Biologia 69/3: 300—310, 2014 Section Botany DOI: 10.2478/s11756-013-0314-z Genetic diversity and relationships among Egyptian Galium (Rubiaceae) and related species using ISSR and RAPD markers Kadry Abdel Khalik 1,2 *, Magdy Abd El-Twab 3 & Rasha Galal 3 1 Botany Department, Faculty of Science, Sohag University, Sohag 82524, Egypt; e-mail: kadry3000@yahoo.com 2 Biology Department, Faculty of Science, Umm-Al-Qura University, Saudi Arabia 3 Botany and Microbiology Department, Faculty of Science, Minia University, Egypt Abstract: Genetic diversity and phylogenetic analyses of 24 species, representing nine sections of the genus Galium (Rubi- aceae), have been made using the Inter Simple Sequence Repeats (ISSR), Randomly Amplified Polymorphic DNA (RAPD), and combined ISSR and RAPD markers. Four ISSR primers and three RAPD primers generated 250 polymorphic ampli- fied fragments. The results of this study showed that the level of genetic variation in Galium is relatively high. RAPD markers revealed a higher level of polymorphism (158 bands) than ISSR (92 bands). Clustering of genotypes within groups was not similar when RAPD and ISSR derived dendrograms were compared. Six clades can be recognized within Galium, which mostly corroborate, but also partly contradict, traditional groupings. UPGMA-based dendrogram showed a close relationship between members of section Leiogalium with G. verum and G. humifusum (sect. Galium), and G. angusti- folium (sect. Lophogalium). Principal coordinated analysis, however, showed some minor differences with UPGMA-based dendrograms. The more apomorphic groups of Galium form the section Leiogalium clade including the perennial sections Galium, Lophogalium, Jubogalium, Hylaea and Leptogalium as well as the annual section Kolgyda. The remaining taxa of Galium are monophyletic. Key words: Galium; genetic diversity; ISSR; RAPD; Rubiaceae Introduction Rubiaceae is the fourth-largest angiosperm family, com- prising approximately 660 genera and 11,500 species and classified into 42 tribes (Robbrecht & Manen 2006; Soza & Olmstead 2010a). Most of the family is trop- ical and woody. Rubieae is the only tribe centered in temperate regions, but has cosmopolitan distribution. Most of its members are herbaceous and adapted to xeric habitats (Robbrecht 1988; Jansen et al. 2000). Rubieae is a monophyletic group, sharing both mor- phological and molecular synapomorphies (Manen et al. 1994; Natali et al. 1995, 1996; Bremer 1996; Andersson & Rova 1999; Bremer & Manen 2000; Nie et al. 2005; Backlund et al. 2007; Bremer & Eriksson 2009). How- ever, classification and identification within Rubieae have been problematic, especially for the larger genera Asperula and Galium. A number of taxa within Aspe- rula appear morphologically similar to Galium, differing only in corolla tube length, and these have been trans- ferred from Asperula to Galium (Ehrendorfer 1958; Na- tali et al. 1995; Ehrendorfer et al. 2005; Abdel Khalik & Bakker 2007). Galium L. is one of the largest genera of Ru- bieae with more than 400 species included into 16 sec- tions containing annual and perennial herb that are distributed in temperate and tropical regions of the world (Willis 1985; Mabberley 1987). Galium itself is problematic taxonomically, because taxa from different sections exhibit similar habit, many species are widely distributed and polymorphic, and species groups often are poorly differentiated both morphologically and geo- graphically (Schischkin 2000). This genus was described by Linnaeus (1753) who established the occurrence of 26 species. He divided them into two groups according to fruit type (glabrous and hispid). Boissier (1881) con- sidered 90 species and divided them into three sections (Eugalium, Aparine and Cruciata) and 11 subsections. Ehrendorfer et al. (1976) recognized 145 species for Eu- ropean flora, classifying into 10 sections. Ehrendorfer & Sch¨ onbeck-Temesy (1982) listed for the flora of Turkey 101 species divided into 10 sections. In Egypt, Tackholm (1974) named 12 species of Galium, Boulos (1995, 2000) recognized only 10 species. Abdel Khalik et al. (2007; 2008a, b, c) studied 13 Egyp- tian taxa of Galium by different means such as mor- phological characters, including vegetative parts, flow- ers, fruits, seeds, pollen grains, anatomical structure. Numerical analysis was conducted, and they classified these species into groups. Molecular markers are useful in identifying the maximally diverse parental genotypes through an eval- * Corresponding author c 2013 Institute of Botany, Slovak Academy of Sciences
Genetic diversity and relationships among Egyptian Galium(Rubiaceae) and related species using ISSR and RAPD markers
Kadry Abdel Khalik1,2*, Magdy Abd El-Twab3 & Rasha Galal3
1Botany Department, Faculty of Science, Sohag University, Sohag 82524, Egypt; e-mail: [email protected] Department, Faculty of Science, Umm-Al-Qura University, Saudi Arabia3Botany and Microbiology Department, Faculty of Science, Minia University, Egypt
Abstract: Genetic diversity and phylogenetic analyses of 24 species, representing nine sections of the genus Galium (Rubi-aceae), have been made using the Inter Simple Sequence Repeats (ISSR), Randomly Amplified Polymorphic DNA (RAPD),and combined ISSR and RAPD markers. Four ISSR primers and three RAPD primers generated 250 polymorphic ampli-fied fragments. The results of this study showed that the level of genetic variation in Galium is relatively high. RAPDmarkers revealed a higher level of polymorphism (158 bands) than ISSR (92 bands). Clustering of genotypes within groupswas not similar when RAPD and ISSR derived dendrograms were compared. Six clades can be recognized within Galium,which mostly corroborate, but also partly contradict, traditional groupings. UPGMA-based dendrogram showed a closerelationship between members of section Leiogalium with G. verum and G. humifusum (sect. Galium), and G. angusti-folium (sect. Lophogalium). Principal coordinated analysis, however, showed some minor differences with UPGMA-baseddendrograms. The more apomorphic groups of Galium form the section Leiogalium clade including the perennial sectionsGalium, Lophogalium, Jubogalium, Hylaea and Leptogalium as well as the annual section Kolgyda. The remaining taxa ofGalium are monophyletic.
