Molecular phylogeny of Muscari (Asparagaceae) inferred from cpDNA sequences

Abstract

In this study, we tried to figure out phylogenetic relationships and taxonomical positions of closely related Muscari species. Four different cpDNA regions including both coding and non-coding ones, namely, matK, trnT(UGU)-trnL(UAA) intergenic spacer (IGS), trnL(UAA) intron and trnL(UAA)-F(GAA) IGS, were employed to determine the exact circumscription of three subgenera, Muscari, Leopoldia and Botryanthus. Seventy Muscari accessions representing 31 Muscari species and different number of previously published sequences retrieved from NCBI database were analyzed. The concatenated and matK data alone were observed to be informative while none of the used non-coding regions was suitable to determine phylogeny of Muscari. Concatenated alignment gave almost the same tree topology with matK sequence. Muscari azureum and M. coeleste phylogenetically separated from all other species of Botryanthus and four main clades were observed in both of the trees even though three subgenera are accepted by Flora of Turkey. After discussing the phylogenetic positions and morphological characters in detail, moving of these two species from Botryanthus to Pseudomuscari subgenus was suggested. Thus, this study proposes that the number of Muscari subgenera should be increased from three to four in Flora of Turkey. The position of M. mirum was also remarkable; it always located distantly to its relatives of Leopoldia. Although this species may also be distinguished based on morphological features such as quite shorter plant length, one or sometimes two relatively wider leaves, and a larger fruit, it needs further studies to resolve its position reliably. Interesting positions of other species were also discussed in detail based on morphological characters in the text.

Full text

1 23
Biologia
Botany, Zoology and Cellular and
Molecular Biology
ISSN 0006-3088
Biologia
DOI 10.2478/s11756-018-00164-0
Molecular phylogeny of Muscari
(Asparagaceae) inferred from cpDNA
sequences
Ayten Dizkirici, Oktay Yigit, Mesut
Pinar & Huseyin Eroglu

1 23
Your article is protected by copyright and all
rights are held exclusively by Plant Science
and Biodiversity Centre, Slovak Academy
of Sciences. This e-offprint is for personal
use only and shall not be self-archived
in electronic repositories. If you wish to
self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

ORIGINAL ARTICLE
Molecular phylogeny of Muscari (Asparagaceae) inferred
from cpDNA sequences
Ayten Dizkirici
1
&Oktay Yigit
1
&Mesut Pinar
2
&Huseyin Eroglu
3
Received: 30 May 2018 /Accepted: 15 November 2018
#Plant Science and Biodiversity Centre, Slovak Academy of Sciences 2018
Abstract
In this study, we tried to figure out phylogenetic relationships and taxonomical positions of closely related Muscari species. Four
different cpDNA regions including both coding and non-coding ones, namely, matK, trnT
(UGU)
-trnL
(UAA)
intergenic spacer
(IGS), trnL
(UAA)
intron and trnL
(UAA)
-F
(GAA)
IGS, were employed to determine the exact circumscription of three subgenera,
Muscari,Leopoldia and Botryanthus. Seventy Muscari accessions representing 31 Muscari species and different number of
previously published sequences retrieved from NCBI database were analyzed. The concatenated and matK data alone were
observed to be informative while none of the used non-coding regions was suitable to determine phylogeny of Muscari.
Concatenated alignment gave almost the same tree topology with matK sequence. Muscari azureum and M. coeleste phyloge-
netically separated from all other species of Botryanthus and four main clades were observed in both of the trees even though
three subgenera are accepted by Flora of Turkey. After discussing the phylogenetic positions and morphological characters in
detail, moving of these two species from Botryanthus to Pseudomuscari subgenus was suggested. Thus, this study proposes that
the number of Muscari subgenera should be increased from three to four in Flora of Turkey. The position of M. mirum was also
remarkable; it always located distantly to its relatives of Leopoldia. Although this species may also be distinguished based on
morphological features such as quite shorter plant length, one or sometimes two relatively wider leaves, and a larger fruit,it needs
further studies to resolve its position reliably. Interesting positions of other species were also discussed in detail based on
morphological characters in the text.
Keywords Muscari .Phylogeny .Pseudomuscari .matK .trnT-F
Introduction
The genus Muscari Miller (Asparagaceae) is a group of pe-
rennial bulbous plants mainly distributed in Caucasus, tem-
perate Europe, Africa, North-western and South-western Asia
(Jafari and Maassoumi 2011). The flowers are in raceme form
with minute bracts. They have cylindrical, bell-shaped or tu-
bular perianth, consisting of six accreted leaves bent at the
edge and the color varies from white to dark blue. (Cullen
et al. 2011). It is characterized by the base chromosome num-
ber x =9, with more bimodal karyotype (Ruiz Rejón and
Oliver 1981). Species of subgenus Leopoldia are generally
diploids (Nersesian 2001; Jafari and Maassoumi 2011)while
species of subgenera Muscari and Botryanthus show a ploidy
series ranging from 2x to 8x (Garbari 2003; Suárez-Santiago
et al. 2007). According to the latest checklist for Muscari,the
genus is represented by 50 taxa (Govaerts 2018) and the center
of diversity is situated in Turkey with 38 species, 25 of which
are endemic (Davis and Stuart 1984; Davis et al. 1988;
Özhatay 2000; Özhatay and Kültür 2006; Özhatay et al.
2009,2011; Demirci et al. 2013; Pirhan et al. 2014; Kaya
2014;Yıldırım2015,2016;Pınar et al. 2018). Infrageneric
taxonomy, specifically boundaries of the subgenera have not
been unambiguously delimited till now. First of all, four
subgenera, Muscari,Leopoldia,Moscharia (Baker) Chouard
Electronic supplementary material The online version of this article
(https://doi.org/10.2478/s11756-018-00164-0) contains supplementary
material, which is available to authorized users.
*Ayte n D izki rici
aytendizkirici@gmail.com
1
Department of Molecular Biology and Genetics, Van Yüzüncü Yıl
University, 65080 Van, Turkey
2
School of Health, Van Yüzüncü Yıl University, 65080 Van, Turkey
3
Department of Biology, Van Yüzüncü YılUniversity,
65080 Van, Turkey
Biologia
https://doi.org/10.2478/s11756-018-00164-0
Author's personal copy

