discriminating species
differences, and also gave a much better resolution in discerning genetic relationships among Secale species as compared with previous studies using other approaches.
Key words: AFLP, Secale, phylogenetic relationship.
Résumé : Le polymorphisme de longueur des fragments amplifiés (AFLP) a été employé pour analyser les relations
phylogénétiques entre 29 accessions représentant 14 des espèces ou sous-espèces les plus communément reconnues au
sein du genre Secale. Les auteurs ont obtenu 789 marqueurs AFLP parmi 1130 fragments en employant 18 combinaisons d’amorces P-/M- and E-/M-. Tous les fragments polymorphes ont été employés pour produire des arbres phénétiques et phylogénétiques. Les phénogrammes et dendrogrammes résultants avaient une topologie semblable. Une analyse
de groupement a montré que le Secale sylvestre était l’espèce la plus éloignée de tous les autres seigles. Les formes
annuelles ont été groupées ensemble et les formes pérennes étaient plus apparentées les unes aux autres. Ceci suggère
que le cycle vital pourrait avoir joué un rôle majeur dans l’établissement des relations au sein des espèces du genre Secale. Le S. sylvestre est considéré comme étant l’espèce la plus ancienne, tandis que le Secale cereale serait l’espèce la
plus récente. L’analyse AFLP a clairement séparé toutes les espèces entre seulement 3 groupes majeurs : S. sylvestre,
Secale montanum (syn. Secale strictum) chez les formes pérennes, et S. cereale chez les formes annuelles. Cette étude
montre que les AFLP constituent un outil utile pour discerner les relations génétiques au sein des espèces du genre Secale par comparaison avec les études antérieures faisant appel à d’autres approches.
Mots clés : AFLP, Secale, relation phylogénétique.
[Traduit par la Rédaction]
Chikmawati et al.
801
Introduction
The genus Secale L. is a typical representative of Mediterranean flora. It has a wide distribution from central Europe
and the western Mediterranean through the Balkans, Anatolia,
Israel and the Caucasus to Central Asia, with an isolated
population in South Africa (Sencer and Hawkes 1980). This
genus includes perennial or annual, self-incompatible or
self-compatible, and cultivated, weedy, or wild species
(Vence et al. 1987). Cultivated rye (Secale cereale L.) is an
Received 26 January 2005. Accepted 3 May 2005. Published on the NRC Research Press Web site at http://genome.nrc.ca on
18 October 2005.
Corresponding Editor: P. Donini.
T. Chikmawati. Department of Agronomy, University of Missouri-Columbia, Columbia, MO 65211, USA; and Department of
Biology, Bogor Agricultural University, Bogor 16144, Indonesia.
B. Skovmand. The Nordic Gene Bank, P.O. Box 41, SE230 53 Alnarp, Sweden.
J.P. Gustafson.1 Department of Agronomy, University of Missouri-Columbia, Columbia, MO 65211, USA; and USDA-ARS, Plant
Genetics Research Unit, University of Missouri-Columbia, Columbia, MO 65201, USA.
1
Corresponding author (e-mail: pgus@missouri.edu).
Genome 48: 792–801 (2005)
doi: 10.1139/G05-043
© 2005 NRC Canada
Chikmawati et al.
important source of bread, especially in parts of Northern
and Eastern Europe with poor soils and severe winters
(Leonard and Martin 1963). In addition, wild and cultivated
ryes also have a great potential as a source for value-added
trait genes such as those for high protein content, disease resistance, and other morphological and biochemical traits, for
wheat (Triticum spp. L.) and triticale (×Triticosecale
Wittmack) improvement.
Despite its economical importance, taxonomy and phylogenetic relationships within the Secale genus have long been
the subjects of controversy. Taxonomists have tried to discriminate each species within Secale using different approaches; however, identification of many taxa has been
difficult because of a lack of diagnostic characters. As a result, many varying classification efforts have been reported.
The first reports based on morphological characteristics, life
cycle, and geographical distribution (Vavilov 1917, 1926)
accepted 4 species in the genus Secale, for example, Secale
africanum Stapf., S. cereale L., Secale fragile Marsch, and
Secale montanum Guss. Roshevitz (1947) distinguished as
many as 14 species based on their crossability. However,
Khush (1962) did not find any cytogenetic support to classify perennial ryes (S. montanum, S. africanum, and Secale
kuprijanovii Grossh.) as different species and proposed that
they should be taken as subspecies of S. montanum, whereas
the weedy ryes (Secale ancestrale, Secale afghanicum Vav.,
Secale dighoricum Vav., and Secale segetale Zhuk.) were to
be considered subspecies of S. cereale. Lastly, Frederiksen
and Petersen (1998) made a taxonomic revision of Secale
based on examination of material in several herbaria and
comments on the application of many of the species names
that were used. They recognized only 3 species: Secale
sylvestre, Secale strictum (syn. S. montanum), and S. cereale.
