PCR-RFLP analysis of the whole chloroplast DNA from three cultivated species of cotton ( Gossypium L

Euphytica (2007) 156:47–56

PCR-RFLP analysis of the whole chloroplast DNAfrom three cultivated species of cotton (Gossypium L.)

Rashid Ismael Hag Ibrahim · Jun-Ichi Azuma · Masahiro Sakamoto

Received: 23 September 2006 / Accepted: 28 December 2006 / Published online: 21 January 2007© Springer Science+Business Media B.V. 2007

Abstract Variations of the chloroplast DNA(cpDNA) from three of the four cultivated spe-cies of cotton (Malvaceae); Gossypium barba-dense L., Gossypium hirsutum L., Gossypiumarboreum L. and its synonym Gossypium nankingMeyen., were analyzed. Using speciWc set of prim-ers, the whole circular cpDNAs from the four testspecies were ampliWed. These were subsequentlydigested with the use of seven restrictionenzymes. The ampliWed fragments of the wholecpDNAs of the diploid cultivated cottonG. arboreum and its synonym G. nanking did notshow any diVerences. However, the allotetraploidcultivated cottons G. barbadense and G. hirsutum,showed some fragment length diVerences directlyvisible after ampliWcation and two types of restric-tion fragment length polymorphism (RFLP), theWrst appeared as slightly lengthened bands andthe other as gain or loss of a restriction site. Theresults also showed that the chloroplast genomesof the allotetraploid cultivated cottons are highly

similar to the diploid cultivated cottons tested interms of length and digestion patterns. Thedetected ampliWed length diVerences, RFLPs andthe restriction sites can be considered as speciesspeciWc markers for the allotetraploid cultivatedcottons, which could be a useful tool for futurestudies of the cpDNA of the genus Gossypium L.

Keywords Chloroplast DNA · Cotton · Gossypium · PCR-RFLP

Introduction

The cotton genus (Gossypium L., Malvaceae)contains about 50 species distributed throughoutthe warm arid regions of the world (Fryxell 1979,1992), it is divided into eight diploid genomegroups, which are designated A through G, inaddition to K and one allotetraploid genomegroup that contains only Wve taxa (Endrizzi et al.1985; Wendel 1989; Stewart 1995).

The genus Gossypium includes only four culti-vated cottons that are morphologically distinctand geographically widespread namely; G. arbor-eum L. and G. herbaceum L. (diploids of 2n = 26)known as the Old World African–Asian cottonsand G. hirsutum L. (Upland cotton) and SeaIsland cotton; G. barbadense L. (allotetraploids of2n = 52) or other known as New World cottons.The New World allotetraploid cottons were

R. I. H. Ibrahim (&) · J.-I. Azuma · M. SakamotoGraduate School of Agriculture, Kyoto University, 606-8502, Kitashirakawa Oiwake-cho, Kyoto, Sakyu-Ku, Japane-mail: [email protected]

R. I. H. IbrahimDepartment of Botany, Faculty of Science, University of Khartoum, P.O. Box 321, PC 11115 Khartoum, Sudan

1 3

48 Euphytica (2007) 156:47–56

found to be allopolyploids that originatedthrough natural hybridization between ancestraldiploid species from the Old World diploid Agenome of African–Asian cotton and the NewWorld diploid D genome (Beasly 1940; Skovsted1934, 1937). Furthermore, the genus Gossypium,both diploid and allotetraploid cottons, has a lin-ear inheritance over generations of cpDNA,which is uni-parentally and especially maternallyinherited. That was also reported for the allotetra-ploid cottons, AD-genome, which contain a chlo-roplast genome like that of the A-genome fromthe Old World diploid cotton (Wendel 1989).

Universal primers capable of amplifying spe-ciWc regions in mitochondrial (Demesure et al.1995) and chloroplast (Taberlet et al. 1991; Bad-enes and ParWtt 1995; Tsumura et al. 1996; Heinze2005; Dhingra and Folta 2005) genomes havemade it possible and easier to use organelle DNAfor taxonomic and phylogenetic studies. On theother hand, cpDNA also provides a possibility forthe identiWcation of species-speciWc markers dueto its slower rate of sequence and structural evo-lution than nuclear DNA plus its uni-parental andlinear inheritance over generations (Palmer 1987;Birky 1988; Clegg and Zurawski 1992).

At present, sequence comparison or sequenceanalysis of fragments ampliWed with the use ofuniversal primers for organelle DNA is widelyused in species identiWcation (Parani et al. 2001;Ridgway et al. 2003; Tsai et al. 2006), geneticdiversity (Parduchi and Szmidt 1999; Huang andSun 2000) and phylogenetic studies in many plantspecies (Gielly and Taberlet 1994; Badenes andParWtt 1995; Tsumura et al. 1995, 1996; Demesureet al. 1996; Parani et al. 2000; Wang et al. 2000;Cronn et al. 2002; Stehlik et al. 2002; Stehlik2002). The same technique was also used to deWnepolymorphic chloroplast microsatellites in manyPinus, Abies, A-genome of rice and Vitis species(Powell et al. 1995; Cato and Richardson 1996;Vendramin and Ziegenhagen 1997; Parducci et al.2001; Ishii et al. 2001; Arroyo-Garcia et al. 2002).Using the same technique, attempts to detectpolymorphism in the cpDNA regions in someplants failed (Boscherini et al. 1994; Vicario et al.1995).

