The molecular investigations on the subgenus Artemisia Less. of the genus Artemisia L. (Asteraceae) in Turkey

Abstract The genus Artemisia L. of the family Asteraceae is systematically very complex. The aim of this study was to evaluate taxonomic positions of taxa of the subgenus Artemisia belonging to the genus Artemisia in Turkey using some molecular techniques. In this molecular study, 44 individuals belong to 14 species of the subgenus Artemisia were examined. Analyses were performed on the combined dataset using maximum parsimony, maximum likelihood and Bayesian inference and Molecular parameters obtained from co-evaluations of sequences of the psbA-trnH, ITS and ETS regions of examined individuals were used in the phylogenetic tree drawing. According to the results of this study, two molecular groups have been formed based on the DNA sequence similarity of the species, but there are no obvious morphological characters corresponding to two molecular groups. There is no also agreement between the two molecular groups and the two morphological groups formed according to the hairiness condition of the receptacle of species. Due to the lack of molecular significance of their receptacles with or without hair, dividing of the subgenus Artemisia species into new subgenera or sections was not considered appropriate. Likewise, it has been found that with or without hair on the corolla lobes of the central hermaphrodite disc flowers have no molecular significance. It was found that there were no gene flow and hybridization between the 14 species of the subgenus Artemisia and these 14 species were found completed their speciation. This study is important as it is the first molecular based study relating with belong to subgenus Artemisia species growing naturally in Turkey. In addition, new haplotypes related to the populations of Turkey belonging to the subgenus Artemisia taxa were reported by us for the first time and added to the GenBank database.


The molecular investigations on the subgenus Artemisia Less. of the genus Artemisia L. (Asteraceae) in Turkey
During the revisionary study of the genus Artemisia L. in Turkey, the subgenus Artemisia did not divide into new subgenera or sections based on the morphological (phenotypical) character of receptacles which are hairy or glabrous (Civelek et al., 2010;Kurşat, 2010).Is this systematic arrangement of the subgenus Artemisia in this revisionary study that made according to only morphological data correct?The biggest question in the minds of the researchers doing the revisionary study is about whether the dividing at new subgeneric and sectional levels of species belonging to the subgenus Artemisia which is the subject of this study.The final decision on whether the subgenus Artemisia should be divided into new subgenera or sections will be made by this molecular study.
In this study, it will be also tried to determine whether this phenotypic variation has a molecular (genotypic) level response.The purposes of this research are as follows: For the species belonging to the subgenus Artemisia, to determine whether there is a molecular significance of the receptacles with or without hairs and accordingly to determine whether the subgenus Artemisia can be divided into new subgenera or sections.To determine whether the hairy and hairless corolla lobes of the central hermaphrodite disc flowers have a molecular significance for species belonging to the subgenus Artemisia and to find whether each species has completed its speciation.Also, obtain original data on species of subgenus Artemisia for use in new scientific molecular studies of Science World, and to provide the haplotypes of psbA-trnH, ITS and ETS regions of species belonging to the subgenus Artemisia in Turkey for the GenBank database.

Plant material
Within the scope of this study, the plants that were gathered during the land survey conducted by M. Kurşat and S. Civelek were used.44 individuals belong to 14 species of the subgenus Artemisia were examined.Depending on the large of populations, between 1 and 3 individuals with the same label information were used for each species.Localities and detailed population information the investigated taxa were gathered, were given in Table 1.

DNA isolation and PCR amplification
For the DNA isolation, was used herbarium material (dried plant leaf).The CTAB method was used in a modified form and DNA isolation was performed manually (Doyle and Doyle, 1987).By measuring the concentrations of the isolated DNAs with a nanodrop spectrophotometer, they were adjusted to 25 ng/ul .The Stock DNA was preserved at -20 C.
In the PCR studies conducted by using some spesific universal primers, ITS, ETS and psbA-trnH region for 44 samples was multiplied.The sequence of primers that were used to amplify all of regions were given in Table 2 (Taberlet et al., 1991).Aiming to make the final concentration for the PCR studies to be 50 µL, by mixing