Rubiaceae is the fourth-largest angiosperm family, com-prising approximately 660 genera and 11,500 speciesand classified into 42 tribes (Robbrecht & Manen 2006;Soza & Olmstead 2010a). Most of the family is trop-ical and woody. Rubieae is the only tribe centered intemperate regions, but has cosmopolitan distribution.Most of its members are herbaceous and adapted toxeric habitats (Robbrecht 1988; Jansen et al. 2000).Rubieae is a monophyletic group, sharing both mor-phological and molecular synapomorphies (Manen et al.1994; Natali et al. 1995, 1996; Bremer 1996; Andersson& Rova 1999; Bremer & Manen 2000; Nie et al. 2005;Backlund et al. 2007; Bremer & Eriksson 2009). How-ever, classification and identification within Rubieaehave been problematic, especially for the larger generaAsperula and Galium. A number of taxa within Aspe-rula appear morphologically similar toGalium, differingonly in corolla tube length, and these have been trans-ferred from Asperula to Galium (Ehrendorfer 1958; Na-tali et al. 1995; Ehrendorfer et al. 2005; Abdel Khalik& Bakker 2007).Galium L. is one of the largest genera of Ru-
bieae with more than 400 species included into 16 sec-tions containing annual and perennial herb that are
distributed in temperate and tropical regions of theworld (Willis 1985; Mabberley 1987). Galium itself isproblematic taxonomically, because taxa from differentsections exhibit similar habit, many species are widelydistributed and polymorphic, and species groups oftenare poorly differentiated both morphologically and geo-graphically (Schischkin 2000). This genus was describedby Linnaeus (1753) who established the occurrence of26 species. He divided them into two groups accordingto fruit type (glabrous and hispid). Boissier (1881) con-sidered 90 species and divided them into three sections(Eugalium, Aparine and Cruciata) and 11 subsections.Ehrendorfer et al. (1976) recognized 145 species for Eu-ropean flora, classifying into 10 sections. Ehrendorfer &Schonbeck-Temesy (1982) listed for the flora of Turkey101 species divided into 10 sections.In Egypt, Tackholm (1974) named 12 species of
Galium, Boulos (1995, 2000) recognized only 10 species.Abdel Khalik et al. (2007; 2008a, b, c) studied 13 Egyp-tian taxa of Galium by different means such as mor-phological characters, including vegetative parts, flow-ers, fruits, seeds, pollen grains, anatomical structure.Numerical analysis was conducted, and they classifiedthese species into groups.Molecular markers are useful in identifying the
maximally diverse parental genotypes through an eval-
Genetic diversity and relationships among Galium 301
Table 1. List of the studied species of Galium sited according to traditional Boissier (1881), more recent traditional (Ehrendorfer &Schonbeck-Temesy 1982; Ehrendorfer et al. 2005) and a recent phylogenetic classification based on molecular data (Soza & Olmstead2010b).
uation of genetic diversity which is useful in culti-var identification, seed purity analysis and breeding.Among the various molecular markers, Random Am-plified Polymorphic DNA (RAPD) and Inter-Simple Se-quence Repeat (ISSR) are simple and quick techniquesand have become popular as their application does notneed any prior information about the target sequencesin the genome, high-efficiency and sharp sensibility, andthese techniques have now been widely used for lineidentification and genetic diversity. These markers havebeen used for DNA fingerprinting, conservation biol-ogy (Martin & Sanchez-Yelamo 2000; Li et al. 2005),to identify and determine relationships at the species,population and cultivar levels in many plants (Pezh-manmehr et al. 2009; Manica-Cattani et al. 2009; Nanet al. 2003; Fracaro et al. 2005; Mattioni et al. 2002;Zhang & Dai 2010), genetic diversity studies (Qian et al.2001; Pradeep et al. 2005; Josiah et al. 2008; Parmaksız& Ozcan 2011; Xavier et al. 2011; Shen et al. 2012;Aghaabasi & Baghizadeh 2012; Buldewo et al. 2012;Abdel Khalik et al. 2012), which utilize the advantagesof the two molecular marker techniques, reduce poten-tial errors connected with each method and hence im-prove the reliability of results.Literature review also revealed that the phyloge-
netic (molecular markers, ISSR and RAPD) investiga-tion of the Galium species have not been documenteduntil now. We focused on Egyptian species and relatedspecies of this genus to compare and align species withgenetic markers.The present study describes the genetic variability
found by using RAPD and ISSR as the combined resultswould be more credible to analyze the genetic structureof Galium species of Egypt and related species.
Material and methods
Plant materialsThe samples of Gallium seeds were taken from wild popu-lations and some herbarium specimens. Voucher specimensof the populations studied are deposited in the herbariumof the Department of Botany of Sohag University (SHG)(Table 1).
Plant genomic DNA extractionTotal genomic DNA was extracted from germinated seeds.These were first ground into a fine powder in liquid nitro-gen using a pestle and mortar following the steps of CTABprotocol (Doyle & Doyle 1990; Doyle 1991).
Random Amplified Polymorphic DNA (RAPD) analysisRAPD was performed as described by Huang et al. (2000)and Abd El-Twab & Zahran (2008). PCR reactions werecarried out with several primers using 25 µL volume PCRmixture containing 2.5 µL of buffer (Taq DNA polymerasecomplete high specificity reaction buffer), 2.5 µL dNTPs(from 10 mM stock, Bioron International, Germany), 12 ngprimers (Operon Nippon EGT Co. Ltd.), 1 U DFS-Taq DNApolymerase (Bioron International, Germany) and 100 ng ofDNA (Table 2). The thermal cycler (Thermo Hybaid) wasoperated as follows: 1 cycle at 95◦C for 5 min followed by40 cycles at 95◦C, 36◦C and 72◦C for 40 sec, 1 min and2 min respectively; and a final amplification was carried outat 72◦C for 10 min.
Genetic diversity and relationships among Galium 303
Fig. 1. A representative agarose gel where PCR products of RAPDs (A: amplified by RAPD primer) and ISSRs (B: amplified by ISSR16 primer) markers, respectively.