(≡Muscarimia)andPseudomuscari (Losinsk.) D. C. Stuart,
were accepted for the genus (Speta 1989,1998; Davis and
Stuart 1980,1984). Karlén (1991) grouped the genus into four
subgenera as Botryanthus (Kunth) Zahar., Leopoldia (Parl.)
Zahar., Muscari,andPseudomuscari Stuart. The genus is di-
vided into three subgenera; Muscari,Leopoldia (Kunth) Parl.
and Botryanthus (Kunth) Rouy in Flora of Turkey (Davis and
Stuart 1984) although the basis of this classification remains
unclear. At present, the Kew World Checklist (Govaerts 2018)
splits Muscari s.l. into three genera (Muscari s.str., Leopoldia
Parl. and Pseudomuscari Garbari & Greuter). Moreover,
Leopoldia and Pseudomuscari subgenera have been consid-
ered as separate genus by some researchers (Garbari and
Greuter 1970; Jafari and Maassoumi 2011; Böhnert and
Lobin 2017). When all these studies and claims are taken into
consideration, the taxonomic situation within Muscari seems
enigmatic and controversial.
A number of studies have been done on distribution
(Yilmaz et al. 2017), anatomy (Herrmann et al. 2005) and
karyology (Valdés and Lifante 1992; Suárez-Santiago et al.
2007;Jafarietal.2008) of the genus in addition to morpho-
logical and taxonomical studies. However, a reliable classifi-
cation of the genus is not still available since morphological
characters (bulb morphology, structures of pistil, petal and
seed) and karyological data are not clear and reliable to make
taxonomic judgements uncontroversial. Because of discrepan-
cies between morphological, karyological and anatomical
characters, we wanted to use molecular data to shed more light
on evolutionary and taxonomic relationships among
subgenera and species within the genus Muscari. There are
no global taxonomic revisions of the genus. Only regional
revisions covering a small number of taxa can be seen
(Garbari and Greuter 1970; Davis and Stuart 1984;
Czerepanov 1995; Johnson et al. 1996). The current study is
invaluable because almost all Muscari species distributed in
Turkey were studied and four regions of chloroplast DNA
(cpDNA), one coding and three non-coding, were used to
demonstrate phylogeny within the Muscari genus.
Regions of cpDNA have been used extensively to infer
plant phylogenies at lower-level (intergeneric, interspecific,
and intraspecific) studies (Taberlet et al. 1991; Gielly and
Taberlet 1994). Non-coding regions of the cpDNA tend to
evolve more rapidly than coding sequences due to the accu-
mulation of insertions/deletions and nucleotide substitutions
(Clegg and Zurawski 1991). Therefore, both coding and non-
coding regions were intentionally selected to figure out phy-
logenetic relationships and taxonomic positions of the species
and to determine the number of subgenera reliably. matK
gene, which encodes a maturase and is involved in splicing
type II introns from RNA transcripts, was selected as coding
region. This gene is encoded by the chloroplast trnK intron
and evolves faster compared to the other coding cpDNA re-
gions due to its relatively fast mutation rate (Hilu and Liang
1997). A large number of different non-coding regions of the
cpDNA have been investigated in angiosperms, some of
which are highly variable while others show relatively little
variation (Shaw et al. 2005). trnT
(UGU)
-trnL
(UAA)
intergenic
spacer (IGS), trnL
(UAA)
intron and trnL
(UAA)
-F
(GAA)
IGS were
selected as non-coding regions (Taberlet et al. 1991; Shaw
et al. 2007). They were selected as DNA barcodes since am-
plification of them across a wide taxonomic range is easy due
to universal primers designed by Taberlet et al. (1991).
Furthermore, they were chosen due to their common use in
recent studies of relative genera such as Bellevalia (Pfosser
and Speta 1999;Buerkietal.2012; Borzatti Von Loewenstern
et al. 2013), Hyacinthella (Buerki et al. 2012), and
Ornithogalum (Martínez-Azorín et al. 2011).
In the present study, we wanted to resolve phylogenetic
relationships among Muscari species and shed more light on
evolutionary and taxonomic relationships among subgenera
and species that show variations from one study to another.
For this purpose, 70 Muscari accessions representing 31
Muscari species and different number of previously published
sequences retrieved from NCBI database were analyzed using
DNA sequences of four different and concatenated cpDNA
regions.
Materials and methods
Plant material
In the present study, a total of 70 accessions representing 31
Muscari species were used; two from subg. Muscari, nine
from subg. Leopoldia and twenty from subg. Botryanthus
(Table 1). Plant specimens were collected from their natural
habitats in the different regions of Turkey. The collected sam-
ples were identified by Dr. Pinar according to the diagnostic
morphological characteristics described in the Flora of Turkey
(Davis and Stuart 1984; Davis et al. 1988;Özhatay2000)and
voucher specimens were deposited in Department of
Biological Sciences, Van Yüzüncü Yil University, Turkey.
For each taxon, at least two samples were collected (Table 1)
and preserved in plastic bags with silica gel until DNA
extraction.
DNA isolation, PCR amplification, sequencing,
sequence alignment and phylogenetic analyses
Total genomic DNA was extracted from 1 g of fresh material
using a standard hexadecyltrimethylammonium bromide
(CTAB) protocol with minor modifications (Doyle and
Doyle 1987). Quality of DNA for each sample was decided
by running them on a 0.8% 1X-TBE (Tris-borate-EDTA) aga-
rose gel stained with EtBr. DNA concentration was measured
Biologia
Author's personal copy