On the other hand, RFLP analysis of cpDNA isnow recognized as a powerful tool for the study of

land-plant evolutionary processes, particularly atthe intra- and inter-speciWc level and molecularsystematics (Strauss and Doerksen 1990; Gov-indaraju et al. 1992; Wang and Szmidt 1994;Krupkin et al. 1996; Soltis et al. 1997; Schaal et al.1998; Ennos et al. 1999; Weins 1999). Anotherway in analyzing cpDNA is the polymerase chainreaction-restriction fragment length polymor-phism (PCR-RFLP). This was used for the identi-Wcation of intra-speciWc variations in the cpDNAof Abies alba (Ziegenhagen et al. 1995), Quercusrobur (Dumolin et al. 1995) and Fagus sylvatica(Demesure et al. 1996). Also it has provided evi-dence for the presence of inter-speciWc variationsin the cpDNA of the genus Phaseolus (Vekemanset al. 1998), Abies (Parducci and Szmidt 1999),Millet (Parani et al. 2000), Mangrove (Paraniet al. 2001), Soybeans (Xu et al. 2001), Papayas(Van Droogenbroeck et al. 2004), Crucifer (Clau-dia et al. 2004) and Houttuynia (Wei et al. 2005).In addition, it was used for inter-generic studies inCajanus (Lakshmi et al. 2000).

Most of the above mentioned studies havefocused on a single or only a few regions of thecpDNA. For the genus Abies, ten diVerentcpDNA regions were used (Parducci and Szmidt1999). Also, Cronn et al. (2002) used only fourdiVerent cpDNA regions of the genus Gossypium.The limited number of regions used in theprevious studies therefore makes the result of thetechniques incomplete, because degrees of poly-morphism can vary substantially across the chlo-roplast genome. Micro- and macro-structuralrearrangements exist in some chloroplastgenomes, for example, small inversions (Hiratsukaet al. 1989), insertions and/or deletions (Ogiharaet al. 1991; Kanno et al. 1993; Maier et al. 1995),base substitutions (Morton and Clegg 1995), andtranslocations (Ogihara et al. 1988), as well asthe large inversions in Oenothera elata (Hupferet al. 2000) and Lotus japonicus (Kato et al.2000).

Hence, the purpose of this study was to inves-tigate the ability of PCR-RFLP to detect inter-speciWc variations in the whole cpDNAs ofcultivated cottons, how the variations were dis-tributed and where, if there, the hyper-variablehotspots in the whole genome are speciWcallylocated.

1 3

Euphytica (2007) 156:47–56 49

Materials and methods

Plant materials

Cotton plants (G. barbadense, G. hirsutum,G. arboreum and G. nanking, the Chinese syno-nym of G. arboreum) were grown under naturalconditions in the experimental farm of the Gradu-ate School of Agriculture, Kyoto University, andNippon Shinyaku Co., LTD, Kyoto, Japan.

DNA isolation

The Plant Genomic DNA Extraction MiniprepSystem (Viogene, USA) was used to extract totalgenomic DNA from leaves collected separatelyfrom at least three plants of each of the testedspecies (four plants of each of G. arboreum,G. nanking and G. barbadense and seven plants ofG. hirsutum). The protocol of the manufacturerwas followed. Extracted DNA was used as a tem-plate for the PCR ampliWcation.

Primers design

Twenty-four pairs of primers were manuallydesigned (Table 1) based on the cpDNAsequence of G. barbadense (Ibrahim et al. 2006).The length of the primer, primer position, C + Gcontent of at least 50%, melting temperature, andprimer-dimer formation factors were considered.These primers were used to amplify the chloro-plast genome of the test species of cotton.

DNA ampliWcation

Long fragments of cpDNA with sizes rangingfrom 5,505 to 14,943 bp (base pair) were ampliWedusing 1.25 units of TaKaRa LA TaqTM polymer-ase (Takara Shuzo, Kyoto, Japan) and speciWcpair of primers. All the recommended protocol ofthe manufacturer was followed except for theamount of reagents, where only half of the recom-mended volumes were used in this experiment.

PCR-ampliWcation of the long fragments wasperformed in an iCycler PCR machine (BIO-RAD, USA) with the following steps: initialdenaturtion step at 94°C for 1 min., followed by30 cycles at two diVerent temperature regimes:

98°C for 10 sec and 68°C for 15 min. The lattercondition was intended for annealing and exten-sion of the primers. The Wnal extension step wascarried out at 72°C for 10 min.

For the ampliWcation of short fragments rangingin size from 952 to 3,328 bp, 1.25 units of KODDash polymerase (TOYOBO, Japan) were used incombination with speciWc pair of primer. Just likein the previous PCR reactions, the manufacturer’sprotocol was followed except for the amount ofreagents where only half of the recommended vol-umes were combined. In contrast to the long frag-ments, short fragments PCR was performed at94°C for 2 min in the initial denaturation step, fol-lowed by 35 cycles at 94°C for 30 sec, and 2 sec at60°C for annealing of primers. The primer exten-sion step was carried out at 74°C for 1–3 mindepending on the length of the fragment and theWnal extension step was done at 74°C for 5 min.