Inter Simple Sequence Repeats (ISSR) analysisISSR procedure was achieved as described by Dogan etal. (2007). PCR amplification was carried out with sev-eral primers using 100 ng of genomic DNA (Table 2). The25 µL PCR mixture contained 2.5 µL of buffer (Taq DNApolymerase complete high specificity reaction buffer (10X)containing 500 mM KCl, 100 mM TrisHCl pH 8, 0.1%20 and 15 mM MgC12, Bioron International, Germany),2.5 µL dNTPs (from 10 mM stock, Bioron International,Germany), 12 ng primers (Operon Nippon EGT CO. LTD.)1 U DFS-Taq DNA polymerase (Bioron International, Ger-many), and 100 ng of DNA. The thermal cycler was operatedas follows: 1 cycle at 94◦C for 1.5 min; 35 cycles at 9, 40 and72◦C for 40, 45 sec and 1.5 min respectively; 1 cycle at 94◦Cfor 45 sec; 1 cycle at 44◦C for 45 s and a final amplificationat 72◦C for 5 min.
Gel-electrophoretic analysisGel electrophoresis following Abd El-Twab & Zahran (2008)was used to determine the presence/absence of the total ge-nomic DNA and size of the DNA fragments after RAPDand ISSR loaded using loading buffer in 1.5% Agarose Gel,which carries DNA from negative to positive side. DNA wasstained in gel by ethidium bromide (0.5 µg mL−1), that com-bines with DNA fragments and gives violet light under UVlight, at that time; photographs were taken using a digitalsystem (Past software) (Fig. 1).
Data analysisRAPD and ISSR markers produce DNA amplification sig-nals that can be converted into measurements of similarity
or dissimilarity (DNA electrophoretic patterns contain vis-ible bands assigned to specific positions in an individuallane). Pairwise similarity of the genotypes or genetic phe-notypes represented in the different lanes can be quantifiedusing indexes or coefficients of similarity. These estimatorsdefine genetic distances that portray DNA divergence be-tween organisms in phenetic and cladistic analyses (Huanget al. 2000). For each primer, the consistent amplified prod-ucts were recorded. The polymorphic fragments (RAPD andISSR) were named by the primer code followed by the sizeof the amplified fragment in base pairs. For phylogeneticanalysis, each amplified band was treated as a unit charac-ter regardless of its intensity and scored in terms of a binarycode, based on presence (1) and absence (0) of bands. Onlyclear and reproducible bands were considered for scoring.For phylogenetic analysis, all the members of Galium wereincluded. To analyse data obtained from the binary ma-trices, the NTSYS-pc version 2.1 statistical package (Rohlf2000) was used. Three datasets were used, viz. RAPD, ISSR,and combined datasets of RAPD and ISSR. The statisti-cal method took into account the presence or absence ofeach band as differential features. The binary qualitativedata matrices were then used to construct similarity matri-ces based on Jaccard similarity coefficients (Jaccard 1908).The similarity matrices were then used to construct dendro-grams using unweighted pair group method with arithmeticaverage (UPGMA).
To compare RAPD- and ISSR-based dendrograms,cophenetic matrices were derived from the dendrograms us-ing the COPH (cophenetic values) program, and the good-ness of fit of the clustering to the 2 data matrices was
304 K. Abdel Khalik et al.
Fig. 2. UPGMA phenogram showing genetic diversity of 24 taxa of Galium based on RAPDs characters.
calculated by comparing the original similarity matriceswith the cophenetic value matrices using the Mantel ma-trix correspondence test (Mantel 1967) in the MXCOMPprogram. Principal co-ordinate analysis (PCOORDA) wasperformed based on the similarity coefficients. Lastly, a com-bined dataset was prepared using both RAPD and ISSRdata and used to calculate the combined similarity matrix,which was ultimately used to construct the phylogenetic treeand PCOORDA. The combined phylogenetic tree was com-pared with RAPD- and ISSR-based trees using the Mantelmatrix correspondence test.
Results
RAPD analysisTwenty six primers were used for the RAPD analysisto investigate the pattern of genetic variation among24 species of the genus Galium growing wild in Egyptand related species. Among the primers tested onlythree revealed a polymorphism (Table 2). Each of theseprimers was tested on all samples studied and were se-lected for genotype analysis because their patterns werereproducible and stable. Polymorphic bands were se-lected for identifying the genetic similarity for the groupof species. One hundred and fifty eight reproduciblepolymorphic bands were produced after 3 RAPD-PCRprimers. The average similarity coefficient ranged from0.49 to 1.00. The highest number of polymorphic am-plification DNA fragments obtained per primer C was61 bands with size ranged from 200 to 1500 bp (Ta-ble 2). Relations between the studied taxa are pre-sented in a dendrogram built on the basis of similar-ity coefficients. For ease of comparison, the 158 bandswere taken together and the number of bands fromeach size of DNA fragments (bp) was scored for ev-ery species. Six main branches and clusters with about0.61 similarity were obtained (Fig. 2). (1) A branch in-cludes Galium setaceum. (2) A branch comprises Gal-ium schultesii. (3) A cluster contains Galium tricor-
nutum, G. aparine and G. lucidum with about 0.63similarity. (4) A cluster with Galium album, G. mol-lugo and G. verum with 0.66 genetic similarities. (5)A cluster comprises Galium spurium, G. obliquum andG. scabrifolium showing about 0.73 genetic similarity.(6) A cluster which is divided into two subgroups. (a)A subgroup includes cluster of G. murale and G. odor-atum with about 0.83 similarity. (b) A subgroup whichcan divided into two subgroups. (I) A subgroup con-tains Galium grande, G. sinaicum and G. angustifoliumsubsp. angustifolium with about 0.80 similarity. (II) Asubgroup comprises Galium parisiense, G. uniflorum,G. humifusum, G. canum,G. asparagifolium, G. circaeand G. glaucum with about 0.78 genetic similarity.
ISSR analysisIn total 18 primers for the ISSR were used to investi-gate the pattern of genetic variation among 24 speciesof the genus Galium growing wild in Egypt and relatedspecies. Among the primers tested only four (Table 2)produced clear bands and had reproducibility, so theywere selected for further analysis. A total number of92 reproducible polymorphic bands were resulted after4 ISSR primers; those bands were used for studyingthe genetic similarity among the species. The averagesimilarity coefficient was ranged from 0.59 to 0.95. Theresults showed that primer ISSR 17 was monomorphicand the rest of the primers polymorphic. The highestnumber of polymorphic amplification DNA fragmentsobtained per primer ISSR 16 was 50 bands with sizeranged from 100 to 1500 bp, while the lowest number of5 bands were generated with primer ISSR 17 (Table 2).The results of the consensus tree from ISSR data indi-cated that tree was divided into 7 main branches andclusters with 0. 77 similarity (Fig. 3). (1) A branch in-cludes Galium aparine. (2) A branch comprises Galiumsetaceum. (3) A branch contains Galium schultesii. (4)A branch with Galium spurium. (5) A branch includes
Genetic diversity and relationships among Galium 305
Fig. 3. UPGMA phenogram showing genetic diversity of 24 taxa of Galium based on ISSRs characters.