Introduction
Artemisia L. is one of the largest genera in the family Asteraceae and the largest genus in the tribe Anthemideae comprises from 200 to more than 500 taxa at the specific or infraspecific level (McArthur, 1979;Mabberley, 1990;Ling, 1991a, b;Bremer and Humpehries, 1993;Ling, 1991a, b;Torrell et al., 1999).Traditionally, the subgeneric classification of the genus Artemisia L. follows a system established by Besser (1829) where in he separated four sections based on various combinations of disc and ray flower occurrences and fertility.These sections are Abrotanum (Duhamel du Monceau) Besser, Absinthium (Miller) Candolle, Dracunculus Besser, and Seriphidium Besser.Besser's four sections have been modified by subsequent workers.Rydberg (1916) (1981).The subgenus Artemisia is included traditional two subgenera Abrotanum Besser and Absinthium (Miller) Lessing.The first detailed phylogenetic interpretation of the sections or subgenera established Artemisia as the more primitive, while Dracunculus, Seriphidium and Tridentatae were considered to be more advanced (Hall and Clements, 1923).Notwithstanding, a major in-depth reclassification of the genus Artemisia (combining molecular and traditional data) has not yet been proposed.
There are 22 species of the genus Artemisia in the 5 th volume of the Flora of Turkey (Cullen, 1975;Davis, 1975;Davis et al., 1988).Later, the species A. verlotiorum Lamotte was added to the 10 th volume of the Flora of Turkey as a new record for Turkey, so species numbers of the genus Artemisia in Turkey became 23 in total (Kurşat et al., 2011a).At the same time, 23 species without any infraspecific taxa that belong to the genus Artemisa in Flora of Turkey are not classified levels of subgenus or section.Civelek et al. (2010) have carried out a revisionary study of the genus Artemisia in Turkey.According to the results of this revisionary study, there are a total of 21 species with 9 infraspecific taxa and 3 subgenera belonging to the genus Artemisia in Turkey (Civelek et al., 2010;Kurşat, 2010;Kurşat et al., 2011aKurşat et al., , b, 2014Kurşat et al., , 2015Kurşat et al., , 2018)).According to the results of this revisionary study of the genus Artemisia, subgenera Artemisia, Dracunculus, and Seriphidium have taxa in Turkey, but the subgenus Tridentatae which is endemic to North America has not taxa in Turkey (Civelek et al., 2010;Kurşat, 2010;Guner et al., 2012).
The subgenus Artemisia has the most species in Turkey that includes 14 species.During in the revisionary study, the species A. alba of the subgenus Artemisia Less.has been removed from Turkey's species list because its distribution is outside of Turkey's borders (Civelek et al., 2010;Kurşat, 2010;Kurşat et al., 2011a;Guner et al., 2012).No conclusive consensus has yet been provided on the classification at subgeneric and sectional levels of species belonging to the subgenus Artemisia which is the subject of this study.
Brazilian Journal of Biology, 2023, vol.83, e252656 3/10 Phylogeny of the subgenus Artemisia (Asteraceae) Table 1.The label information of examined individuals of specimens belonging to species of the subgenus Artemisia used in the study and detailed localities of populations (Civelek et al., 2010;Kurşat et al., 2011b).

Sequence and phylogenetic analysis
A two-way reading was applied to the amplification products.The PCR purification process was realized before the sequence analysis.The purification and sequencing process was realized by the Macrogen Company.In order to evaluate the data of the chromatograms, a Finch TV 1.4 version was used.The variable sites, number of parsimony informative sites, genetic distance, nucleotide diversity, and divergence within the species were computed as molecular diversity statistics for each dataset using Molecular Evolutionary Genetics Analysis software (MEGA X) (Kumar et al., 2018).The DNA sequence alignments of the 44 individuals were evaluated by using the X version of the MEGA program.The DNA sequence alignments of all the individuals was subject to the statistical analysis within the scope of this program.Ultimately, the phylogenetic trees were constructed by using the Maximum Likelihood Method with 100 bootstrap replicates.The GenBank accession numbers, where the investigated taxa were gathered, are given in Table 3.

Results
In this research, 44 individuals belong to 14 species of the subgenus Artemisia were examined.Depending on the numbers of individuals in the herbarium specimens, between 1 and 3 individuals with the same label information were used for each specimen.The sequences of the cpDNA psbA-trnH, r-DNA ITS and ETS regions for all individuals examined, were used to determine phylogenetic kinship relationships.
The peak results of the bi-directional sequences sent from the Macrogen were evaluated using Version 1.4 of the Finch TV program.Using the "Multiple Alignment Blast System" of the automatic sequencing systems, sequences were aligned.The differences in the noticeable were manually corrected.
As a result of scans performed in the NCBI (National Center for Biotechnology Information) database site, for seven species of the subgenus Artemisia, three reference regions were obtained.These reference regions were an ITS with 700-750 bases (ITS1-5.8gene -ITS2), ETS with 400-450 bases and psbA-trnH with 450-500 bases.For a more accurate visualization of the results of the alignment, about 50-100 base from the head and the end was not evaluated by us.For this reason, approximately 684 base pairs for the ITS regions, 506 base pairs for the ETS regions Table 2.The base sequences of the primers used (Taberlet et al., 1991).

Primers
Base sequences (5' -3') Table 3. GenBank accession numbers for the r-DNA ITS, ETS and psbA-trnH regions of the studied samples.and 466 base pairs for the psbA -trnH regions were used and the Genbank accesion number of these regions were given in the Table 3.