using Thermo Multiskan GO spectrophotometer, and DNA
stocks were diluted to 10 ng/μLforPCRstudies.
We amplified and sequenced a large part of the matK re-
gion by using primers F1/R3 (Li et al. 1997). The trnT-L IGS,
trnL intron and trnL-F IGS regions were amplified by using
primer pairs a (forward)/b (reverse), c/d and e/f, respectively
(Taberlet et al. 1991). PCR reactions were carried out in 50 μL
volumes with the following reaction components: genomic
DNA (10 ng/μl), 10X PCR Buffer [750 mM Tris-HCL
(pH 8.8), 200 mM (NH
4
)
2
SO
4
, 0.1% Tween 20], MgCl
2
(25 mM), dNTP mixture (10 mM), selected primer pair
(10 μM), Taq polymerase (5u/μl) and sterile water. The PCR
condition for matK region was optimized as 95 °C for 5 min
followed by 35 cycles of denaturation at 94 °C for 1 min,
annealing at 58 °C for 1 min, and elongation at 72 °C for
1.5 min, and final extension at 72 °C for 10 min to complete
the primer-template extensions. PCR conditions for non-
coding regions were as follow: initial denaturation at 94 °C
for 2 min, followed by 30 cycles of 30 s at 94 °C, 25 s at 53 °C,
45 s at 72 °C, ending with a single final elongation of 7 min at
72 °C. Purified PCR products were sequenced (see Table 1for
accession numbers) in both directions using ABI 3730XL
automated sequencer (Applied Biosystems, Foster City, CA).
Alibee Multiple Alignment 3.0 software from the GeneBee
website (www.genebee.msu.su/genebee.html)wasusedto
assemble complementary strands and verify software base-
calling. In addition to our sequences, three sequences of
matK, five sequences of trnL intron and four sequences of
trnL-F IGS region were retrieved from NCBI database to in-
crease the inter-specific sampling (Appendix 1). No sequence
of trnT-L IGS region was found in NCBI database. All se-
quences were concatenated into a super-gene alignment,
Table 1 Samples of Muscari taxa used in the study
Subgenus Muscari species # of Sample Locality Accession Number (matK) Accession Number (trnT-L-F)
Muscari M. racemosum Mill. 2 Burdur MH152601 MH152632
M. macrocarpum Sweet 2 Muğla MH152602 MH152633
Leopoldia M. comosum (L.) Mill. 5 Mersin MH152603 MH152634
M. weissii Freyn 2 Antalya MH152604 MH152635
M. caucasicum (Griseb.) Baker 2 Van MH152605 MH152636
M. tenuiflorum Tausch 2 Adana MH152606 MH152637
M. babachii Eker & Koyuncu 2 Hatay MH152607 MH152638
M. erdalii Özhatay & Demirci 2 Mersin MH152608 MH152639
M. longipes Boiss. 2 Konya MH152609 MH152640
M. massayanum C. Grunert 2 Adana MH152610 MH152641
M. mirum Speta 2 Burdur MH152611 MH152642
Botryanthus M. aucheri (Boiss.) Baker 2 Kayseri MH152612 MH152643
M. armeniacum Leichtlin ex Baker 2 Karaman MH152613 MH152644
M. sivrihisardaghlarensis Yıld. & B. Selvi 2 Eskişehir MH152614 MH152645
M. neglectum Guss. ex Ten. 7 Gaziantep MH152615 MH152646
M. anatolicum Cowley & Özhatay 2 Adana MH152616 MH152647
M. tuzgoluensis Yıld. 2 Konya MH152617 MH152648
M. discolor Boiss. & Hausskn. ex Boiss 2 Mardin MH152618 MH152649
M. turcicum Uysal, Ertugrul & Dural 2 Konya MH152619 MH152650
M. inconstrictum Rech. f. 2 Hatay MH152620 MH152651
M. latifolium J. Kirk 2 Çanakkale MH152621 MH152652
M. adilii M.B.Güner & H. Duman 2 Ankara MH152622 MH152653
M. bourgaei Baker 2 Bursa MH152623 MH152654
M. sandrasicum Karlen 2 Muğla MH152624 MH152655
M. microstomum P.H.Davis & D.C.Stuart 2 Kayseri MH152625 MH152656
M. azureum Fenzl 2 Sivas MH152626 MH152657
M. coeleste Fomin 2 Erzincan MH152627 MH152658
M. macbeathianum Kit Tan 2 Adana MH152628 MH152659
M. vuralii Bağcı&Doğu 2 Karaman MH152629 MH152660
M. parviflorum Desf. 2 Mersin MH152630 MH152661
M. serpentinicum Yıldırım, Altıoğlu & Pirhan 2 Muğla MH152631 MH152662
Number of sample, location, and accession number for matK and trnT-L-F cpDNA regions
Biologia
Author's personal copy

which was then analyzed to generate the species tree.
Furthermore, phylogenies were inferred separately from each
studied region (Supplementary Material). The samples
downloaded from NCBI could only be added when regions
were analyzed separately because whole sequences of studied
regions could not be found for a retrieved species. As
outgroup, complete chloroplast DNA sequence of Yucca
brevifolia (NC_032711) was selected to be able to find se-
quence of each studied region from one sample. Borders of
the matK, trnL intron, and trnL-F IGS regions were decided
based on sequences [HM640628 (M. aucheri)formatK,
AJ508003 (M. parviflorum) for trnL intron, and FJ423215
(M. comosum), AJ232668 (M. botryoides), AJ232667 (M.
macrocarpum) for trnL-F IGS] retrieved from NCBI database.
The border of trnT-L IGS region was decided using a species
(EU092532; Yucca schottii)fromsamefamily
(Asparagaceae) due to the absence of close relative of same
genus.
Total nucleotide length (base pair, bp), GC content (%),
number of deletion/insertion, parsimony informative
(variable) sites were calculated using Molecular
Evolutionary Genetics Analysis software (MEGA 7.0;
Kumar et al. 2016). The phylogenetic relationships of the
species were analyzed via MEGA and MrBayes v.3.1.2
(Huelsenbeck and Ronquist 2003; Ronquist and
Huelsenbeck 2003) analyses. All positions containing gaps
and missing data were eliminated. The sequence data was
analyzed by using the Maximum Likelihood (ML) method
based on the T92: Tamura 3-parameter model (Tamura
1992) and bootstrap analysis with 1000 replications
(Felsenstein 1985) in the MEGA analysis. To find best-fit
substitution model, the concatenated data was analyzed in
the MEGA 7.0 and the model with the lowest BIC score
(Bayesian Information Criterion) was selected. To obtain phy-
logenetic ML tree, the tree with the highest log likelihood was
selected. Initial tree(s) for the heuristic search were obtained
automatically by applying Neighbor-Join and BioNJ algo-
rithms to a matrix of pairwise distances estimated using the
Maximum Composite Likelihood (MCL) approach, and then
the topology with superior log likelihood value was selected.
The bootstrap values lower than 50% were not given in the
phylogenetic trees. The concatenated dataset was also ana-
lyzed using Bayesian inference (BI) analysis. Markov chain
Monte Carlo (MCMC) analyses were run for 1,000,000 gen-
erations, saving one tree each 1000 generations. A conserva-
tive burn-in (25%) was applied after checking for stability on
the log-likelihood curves and split variances being <0.01. A
majority rule consensus tree of the remaining trees was calcu-
lated. Branch support was determined by Bayesian Posterior
Probabilities (BPP).
Results
Successful DNA amplifications and sequences were obtained
from each specimen of the species. Alignment of the matK
gene sequences was straightforward while non-coding regions
required the inclusion of gaps with different lengths.
Ambiguous parts at the both end of sequences were trimmed
prior to analyses. The sequence length of the each region
including previously published sequences retrieved from
NCBI was as follows; matK: 1.238 bp, trnT-L: 695 bp, trnL
intron: 560, trnL-F: 354 bp while concatenated phylogenetic
data length was shorter (2.840 bp) than the expected total
length due to missing NCBI sequences in the analyzed data
(Table 2).
The matK region consisted of 70 accessions of 31 native
Muscari species and 3 accessions of 3 different species re-
trieved from database. Total length of the region (with species
obtained from NCBI) was calculated as 1238 bp with 48 poly-
morphic sites. Mean divergence was calculated as 0.008 and
the highest value (0.019) was observed between sequences of
M. macrocarpum and M. aucheri* (HM640628, * indicates
the sample taken from NCBI). GC content of the region was
calculated as 30.3% (Table 2). The matrix obtained from the
trnT-L alignment was composed of 70 accessions of 31 native
Table 2 Estimated molecular
diversity parameters for matK,
trnT-L IGS, trnL intron, trnL-F
IGS regions and concatenated
sequences
matK trnT-L IGS trnL intron trnL-F IGS Concatenated sequences
Number of taxa 31 31 32 32 31
Number of sequences 73 70 75 74 70
Total length (bp) 1.238 695 560 354 2.840
Variable sites 48 40 20 25 118
P. informative sites 48 40 20 25 118
Number of indels (bp) –92 31 54 177
G/C content (%) 30.3 29.2 34.3 31.5 31.0
Mean Distance (Divergence) 0.008 0.008 0.003 0.008 0.007
All the values were estimated with the previously published sequences retrieved from NCBI database
Biologia
Author's personal copy