Fig. 4. UPGMA phenogram showing genetic diversity of 24 taxa of Galium based on combination of RAPDs and ISSRs characters.
Galium parisiense. (6) A major cluster includes Gal-ium tricornutum, Galium album subsp. pycnotrychum,G. grande, G. uniflorum, G. cannum, G. humifusumand G. murale with about 0.80 similarity. (7) A majorcluster divided into 2 sub-clusters with about 0.82 sim-ilarity: first sub-cluster with Galium sinaicum, G. an-gustifolium subsp. angustifolium, G. odoratum, G. as-paragifolium, G. circae and G. glaucum with about 0.86genetic similarity; second sub-cluster includes Galiummollugo, G. verum, G. obliquum, G. lucidum, G. scabri-folium and G. album subsp. album with 0.83 similar-ity.
Combined RAPD and ISSR analysisThe UPGMA dendrogram obtained from the clusteranalysis of RAPD and ISSR combined data gave nearsimilar clustering pattern, with Jaccard’s similarity co-efficient ranging from 0.58 to 0.95. The results indicatedthat the consensus tree was divided into 6 major clus-ters and branches with 0.66 similarity and confirmedby PCOA (Figs 4, 5). The groups corresponding withspecies were clearly defined by the first and second prin-cipal coordinates which represented 16% and 11% oftotal variation, respectively. (1) A branch includes Gal-ium aparine. (2) A branch comprises Galium setaceum.
306 K. Abdel Khalik et al.
Fig. 5. Principal Coordinates Analysis of RAPDs and ISSRs combined constructed using 250 variable DNA bands from 24 species ofGalium.
(3) A branch contains Galium sinaicum. (4) A clusterwith Galium grande and G. tricornutum. (5) A clusterincludes Galium mollugo, G. verum, G. album subsp.pycnotrychum, G. lucidum and G. scabrifolium. (6) Amajor cluster divided into 2 sub-clusters with about0.70 similarity: first sub-cluster with Galium odoratum,G. album subsp. album, G. obliquum and G. parisiensewith about 0.75 genetic similarity; second sub-clusterincludesGalium uniflorum, G. spurium, G. humifusum,G. canum, G. murale, G. angustifolium subsp. angusti-folium, G. circae, G. schultesii, G. asparagifolium andG. glaucum with 0.71 similarities.
Discussion
Morphological characters in plants may be affected byenvironmental conditions. Thus, the use of morpholog-ical characters for classification may result in discrep-ancies. Productivity of a molecular marker techniquedepends on the amount of polymorphism it can de-tect among the set of accessions under investigation.RAPD and ISSR markers have been used in many stud-ies for DNA fingerprinting and phylogenetic analyses.Galvan et al. (2003) concluded that ISSR would be abetter tool than RAPD for phylogenetic studies. Thepresent study, however, has demonstrated that bothRAPD and ISSR technique along with suitable statis-tical tools could be successfully applied to assess thegenetic diversity and phylogenetic analysis in Galium.Although RAPD and ISSR markers showed consider-able differences in detecting polymorphism and discrim-inating capacity, they showed nearly similar topology in
dendrograms generated on the basis of similarity matri-ces. A highly significant correlation between these twodendrograms suggested that both markers were equallyefficient for assessing phylogenetic relationships amongthe investigated taxa. Moreover, the genotype distri-bution on the consensus tree based on the combinedbanding patterns of RAPD and ISSR may significantlydiffer because it is possible that each technique amplifydifferent parts of the genome. The RAPD markers coverthe whole genome for amplification, ISSR amplify theregion between two micro satellites. Hence, the poly-morphisms reflect the diversity of these regions of thegenome. It is therefore better to use the combinationof banding patterns of the two techniques in order touse more segments sites of the genome that will increasethe validity of the consensus tree. In general, our resultsobtained from the RAPD and ISSR analyses, suggestedgroups and partially confirmed the sectional classifica-tion of Galium by the most recent traditional (Ehren-dorfer & Schonbeck-Temesy 1982; Ehrendorfer et al.2005) and a recent phylogenetic classification based onmolecular data (Soza & Olmstead 2010b).