To conWrm successful ampliWcation and todetermine the size of the ampliWed fragments, 2 �lof each of the PCR products were separated byelectrophoresis in 0.8% agarose gels, in0.5 £ TAE buVer. The DNA fragments werevisualized by UV Xuorescence after staining withethidium bromide.

DNA digestion

After ampliWcation, 2 �l of the PCR products weredigested separately in total volumes of 10 �l with:EcoRI, HindIII, MfeI, SphI, PvuII, BglII, and NcoIrestriction enzymes. The manufacturer’s protocolwas followed (New England BioLabs, UK). TheampliWed and digested cpDNA fragments werethen separated in 1.5% agarose gels in 1 £ TAEbuVer. The gels were run at a constant current of500 mA and a voltage of 100 V, for about an hour.The digested products were visualized by UV Xuo-rescence after staining with ethidium bromide. Thesize of the individual fragments was estimated usinga standard 1-kbp (Kilo-base pair) molecularmarker (New England BioLabs, UK).

Results

PCR-RFLP of four cpDNAs from G. arboreum,G. nanking, G. barbadense and G. hirsutum were

1 3

50 Euphytica (2007) 156:47–56

carried out. Each pair of the twenty-four designedprimers successfully ampliWed a single band of thecpDNA from each of the tested Gossypium spe-cies. These ampliWed PCR products were found

similar in size among the four species. Moreover,these bands were found comparable to the calcu-lated size of the corresponding regions in thecpDNA sequence of the standard G. barbadense

Table 1 DNA sequences of the primers used in this study, genes they were situated on, their positions on G. barbadense L.cpDNA, and size of the expected PCR products in base pairs (bp)

Primer Sequence Gene Position Expected size (bp)

F1 GGATCAACTAGGACAGAAATAAAGCATTGGG rpl23 (158,533)R1* GGAGAGATGGCTGAGTGGACTAAAGCGTCGGA S-(GCU) (08,247) 10,031F1* CTGGGACGGAAGGATTCGAACCTCCGCATAGCG Q-UUG (06,983)R1 GGCGCCATATATCTCTGCAAAACGTAAGGG rps2 (17,009) 10,026F2 CGCAGAGCCGTATATTCTTCTTGCTGATCAACG rps2 (17,362)R2 GCTACAGCGCTGTTCATTCTAGTTCCTACCGC psbM (30,821) 13,459F3* CCTTCAGACGTGTCAATTGGGCAAATGCGTCC rpoB (26,411)R3* CGGTAGCACCATGAATAGCGCATAGCAGAGCCGCG psbD (35,153) 08,742FD-LA ACTCGTTGGTGTCAATTAGGTGGTCTGTGGAC psbD (34,838)Sf (r) AACCACTCGGCCATCTCTCCTA S-UGA (37,288) 02,4505-f4 GAGAGAGAGGGATTCGAACC S-UGA (37,217)295Ar GGTAGCTCGCAAGGCTCATA fM-CAU (38,956) 01,7395-f5 GAACCCGTGACCTCAAGGTTATG fM-CAU (38,937)5-r5-1 CGCAGGATTCATCATGACAG psaB (40,635) 01,698F5-5-1(f*) TTCTTGCCAATACATTATCCTCATT psaB (40,557)F5-5-3(r) GTTTGATACCGGATAAAGCAAATCT psaA (42,366) 01,809F4** CAATGCTCAAGGCTCTAGGCTGAGTAGCAGGAGC psaA (41,908)Sr GAGAGGGATTCGAACCCTCG S-GGA (47,413) 05,505Sf TAGGAGAGATGGCCGAGTGGTT S-GGA (47,350)rpr TCCTTCAGTCGATTTCGCTTCA rp-4»T-UGU (48,409) 01,059rpf GAGGAATAGTCGAAGCCATACC rps4 (48,009)Tr GCTGAATCCCTATGGATACC T-UGU»L-UAA (48,961) 00,952Tf CATTACAAATGCGATGCTCT T-UGU (48,904)Or CCGTCTTTCGTCTCTATCGG T-UGU»L-UAA (50,080) 01,176Of CGACCCTTCGAGTATTCCACC T-UGU»L-UAA (49,935)RK-LA GGTACCAAGATCGAGTCCTAGACAGGCAGACC ndhK (53,263) 03,328F4* GGCTAACAGTCCGCCTGGTGCTACATCATAGGC ndhJ (52,446)R4 GAAACGGTCTCTCCAGCGCATAAATGGTTGG rbcL (59,406) 06,960F5 CTGGTACATGGACAACCGTGTGGACCGATGG rbcL (58,970)R5* CCTATTCATCGCGGGTTGGTTATTCGTCAGCAC psbE (68,922) 09,952F5** CCGTGGATATTGAGGTTCCACAAGCGGTACTTCC petA (66,372)R5 GCTTTGGTTTCGGCTCATCGGGATAGAGATAGG rpl20 (73,028) 06,656F6 CTGCATTTATTCGAGTGATCCACAAACGACG rpl20 (72,982)R6 GACTATCTTCTGGTGTCGAGGGAATTGCACG rps3 (87,802) 14,820F7 CGCTCCAATGGTTTGTAGAGGAACCCTACC rps3 (87,414)R7** GCCTTGGTGGTGAAATGGTAGACACGCGAGACTC L-CAA (98,935) 11,521F7* GTTCTGCCGGATCGGTCGCTCAAGACCTTTGG ycf2 (96,942)R7 CGGCTACCTTGTTACGACTTCACTCCAGTC rrn16 (107,119) 10,177F8 GGAGAGTTCGATCCTGGCTCAGGATGAACGC rrn16 (105,695)R8 GAATTCCCCGTAGAAAGAGATTTGGCTAATG rpl32 (118,167) 12,472F9 ACGCACTTCGACATCAAAAAAGCGTATTCG rpl32 (118,087)R9 GGTGTTCTTCGTCTCATCGTTACTCTAGATGG ndhH (128,277) 10,190F10 CCACTTCTCCCATAATGATATCTATGCTACC ndhH (127,932)R10 GCCGAAAGCATCACTAGCTTACGCTCTGAC rrn23 (139,163) 11,231F11 GGACCGAACTGTCTCACGACGTTCTGAACCC rrn23 (136,913)R11 GGAACTTGATGAGTCGATCCGCCTACACGTC rps7 (146,795) 09,882F12 CGTAATCGATTAGTTAACATGTTGGTTAACCG rps7 (146,617)R12 GGTGCCATTATTCCTACTTCTGCAGCTATAGG psbA (01,243) 14,943