Muscari species (no sequence retrieved from NCBI). The
length of the region ranged from 607 bp
(M. sivrihisardaghlarensis) to 637 bp (M. racemosum)with
a GC content from 27.8% (M. serpentinicum) to 30%
(M. mirum). Forty polymorphic sites, all parsimony informa-
tive, were detected in 695 bp aligned sequence (Table 2). In
the aligned sequence about 92 bp indel positions were detect-
ed. Most of them were composed of very big deleted areas
such as 30 bp length. Mean divergence was calculated as
0.008 and most of the species pairs showed no genetic
divergence.
The trnL intron region comprised of 70 accessions of 31
native Muscari species and 5 accessions of 4 different species.
Aligned sequences (with species obtained from NCBI)
yielded 560 characters; M. tenuiflorum showed the shortest
length (543 bp) while M. sivrihisardaghlarensis had the lon-
gest sequence. Twenty polymorphic sites were detected and
average GC content was calculated as 34.3%. In the aligned
sequence, there were about 31 bp indel positions (Table 2).
Mean divergence was calculated as 0.003 and the highest
value (0.011) was observed between sequences of
M. turcicum,M. bourgaei and M. comosum* (AJ232546).
The trnL-F IGS region was composed of 70 accessions of
31 native Muscari species and 4 accessions of 3 different
species. Total length of the region (with species obtained from
NCBI) was calculated as 354 bp with 25 polymorphic sites.
Four species (M. inconstrictum, M. parviflorum, M. azureum,
M. coeleste) were shorter due to big deletion (53 bp length) in
their sequences. Mean divergence was calculated as 0.008 and
the highest value (0.027) was observed between sequences of
M. erdalii,M. babachii, M. longipes and M. inconstricum.
Finally, GC content of the region was calculated as 31.5%.
The concatenated data included 70 accessions of 31 native
Muscari species (no sequence retrieved from NCBI). Total
length of the region was 2.840 bp with 118 polymorphic sites.
Mean divergence was calculated as 0.007 and GC content of
the region was 31% (Table 2).
The length of the regions was variable among species due
to mononucleotide (poly-A/T motif) and dinucleotide (TA
motif) repeats. In the sequence of trnT-L IGS region,
BTTATATATAT TTATA^repeats located between 116th –
202nd bases and BA^repeats located between 366th –376th
bases caused intricate indels among samples. A similar situa-
tion was also observed in the sequence of trnL intron region;
polyT bases observed between 102nd –117th bases caused
intricate indels. Therefore, indels were not included in the
analyses when the phylogenetic trees were constructed.
Indels were not observed among sequences of matK. No ge-
netic variation among the sequence of different individuals
from the same species was observed regardless of the region
analyzed. Therefore, only one representative sample for each
studied species was used while constructing phylogenetic
trees.
The ML and BI analyses based on concatenated dataset
uncovered almost the same topologies and relationships
among studied Muscari species. The vast majority of nodes
in both phylogenies achieved moderate to strong statistical
support (Fig. 1; S(supplementary)1). The concatenated and
matK data gave almost same tree topologies with the excep-
tion of higher bootstrap values in the concatenated one [Fig. 1
and S2)]. The evolutionary relationship between the species of
subgenera Botryanthus and Leopoldia was complicated. The
nested position of subgenus Leopoldia within Botryanthus
suggested that the latter one was paraphyletic. This relation
was clearly observed in both concatenated sequence and matK
trees; clade A coincided most of species included in
Botryanthus subgenus and all species of Leopoldia (Clade
B) except M. mirum clustered with %98 bootstrap value and
connected to the Clade A (Fig. 1and S2). Close relation
among M. bourgaei, M. latifolium and M. turcicum was ob-
served in each studied region except trnL-F and concatenated
data (Fig. 1,S2-S4). Muscari neglectum, M. tuzgoluensis, M.
adilii clustered together in the all trees but trnL intron (Fig. 1,
S2,S3,S5). Muscari azureum, M. coeleste, M. parviflorum
and M. incostrictum phylogenetically separated from relative
species of Botryanthus subgenus and grouped distantly (Clade
C, Fig. 1,S2,S3). Especially phylogenetic separation of both
M. azureum and M. coeleste species was one of the most
interesting and unexpected separations observed in all trees.
No genetic variation between M. serpentinicum and
M. sandrasicum was identified no matter which region was
analyzed, and they grouped together without any genetic dis-
tance (Fig. 1,S1-S5). Muscari erdalii,M. longipes and
M. babachii phylogenetically separated from the other species
of Leopoldia subgenus (Clade B) in all trees except trnL intron
tree (Figure S4). The other interesting phylogenetic position
was observed at M. mirum that is represented as endangered in
the Red Data Book of Turkish Plant (Ekim et al. 2000).
Although this species is a member of Leopoldia, it always
located distantly to its close relatives in all constructed trees
(Fig. 1,S1-S5). Finally, M. racemosum and M. macrocarpum
found in Muscari subgenus separated and caused Clade D in
the most of trees (Fig. 1,S2,S3). Muscari aucheri,
M. armeniacum and M. neglectum clustered with the repre-
sentatives (HM640628, MF349951 and JX090381, respec-
tively) in the matK tree (Figure S2). However, close relations
between studied and representative samples downloaded from
database were not observed in the trnL intron tree carrying low
nucleotide variation (Figure S4).
Although, matK and concatenated data gave same and
reliable resolutions among species, the genetic variations
of non-coding regions were not sufficient to figure out
taxonomical relations of subgenera. The most informative
non-coding region was trnT-L, which carried more varia-
tions. Paraphyletic structure of the Botryanthus subgenus
observed in matK and concatenated data was also proved
Biologia
Author's personal copy