Sections Leiogalium, Galium and Lophogalium (groups1 and 4 B)According to the combined RAPD and ISSR tree(group 1) there is a close relationship between fourspecies of section Leiogalium (G. mollugo, G. albumsubsp. pycnotrychum, G. lucidum and G. scabrifolium)and one species (G. verum) of section Galium with 0.70genetic similarities. These species are morphologicallysimilar in the leaves structure by having consistently
Genetic diversity and relationships among Galium 307
whorled leaves with six or more organs. Kliphuis (1986)presented a cytological and morphological analysis of19 species of Galium from the Balkans, and he classi-fied the species of sectionLeiogalium into four groupsand counted Galium mollugo (2n = 22), G. album(2n = 44), G. lucidum (2n = 44) and G. scabrifolium(2n = 22). Also, he concluded that G. mollugo andG. album constituted a polyploid complex. Moreover,Manen et al. (1994) investigated phylogenetic analy-sis of 25 species of the tribe Rubieae using sequencedata from the atpB-rbcL intergene region. They showedvery little change in the intergene cpDNA region in theperennials studied from section Galium (G. verum, 2x,4x, Eurasian) and section Leiogalium which is a poly-morphic Mediterranean-European polyploidy complexwith [G. mollugo (2x), G. album (4x), G. lucidum (4x),and G. corrudifolium (2x)]. Natali et al. (1995) pre-sented phylogenetic analysis of 39 species of the tribeRubieae using sequence data from the atpB-rbcL in-tergene region. They showed that the section Leiogal-ium including G. mollugo, G. album, G. corrudifoliumand G. lucidum was closely related to the section Gal-ium (with G. verum) and exhibited practically no se-quence differences. Furthermore, Mitova et al. (2002)presented iridoid patterns in 19 species ofGalium. Theytreated G. mollugo and G. album as one group and in-cluded secogalioside 18 that considered it as an impor-tant chemotaxonomic marker. Also, they showed theaffinity of G. verum to the groups of G. mollugo andG. album by presence of loganin 13 and 6 acetylscan-doside 8. Likewise, Soza & Olmstead (2010b) studiedphylogenetically 126 old and new world taxa of Ru-bieae using sequence data from three chloroplast re-gions. They indicated seven major clades and identi-fied one clade (Clade D) comprising members of sectionGalium (G. verum) and sect. Leiogalium (G. mollugo,G. album, and G. lucidum).Within the subgroup 4B, species of section Leio-
galium (G. glaucum, G. asparagifolium, G. schultesiiand G. circae) and one species of section Lophogalium(G. angustifolium) have been recognized as a distinctsub-group with 0.84 genetic similarities. These speciescan be clearly defined on the basis of various features:perennials, forming rhizomes with sexual and vegeta-tive reproduction, polyploids, and with narrow leaves,mesoxerophylous plants. Ančev & Krendl (2011) inves-tigated the 18 species of section Leiogalium in Bulgariabased on morphology and chromosome numbers andconsidered G. glaucum (4x = 44), G. asparagifolium(4x = 44) and G. schultesii (6x = 66) as polyploids.Within section Leiogalium reticulate relationships andaffinities linked to hybridization, polyploidy, and ac-tive recent evolutionary differentiation are obvious. Onthe other hand, G. angustifolium (sect. Lophogalium) ischaracterized as a perennial herb, bearing linear leaveswith apically directed, short hairs along the marginsand polyploid (Ehrendorfer 1956; Dempster 1993).Additionally, Galium humifusum (sect. Galium)
separated with above members with 0.77 genetic sim-ilarities. Natali et al. (1995) suggested the monophyly
of the clade represented by perennial sections Galium,Leiogalium, Leptogalium, Hylaea and the annual Kol-gyda. Mitova et al. (2002) have found iridoid esterswith p-hydroxyphenylpropoionic acid, loganin 13 and6-acetylscandoside 8 which are found also in G. verumand members of section Leiogalium. These results re-flect the close relationship between G. humifusum withmembers of section Leiogalium. Since our results are in-herited data, and we suggest that (1) species of sectionLeiogalium form a polyphyletic group, and (2) thereare close relationships between members of this sectionwith G. verum and G. humifusum (sect. Galium) andG. angustifolium (sect. Lophogalium). Therefore, thesedata agree with those of Manen et al. (1994), Natali etal. (1995), Mitova et al. (2002) and Soza & Olmstead(2010b).
Section Jubogalium (groups 2, 4B and 5)Concerning the clades of G. canum, G. setaceum andG. sinaicum, our results do not support the mono-phyly of the artificial section Jubogalium. This is dueto the placement of Galium sinaicum, G. canum andG. setacum within three separate clades with 0.84 ge-netic similarities. Boissier (1881) classified these speciesinto two sections on the basis of annual or perennialhabit, flowers hermaphrodite or polygamous, and pe-duncle erect or recurved: sect. Aparine and sect. Eugal-ium. He placed G. setaceum in the former, with annualhabit, flower hermaphrodite or polygamous and pedun-cle erect or recurved, while G. canum and G. sinaicumwere placed in the second section on the basis of peren-nial habit, flower hermaphrodite and erects peduncle.Furthermore, Ehrendorfer et al. (1976) and Ehrendorfer& Schonbeck-Temesy (1982) placed those three speciesin a separate section (Jubogalium). Abdel Khalik etal. (2007) studied the pollen morphology of Galiumin Egypt and indicated that G. canum has 5–7 colpi,G. sinaicum 5–6 colpi and G. setaceum 6–7 colpi. Fur-thermore, Abdel Khalik et al. (2008b) investigated thefruit and seed morphology of 13 Egyptian species ofGalium and showed that mericarp surface is micropapil-late inG. sinaicum, with hooked or depressed hairsin G. setaceum and with long white straight hairs inG. canum. Moreover, Abdel Khalik et al. (2008c) in-vestigated 50 morphological characters, including veg-etative parts, flowers, fruits, seeds, pollen grains, andanatomical structure by means of numerical analysis, of13 taxa belonging to genus Galium from Egypt. Theyconcluded that species of section Jubogalium are het-erogeneous. Our results disagree with those of Boissier(1881), Ehrendorfer et al. (1976) and Ehrendorfer &Schonbeck-Temesy (1982), and agree with Abdel Kha-lik et al. (2007, 2008a, b, c).
Section Kolgyda (groups 3, 4, and 6)Within Kolgyda group, three major clades were iden-tified (Fig. 4). The first clade includes Galium tri-cornutum; the second clade includes G. aparine andthe third clade contains G. parisiense, G. murale andG. supurum with 0.70 genetic similarities. Boissier
308 K. Abdel Khalik et al.
(1881) treatedG. aparine, G. tricornutum, G. spurium,G. murale and G. parisiense as members of sectionAparine, and classified them in different subsections.However, Ehrendorfer et al. (1976) and Ehrendorferand Schonbeck-Temesy (1982) considered these taxa asgood members in section Kolgyda (synonym of sect.Aparine). Ehrendorfer (1971) considered the autoga-mous G. aparine complex as probably originating byallopolyploidy from three racial stocks of SouthwestAsian origin. The species seems to include tetraploid,hexaploid and octaploid cytotypes (2n = 42, 44, 48,62, 66 and 68). Hanf (1983) showed that variants ofG. spurium with setose fruits, besides the flower char-acters and the diploid chromosome number are oftennot easy to distinguish from G. aparine. Also, Kli-phuis (1986) investigated the section Kolgyda and in-cluding Galium aparine (2n = 42, 44, 48, 64, 66) andG. parisiense (2n = 22, 44, 66), and concluded thatboth species constituted a polyploid complex. Natali etal. (1995, 1996) exhibited that G. aparine, G. spuriumand G. tricornutum were joined together in the sameclade.With Egyptian material, Abdel Khalik et al. (2007)
indicated that G. aparine has 7–9 colpi, G. spurium 6–8colpi, G. murale 6–7 colpi, G. parisiense 8–10 colpi, andG. tricornutum 8–9 colpi. Moreover, Abdel Khalik et al.(2008c) settled that results of both cluster and PCO-ORDA analyses confirmed the group of G. aparine,G. tricornutum, G. ceratopodum and G. spurium asa well-distinguished group and showed that G. aparineand G. tricornutum form a subgroup, and another sub-group includes G. ceratopodium and G. spurium. Sim-ilarly, Soza & Olmstead (2010b) identified one clade(Clade III) comprising members of section Kolgydawith other sections, but under this clade (III) distin-guished two subclades (A and B); subclade A includedG. aparine and G. tricornutum and subclade B com-prised G. murale and G. parisiense; unfortunately, theydid not included G. spurium in their study. In general,our results mainly do not support the taxonomic sys-tem of the section Kolgyda proposed by Boissier (1881),Ehrendorfer et al. (1976), Ehrendorfer & Schonbeck-Temesy (1982), Natali et al. (1995, 1996) and AbdelKhalik et al. (2008c), but partially agree with Soza &Olmstead (2010b) in the classification of the species ofthis section into two sub-clades.