1 3

Euphytica (2007) 156:47–56 51

as shown in Table 1. The data further revealedthat regions ampliWed with rpf/Tr primer pair pro-duced the shortest fragment, »952 bp, while thoseampliWed with F12/R12 produced the longestfragment, »14,943 bp. Based on the above results,it is calculated that the total length of the twenty-four ampliWed regions of each of the tested spe-cies was 160 kbp § 500 bp for each species, whichperfectly represent the whole cotton cpDNA asindicated in previous works (Wendel 1989; Leeet al. 2006; Ibrahim et al. 2006).

One visually detectable ampliWed fragmentvariation was observed in the undigested PCRproducts from the four cotton species after sepa-rating them on agarose gel, which was the produc-tion of 5-f4/295Ar primers. G. barbadense gavethe longest ampliWed fragment while G. hirsustumproduced the shortest (Fig. 1). This 5-f4/295ArPCR product was indigestible when exposed tothe seven tested enzymes.

When digested with seven restriction endonuc-leases, EcoRI, HindIII, MfeI, SphI, PvuII, BglII

and NcoI, the ampliWed PCR products from eachof the four tested species of cotton producedabout 555 fragments. Sixteen diVerent fragmentswere detected in the four tested species, while therest of the fragments were similar. The diVerentfragments were detected in the ampliWed anddigested products of primer pairs F1/R1*, F2/R2,F3*/R3* and F5/R5*, which gave PCR productsof 10,031, 13,459, 8,742 and 9,952 bp long, respec-tively.

The amplicon of F1/R1* primer pair producedpolymorphic pattern only when digested withEcoRI but not with the other six restriction endo-nucleases. On the other hand, the ampliWed prod-ucts of F2/R2 and F3*/R3* primer pairs resultedin polymorphic band patterns when digested withEcoRI, HindIII, MfeI and BglII enzymes. For thePCR product of F5/R5* primer pair, only MfeIenzyme gave polymorphic proWle, while the restof the restriction enzymes did not show any poly-morphism.

The digestion of the fragment ampliWed by F2/R2 primers with the use of EcoRI, MfeI or BglIIenzymes resulted in the production of 7, 7 and 5fragments, respectively. In either of the PCRproduct/enzyme combination, only one polymor-phic fragment was noticed (Fig. 2). This polymor-phic fragment showed slight-length diVerencewith the other fragments and this was onlyobserved in G. hirsutum. It is noteworthy that theslight length-diVerence in the fragment ampliWedby F2/R2 could not be detected in the undigestedproducts, probably due to the long size of thefragments; »13,459 bp.

The digestion of the fragment ampliWed byF3*/R3* primers with EcoRI, MfeI and BglII pro-duced 6, 5 and 6 fragments, respectively, whichshowed one diVerent band each. This polymor-phic band is also due to slight-length diVerencein G. hirsutum that seemed bigger than in otherspecies (Fig. 2). Considering these two slightlonger parts in the cpDNA from G. hirsutum andthe loss in the shorter fragment that was earlierdetected in the PCR product ampliWed by 5-f4/295Ar primers, the total cpDNA obtained fromthis species was comparable with the totalcpDNA of the other tested species, which were160 kbp § 500 bp, with no detectable diVerencesin fragment length. That result is in consistence

Fig. 1 PCR products ampliWed by 5-f4/295Ar primers fromG. babrbadense (b), G. hirsutum (h), G. arboreum (a) andG. nanking (n), respectively. The arrow head shows thediVerent ampliWed fragment from G. hirsutum (h)

(1kb) M b h a n

Primers 5-f4/295Ar

10

8

6

5

4

3

2

1.5

1

0.5

1 3

52 Euphytica (2007) 156:47–56

with previous reports (Wendel 1989; Lee et al.2006; Ibrahim et al. 2006).