by trnT-L tree (Figure S3). The distant location of
M. mirum from Leopoldia species and separation of
M. azureum and M. coeleste (Clade C) were also observed
in the trnT-L tree. However, Leopoldia subgenus could
not be phylogenetically separated from Botryanthus so
that the Clade including species of both subgenera named
asCladeA+Binthetree(FigureS3). In this clade, sev-
eral clusters with different sizes were observed. For in-
stance, M. neglectum, M. tuzgoluensis, M. adilii
(Botryanthus) grouped without genetic distance. Muscari
erdalii,M. longipes and M. babachii (Leopoldia) clus-
tered together. Phylogenetically distant location of
M. racemosum and M. macrocarpum (Muscari) was also
observed in the trnT-L tree as observed in the matK and
concatenated trees. The resolutions of trnL intron and
trnL-F regions were not proper to see these relations with-
in the genus. Therefore, most of mentioned relationships
abovewerenotobservedinthesetrees(FigureS4and
S5). However, clustering of M. serpentinicum /
M. sandrasicum, M. azureum / M. coeleste and distant
location of M. mirum were identified.
Discussion
Phylogeny and taxonomy
This study demonstrated that both concatenated and matK
sequences gave almost same results and resolved phylogenetic
relationships at the subgenus and species levels. In Flora of
Turkey (Vol. 8), the genus is divided into three subgenera as
Muscari, Leopoldia and Botryanthus (Davis and Stuart 1984)
and species of Pseudomuscari subgenus (M. azureum and
M. coeleste) are included in Botryanthus subgenus.In the
trees, these two species phylogenetically separated from all
other species of Botryanthus with high bootstrap values.
Therefore, moving of M. azureum and M. coeleste from
Botryanthus to Pseudomuscari subgenus is suggested based
on the results of the current study. Thus, we propose that the
number of subgenera of Muscari should be increased from
three to four in Flora of Turkey. Subgenus Pseudomuscari
and its species are also morphologically separated from the
other subgenera by the absence of sterile flowers (if they exist,
smaller and fewer than fertile), subcampanulate or campanu-
late actinomorphic flowers with sky-blue color and deep blue
Fig. 1 Phylogenetic tree for
representatives of the genus
Muscari constructed on the basis
of comparison of the sequences of
concatenated regions (matK +
trnT-L IGS + trnL intron + trnL-F
IGS) by the ML method. The
numbers above branches refer to
the bootstrap support and the
values lower than 50% were not
given. Triangle: Muscari
subgenus, Circle: Leopoldia
subgenus, Square: Botryanthus
subgenus
Biologia
Author's personal copy

fascia running down tube into lobes. Pseudomuscari may
show similarity with Muscari subgenus based on the absence
of sterile flower but it can be differentiated easily by the lack-
ing of corona (floral tube) (Garbari and Greuter 1970;Jafari
and Maassoumi 2011).
One of the most interesting phylogenetic positions was
observed between M. inconstrictum and M. parviflorum spe-
cies. These two species located distantly to the relative species
of same subgenus (Botryanthus) even though they are not
morphologically similar. In the trees, these two species located
closely with M. coeleste and M. azureum, which are now
considered as the members of Pseudomuscari subgenus.
There are lots of morphological differences among these four
species when structures of flowers or leaves are taken into
consideration. For instance, M. azureum and M. coeleste have
campanulate flowers while M. parviflorum and
M. inconstrictum carry urceolate and tubular flowers, respec-
tively. They only show similarities with the absence (if they
exist, few and small) of sterile flowers in the inflorescence.
Muscari parviflorum is the only Muscari species that
flowering in autumn while M. inconstrictum gives the first
flowers in early spring (February). Flowers of
M. parviflorum are pale blue-violet and urceolate, separating
at the mouth into individual flared petals or small teeth-like
flaps. Muscari inconstrictum is characterized by open tubular
flowers with dark violet-blue to blackish-blue color. The only
similaritybetween these two species is the narrow canaliculate
leaves but they are shorter than scape in M. parviflorum while
longer in M. inconstrictum. In spite of such morphological
differences, close relationship is an unexpected situation and
needs further studies for phylogenetic and taxonomic
resolutions.
No genetic difference was observed among
M. tuzgoluensis, M. adilii and M. neglectum species of
Botryanthus subgenus and they caused a small cluster in all
of the trees but trnL intron (99% bootstrap value in the
concatenated tree). Muscari tuzgoluensis and M. adilii are
not morphologically similar but they live under similar intense
stress condition (saline/alkaline soil) in terms of habitats.
Their abilities to survive in the extreme environment may be
originated from the same evolutionary adaptation. The lack of
genetic distinction among these morphologically distinct taxa
might indicate their young age and rapid diversification ac-
companied by environmental adaptation. Muscari
tuzgoluensis has similar morphological characters with
M. neglectum but it can be separated from M. neglectum by
the narrowerleaves, lax raceme, and erect (not recurved) lobes
of fertile flowers (Yıldırımlı2011).
Leopoldia subgenus is phylogenetically separated from the
other studied subgenera in the concatenated and matK tree
(98% and 97% bootstrap values, respectively) but its position
was not clearly recognized in the trees constructed based on
non-coding regions. Leopoldia species were paraphyletic and
form two sister clades (Clade C and D, Fig. 1), with the clade
C appeared in sister position to M. mirum. Leopoldia may be
distinguished by using several morphological characters.
Fertile flowers are oblong-urceolate or tubular and strongly
constricted at tip. They are usually brownish, dirty yellowish
or greenish and lacking a corona. Especially, the prominent
bluish to violet sterile flowers as well as the more or less
zygomorphic fertile flowers with apophyses on the shoulders
were used to separate Leopoldia subgenus (Jafari and
Maassoumi 2011). In the trees, M. erdalii, M. babachii and
M. longipes separated from close relatives of the same subge-
nus with 99% bootstrap and grouped without almost any ge-
netic divergence. Demirci and his colleagues (Demirci et al.
2013) indicated that M. erdalii is closely related to
M. tenuiflorum and M. babachii, but it is separated from them
based on several morphological characters such as shorter
scape, scabrid leaf margins, glaucous greenish ivory fertile
flowers and larger capsule. They also claimed that all studied
samples carried same diploid chromosome number but
M. erdalii was different since it includes 4 submetacentric
and 4 subtelocentric chromosome pairs. In the present study,
no genetic variation was observed between the DNA se-
quences of M. erdalii and M. babachii but they were phylo-
genetically separated from M. tenuiflorum in all trees except
trnL intron. Actually, the Leopoldia subgenus may be divided
into two main groups according to the color of flower lobes.
M. erdalii, M. babachii, M. longipes, M. tenuiflorum, and
M. massayanum have blackish flower lobes color. Muscari
erdalii, M. babachii, M. longipes and M. tenuiflorum separat-
ed from M. massayanum by longer pedicel of fertile flowers,
lax raceme, pinkish to violaceous color of sterile flowers, and
smaller capsules. The pedicels of fertile flower of
M. massayanum are clearly shorter, sterile flowers color are
pink, raceme is dense, and capsules are at least twice as large
(Demirci et al. 2013). Therefore, phylogenetic separation of
M. massayanum from M. erdalii, M. babachii and M. longipes
may be expected. When all these morphological features were
considered, we also anticipated M. tenuiflorum to be in the
same cluster with these three species. The distant position of
M. tenuiflorum remains as a questionable part of the present
study.
Even though M. mirum is found in Leopoldia subgenus,
its position was distant to its relative species in all trees. A
similar situation is also observed when only morphological
characters are taken into account. This species has yellow-
ish lobes like M. comosum, M. caucasicum, M. weisii.
However, M. mirum can be differentiated from species of
Leopoldia by quite short plant length compared to relatives,
one or sometimes two relatively wider leaves, mostly flat-
tened and always longer than the sum of the scape and
inflorescence. Moreover, the fruit of M. mirum is larger
and has a shorter pedicel than its close relatives. This spe-
cies grows in serpentine soil and phylogenetic separation
Biologia
Author's personal copy