Phylogenetic relationships among Secale species have also
been studied using many different approaches, including
morphological analyses (Frederiksen and Petersen 1997),
isozymes (Vence et al. 1987), thin-layer chromatography
patterns (Dedio et al. 1969), ribosomal DNA spacer lengths
(Reddy et al. 1990), restriction fragment length polymorphism (RFLP) of plastid genome (Murai et al. 1989), and
chloroplast DNA variation (Petersen and Doebley 1993), as
well as the internal transcribed spacer sequences of the 18S–
5.8S rDNA (ITS-1) region of cultivated and wild species (De
Bustos and Jouve 2002). In contrast with the classification
systems, all phylogenetic studies demonstrated similar results. Secale sylvestre showed distinct characteristics and
was the most distant species, whereas other taxa have been
and continue to be more difficult to distinguish from each
other.
Amplified fragment length polymorphism (AFLP) analysis is a technique through which selected fragments from the
digestion of total plant DNA are amplified by PCR (Vos et
al. 1995). This technique produces DNA fingerprints that
provide a large number of genetic markers, which can be
used as a satisfactory alternative to morphological and biochemical trait analyses. Amplified fragment length polymorphism markers have been used for phylogenetic studies to
uncover closely related taxa that had been impossible to resolve with morphological or other molecular systematic characteristics (Russell et al. 1997; Janssen et al. 1997; Mueller
and Wolfenbarger 1999; Zhang et al. 2001; Hodkinson et al.
793
2000; Beardsley et al. 2003; Laurence et al. 2003). This
method could be suitable for the analysis of relatedness, especially because AFLP markers are virtually free of artifacts, which is an acute problem of anonymous markers for
relatedness estimation, and because comigration of nonallelic fragments occurs at extremely low levels (Waugh et
al. 1997). Previous studies showed that AFLP technology
also provided better resolution in discerning phylogenetic relationships as compared with isozymes, nuclear RFLPs and
chloroplast DNAs (Sharma et al. 1996). However, the use of
AFLPs for studying phylogenetic relationships in the genus
Secale has not been previously reported. The present
research was designed to demonstrate the use of AFLPs in
discriminating Secale species and to understand the phylogenetic relationships among them.
The objectives of the present study were to (i) apply
AFLP technology to perform detailed analyses of polymorphism in the genus Secale, (ii) discriminate Secale species,
and (iii) describe the phylogenetic relationships among
Secale species based on AFLPs.
Materials and methods
Plant materials and DNA isolation
Twenty-nine accessions of weedy/wild rye and cultivated
rye were used (Table 1), representing 14 of the most commonly recognized taxa ranked as species or subspecies in the
genus Secale. Seeds were kindly provided by the germplasm
collections of the Plant Breeding and Acclimatization Institute (IHAR), Poland; Germplasm Resources Information
Network (GRIN) of the United States Department of Agriculture (USDA); and International Maize and Wheat Center
(CIMMYT), Mexico. Different numbers of accessions per
species were used, because that is what was available from
the germplasm collections. One accession from each of
Agropyron cristatum (L.) Gaertn and Triticum monococcum L. were used as outgroups (Monte et al. 1995) and were
obtained from the USDA-Sears collection, University of
Missouri. All of the accessions are carefully maintained in
seed form by the various collections and can be obtained
from the collection’s curator. DNA was extracted from
young freeze-dried tissue collected from 5–10 plants of each
accession using a cetyltrimethylammonium bromide method
(Saghai-Maroof et al. 1994). Five to ten plants were used
based on preliminary results, which showed that 5–10 samples per accession represented the diversity within each species.
AFLP analyses
Amplified fragment length polymorphism analyses were
carried out according to the technique developed by Vos et
al. (1995). From each accession, a 500-ng sample of total
genomic DNA was digested with the combination of EcoRI
and MseI or PstI and MseI restriction enzymes and ligated
with the respective adaptors for each enzyme. Preamplification reactions were performed using EcoRI and
MseI or PstI and MseI primer combinations with 1 nucleotide extension being added to the 3′ end of each primer. The
pre-amplification products were used as templates for selective amplification, which involved 3 additional nucleotides
being placed at the 3′ end of EcoRI or PstI and MseI prim© 2005 NRC Canada
794
Genome Vol. 48, 2005
Table 1. List of species name, accessions number (existing germplasm bank location), country of origin, type, life cycle of Secale species studied.
No.