Another feature of the F3*/R3* PCR ampliWedproduct is the presence of two HindIII sites in theallotetraploid cultivated cotton species G. barba-dense and G. hirsutum, while only one of the twosites was observed in the diploid cultivated spe-cies G. arboreum and G. nanking (Fig. 3). Thegain of the extra HindIII site in the allotetraploidspecies as shown above could possibly be attrib-uted to a point mutation that occurred by base

substitution, or small insertions/deletions, beforethe divergence of the two allotetraploid cottonspecies from each other during the evolution andspeciation process of the genus Gossypium. Onthe other hand, the loss of one HindIII site in thediploid species could possibly also be an occur-rence in the diploid cultivated cottons after thedivergence of the cultivated allotetraploid spe-cies.

Aside from the restriction sites detected above,there were other two instances observed in the

Fig. 2 The PCR products of F1/R1*, F2/R2, F3*/R3* and F4**/R4-1 prim-ers digested with MfeI. The arrow heads show the bands of short length diVerences from G. hirsu-tum (h) in the PCR prod-ucts of F2/R2 and F3/R3*

(1Kb) M b h a n b h a n b h a n b h a n M

Primers F1/R1* F2/R2 F3*/R3* F4**/R4-1

Restriction Enzyme MfeI

10

8

6

5

4

3

2

1.5

1

0.5

Fig. 3 Lanes 1–5 show the PCR product of F1/R1* primersdigested with EcoRI; arrow heads show the fragments pro-duced by the internal EcoRI site in G. barbadense (b) whilethe arrow shows the fragment that lacks the internal site inG. hirsutum (h), G. arboreum (a) and G. nanking (n).Lanes 6–11 show the PCR product of F3*/R3* digestedwith HindIII; the arrow heads refer to the bands resulted

from the internal HindIII site in G. barbadense (b) and G.hirsutum (h) of the band indicated by arrow in G. arboreum(a) and G. nanking (n). Lanes 12–16 show the PCR productof F5/R5* digested with MfeI; arrow heads refer to thediVerent bands in G. hirsutum (h), which were a result of aninternal MfeI site in the band indicated by arrow in G. bar-badense (b), G. arboreum (a) and G. nanking (n)

(1Kb) M b b h a n b b h h a n b h h a n M

Primers F1/R1* F3*/R3* F5/R5*

Restriction Enzyme EcoRI HindIII MfeI

10

8

6

5

4

3

2

1.5

1

0.5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

1 3

Euphytica (2007) 156:47–56 53

PCR products of F1/R1* and F5/R5* primerpairs. In the PCR product of F1/R1* primer pair,an EcoRI site was detected in the allotetraploidG. barbadense but not in the other allotetraploidG. hirsutum or the two diploid species G. arbor-eum and G. nanking (Fig. 3). In contrast, in thePCR product of F5/R5* primer pair, an MfeI sitewas found in the allotetraploid G. hirsutum butnot in the other allotetraploid G. barbadense orthe two diploid species G. arboreum and G. nan-king (Fig. 3). These two restriction sites justdescribed seem to be gained after the divergenceof the two allotetraploid cotton species, becausethey were not detected in the diploid cottonG. arboreum and G. nanking.

When the experiment was repeated three timeswith the use of at least three diVerent plants fromeach species, the same results were observed asshown in Fig. 3. This therefore strongly suggeststhat these restriction sites were not at all due toerrors by the DNA polymerase rather they arenormal occurrence during the evolution process.Three justiWcations can be oVered for this claim:Wrst, the study used high Wdelity DNA polyme-rases; second, no extra restriction sites wereobserved in one of G. arboreum and its synonymG. nanking (Fig. 1, 2, 3 and Wgures not shown);and third, the reproducible results obtained inthree series of experiments and three or morediVerent plants of each species. Furthermore, theclaim can be further supported by the detection ofHindIII site in the allotetraploid cotton speciesand its absence in the diploid cotton species.

Discussion

The use of PCR-RFLP for cpDNA detecteddiVerent features in the four investigated species;G. barbadense L., G. hirsutum L., G. arboreum L.and G. nanking Meyen. Three cpDNA variationswere observed, one ampliWed fragment lengthdiVerence and RFLPs in the form of small frag-ment length diVerences and restriction sites.Therefore it is possible to say the present methodwas capable to detect some inter-speciWc varia-tions of cpDNA in the genus Gossypium L. How-ever, the limited number of species used in thisstudy can only give a Wrst approximation of the

potential of the suggested approach to carry out aphylogenetic analysis. Usually, in phylogeneticstudies, one or few individuals from each speciesare required for intra- and/or inter-speciWc com-parisons of cpDNA variation (Strauss and Doerk-sen 1990; Ziegenhagen and Fladung 1997; Paducciand Szmidt 1999; Lakshmi et al. 2000). In manyprevious studies where large sample sizes wereused, intra-speciWc cpDNA polymorphism couldnot be detected in diVerent species of Pinus andAbies (Wang and Szmidt 1994; Boscherini et al.1994; Vicario et al. 1995).

In this study, twenty-four pairs of primers wereproved capable of amplifying the cpDNAs fromthree cultivated species of the genus Gossypium,they are also expected to amplify the cpDNAfrom the fourth cultivated species, G. herbaceumL. The digestion of the twenty-four resulted PCRproducts with seven restriction enzymes revealedsixteen diVerent fragments in the cotton speciestested. This inter-speciWc comparison result showsthat the cpDNA in the genus Gossypium is highlyconserved, although it may be more polymorphicat least at the level of the same genomic group.