maybecausedbyspeciation,whichiscommoninserpen-
tine habitats (Anacker et al. 2011; Anacker 2014).
Muscari racemosum and M. macrocarpum, which are
found in Muscari subgenus, were always phylogenetically
separated and caused a different cluster in the trees. This tax-
onomic position is expected since their morphologies are also
very different from those of other studied species. For in-
stance, flowers are very distinct by carrying a corona structure
at the apex and this structure is not observed at any flowers of
the other species. Presence of this structure is an important
character to identify Muscari subgenus. Additionally, these
species carry almost no or weakly developed sterile flower
and by this feature show similarities with species of
Pseudomuscari subgenus. However, color of flower is
sky-blue within Pseudomuscari subgenus whereas it is
yellow or greyish-white within Muscari subgenus
(Garbari and Greuter 1970).
Gürsoy and Şık(2010) tried to distinguish M. armeniacum
and M. neglectum distributing in west Anatolia based on ana-
tomical structures and indicated high amount of similarities
except leaf structures. In the current study, these two species
phylogenetically separated not matter what region was ana-
lyzed. They can be differentiated based on morphological
characters as well. For instance, M. neglectum has flowers that
are sharply constricted distally, very dark to blackish (navy)
blue, and leafs that are strongly canaliculate to pseudo-terete.
Muscari armeniacum has flowers that are sky-blue or violet-
tinged, and leafs that are linear to weakly canaliculated.
Furthermore, inflorescence structure of M. neglectum is al-
most cylindrical with sterile flowers covering mostly one-
third of inflorescence while M. armeniacum carries conical
inflorescence with sterile flowers covering only tip of flower.
Uysal et al. (2007)studiedonM. turcicum by using mor-
phological characters and found close relations with
M. discolor,M. anatolicum and M. macbeathianum.
Muscari turcicum differs from M. discolor by its smaller
flowers, shorter scape, white and tubular-urceolate fertile
flowers, and from M. anatolicum by the presence of bulblets,
smaller bulb and linear-lanceolate leaves. M. turcicum and
M. macbeathianum have white flowers, but M. turcicum is
separated by its subuniseriate stamens and the presence of
sterile flowers. In the current study, M. turcicum was phylo-
genetically differentiated from closely located M. discolor, M.
anatolicum and M. macbeathianum in the concatenated, matK
and trnT-L trees.
Pirhan et al. (2014) described morphological similarities
between M. serpentinicum and M. latifolium. They separated
M. serpentinicum from M. latifolium by shorter scape, narrow-
ly linear-lanceolate leaves, less dense flowered raceme,
strongly papillate seed surface. These two species were also
phylogenetically distinguished from each other when nucleo-
tide variations of concatenated, matK and trnT-L regions were
analyzed. Close relationship between M. serpentinicum and
M. sandrasicum was observed in the phylogenetic trees.
However, these two species are morphologically very differ-
ent according to Pirhan et al. (2014) even though they grow in
serpentine soil. This type of habitat has fewer nutrients in the
soil and high level of heavy metals such as nickel, chrome,
cobalt, and iron. Therefore, serpentine soil gives rise to edaph-
ic stress during plant growth. Diversification and speciation of
this couple of species (M. serpentinicum and M. sandrasicum)
is most probably linked with serpentine bedrock.
The resolutions provided by trnL intron and trnL-F
genetic markers
The sequences of trnL intron region were almost identical
across the species even they found in different subgenus.
This low variation may reflect the presence of conserved ele-
ments within the noncoding regions such promoter or regula-
tory motifs in intergenic spacers, or conserved secondary
structures in introns (Small et al. 2005). Because of the nature
of trnL intron (a group I intron), this region consists of DNA
sequences that is evolutionary well-preserved due to its sec-
ondary structure (nine stem-loop structures; P1-P9) and its
catalytic properties (Bogler and Francisco-Ortega 2004;
Taberlet et al. 2007; Turktas et al. 2012). Although this region
alone was not sufficient to understand phylogeny of Muscari,
concatenation with other regions may be beneficial to get the
synergic effect and higher resolution.
The trnL-F region had more nucleotide variations than trnL
intron even though the length of the region is shorter.
However, amount of variation was not enough to determine
phylogeny of Muscari. Low variation of this region was also
indicated by Heed (2010) for Muscari genus. He used the
trnL-F region of the maternally inherited chloroplast genome,
ITS and CytP450 regions from the nuclear genome in order to
explore the relationships between M. armeniacum and
M. polyanthum. Values of genetic variations differed very
much among the studied regions; chloroplast data included
very small diversity compared to ITS and CytP450 data.
This result may be expected because of low mutation rate in
chloroplast genes (Yang et al. 2017).
In this particular example, the cladistic analysis results in
significant progress to figure out evolutionary and taxonomic
relationships among subgenera and species within the genus
Muscari. The study is valuable to figure out taxonomic struc-
ture of the genus because there is not a comprehensive phylo-
genetic study to indicate relationships of Muscari species ex-
plicitly. Muscari azureum and M. coeleste, which are within
Botryanthus subgenus in Flora of Turkey, ought to be moved
to Pseudomuscari subgenus when both molecular and mor-
phological data are considered. Thus, the number of
subgenera of Muscari should be increased from three to four
in the Turkish Flora. Controversial position of few species
such as M. mirum,M. tenuiflorum needs further studies that
Biologia
Author's personal copy