Accession No.*
Species†
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
31209 (IHAR)
31082 (IHAR)
31319 (IHAR)
31344 (IHAR)
CIse 107
PI 283971 (USDA)
31322 (IHAR)
31323 (IHAR)
CIse 1
PI 534929 (USDA)
PI 534962 (USDA)
PI 534965 (USDA)
PI 535008 (USDA)
30186 (IHAR)
31211 (IHAR)
31083 (IHAR)
31328 (IHAR)
31351 (IHAR)
CIse 105
PI 283982 (USDA)
31359 (IHAR)
31366 (IHAR)
31367 (IHAR)
PI 573649 (USDA)
PI 383756 (USDA)
PI 383757 (USDA)
PI 205222 (USDA)
PI 401404 (USDA)
PI 440654 (USDA)
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
S.
anatolicum Boiss.
afghanicum (Vav.) Roshev.
africanum Stapf
ancestrale (Zhuk.) Zhuk.
ancestrale (Zhuk.) Zhuk.
ancestrale (Zhuk.) Zhuk.
chaldicum Fed.
ciliatoglume (Boiss.) Grossh.
cereale L.
cereale L.
cereale L.
cereale L.
cereale L.
cereale L.
dighoricum (Vav.) Roshev.
dighoricum (Vav.) Roshev.
kuprijanovii Grossh.
segetale (Zhuk.) Roshev.
segetale (Zhuk.) Roshev.
segetale (Zhuk.) Roshev.
sylvestre Host
turkestanicum Bensin
vavilovii Grossh.
vavilovii Grossh.
montanum Guss.
montanum Guss.
montanum Guss.
montanum Guss.
montanum Guss.
Country of origin
Type‡
Life cycle
Poland
Afghanistan
South Africa
Turkey
Japan
Algeria
Russia Federation
Poland
Sweden
Italy
US (Mississipi)
US (Florida)
Canada (Alberta)
Poland
Poland
Poland
Poland
Russian Federation
Italy
Former USSR
Poland
Turkey
Poland
Afghanistan
Turkey
Turkey
Turkey
Iran
Hungary
Wi
We
Wi
We
We
We
Wi
Wi
C
C
C
C
C
C
We
We
Wi
We
We
We
Wi
C
Wi
Wi
Wi
Wi
Wi
Wi
Wi
P
A
P
A
A
A
P
P
A
A
A
A
A
A
A
A
P
A
A
A
A
A
A
A
P
P
P
P
P
*Accession numbers followed by existing germplasm bank location.
†Botanical names of accessions that originated from Poland were verified using Secale monographs (Hammer et al. 1987; and Kobylyanskyi 1989). Botanical names of accessions that originated from GRIN were based on old names.
‡Wi, wild; We, weedy; A, annual; C, cultivated; and P, perennial.
ers. The EcoRI or PstI primers were labeled with 33P prior to
amplification. The amplification products were resolved on a
5% polyacrylamide sequencing gel in 1× Tris–borate–EDTA
buffer for 2.5–3 h. The gels were dried for 2 h and exposed
to X-ray film for 48 h.
Data analysis
Polymorphism was scored manually. DNA fragments located on the same migration position were read as 1 locus
and scored 1 for present or 0 for absent. The data matrices
for each primer combination were converted into Nei–Li’s or
Dice similarity matrix and then compared among each other
using Mantel’s test (Mantel 1967). The data matrices showing good relationships were then combined and used for
cluster analysis. To detect potentially weakly supported lineages, 2 different algorithms were used in the data analysis.
The 1st approach utilized phenetic analysis involving various
similarity coefficients and clustering methods to obtain the
shortest tree. Similarity coefficients and clustering methods
were tested using similarity of qualitative data (SIMQUAL),
sequential agglomerative hierarchical nested (SAHN) clus-
tering routine, and TREE from program NTSYSpc version
2.1 (Rohlf 2000). The unweighted pair group method with
arithmetic means (UPGMA), weighted pair group method
with arithmetic means (WPGMA), complete-link, and singlelink clustering methods were applied in all possible combinations with both the Dice and Jaccard similarity coefficients.
Cophenetic correlation coefficients (r) were calculated and
compared for each of the combinations using COPH and
MXCOMP from NTSYSpc 2.1 procedures. Principal coordinate analysis (PCO) was performed using the procedures
DCENTER, EIGEN, and MXPLOT from NTSYSpc 2.1k.
The 2nd approach involved a phylogenetic analysis using a
maximum parsimony method that was performed with the
PAUP* (phylogenetic analysis using parsimony) program
(Swofford 1998). Bootstrap tests were performed using 5000
replications for phenetic analysis and 1000 replications for
phylogenetic analysis to assess confidence in producing tree
topologies (Felsenstein 1985). The results from both phenetic
and phylogenetic analyses were compared to determine the
best phylogenetic relationship among the species. Fisher’s
exact test was used to test the correlation between life cycle
© 2005 NRC Canada
Chikmawati et al.