Section Baccogalium (group 3)This section includes only G. grande. It has been rec-ognized as a distinct clade with 0.67 genetic similari-ties and can be clearly defined on the basis of variousfeatures: perennial, fleshy-fruited, dioecious and polyg-amous species. Soza and Olmstead (2010a) studied thefruit morphology and evolution ofGalium and they con-cluded that section Baccogalium is a monophyletic. Ourresults agree with those authors.
Sections Hylaea and Leptogalium (group 4 A)This group includes G. odoratum (sect. Hylaea) andG. obliquum (sect. Leptogalium) with 0.78 genetic sim-
ilarities. Natali et al. (1995) recommended a close rela-tionship between G. odoratum and G. corsicum (sect.Leptogalium). These species can be defined on the ba-sis of perennial habit. Soza & Olmstead (2010b) iden-tified Clade III including G. odoratum (sect. Haylaea,sub-clade A) and species of both section Leptogalium(sub-clade B). Our results suggest close relationshipsbetween species of both sections and agree with thoseof Natali et al. (1995) and Soza & Olmstead (2010).
Section Bataparine (group 4B)This section comprises only G. uniflorum. It has beenrecognized as a distinct clade with 0.77 genetic similar-ities. This species can be clearly defined on the basis ofvarious features: perennial, hermaphroditic, glabrous,fleshy fruits to hooked-hairy. Soza & Olmstead (2010b)settled that section Bataparine is a paraphyletic grade.Our results show that G. uniflorum belongs to the cladeincludes species from different sections as sister to thisclade. More data are necessary for deciding this assump-tion that the section Bataparine is paraphyletic or not.In conclusion, the present study demonstrated that
RAPD, ISSR and cost-effective markers such as RAPDand ISSR together with good statistical tools can besuccessfully applied to study phylogenetic relationshipsat the intraspecific level in Galium. Additionally, thelarge number of polymorphic bands obtained in thepresent study signifies the power of RAPD and ISSRmarkers in fingerprinting and diversity analyses. Sixclades can be recognized within Galium, which mostlycorroborate, but also partly contradict, traditional tax-onomic treatments. A remarkable result from this studywas to identify a close relationship between membersof section Leiogalium and with G. verum and G. hu-mifusum (sect. Galium) and G. angustifolium (sect.Lophogalium). Further support comes from the molec-ular data of RAPD and ISSR which indicate that themore apomorphic groups of Galium form the sectionLeiogalium clade including the perennial sections Gal-ium, Lophogalium, Jubogalium, Hylaea and Leptogal-ium as well as the annual section Kolgyda. However, webelieve that molecular and morphological approachesshould be combined in order to arrive at a broadly ac-cepted phylogenetic reconstruction of the genus Gal-ium. Moreover, comprehensive study covering furtherspecies from different sections of Galium would be nec-essary to make a more thorough classification.
Acknowledgements
We are grateful to the Director and Curator of Kew herbar-ium (K), Berlin herbarium (B), Bonn botanical garden’sherbarium (BONN), botanical gardens of the University ofUlm (ULM), Rancho Santa Ana Botanic Garden’s herbar-ium, USA (RSABG) and Wageningen University herbar-ium (WAG). We are indebted to Prof. Dr. Ana OrtegaOlivencia, Area of Botany, Faculty of Science, University ofExtremadura, Badajoz, Spain for going through the man-uscript and making valuable suggestions. Special thanksto Dr. Valerie Soza, Department of Biology, University of
Genetic diversity and relationships among Galium 309
Washington and Prof. Dr. Klaus Mummenhoff, Plant Tax-onomy, Biology Department, Botany, Osnabruck University,Germany for their valuable suggestions and comments onthe manuscript
References
Abd El-Twab M.H. & Zahran F.A. 2008. Extracting total ge-nomic DNA of Chrysanthemum sensu lato by CTAB and SDSwithout both liquid nitrogen and phenol. Chromos. Bot. 3:83–88.
Abdel Khalik K., Abd El-Ghani M.M. & El Kordy A. 2007. Apalynological study of Galium L. (Rubiaceae) in Egypt andits systematic implication. Feds. Rep. 118: 311–326.
Abdel Khalik K., Abd El-Ghani M.M. & El Kordy A. 2008a.Anatomical findings of the genus Galium L. (Rubiacaeae) inEgypt and their systematic implications. Turk. J. Bot. 32:353–359.
Abdel Khalik K., Abd El-Ghani M.M. & El Kordy A. 2008b.Fruit and seed morphology in Galium L. (Rubiaceae) andimportance for taxonomic identification. Act. Bot. Croat. 67:1–20.
Abdel Khalik K., Abd El-Ghani M.M. & El Kordy A. 2008c.Numerical taxonomy of the genus Galium L. (Rubiaceae) inEgypt. Phyto. Balcn. 14: 245–253.
Abdel Khalik K. & Bakker F. 2007. Nasturtiopsis integrifolia(Boulos) Abdel Khalik and Bakker (Brassicaceae), a newcombination, and Cruciata articulata (L.) Ehrend. (Rubi-aceae), a new record for the Flora of Egypt. Turk. J. Bot.31: 571–574.