include more taxa and numbers of loci to evaluate the situation
in greater detail.
Acknowledgements This research was supported by The Scientific and
Technological Research Council of Turkey (TUBİTAK; grant no.
114Z736). We sincerely thank Banonymous^reviewers and the managing
editor, Katarina Hegedusova, for their constructive criticisms that im-
proved the manuscript.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Appendix 1. Species of which sequences were
retrieved from NCBI
Region matK: M. aucheri (HM640628), M. armeniacum
(MF349951), M. neglectum (JX090381).
Region trnL intron: Muscari botryoides (AJ232545),
M. macrocarpum (AJ232544), M. comosum (AJ232546,
FJ423215), M. parviflorum (AJ508003).
Region trnL-F IGS: Muscari comosum (FJ423215,
AJ232669), M. botryoides (AJ232668), M. macrocarpum
(AJ232667)
References
Anacker BL (2014) The nature of serpentine endemism. Am J Bot
101(2):219–224. https://doi.org/10.3732/ajb.1300349
Anacker BL, Whittall JB, Goldberg EE, Harrison SP (2011) Origins and
consequences of serpentine endemism in the California flora.
Evolution 65(2):365–376. https://doi.org/10.1111/j.1558-5646.
2010.01114.x
Bogler D, Francisco-Ortega J (2004) Molecular systematic studies in
cycads: evidence from trnL intron and ITS2 rDNA sequences. Bot
Rev 70:260–273. https://doi.org/10.1663/0006-
8101(2004)070[0260:MSSICE]2.0.CO;2
Böhnert T, Lobin W (2017) Leopoldia neumannii sp. nov.
(Asparagaceae,Scilloideae): a new species of Muscari sensu lato
from Greece. Willdenowia 47:179–185. https://doi.org/10.3372/wi.
47.47210
Borzatti Von Loewenstern A, Giordani T, Astuti G, Andreucci A, Peruzzi
L (2013) Phylogenetic relationships of Italian Bellevalia species
(Asparagaceae), inferred from morphology, karyology and molec-
ular systematics. Plant Biosyst 147:776–787. https://doi.org/10.
1080/11263504.2013.829884
Buerki S, Jose S, Yadav SR, Goldblatt P, Manning JC, Forest F (2012)
Contrasting biogeographic and diversification patterns in two
Mediterranean-type ecosystems. PLoS One 7(6):e39377. https://
doi.org/10.1371/journal.pone.0039377
Clegg MT, Zurawski G (1991) Chloroplast DNA and the study of plant
phylogeny. In: Soltis PS, Soltis DE, Doyle JJ (eds) Molecular sys-
tematics of plants. Chapman and Hall, New York, pp 1–13
Cullen J, Knees SG, Cubey SH, Shaw JMH (2011) The European Garden
Flora - flowering plants: a manual for the identification of plants
cultivated in Europe both out-doors and under glass, 2nd Edn.
Cambridge University Press, New York, pp 124–125
Czerepanov SK (1995) Vascular plants of Russia and adjacent states (the
former USSR). Cambridge University Press, Cambridge, New York
Davis PH, Stuart DC (1980) Muscari Mill. In: Tutin TG, Heywood VH,
Burges NA, Moore DM, Valentine DH, Walters SM, Webb DA
(eds) Flora Europaea 5. Cambridge University Press, Cambridge,
pp 46–49
Davis PH, Stuart DC (1984) Muscari Mill. In: Davis PH (ed) Flora of
Turkey and the East Aegean Islands 8. Edinburgh University Press,
Edinburgh, pp 245–263
Davis PH, Mill RR, Kit T (1988) Flora of Turkey and the East Aegean
Islands, vol 10. Edinburgh University Press, Edinburgh
Demirci S, Özhatay N, Koçyiğit M (2013) Muscari erdalii
(Asparagaceae, Scilloideae), a new species from southern Turkey.
Phytotaxa 154:38–46. https://doi.org/10.11646/phytotaxa.154.1.2
Doyle JJ, Doyle JL (1987) A rapid DNA procedure for small quantities of
fresh leaf tissue. Phytochem Bull 19:11–15 http://ci.nii.ac.jp/naid/
10021087108/. Accessed 15 Aug 2018
Ekim T, Koyuncu M, Vural M, Duman H, Aytaç Z, Adıgüzel N (2000)
Pteriophyta and Spermatophyta. Red Data Book of Turkish plants.
Barışcan Ofset, Ankara
Felsenstein J (1985) Confidence limits on phylogenies: an approach using
the bootstrap. Evolution 39:783–791. https://doi.org/10.1111/j.
1558-5646.1985.tb00420.x
Garbari F (2003) Muscari neglectum Guss. e M. atlanticum Boiss. et
Reuter (Hyacinthaceae). Tipi, caratteristiche e considerazioni sulle
due specie. Inform Bot Ital 35(2):329–336 http://www.
societabotanicaitaliana.it/uploaded/60.pdf. Accessed 15 Aug 2018
Garbari F, Greuter W (1970) On the taxonomy and typification of
Muscari Miller (Liliaceae) and allied genera, and on the typification
of generic names. Taxon 19:329–334. https://doi.org/10.2307/
1219056
Gielly L, Taberlet P (1994) The use of chloroplast DNA to resolve plant
phylogenies: noncoding versus rbcL sequences. Mol Biol Evol 11:
769–777. https://doi.org/10.1093/oxfordjournals.molbev.a040157
Govaerts R (2018) World checklist of Asparagaceae. Facilitated by the
Royal Botanic Gardens, Kew. http://apps.kew.org/wcsp/ Accessed
15 August 2018
Gürsoy M, Şık L (2010) BatıAnadolu’daki Muscari armeniacum
Leichtlin ex Baker ve Muscari neglegtum Guss. türleri üzerinde
karşılaştırmalıanatomik araştırmalar. Celal Bayar Üniversitesi Fen
Bilimleri Dergisi 6(1):61–72 http://dergipark.gov.tr/cbayarfbe/issue/
4049/53375. Accessed 15 Aug 2018
Heed Y (2010) Muscari armeniacum and M. polyanthum - one or two
species. Dissertation, Department of Plant and Environmental
Sciences University of Gothenburg
Herrmann H, Weiss G, Durka W (2005) Biological flora of Central
Europe: Muscari tenuiflorum Tausch. Flora 201:81–101. https://
doi.org/10.1016/j.flora.2005.03.002
Hilu KW, Liang H (1997) The matK gene: sequence variation and appli-
cation in plant systematics. Am J Bot 84:830–839. https://doi.org/
10.2307/2445819
Huelsenbeck JP, Ronquist F (2003) MrBayes: Bayesian inference of phy-
logeny. Bioinformatics 17:754–755. https://doi.org/10.1093/
bioinformatics/17.8.754
Jafari A, Maassoumi AA (2011) Synopsis of Leopoldia, Muscari and
Pseudomuscari (Hyacithaceae)in Iran, with Leopoldia
ghouschtchiensis sp. nova. Ann Bot Fen 48:396–400. https://doi.
org/10.5735/085.048.0502
Jafari A, Maassoumi AA, Farsi M (2008) Karyological study on
Bellevalia and Muscari (Liliaceae) species of Iran. Asian J Plant
Sci 7:50–59. https://doi.org/10.3923/ajps.2008.50.59
Johnson M, Özhatay N, Garbari F (1996) The genus Muscari
(Hyacinthaceae) in Turkey: taxonomy, distribution and chromo-
some analysis. In: Öztürk MA et al (eds) Plant life In Southwest
and Central Asia. Ege Univ. Press, İzmir, pp 34–53
Biologia
Author's personal copy