795
Table 2. Primer combination, total bands, polymorphic bands, level of polymorphisms, and average of r value with respect to the genus Secale.
Primer combination
Total
bands
Polymorphic
bands
Polymorphism
(%)
Average of
r value*
P-ACA/M-AAC
P-ACA/M-ACT
P-ACA/M-ATC
P-ACA/M-CCA
P-ACA/M-CCT
P-AGT/M-AAC
P-AGT/M-ACG
P-AGT/M-AGA
P-AGT/M-ATC
P-AGT/M-ATG
P-AGT/M-CCT
Average
86
42
77
63
38
53
59
67
67
92
57
64
73
32
68
59
31
36
48
46
36
60
43
48
85
76
88
94
82
68
81
69
54
65
75
76
0.80
0.65
0.74
0.77
0.69
0.79
0.71
0.79
0.70
0.77
0.79
—
E-ACA/M-ACG
E-AGC/M-ACG
E-CCT/M-CCA
E-CCT/M-CAC
E-CCT/M-CAG
E-CCT/M-CCT
E-CCT/M-CTG
Average
54
64
75
72
51
63
50
61
33
39
48
40
33
34
30
38
61
61
64
56
65
56
60
60
0.72
0.73
0.80
0.75
0.67
0.77
0.60
—
1130
789
—
—
Total
*r value from Mantel’s test.
character and AFLP data (Fisher 1970). Genetic structure of
the Secale genus was analyzed using Wright’s F test (Hartl
and Clark 1997).
Results
Eighteen primer combinations were utilized in analyzing
phylogenetic relationships among the various Secale species
(Table 2). Amplified fragment length polymorphism analysis
revealed a very large distinct number of fragments per
primer pair in the members of the genus Secale (Fig. 1). The
number of fragments present varied with each primer and
ranged from 38 to 86. Each primer combination demonstrated approximately 54%–94% polymorphism. In total,
789 polymorphic markers of 1130 fragments were observed.
Primer combinations P-/M- and E-/M- generally yielded
similar numbers of fragments, 64 and 61 bands, respectively;
however, P-/M- primer combinations produced a higher level
of polymorphism (76%) than the E-/M- primer combinations
(60%). The most efficient primer combination, which showed
the largest number of polymorphic fragments, was the PACA/M-AAC (Table 2).
The Mantel test was performed to establish the correlation
among primer combinations before constructing phenetic
and phylogenetic trees. The correlation values ranged from
0.60 to 0.80 (Table 2) and were significant in all primer
combinations. The 789 polymorphic markers resulting from
the 18 primer combinations were then used to construct
phenetic and phylogenetic trees.
Cophenetic correlation coefficients measured the goodness of fit of the cluster analysis to the similarity matrix, and
the results showed that cophenetic correlation coefficients
were >0.90 (Table 3) indicating that the goodness of fit for
all combinations of the similarity matrix and cluster analysis
were very good. Therefore, all those combinations were appropriate for use in analyzing the phenetic relationships.
Among the cluster methods, UPGMA yielded the highest
cophenetic correlation in all cases. The combinations of
UPGMA with the Dice and Jaccard coefficients also yielded
identical tree topologies. Therefore, they were considered
the most suitable combinations for data analysis.
The phenogram and the cladogram, from combinations of
UPGMA and Dice similarity, had similar tree topologies
(Figs. 2 and 3). In both cluster analyses, S. sylvestre was
first separated (100% support), followed by Secale ciliatoglume (70% and 62% support in dendrogram and cladogram,
respectively). The rest of the accessions were clustered into
2 major groups. Group III, S. montanum, Secale anatolicum,
S. kuprijanovii, Secale chaldicum, and S. africanum, consisted of all perennial taxa (42% and 72% support in dendrogram and cladogram, respectively). Group II, S. ancestrale,
S. afghanicum, S. cereale, S. dighoricum, Secale turkestanicum, S. segetale, and Secale vavilovii, consisted of annual
taxa (91% and 37% support in dendrogram and cladogram,
respectively).
The results of PCO analysis showed that the first 2 axes
accounted for 16.38% and 7.24% of the data variance
(Fig. 4). Cumulatively, the PCO results represented 24% of
the data, which were sufficient to resolve all the analyzed
accessions into 3 distinct groups. Group 1 contained
S. sylvestre and the 2 outgroups, group II contained the annual taxa, and group III contained the perennial taxa. The
© 2005 NRC Canada
796
Genome Vol. 48, 2005
Fig. 1. An example of the AFLP polymorphism among Secale species, detected with the PstI-ACA and MseI-AAC primer combination.
Table 3. Cophenetic correlation coefficients for amplified fragment length polymorphism data of all Secale species used.