Abdel Khalik, K., G. Osman and W. Al-Amoudi. 2012. Ge-netic diversity and taxonomic relationships of some Ipomoeaspecies based on analysis of RAPD-PCR and SDS-PAGE ofseed proteins. Aust. J. Crop. Sci. 6: 1088–1093.
Aghaabasi K. & Baghizadeh A. 2012. Investigation of geneticdiversity in flixweed (Descurainia sophia) germplasm fromKerman province using inter-simple sequence repeat (ISSR)and random amplified polymorphic DNA (RAPD) molecularmarkers. Afr. J. Biotechnol. 11: 10056–10062.
Ančev M. & Krendl F. 2011. Galium sect. Leiogalium (Rubiaceae)in the Bulgarian flora. Phyto. Balcn. 17: 291–314.
Andersson L. & Rova J.H.E. 1999. The rps16 intron and thephylogeny of the Rubioideae (Rubiaceae). Pl. Syst. Evol. 214:161–186.
Backlund M., Bremer B. & Thullin M. 2007. Paraphyly of Paed-erieae, recognition of Putorieae and expansion of Plocama(Rubiaceae-Rubioideae). Taxon 56: 315–328.
Boissier E. 1881. Flora Orientalis. Vol. 5. Geneva and Basel.Boulos L. 1995. Flora of Egypt, Checklist. Al Hadara Publishing,Cairo, 287 pp.
Boulos L. 2000. Flora of Egypt. Vol. 2. Al Hadara Publishing,Cairo, 352 pp.
Bremer B. 1996. Phylogenetic studies within Rubiaceae and re-lationships to other families based on molecular data. Opera.Bot. Belg. 7: 33–50.
Bremer B. & Eriksson T. 2009. Time tree of Rubiaceae: Phy-logeny and dating the family, subfamilies, and tribes. Int. J.Pl. Sci. 170: 766–793.
Bremer B. & Manen J.F. 2000. Phylogeny and classification ofthe subfamily Rubioideae (Rubiaceae). Pl. Syst. Evol. 225:43–72.
Buldewo S., Pillay M. & Jaufeerally-Fakim Y. 2012. Genetic di-versity in Anthurium andraeanum cultivars in Mauritius. Afr.J. Biotechnol. 11: 16737–16744.
Dempster L.T. 1993. Rubiaceae, pp. 976–986. In: Hickman J.C.(ed.), The Jepson manual, higher plants of California. Uni-versity of California Press, Berkeley, California, USA.
Dogan B., Duran A. & Hakki E.E. 2007. Phylogenetic analysisof Jurinea (Asteraceae) species from Turkey based on ISSRamplification. Ann. Bot. Fenn. 44: 353–358.
Doyle J.J. 1991. DNA protocols for plants, pp. 283–293. In: He-witt G., Johnson A.W.B. & Young J.P.W. (eds), Molecular
Techniques in Taxonomy, NATO ASI Series H, Cell BiologyVol. 57.
Doyle J.J. & Doyle J.L. 1990. A rapid total DNA preparationprocedure for fresh plant tissue. Focus 12: 13–15.
Ehrendorfer F. 1956. Survey of the Galium multi?orum complexin western North America. Cont. Dud. Herb. 5: 1–21.
Ehrendorfer F. 1958. Critical notes on Turkish Rubiaceae. Notes.Roy. Bot. Gard. Edinb. 22: 323–401.
Ehrendorfer F., Krendl F. & Puff C. 1971. Evolution and eco-geographical differentiation in some South-West Asiatic Ru-biaceae, pp. 195–215. In: Davis, P.H., Harper P.C. & HedgeI.C. (eds), Plant life of South-West Asia. Botanical Society ofEdinburgh, Edinburgh, UK.
Ehrendorfer F., Krendl F. & Puff C. 1976. Galium L., pp. 14–36.In: Tutin T.G. et al. (eds), Flora Europaea. Vol. 4, CambridgeUniversity Press, Cambridge.
Ehrendorfer F. & Schonbeck-Temesy E. 1982. Galium L., pp.767–823. In: Davis P.H. (ed.), Flora of Turkey and the EastAegean Islands. Vol. 7. Edinburgh University Press, Edin-burgh.
Ehrendorfer F., Schonbeck-Temesy E., Puff C. & Rechinger W.2005. Rubiaceae. Part 176. In: Rechinger K.H. (ed.), FloraIranica. Verlag des Naturhistorischen Museums Wien, Vi-enna.
Fracaro F., Zacaria J. & Echeverrigaray S. 2005. RAPD based ge-netic relationships between populations of three chemotypesof Cunila galioides Benth. Bioch. Syst. Ecol. 33: 409–417.
Galvan M.Z., Bornet B., Balatti P.A. & Branchard M. 2003. In-ter simple sequence repeat (ISSR) marker as a tool for theassessment of both genetic diversity and gene pool origin incommon bean (Phaseolus vulgaris L.). Euphytica 132: 297–301.
Hanf M. 1983. The Arable Weeds of Europe. BASF UK Ltd,Ipswich, UK.
Huang S.C., Tsai C.C. & Sheu C.S. 2000. Genetic analysis ofChrysanthemum hybrids based on RAPD molecular markers.Bot. Bull. Acad. Sin. 41: 257–262.
Jaccard P. 1908. Nouvelles recherches sur la distribution florale.Bull. Soc.Vaud. Sci. Natur.44: 223–270.
Jansen S., Dessein S., Piesschaert F., Robbrecht E. & Smets E.2000. Aluminum accumulation in leaves of Rubiaceae: Sys-tematic and phylogenetic implications. Ann. Bot. 85: 91–101.
Josiah C.C., George D.O., Eleazar O.M. & Nyamu W.F. 2008.Genetic diversity in Kenyan populations of Acacia senegal(L.) willd revealed by combined RAPD and ISSR markers.Afr. J. Biotechnol. 7: 2333–2340.
Kliphuis E. 1986. Cytotaxonomic investigations on some speciesof the genus Galium (Rubiaceae) from the Balkans. Nord. J.Bot.6: 15–20.
Li Y.Y., Chen X.Y. & Zhang X. 2005. Genetic differences be-tween wild and artificial populations of Metasequoia glyp-tostroboides: implications for species recovery. Cons. Biol. 19:224–231.