ITS region psbA-trnH region ETS region
The DNA sequences of the ITS, ETS and the psbA-trnH regions of 44 individuals belong to subgenus Artemisia were analyzed statistically.In the both separate and co-evaluations of sequences of the ITS, ETS and psbA-trnH regions of examined individuals, some parameters of molecular diversity like conservated regions (C), variation regions (V), parsimony informative regions (Pi), single parts (S), homologous base pairs (ii), transitional base pairs (si), transversional base pairs (sv) and R value (si / sv) were calculated and the values obtained are given in Table 4.These parameters determine the distribution of individuals in the phylogenetic tree and thus gave us information about their phylogenetic relationships.
The nucleotide composition of the individuals was determined as a result of the statistical analyzes performed by cutting the excess parts at the head and end of the DNA sequences.When the values given in Table 4 are examined, it is seen that individuals are not very different in terms of the ratio of T, C, A, and G bases.At the same time, it was found that when the average values of the base contents of the individuals were calculated, the A-T ratio was 56% and the G-C ratio was 44%, and the A-T base pair was richer than G-C base pair.
In order to determine the phylogenetic relationships between individuals belonging to subgenus Artemisia species from both the same and different populations, in total 52 individuals were studied, 7 of them were control group and 1 of which was outgroup were arranged in Mega program.
Then, by using the Best DNA / Protein step in the Models menu of this program, we determined the methods that best express the phylogenetic relationship between individuals.In the list of methods given, the lowest value of BIC (Bayesian Information Criterion) was found in the T92 + G (Tamura-3-parameter) method.According to T92 + G, to see phylogenetic relationships between individuals, we have seen that phylogenetic trees can be plotted using any of the methods such as Maximum Parsimony, Neighbor-Joining, UPGMA, and Maximum Likelihood.
Maximum Likelihood, Neighbor-Joining, UPGMA, and Maximum Parsimony methods were applied separately, but it was decided that the method that best illustrates the evolutionary and phylogenetic relations among the examined individuals, we work with is the Maximum Likelihood method.In the phylogenetic tree drawing, the DNA sequences of ITS and ETS regions in the nuclear genome and psbA-trnH in the chloroplast genome were co-evaluated by using version 7.1 of the Mega program.In the Maximum Likelihood method, by entering the boostrap value 100, a single phylogenetic tree for a total of 52 individuals were obtained (Figure 1).

Discussions
There are 3 subgenera, 24 species with 5 infraspecific taxa belong to the genus Artemisia in Turkey.The subgenus Artemisia includes 14 species, the subgenus Dracunculus includes 4 species and the subgenus Seriphidium includes 6 species with 5 taxa that settled in the categories of subspecies and variety.Three species (A.santonicum, A. taurica, and A. sieberi) of the subgenus Seriphidum have five infraspecific taxa that settled in the categories of subspecies and variety (Table 5) (Civelek et al., 2010;Kurşat, 2010;Kurşat et al., 2011aKurşat et al., , b, 2014Kurşat et al., , 2015Kurşat et al., , 2018)).Although the subgenus Artemisia has been divided into new second subgenera or sections by some authors (Bremer and Humpehries, 1993;Schinskin and Bobrov, 1995;Ling et al., 2006;Shultz, 2006a, b), the subgenus Artemisia did not divide to new subgenera or sections during the revisionary study of the genus Artemisia in Turkey (Civelek et al., 2010;Kurşat, 2010).The most importanat of the purposes of this molecular study is to make the final decision whether the subgenus Artemisia should be divided into new subgenera or sections.During the revisionary study of the genus, Artemisia in Turkey, a new diagnostic character (a new phenotypic variation) had been determined for taxa belonging to subgenus Artemisia.This diagnostic character is that with or without hair on the corolla lobes of the central hermaphrodite disc flowers.Four species A. arborescens, A. armeniaca, A. austriaca, and A. incana of the subgenus Artemisia have hairs on the corolla lobes of central hermaphrodite disc florets.However, ten species A. vulgaris, A. verloitorum, A. abrotanum, A. chamaemelifolia, A. annua, A. tournefortiana, A. absinthium, A. splendens, A. alpina and .A. haussknechtii of the subgenus Artemisia have not hairs speciation and there is no hybridization between them (Civelek et al., 2010;Kurşat, 2010).These molecular study data confirmed this generalization obtained only from morphological data.
The fact that the individuals of the seven species used as the control group are only in the same branches with individuals belonging to their own species makes our study 100% reliable.
As a result of this research, it was obtained original data for use in new scientific molecular studies on the taxa of genus Artemisia and provided the haplotypes of ETS, ITS and psbA-trnH regions for 14 species of the subgenus Artemisia in Turkey for the GenBank database (Table 3).
This study reports for the first time, molecular phylogeny of subgenus Artemisia from Turkey by using r-DNA (ITS and ETS) and cpDNA (psbA-trnH) sequences.
Based on the current data and all available in literature, it is concluded that the morphological studies coupled with modern molecular techniques may lead to the clear infrageneric classification of the genus Artemisia.

Figure 1 .
Figure 1.Phylogenetic tree obtained from the co-evaluation of sequences of the psbA-trnH, ETS and ITS regions of individuals.
created subordinate sections including new section Tridentatae for the North American members of the section Seriphidium and elevated these sections to subgenera.Nowadays, the most commonly accepted subdivisions of the genus Artemisia are separated into 4 subgenera as Artemisia Lessing, Dracunculus (Besser) Rydberg, Seriphidium Besser ex Lessing and Tridentatae (Rydberg) McArthur et al.

Table 4 .
Molecular diversity parameters obtained from both separate and co-evaluations of sequences of the ITS, ETS and psbA-trnH regions of examined individuals.