UPGMA, unweighted pair group method with arithmetic means;
WPGMA, weighted pair group method with arithmetic means.
Clustering/Similarity
Jaccard
Dice
UPGMA
WPGMA
Complete-link
Single-link
0.98
0.96
0.95
0.97
0.98
0.96
0.96
0.97
PCO results were congruent with the results of cluster
analysis.
Genetic differentiation among annual and perennial taxa
was analyzed using the Wright’s F test (Table 4). Since
S. sylvestre showed really distinct characters, this taxon was
excluded from the analysis. The genetic differentiation level
of annual taxa was smaller than that of perennial taxa; however, the genetic diversity within the annual taxa was much
higher than that of perennial taxa.
Discussion
Phylogenetic relationships among Secale species
The AFLP technique produced a wide range of variability
among Secale taxa that was sufficient to clearly resolve all
analyzed accessions into 3 major groups: group 1 contained
of S. sylvestre, group II contained all of the perennial taxa,
and group III contained all of the annual taxa. This indicated
that the perennial versus annual life cycle probably played
an important role in determining the relationships among the
Secale species. Further analysis using Fisher’s exact test
supported this by showing that 24% of the AFLPs detected
were associated with the character life cycle. Our results
supported the validity of 3 major series within Secale as recognized by Roshevitz (1947). Secale montanum and all perennial forms constituted 1 major group, the series
Kuprijanovia Roshev.; S. cereale and all weedy annual relatives constituted the series Cerealia Roshev.; and S. sylvestre
stood alone as an annual and constituted the series Silvestria
Roshev.
Among the annual taxa, S. cereale had a closer relationship to S. ancestrale, S. afghanicum, S. dighoricum, and
S. segetale than to S. vavilovii. Even though both S. turkestanicum and S. cereale are cultivated plants, they had a
distant relationship with each other. The breeding system
differences between the 2 taxa (S. turkestanicum is selfpollinated and S. cereale is cross-pollinated) may offer a
possible explanation.
Among perennial taxa, S. ciliatoglume showed the most
distant relationship from the others. Secale ciliatoglume is
an isolated weedy population with pubescent culms that appear to be endemic to orchards and vineyards near Mardin,
Turkey. It is possible that the very limited distribution of this
taxon allowed it to maintain a distinct identity from the others.
Within group II, S. africanum had the furthest relationship from
S. montanum, whereas S. anatolicum and S. kuprijanovii had
the closest relationship to S. montanum. Somewhat surprisingly, S. sylvestre was closer to the perennial taxa than to the
annual taxa. Since S. sylvestre had the closest relationship to
the outgroups, it can be considered as the most ancient
among Secale species, and S. cereale can be considered as
the youngest of the Secale species.
Based on cytological, ecological, and morphological
© 2005 NRC Canada
Chikmawati et al.
797
Fig. 2. A dendrogram constructed from the AFLP data, using Nei–Li’s distance and UPGMA clustering. Numbers on the branches are
bootstrap and range from 16% to 100%.
studies, Stutz (1972) demonstrated that cultivated rye
(S. cereale L.) originated from the weedy progeny derived
from introgression of S. montanum (syn. S. strictum) into
S. vavilovii, S. africanum, Secale dalmaticum, S. ciliatoglume,
and S. kuprijanovii as they had close relationships with each
other and appeared to be only slightly modified isolated
populations of S. montanum. Populations of S. anatolicum
were thought to be weedy forms of S. montanum, genetically
and chromosomally distinct from the weedy annual forms.
In general, the species relationships within genus Secale as
based on AFLP data were consistent with Stutz (1972).
However, Stutz (1972) also suggested that S. montanum was
the common ancestor of all the Secale species, which conflicts with the present AFLP data. The present data clearly
demonstrated that S. sylvestre was the most ancient species
that split off first from the common ancestor, whereas
S. montanum split off after the separation of S. sylvestre.
Molecular taxonomy of Secale based on AFLP
Secale sylvestre is a low growing plant with a fragile
© 2005 NRC Canada
798
Genome Vol. 48, 2005
Fig. 3. A cladogram resulting from the maximum parsimony method that was conducted using heuristic search methods with TBR
branch swapping, collapse of zero-options, and weighting of all characters equally. Numbers on branches are bootstrap values and
range from 5% to 100%. Consistency index (CI) = 0.18, retention index (RI) = 0.35, rescale consistency index (RCI) = 0.06.
rachis, widely distributed from central Hungary eastward
throughout the sandy steppes of southern Russia. This taxon
can be easily distinguished from other taxa by its long
awned glumes (Stutz 1972). Khush and Stebbins (1961)
showed that S. sylvestre was cytogenetically very distant
from S. cereale, and was geographically, ecologically, and
reproductively isolated from S. montanum (Sencer and
Hawkes 1980). In addition, S. sylvestre also has other unique
characteristics, such as distinct chloroplast DNA (Petersen
and Doebley 1993), a spacer length variant of the ribosomal
DNA (Reddy et al. 1990), and an internal transcribed spacer
of the 18S–5.8S–26S rDNA (ITS-1) region, as compared
with the other Secale species. Given the strong distinction of
S. sylvestre from other taxa, it is easy to consider S. sylvestre
as a distinct species. Amplified fragment length polymorphism analysis showed that S. sylvestre also demonstrated a
© 2005 NRC Canada
Chikmawati et al.