Linnaeus C. 1753. Species Plantarum. Laurentius Salvius, Hol-miae.
Mabberley D.J. 1987. The plant-book. Cambridge UniversityPress, Cambridge.
Manen J.F., Natali A. & Ehrendorfer F. 1994. Phylogeny ofRubiaceae-Rubieae inferred from the sequence of a cpDNAintergene region. Pl. Syst. Evol. 190: 195–211.
Manica-Cattani M.F., Zacaria J., Pauletti G., Atti-Serafini L.& Echeverrigaray S. 2009. Genetic variation among SouthBrazilian accessions of Lippia alba Mill. (Verbenaceae) de-tected by ISSR and RAPD markers. Braz. J. Biol. 69: 375–380.
Mantel N. 1967. The detection of disease clustering and a gener-alized regression approach. Can. Res. 27: 209–220.
Martín J.P. & Sánchez-Yélamo M.D. 2000. Genetic relationshipsamong species of the genus Diplotaxis (Brassicaceae) usinginter-simple sequence repeat markers. Theo. Appl. Genet.101: 1234–1241.
Mattioni C., Casasoli M., Gonzalez M. & Ipinza R. 2002. Compar-ison of ISSR and RAPDmarkers to characterize three ChileanNothofagus species. Theo. Appl. Genet. 104: 1064–1070.
310 K. Abdel Khalik et al.
Mitova M., Ančev M., Handjieva N. & Popov S. 2002. Iridoidpatterns in Galium L. and some phylogenetic considerations.Z. Naturf. 57c: 226–234.
Nan P., Peng S., Shi S., Ren H., Yang J. & Zhong Y. 2003. Interpopulation congruence in Chinese Primula ovalifolia revealedby chemical and molecular markers using essential oils andISSRs. J. Biosci. 58: 57–61.
Natali A., Manen J.F. & Ehrendorfer F. 1995. Phylogeny of theRubiaceae-Rubioideae, in particular the tribe Rubieae: ev-idence from a non-coding chloroplast DNA sequence. Ann.Miss. Bot. Gard. 82: 428–439.
Natali A., Manen J.F., Kiehn M. & Ehrendorfer F. 1996. Tribalgeneric and specific relationships in the Rubioideae–Rubieae(Rubiaceae) based on sequence data of a cpDNA intergeneregion. Opera. Bot. Belg. 7: 193–203.
Nie Z.L., Wen J., Sun H. & Bartholomew B. 2005. Monophyly ofKelloggia Torrey ex Benth. (Rubiaceae) and evolution of itsintercontinental disjunction between western North Americaand eastern Asia. Am. J. Bot. 92: 642–652.
Parmaksız I. & Ozcan S. 2011. Morphological, chemical, andmolecular analyses of Turkish Papaver accessions (sect. Oxy-tona). Turk. J. Bot. 35: 1–16.
Pezhmanmehr M., Hassani M.S., Jahansooz F., Najafi A.A., Se-fidkon F., Mardi M. & Pirseiedi M. 2009. Assessment of ge-netic diversity in some Iranian populations of Bunium per-sicum using RAPD and AFLP markers. Afr. J. Biotechnol.7: 93–100.
Pradeep A.R., Chatterjee S.N. & Nair C.V. 2005. Genetic differ-entiation induced by selection in an inbred population of thesilkworm Bombyx mori, revealed by RAPD and ISSR markersystems. J. Appl. Gen. 46: 291–298.
Qian W., Ge S. & Hong D.Y. 2001. Genetic variation within andamong populations of a wild rice Oryza granulata from Chinadetected by RAPD and ISSR markers. Theo. Appl. Gen. 102:440–449.
Robbrecht E. 1988. Tropical woody Rubiaceae. Characteristic fea-tures and progressions. Contributions to a new subfamilialclassification. Opera. Bot. Belg. 1: 1–271.
Robbrecht E. & Manen J.F. 2006. The major evolutionary lin-eages of the coffee family (Rubiaceae, angiosperms). Com-bined analysis (nDNA and cpDNA) to infer the position ofCoptosapelta and Luculia, and supertree construction basedon rbcL, rps16, trnLtrnF and atpB-rbcL data. A new clas-sification in two subfamilies Cinchonoideae and Rubioideae.Syst. Geog. Pl. 76: 85–146.
Rohlf F.J. 2000. NTSYS-pc: Numerical Taxonomy and Multi-variate Analysis System. Ver. 2.1. Setauket, New York, USA:Exeter Publishing Ltd.
Rydin C., Kainulainen K., Razafimandimbison S.G., SmedmarkJ.E.E. & Bremer B. 2009. Deep divergences in the coffee fam-ily and the systematic position of Acranthera. Pl. Syst. Evol.278: 101–123.
Schischkin B.K. 2000. Flora of the USSR, vol. 23. Delhra Dunn,India: Bishen Singh Mahendra Pal Singh; Koenigstein: KoeltzScientific Books.
Shen W., Xi P., Li M., Liu RSun., L., Jiang Z. & Zhang L.2012. Genetic diversity of Ustilago scitaminea Syd. in South-ern China revealed by combined ISSR and RAPD analysis.Afr. J. Biotechnol. 11: 11693–11703.
Soza V.L. & Olmstead R.G. 2010a. Evolution breeding systemsand fruits in New World Galium and relatives (Rubiaceae).Am. J. Bot. 97: 1630–1646.
Soza V.L. & Olmstead R.G. 2010b. Molecular systematics of tribeRubieae (Rubiaceae): Evolution of major clades, developmentof leaf-like whorls, and biogeography. Taxon 59: 755–771.
Tackholm V. 1974. Students’ Flora of Egypt. ed. 2. PublisherCairo University, Beirut, 888 pp.
Willis J.C. 1985. A Dictionary of the Flowering Plants and Ferns.Cambridge University Press, Cambridge, 1294 pp.
Xavier J.R., Kumar J. & Srivastava R.B. 2011. Characteriza-tion of genetic structure of alfalfa (Medicago sp.) from trans-Himalaya using RAPD and ISSR markers. Afr. J. Biotechnol.10: 8176–8187.
Zhang L.J. & Dai S.L. 2010. Genetic variation within and amongpopulations of Orychophragmus violaceus (Cruciferae) inChina as detected by ISSR analysis. Genet. Res. Crop. Evol.57: 55–64.