Most of the recent PCR-RFLP studies havefocused on only few polymorphic regions of thecpDNA (Powell et al. 1995; Cato and Richardson1996; Vendramin and Ziegenhagen 1997; Veke-mans et al. 1998; Lakshmi et al. 2000; Wei et al.2005). This study explored the possibility of Wnd-ing polymorphic regions other than what has beenreported in earlier studies. Based on the results, itcan therefore be concluded that the whole chloro-plast genome PCR-RFLP can be adopted. UsingPCR-RFLP it was possible to detect polymor-phism at the inter-speciWc level suggesting thatthe chloroplast genome of the genus Gossypiummay contain variable regions that appear to behidden throughout the whole genome. The use ofmore sensitive gel resolutions, higher gel concen-trations, smaller markers and more restrictionenzymes, may give more details of polymorphism.The pattern of cpDNA variations found in thisstudy reveals a close relationship between G. bar-badense, G. hirsutum, G. arboreum and G. nan-king, which is concordant with previous studies(Wendel 1989).

Referring to this work, it is possible to sayPCR-RFLP of the whole cpDNA method has the

1 3

54 Euphytica (2007) 156:47–56

ability and reliability to detect approximate varia-tions at least at the inter-speciWc level. Moreover,small length-diVerences in long PCR products areeasier to detect with the use of restriction diges-tion. There is also a possibility to track other vari-ations in other parts of the cpDNA with theadoption of this method and more sensitivity.

Acknowledgements This work was supported by aGrant-in-Aid (No 020518) from the Ministry of Education,Science, Sports and Culture of Japan for R. I. Cotton seedswere a gift from Nippon Shinyaku Co., LTD (Kyoto,Japan) to whom we send sincere gratitude. We would liketo thank Dr. Abdelbagi Mukhtar, Gezira Research Station,Sudan, and Dr. Shirley Agrupis, Mariano Marcos StateUniversity, Philippines, for reading the manuscript.

Reference

Arroyo-Garcia R, Lefort F, de Andres MT, Ibanaez J,Borrego J, Jouve N, Cabello F, Martinez-Zabater JM(2002) Chloroplast microsatellite polymorphisms inVitis species. Genome 45:1142–1149

Badenes ML, ParWtt DE (1995) Phylogenetic relationshipsof the cultivated Prunus species from an analysis ofchloroplast DNA variation. Theor Appl Genet90:1035–1041

Beasley JO (1940) The origin of American tetraploid Gos-sypium species. Am Nat 74:285–286

Birky CW (1988) Evolution and variation in plant chloro-plast and mitochondrial genomes. In: Gottlieb L, JainS (eds) Plant evolutionary biology. Chapman and Hall,London pp 23–53

Boscherini G, Morgante M, Rossi P, Vendramin G (1994)Allozyme and chloroplast DNA variation in Italianand Greek populations of Pinus leucodermis. Heredity73:284–290

Cato SA, Richardson TE (1996) Inter- and intra-speciWcpolymorphism at chloroplast SSR loci and the inheri-tance of plastids in Pinus radiata D Don. Theor ApplGenet 93:587–592

Claudia C, Tonguc M, GriYths PD (2004) Discriminationof diploid crucifer species using PCR-RFLP of chloro-plast DNA. HortScience 39:1575–1577

Clegg MT, Zurawski G (1992) Chloroplast DNA and thestudy of plant phylogeny: present status and futureprospects. In: Soltis PE, Soltis DE, Doyle JJ (eds)Molecular systematics of plants. Chapman and Hall,New York, USA, pp 1–13

Cronn RC, Small RL, Haselkorn T, Wendel J (2002) RapiddiversiWcation of the cotton genus (Gossypium: Malv-aceae) revealed by analysis of sixteen nuclear andchloroplast genes. Am J Bot 89:707–725

Demesure B, Comps B, Petit RJ (1996) Chloroplast DNAphylogeography of the common beech (Fagus sylvati-ca L.) in Europe. Evolution 50:2515–2520

Demesure B, Sodzi N, Petit RJ (1995) A set of universalprimers for ampliWcation of polymorphic non-codingregions of mitochondrial and chloroplast DNA inplants. Mol Ecol 14:129–131

Dhingra A, Folta MK (2005) ASAP: ampliWcation,sequencing & annotation of plastomes. BMC Genom-ics 6:176 doi:10.1186/1471-2164-6-176

Dumolin S, Demesure B, Petit RJ (1995) Inheritance ofchloroplast and mitochondrial genomes in peduncu-late oak investigated with an eYcient PCR method.Theor Appl Genet 91:1253–1256

Endrizzi JE, Turcotte EL, Kohel RJ (1985) Genetics, cytol-ogy, and evolution of Gossypium. Adv Genet 23:271–375

Fryxell PA (1979) The natural history of the cotton tribe.Texas A&M Univ Press, College Station

Fryxell PA (1992) A revised taxonomic interpretation ofGossypium L. (Malvaceae). Rheedea 2:108–165

Gielly L, Taberlet P (1994) The use of chloroplast DNA toresolve plant phylogenies: noncoding versus rbcL se-quence. Mol Biol Evol 11:769–777