Karlén T (1991) Muscari Miller. In: Strid A, Tan K (eds) Mountain Flora
of Greece, vol 2. Edinburgh Univ. Press, Edinburgh, pp 697–701
Kaya E (2014) Muscari Mill. In: Türkiye Geofitleri, vol. 2. Atatürk Bahçe
Kültürleri Merkez Araştırma Enstitüsü, Yayın No: 96, Yalova,
Turkiye, pp 350–411
Kumar S, Stecher G, Tamura K (2016) MEGA7: Molecular evolutionary
genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 33:
1870–1874. https://doi.org/10.1093/molbev/msw054
Li JH, Bogle AL, Klein AS (1997) Interspecific relationships and genetic
divergence of the disjunct genus Liquidambar (Hamamelidaceae).
Rhodora 99:229–241 https://www.jstor.org/stable/23313320.
Accessed 15 Aug 2018
Martínez-Azorín M, Crespo MB, Juan A, Fay MF (2011) Molecular
phylogenetics of subfamily Ornithogaloideae (Hyacinthaceae)
based on nuclear and plastid DNA regions, including a new taxo-
nomic arrangement. Ann Bot 107:1–37. https://doi.org/10.1093/
aob/mcq207
Nersesian A (2001) A karyosystematic study of Armenian Muscari and
Bellevalia (Hyacinthaceae). Bocconea 13:383–389 http://147.163.
105.223/bocconea/13-383.pdf. Accessed 15 Aug 2018
Özhatay N (2000) Muscari Miller. In: Güner A, Özhatay N, Ekim T,
Başer KHC (eds) Flora of Turkey and the East Aegean Islands,
vol 11. Edinburgh University Press, Edinburgh, pp 237–240
Özhatay N, Kültür Ş(2006) Check-list of additional taxa to the supple-
ment Flora of Turkey III. Turk J Bot 30:281–316 http://journals.
tubitak.gov.tr/botany/abstract.htm?id=8420. Accessed 15 Aug 2018
Özhatay N, Kültür Ş, Aslan S (2009) Check-list of additional taxa to the
supplement Flora of Turkey IV. Turk J Bot 33:191–226. https://doi.
org/10.3906/bot-0805-12
Özhatay FN, Kültür Ş,GürdalMB(2011)Check-Listofadditionaltaxa
to the supplement Flora of Turkey V. Turk J Bot 35:589–624. https://
doi.org/10.3906/bot-1101-20
Pfosser M, Speta F (1999) Phylogenetics of Hyacinthaceae based on
plastid DNA sequences. Ann Mo Bot Gard 86:852–875. https://
doi.org/10.2307/2666172
Pınar SM, Fidan M, Eroğlu H (2018) Muscari botryoides (L.) Mill.: a
new record for the family Asparagaceae from Turkey. Turk. J Agric
Res 5(2):116–119. https://doi.org/10.19159/tutad.368374
Pirhan AF, Yıldırım H, Altıoğlu Y (2014) Muscari serpentinicum sp.
nova (Asparagaceae): a new species from western Anatolia,
Turkey. Ot Sistematik Botanik Dergisi 21(1):1–14
Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic
inference under mixed models. Bioinformatics 19(12):1572–1574.
https://doi.org/10.1093/bioinformatics/btg180
Ruiz Rejón M, Oliver JL (1981) Genetic variability in Muscari comosum
(Liliaceae). I. A comparative analysis of chromosome polymor-
phisms in Spanish and Aegean populations. Heredity 47(3):403–
407. https://doi.org/10.1038/hdy.1981.101
Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, Siripun KC,
Winder CT, Schilling EE, Small RL (2005) The tortoise and the hare
II: relative utility of 21 noncoding chloroplast DNA sequences for
phylogenetic analysis. Am J Bot 92(1):142–166. https://doi.org/10.
3732/ajb.92.1.142
Shaw J, Lickey EB, Schilling EE, Small RL (2007) Comparison ofwhole
chloroplast genome sequences to choose noncoding regions for phy-
logenetic studies in angiosperms: the tortoise and the hare III. Am J
Bot 94(3):275–288. https://doi.org/10.3732/ajb.94.3.275
Small RL, Lickey EB, Shaw J, Hauk WD (2005) Amplification of non-
coding chloroplast DNA for phylogenetic studies in lycophytes and
monilophytes with a comparative example of relative phylogenetic
utility from Ophioglossaceae. Mol Phylogenet Evol 36(3):509–522.
https://doi.org/10.1016/j.ympev.2005.04.018
Speta F (1989) Muscari (subg. Leopoldia) mirum Speta, spec. nova, im
Kreise seiner nächsten Verwandten. Phyton (Austria) 29(1):105–117
Speta F (1998) Hyacinthaceae. In: Kubitzki K (ed) The families and
genera of vascular plants III. Springer, Berlin, pp 261–285
Suárez-Santiago VN, Salinas JS, Romero-García AT, Garrido-Ramos
MA,delaHerranR,Ruiz-RejónC,Ruiz-RejónM,BlancaG
(2007) Polyploidy, the major speciation mechanism in Muscari sub-
genus Botryanthus in the Iberian Peninsula. Taxon 56:1171–1184.
https://doi.org/10.2307/25065910
Taberlet P, Gielly L, Pautou G, Bouvet J (1991) Universal primers for
amplification of three non-coding regions of chloroplast DNA. Plant
MolBiol17:1105–1109. https://doi.org/10.1007/bf00037152
Taberlet P, Coissac E, Pompanon F, Gielly L, Miquel C, Valentini A,
Vermat T, Corthier G, Brochmann C, Willerslev E (2007) Power
and limitations of the chloroplast trnL (UAA) intron for plant
DNA barcoding. Nucleic Acids Res 35(3):e14. https://doi.org/10.
1093/nar/gkl938
Tamura K (1992) Estimation of the number of nucleotide substitutions
when there are strong transition-transversion and G + C content
biases. Mol Biol Evol 9:678–687. https://doi.org/10.1093/
oxfordjournals.molbev.a040752
Turktas M, Aslay M, Kaya EF (2012) Molecular characterization of phy-
logenetic relationships in Fritillaria species inferred from chloro-
plast trnL-trnF sequences. Turk J Biol 36:552–560. https://doi.org/
10.3906/biy-1201-30
Uysal T, Ertuğrul K, Dural H, Küçüködük M (2007) Muscari turcicum
(Liliaceae/Hyacinthaceae), a new species from South Anatolia. Bot J
Linn Soc 154:233–236. https://doi.org/10.1111/j.1095-8339.2007.
00646.x
Valdés B, Lifante ZD (1992) Karyology and reproductive biology of
Muscari matritensis M. Ruíz Rejón et al. (Liliaceae). Bot J Linn
Soc 109:125–130. https://doi.org/10.1111/j.1095-8339.1992.
tb00261.x
Yang J, Vázquez L, Chen X, Li H, Zhang H, Liu Z, Zhao G (2017)
Development of chloroplast and nuclear DNA markers for
Chinese Oaks (Quercus Subgenus Quercus) and assessment of their
utility as DNA barcodes. Front Plant Sci 8:816. https://doi.org/10.
3389/fpls.2017.00816
Yıldırım H (2015) Muscari atillae (Asparagaceae): a new species from
eastern Anatolia, Turkey. Phytotaxa 213(3):291–295. https://doi.
org/10.11646/phytotaxa.213.3.9
Yıldırım H (2016) Muscari elmasii sp. nova (Asparagaceae): a new spe-
cies from western Anatolia, Turkey. Turk J Bot 40:380–387. https://
doi.org/10.3906/bot-1507-17
YıldırımlıŞ(2011) Three new species from Turkey. Ot Sistematik
Botanik Dergisi 18(1):1–12
Yilmaz H, Yilmaz OY, Akyüz YF (2017) Determining the factors affect-
ing the distribution of Muscari latifolium, an endemic plant of
Turkey, and a mapping species distribution model. Ecol Evol 7(4):
1112–1124. https://doi.org/10.1002/ece3.2766
Biologia
Author's personal copy