799
Fig. 4. Principal coordinate analysis of Secale species based on AFLPs. I = S. sylvestre and 2 outgroups; II = annual taxa; and III =
perennial taxa.
Table 4. Total genetic diversity, genetic diversity within population, genetic differentiation and genetic identity of Secale accessions. Ht, total genetic diversity; Hs, genetic diversity within
population; Gst, genetic differentiation; GI, genetic identity.
Groups
Ht
Hs
Gst
GI
Annual taxa
Perennial taxa
0.25±0.03
0.30±0.03
0.09±0.0062
0.02±0.0009
0.63
0.90
0.68–0.89
0.24–0.48
very distinct profile in all primer combinations, and it was
well separated from others in all analyses. Thus, AFLP analyses confirm S. sylvestre as a distinct species.
The presence of a high degree of similarity among wild,
weedy annual forms and cultivated rye has been demonstrated. Khush (1963) showed that there was no evidence of
structural differences between the genome of cultivated rye
and several weedy ryes (S. cereale, S. vavilovii, S. ancestrale,
S. afghanicum, S. dighoricum, and S. segetale), which had
previously been recognized as varieties, subspecies, or even
species. They all readily crossed and produced vigorous F1s,
which had similar chromosome arrangements, breeding
habit, periodicity, and crossability, and they also demonstrated geographical continuity. Therefore, Khush (1963)
proposed all annual forms are subspecies of S. cereale. Further study based on morphometrical analyses concluded that
it was impossible to recognize each annual taxon based on
their morphology (Frederiksen and Petersen 1997). They proposed 2 intraspecific taxa within a single species (S. cereale),
which are S. cereale subsp. cereale for cultivated rye and
S. cereale subsp. ancestrale for weedy and wild annual rye
taxa. The most recent study based on ITS-1 region also
found no differences between the weedy forms and cultivated rye (Bustos and Jouve 2002). Thus, all previous studies demonstrated that morphologically and genetically all
annual taxa were similar to each other and that it was impossible to discriminate them clearly.
The AFLP results showed that 6 accessions of S. cereale,
which originated from different locations, composed a monophyletic group. The S. dighoricum accessions also clustered
together, which was not too surprising since they both originated from the same location. Thus, it is highly possible that
those accessions are duplicated accessions, whereas other
annual taxa that were represented by more than 1 accession
did not cluster together. Principal coordinate analyses intermingled them with each other. Nei–Li’s distances yielded
very small genetic-distance values among them indicating
they had a high degree of AFLP similarity. This was also
shown from the intermediate level of genetic differentiation
(Gst = 0.63). Except for cultivated rye, it was still difficult to
discriminate between wild and weedy rye using AFLPs,
therefore, AFLP analysis supported Frederiksen and Petersen
(1997).
Among perennial species, only S. ciliatoglume did not
cluster together with the others. Secale ciliatoglume stood
alone between annual and perennial taxa in cluster analyses,
but the separation was intermediately supported (62% in
phylogenetic and 70% in phenetic analysis). The PCO analysis placed this accession in the same quadrant with the other
perennial taxa. The information about S. ciliatoglume from
previous studies was very limited, and only dealt with morphological data. This taxon has been shown to be morphologically similar to S. montanum, and only deviated by having
a dense cover of hairs over the internodes, leaf sheaths, and
blades (Frederiksen and Petersen 1998). Frederiksen and
Petersen (1997) suggested that S. ciliatoglume should be given
an intraspecific rank.
Stutz (1972) demonstrated that several perennial forms
(S. anatolicum, S. africanum, S. dalmaticum, and S. mon© 2005 NRC Canada
800
tanum) readily crossed to each other, and that crossing
among them yielded normal chromosome configurations indicating no reproductive barrier. Khush (1962) proposed all
perennial taxa as subspecies of S. montanum. Furthermore,
Sencer and Hawkes (1980) showed that all the wild perennial forms had a fairly common morphological resemblance.