Govindaraju D, Lewis P, Cullis C (1992) Phylogenetic anal-ysis of pines using ribosomal DNA restriction frag-ment length polymorphisms. Plant Syst Evol 179:141–153

Heinze B (2005) A database for PCR primers in the chlo-roplast genome. http://www.bfw.ac.at/200/1859.html

Hiratsuka J, Shimada H, Whittier R et al (1989) The com-plete sequence of the rice (Oryza sativa) chloroplastgenome: intermolecular recombination between dis-tinct tRNA genes accounts for a major plastid DNAinversion during the evolution of the cereals. Mol GenGenet 217:185–194

Huang JC, Sun M (2000) Genetic diversity and relation-ships of sweetpotato and its wild relatives in Ipomoeaseries Batatas (Convolvulaceae) as revealed by inter-simple sequence repeat (ISSR) and restriction analysisof chloroplast DNA. Theor Appl Genet 100:1050–1060

Hupfer H, Swiatek M, Hornung S, Herrmann RG, MaierRM, Chiu WL, Sears B (2000) Complete nucleotidesequence of the Oenothera elata plastid chromosome,representing plastome I of the Wve distinguishableEuoenothera plastomes. Mol Gen Genet 263:581–585

Ibrahim RIH, Azuma JI, Sakamoto M (2006) Completenucleotide sequence of the cotton (Gossypium barba-dense L.) chloroplast genome with a comparative anal-ysis of sequences among 9 dicot plants. Genes GenetSys 81:311–321

Ishii T, Xu Y, McCouch SR (2001) Nuclear- and chloro-plast-microsatellite variation in A-genome species ofrice. Genome 44:658–66

Kanno A, Watanabe N, Nakamura I, Hirai A (1993) Vari-ation in chloroplast DNA from rice (Oryza sativa):diVerences between deletions mediated by short di-rect-repeat sequences within a single species. TheorAppl Genet 86:579–584

Kato T, Kaneko T, Sato S, Nakamura Y, Tabata S (2000)Complete structure of the chloroplast genome of alegume, Lotus japonicus. DNA Res 7:323–330

1 3

Euphytica (2007) 156:47–56 55

Krupkin AB, Liston A, Strauss SH (1996) Phylogeneticanalysis of the hard pines (Pinus subgenus Pinus, Pin-aceae) from chloroplast DNA restriction-site analysis.Am J Bot 83:489–498

Lakshmi M, Senthilkumar P, Parani M, Jithesh MN, ParidaAK (2000) PCR-RFLP analysis of chloroplast gene re-gions in Cajanus (Leguminosae) and allied genera.Euphytica 116:243–250

Lee SB, Kaittanis C, Jansen RK, Hostetler JB, Tallon LJ,Town CD, Daniell H (2006) The complete chloro-plast genome sequence of Gossypium hirsutum:organization and phylogenetic relationships to otherangiosperms. BMC Genomics 7:61 doi:10.1186/1471-2164-7-61

Maier RM, Neckermann K, Igloi GL, Kössel H (1995)Complete sequence of the maize chloroplast genome:Gene content, hotspots of divergence and Wne tuningof genetic information by transcript editing. J Mol Biol251:614–628

Morton BR, Clegg MT (1995) Neighboring base composi-tion is strongly correlated with base substitution bias ina region of the chloroplast genome. J Mol Evol 41:597–603

Ogihara Y, Terachi T, Sasakuma T (1988) Intramolecularrecombination of chloroplast genome mediated byshort direct-repeat sequences in wheat species. ProcNatl Acad Sci USA 85:8573–8577

Ogihara Y, Terachi. T, Sasakuma T (1991) Molecular anal-ysis of the hot spot region related to length mutationsin wheat chloroplast DNAs: I. Nucleotide divergenceof genes and intergenic spacer regions located in thehot spot region. Genetics 129:873–884

Palmer JD (1987) Chloroplast DNA evolution and biosys-tematic uses of chloroplast DNA variation. In: CleggMT (ed) Plant molecular evolution. Am Nat 130(Sup-plement):6–29

Parani M, Lakshmi M, Ziegenhagen B, Fladung M, Sent-hilkumar P, Parida A (2000) Molecular phylogeny ofmangrove VII. PCR-RFLP of trnS-psbC and rbcLgene regions in 24 mangrove and mangrove associatespecies. Theor Appl Genet 100:454–460

Parani M, Rajesh K, Lakshmi M, Parducci L, Szmidt AE,Parida A (2001) Species identiWcation in seven smallmillet species using polymerase chain reaction restric-tion fragment length polymorphism of trnS-psbC generegion. Genome 44:495–499

Parducci L, Szmidt AE (1999) PCR-RFLP analysis of cpD-NA in the genus Abies. Theor Appl Genet 98:802–808

Parducci l, Szmidt AE, Madaghiele A, Anzidei M, Vendr-amin GG (2001) Genetic variation at chloroplast mi-crosatellites (cpSSRs) in Abies nebrodensis (Lojac.)Mattei and three neighboring Abies species. TheorAppl Genet 102:733–740

Powell W, Morgante M, McDevitt R, Vendramin G, Rafal-ski J (1995) Polymorphic simple sequence regions inchloroplast genomes: applications to the populationgenetics of pines. Proc Natl Acad Sci USA 99:7759–7763