Our analysis, based on AFLPs, clustered the 5 accessions of
S. montanum together with low bootstrap support (47% in
cladogram). The separation of the other perennial taxa was
with intermediate bootstrap support (47%–74% in cladogram). However, the genetic differentiation level among perennial taxa was very high (Gst = 0.90). This result suggests
that AFLP marker polymorphism levels within the perennial
taxa were sufficient to discriminate and place them in an
intraspecific rank, instead of in an interspecific rank.
In summary, this is the first phylogenetic study of the genus Secale based on AFLP analyses. Analyzing AFLPs
among 29 accessions representing 14 of the most recognized
ranked species or subspecies in the genus Secale demonstrated that AFLP marker technology is a better tool for analyzing phylogenetic relationships among Secale species,
because it produced high levels of polymorphism that were
sufficient to resolve all accessions into 3 distinct groups.
Group I contained S. sylvestre, group II contained all perennial taxa, and group III contained all annual taxa. (Secale
ciliatoglume stood alone between annual and perennial taxa
in cluster analyses.) The phylogenetic relationships among
Secale taxa based on AFLPs strongly supported Stutz
(1972). The AFLP data were clearly able to distinguish 3
Secale species: S. sylvestre, S. cereale, and S. montanum. In
addition, AFLPs were also able to recognize 2 intraspecific
taxa of S. cereale, namely S. cereale subsp. cereale for cultivated rye and S. cereale subsp. ancestrale for wild and
weedy rye annual taxa, and 6 intraspecific taxa of S. montanum, namely S. montanum subsp. montanum; S. montanum
subsp. africanum; S. montanum subsp. anatolicum; S. montanum subsp chaldicum; S. montanum subsp. ciliatoglume;
and S. montanum subsp. kuprijanovii. These results were
consistent with the last revision of the genus Secale by
Frederiksen and Petersen (1998) who recognized only 3 species within the genus Secale: S. sylvestre, S. strictum (syn. S.
montanum), and S. cereale.
Acknowledgements
The authors thank Miftahudin for help and advice on data
analysis and Kathleen Ross for helpful discussions on plant
preparation. We are also grateful to Dr. J. Bockelman from
Germplasm Resources Information Network (GRIN) of the
United States Department of Agriculture, Dr. M. Niedzielski
and Dr. W. Podima from the Plant Breeding and Acclimatization Institute (IHAR), Poland, for supplying us with seeds.
Tatik Chikmawati is a Ph.D. student funded by the Quality
for Undergraduate Education (QUE) project from the Department of Biology, Bogor Agricultural University, Indonesia.
References
Beardsley, P.M., Yen, A., and Olmstead, R.G. 2003. AFLP phylogeny of Mimulus section Erythranthe and the evolution of hummingbird pollination. Evol. 57(6): 1397–1410.
Genome Vol. 48, 2005
De Bustos, A., and Jouve, N. 2002. Phylogenetic relationships of
the genus Secale based on the characterization of rDNA ITS sequences. Plant Syst. Evol. 235: 147–154.
Dedio, W., Kaltsikes, P.J., and Larter, E.N. 1969. Numerical
chemotaxonomy in the genus Secale. Can. J. Bot. 47: 1175–
1180.
Felsenstein, J. 1985. Confidence limit on phylogenies: an approach
using the bootstrap. Evolution, 39: 783–79.
Fisher, R.A. 1970. Statistical methods for research workers. 14th
ed. Oliver and Boud, Edinburgh, U.K.
Frederiksen, S., and Petersen, G. 1997. Morphometrical analyses
of Secale (Triticeae, Poaceae). Nord. J. Bot. 17: 185–197.
Frederiksen, S., and Petersen, G. 1998. A taxonomic revision of
Secale. Nord. J. Bot. 18: 399–420.
Hammer, K., Skolimowska, E., and Knueffer, H. 1987. Vorarbeiten
zur monographischen Darstellung von Wildpflanzensortimenten:
Secale L. Kulturplanze, 35: 135–177. [In German.]
Hartl, D.L., and Clark, A.G. 1997. Principles of population genetics. Sinauer Associates, Inc., Suntherland, Mass., U.S.A.
Janssen, P., Maquelin, K., Coopman, R., Tjernberg, I., Bouvet, P.,
Kersters, K., and Dijkshoorn, L. 1997. Discrimination of
Acinetobacter genomic species by AFLP fingerprinting. Int. J.
Syst. Bacteriol. 47: 1179–1187.
Khush, G. 1962. Cytogenetic and evolution studies in Secale II: interrelationships of the wild species. Evolution, 16: 484–496.
Khush, G. 1963. Cytogenetic and evolutionary studies in Secale
III: cytogenetics of weedy ryes and origin of cultivated rye.
Econ. Bot. 17: 60–71.
Khush, G., and Stebbins, G.L. 1961. Cytogenetic and evolution in
Secale I: some new data on the ancestry of S. cereale. Am. J.