Ridgway KP, Duck JM, Young JPW (2003) IdentiWcationof roots from grass swards using PCR-RFLP and FFLP

of the plastid trnL (UAA) intron. BMC Ecol 3:8. doi:10.1186/1472-6785-3-8

Skovsted A (1934) Cytological studies in cotton. II. TwointerspeciWc hybrids between Asiatic and New Worldcottons. J Genet 28:407–424

Skovsted A (1937) Cytological studies in cotton. IV. Chro-mosome conjugation in interspeciWc hybrids. J Genet34:97–134

Soltis DE, Gitzendanner MA, Strenge DD, Soltis PS (1997)Chloroplast DNA intraspeciWc phylogeography ofplants from the PaciWc Northwest of North America.Plant Syst Evol 206:353–373

Stehlik I (2002) Glacial history of the alpine herb Rumexnivalis (polygonaceae): a comparison of common phy-logeographic methods with nested clade analysis. AmJ Bot 89:2007–2016

Stehlik F, Blattner R, Holderegger R, Bachmann K (2002)Nunatak survival of the high Alpine plant Eritrichiumnanum (L.) Gaudin in the central Alps during the iceages. Mol Ecol 11:2027–2036

Stewart J McD (1995) Potential for crop improvement withexotic germplasm and genetic engineering. In: Consta-ble GA, Forrester NW (eds) Challenging the future.Proceedings of the World Cotton Research Confer-ence-1. CSIRO, Melbourne Australia, pp 313–327

Strauss SH, Doerksen AH (1990) Restriction fragmentanalysis of pine phylogeny. Evolution 44:1081–1096

Taberlet P, Gielly L, Pautou G, Bouvet J (1991) Universalprimers for ampliWcation of three non-coding regionsof chloroplast DNA. Plant Mol Biol 17:1105–1109

Tsai LC, Yu YC, Hsieh HM, Wang JC, Linacre A, Lee JCI(2006) Species identiWcation using sequences of thetrnL intron and the trnL-trnF IGS of chloroplast ge-nome among popular plants in Taiwan. Forensic SciInt (in press) doi:10.1016/j.forsciint.2006.01.007

Tsumura Y, Yoshimura K, Tomaru N, Ohba K (1995)Molecular phylogeny of conifers using RFLP analysisof PCR ampliWed speciWc chloroplast genes. TheorAppl Genet 91:1222–1236

Tsumura Y, Kawahara T, Wickneswari R, Yoshimura K(1996) Molecular phylogeny of Dipterocarpaceae insoutheast Asia using RFLP of PCR-ampliWed chloro-plast genes. Theor Appl Genet 93:22–29

Van Droogenbroeck B, Kyndt T, Maertens I, Romeijn-Pe-eters E, Scheldeman X, Romero-Motochi JP, VanDamme P, Goetghebeur P, Gheysen G (2004) Phylo-genetic analysis of the highland papayas (Vasconcel-lea) and allied genera (Caricaceae) using PCR-RFLP.Theor Appl Genet 108:1473–86

Vekemans X, Hardy H, Berken B, Fofana B, Baudoin JP(1998) Use of PCRRFLP on chloroplast DNA toinvestigate phylogenetic relationships in the genusPhaseolus. Biotechnol Agron Soc Environ 2 (2):128–134

Vendramin G, Ziegenhagen B (1997) Characterization andinheritance of polymorphic plastid microsatellites inAbies. Genome 40:857–864

Vicario F, Vendrami G, Rossi P, Liò P, Giannini R (1995)Allozyme, chloroplast DNA and RAPD markers fordetermining genetic relationships between Abies alba

1 3

56 Euphytica (2007) 156:47–56

and the relic population of Abies nebrodensis. TheorAppl Genet 90:1012–1018

Wang XR, Szmidt AE (1994) Hybridization and chloro-plast DNA variation in a Pinus species complex fromAsia. Evolution 48:1020–1031 807

Wang GZ, Matsuoka Y, Tsunewaki K (2000) Evolutionaryfeatures of chondriome divergence in Triticum(wheat) and Aegilops shown by RFLP analysis ofmitochondrial DNAs. Theor Appl Genet 100:221–231

Wendel JF (1989) New World tetraploid cottons containOld World cytoplasm. Proc Natl Acad Sci USA86:4132–4136

Wei W, Youliang Z, Li C, Yuming W, Zehong Y, Ruiwu Y(2005) PCR-RFLP analysis of cpDNA and mtDNA inthe genus Houttuynia in some areas of China. Hered-itas 142:24–32

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotidesubstitution vary greatly among plant mitochondrial,chloroplast, and nuclear DNAs. Proc Natl Acad SciUSA 84:9054–9058

Xu DH, Abe J, Kanazawa A, Gai JY, Shimamoto Y (2001)IdentiWcation of sequence variations by PCR-RFLPand its application to the evaluation of cpDNA diver-sity in wild and cultivated soybeans. Theor Appl Gen-et 102:683–688

Ziegenhagen B, Fladung M (1997) Variation in the psbCgene region of gymnosperms and angiosperms as de-tected by a single restriction-site polymorphism. TheorAppl Genet 94:1065–1071

Ziegenhagen B, Kormutak A, Schauerte M, Scholz F (1995)Restriction-site polymorphism in chloroplast DNA ofSilver Fir (Abies alba Mill.). For Genet 2:99–107

1 3