Bot. 48: 723–730.
Kobylyanskyi, V.D. 1989. Flora of cultivated plants of USSR. Vol.
II. Part 1. Agronomizdat, Leningrad, Russia.
Laurence, D., Ludovic, G., Benedicte, R., and Pierre, T. 2003.
Using AFLP to resolve phylogenetic relationships in a morphologically diversified plant species complex when nuclear and
chloroplast sequences fail to reveal variability. Mol. Phylogenet.
Evol. 27: 185–196.
Leonard, W.H., and Martin, J.H. 1963. Cereal crops. The
Macmillan Company, New York.
Mantel, N. 1967. The detection of disease clustering and a generalized regression approach. Cancer Res. 27: 209–220.
Monte, J.V., McIntyre, C.L., and Gustafson, J.P. 1995. Analysis of
phylogenetic relationships in the Triticeae tribe using RFLPs.
Theor. Appl. Genet. 86: 649–655.
Mueller, U.G., and Wolfenbarger, L.L. 1999. AFLP genotyping and
fingerprinting. Trends Ecol. Evol. 14: 389–394.
Murai, K., Naiyu, X., and Tsunewaki, K. 1989. Studies on the origin of crop species by restriction endonuclease analysis of
organellar DNA. III. Chloroplast DNA variation and
interspecific relationships in the genus Secale. Japan. J. Genet.
64: 36–47.
Petersen, G., and Doebley, F. 1993. Chloroplast DNA variation in
the genus Secale (Poaceae). Plant Syst. Evol. 187: 115–125.
Reddy, P., Appel, R., and Baum, B.R. 1990. Ribosomal DNA
spacer-length variation in Secale spp. (Poaceae). Plant Syst. Evol.
171: 205–220.
Rohlf, F.J. 2000. NTSYSpc. Numerical taxonomy and multivariate
analysis system. Version 2.1. Exeter Software, Setauket, New
York.
Roshevitz, R.I. 1947. Monographie of the genus Secale L. Acta
Inst. Botanici Acad. Scientiarum U.R.S.S. I(6): 105–163.
Russell, J.R., Fuller, J.D., Macaulay, M., Hatz, B.G., Jahoor, A.,
Powell, W., and Waugh, R. 1997. Direct comparison of levels of
© 2005 NRC Canada
Chikmawati et al.
genetic variation among barley accessions detected by RFLPs,
AFLPs, SSRs, and RAPDs. Theor. Appl. Genet. 95: 714–722.
Saghai-Maroof, M.A., Biyashev, R.M., Yang, G.P., Zhang, Q., and
Allard, R.W. 1994. Extraordinarily polymorphic microsatellite
DNA in barley: species diversity, chromosomal locations, and population dynamics. Proc. Natl. Acad. Sci. U.S.A. 91: 5466–5470.
Sencer, H.A., and Hawkes, J.G. 1980. On the origin of cultivated
rye. Biol. J. Linn. Soc. 13: 299–313.
Sharma, S.K., Knox, M.R., and Ellis, T.H.N. 1996. AFLP analysis
of the diversity and phylogeny of Lens and its comparison with
RAPD analysis. Theor. Appl. Genet. 93: 751–758.
Stutz, H.C. 1972. On the origin of cultivated rye. Am. J. Bot. 59:
59–70.
Swofford, D.L. 1998. Phylogenetic Analysis Using Parsimony
(PAUP*). Version 4.0b10 [computer program]. Sinauer Associates, Sunderland, Mass., U.S.A.
Vavilov, N.I. 1917. On the origin of cultivated rye. Bull. Appl. Bot.
Genet. Plant Breed. 10: 561–590.
801
Vavilov, N.I. 1926. Studies on the origin of cultivated plants. Bull.
Appl. Bot. Genet. Plant Breed. 16: 1–248.
Vence, F.J., Vaquero, F., and Perez de la Vega, M. 1987. Phylogenetic relationships in Secale (Poaceae): an isozymatic study. Plant
Syst. Evol. 157: 33–47.
Vos, P., Hogers, R., Blecker, M., Reijans, M., van de Lee, T.,
Hornes, M., Frijters, A., Peleman, J., Kuiper, M., and Zabeau,
M. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res. 23: 4407–4414.
Waugh, R., Bonar, N., Baird, E., Thomas, B., and Graner, A. 1997.
Homology of AFLP products in three mapping populations of
barley. Mol. Gen. Genet. 255: 311–321.
Zhang, L.-B., Comes, H.P., and Kadereit, J.W. 2001. Phylogeny
and quaternary history of the European montane/alpine endemic
Soldanella (Primulaceae) based on ITS and AFLP variation.
Am. J. Bot. 88: 2331–2345.
© 2005 NRC Canada