TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics ISSN 1346-7565 ActaPhytotax,Geobot.62(1):1-(124011) Genetic Variation inthe Apogamous Fern Cyrtomium fo rtunei (Dryopteridaceae) Ryo OoTsuKII･"W,AI[ARuSHINOHARA2,TAKESHISUZUKI3AND NoRIAKIMuRAKAM]i ifbk.vo Mbtropotitan Uhiversic" Mdkino Herbarium, Minamiosai-, 1a-i ,Hinchioj 7ib,kyo i92-039Z .1apan. "[email protected],ac,ip (tiuthorf eoorrrespondencel; 22k3poto Uitivenyit pDEuJartmen qifBotaJu Graduate School qfScience, Kitashirakaiva-Oi)vake cShaok,yoku ,k5,ot o606-8502, lapan; `thtseum of ,Nintu raned HumanAetivities, Iij,ogo ,Division ofBiolagicalResou rIZcreysoi,gaoka ,Sanda 669-i546 l.apan In apogamous ferns a,ll offspring frem aparent are expected to be e]onal, However, apogamous specics fi'equentl yshow a !arge amount ofmorpholegical and geneti vcariatlon, Cyrtomiumfortunei and it srela- tives ,which are distributed threughout Japan, are reported to be apogamous tripleids b,ut shew 1arg eand continueus morphological variation. Four varieties of C. fortun heaive been recogmized, We sought to determine whether an apogamous specics has genetic variation, and if so, whether this variation relates to morphologieal variation within loca lpopulation sA.meng 224 individua lgsrowing in four distantly located populations in Japan, where several varieties grew together ,two rbeL types (a and B) and eight allozymetypes (A-H )were identifi eSde.vera ]differe ngteneti cclones "Tere detected in the localpopula- tions examined, Only individuals that could be morphologically identifi easd C. fortun vear.i interme- dium, based on bjcolore idndusi ah,ac lrbcL B and allozyme type H, and thus, were genetical dliy fferenti- ated by their nuclear and plasti gdenomes fi:o mthe other three varieties ef C. fortttn Tehie. other three varieties shared the same rbcL (t tma)k,ing correlation between allezyme types and varieties, especjally geneti cdifferenc ebsetween var. fortun anedi var. ctivicola, unclear, Key words: allozyme, apogamy, c]onal diversit Cy,yrtotniunz.f goerntetuin cveariia,tion, populatio gne- 'netics, pbcL ApogamM in the broad sense, or agamospory Darnaedi et al. (199 r0ep)erted the absence of in ferns ,is a type of asexual reproduction in allozyme variation within Dr:yopter iysakusilvi- which unreduced spores are fbrmed. The resul- cola Sa. Kurata ,a triplojd apogamous species of tant gametophytes produce sporophytes of the recent hybrid origin endemic to Yaku Island .Ab- next generatio nwithout fertilizat i(onMantonsence of allozyme variation, however ,is uncom- 1950) .Apogamous reproduction is common in mon. In spite of the clonal nature of apogamous ferns .Approximately 10% of the world's fern reproduction, many apogamous fer nspecies species (Lovi 1'9s77) and 15% of Japanes efern show larg emorphological and some geneti cvari- taxa have apogamous rcproduct'ion (1fakamiyaation (Watan o& Iwatsuki 1988, Suzuki & Iwat- 1996) .Jn apogamous ferns ,all offspring from a su k i 1 9 9 0 , L in e t a l. 1 9 9 5 , T ak a m i y a et /al. 2001). sporophyte are genetical ltyhe same unless muta- Previous studies analyzing genetic variation in tions occur during reproduction. The amount of apogamous ferns using enzyme elcctrophoresis geneti cvariation within an apogamous species is reported 45, 4, and 14 ditfere nctlones from D. therefbre cxpected to be very low, especially in nipponensis (Ishika ewta aL 2003al ,Aspieniune apogamous species not of recent and recurring unilaterale (Watan o& Iwatsuk i1988) ,and D. 'l Orlglll. bissetiana (Li net al. 995), respectively. Further- NII-Electronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics 2 ActaPhytotax.Geobot. Vol.62 more, geneti cvariation in combination with cyto- anterior base of the pinnae (Tab ll)e ,Among the logica vlariation has been reported in apogamous fbur varieties of C. fortun evair., fortun heais the ferns ,Six differen ttriploi dand fiv ediplo icdlones highes tnumber of pair s(15-3 o0f) pinnae .Vari- have been reported in Pteris cretica (Suzu k&i eties clivicola, intermedit{m and atropunctatum Iwatsuki 1990) and one tetraploi dand four trip- have 5-20, 10-15, and 10-20 pairs ,respectjve]y. loid clones have been reported in Diplazium do- The color ofthe indusi aof varg. intermediutn and eclerleiitii (Takami yeta al. 2001). atropunctatum is grayis hwhite with a dark Cyrtomiu Cm, Pres l(Dryopterida ccehaar-e),brown center, The indusi ain the other two variet- acterized by pcltate indusi aand anastomosing ics are comptetely grayis hwhite. Var. intermedi- veins that fbrm areolae with include dvein]ets, {s um has an auricle at the anterior base of eaeh pin- a genus of Asian fbrn scomprising approximately na, which is absent in var, atropunctatune. Vars. 40 q,rtomium fortunei have species. contains many apoga- and clivicola sometimes auricles, C. [fagawa but mous species, such as macroph.yllum not always. (Kuri t1a96Z Hirabayash i1970) ,C, ca,:yotideum The numbers ofpajrs ofpinnae overlap among C, Presl (Kuri !t9a66, Mitsuta 1.98 6M,atsumoto thc varieties. Because morphological variation is 1976) ,anct C fortu nJe. Sim. (Kuri 1t9a60, Mitui continuous and individua lswith intermediate 1968, Mitui 1980, Hirabayash i1970 ,Matsumoto morphology or various combinations of the & Shimura l985) .In terms ofreproductive modes above-mentioned merphotogical characteristics and cytotypes of C. ,fbrtunei ,Nakato et al, (1995)exist, it is very difficu tlot distingui sahmong the reported a diplo isdexual type from China that four varieties. Furthermore ,even within local has 82 somatic chromosomes and 64 spores in populations of C)?rfomi fuomrtun eiin Japan, each sporangium, in addition to a triploi dapoga- larg eand continuous morphological variation js mous type, which has 123 somatic chromosomes observed. Since C. fortt tins eapoigamous, it is and 32 spores in each sporangium. Unti lnow, expected to have low or discontinue uisntraspe- however, only apogamous triploid shave been re- cific variation, but the actual situation is very difi inC,,fortuneiinjapan(Takamiy1a996), ferentfrom corded expectations. In the most recent Flora of.lap abyn lwatsuki Because it is difficu tlot identif tyhe varieties (1992C)l,yrtomium.foTtuHei Ortomaiumfortunei s.lat, was character- of on morphological charac- ized by thin chartaceous to papyraceous pinnae teristics ,we sought to determine the geneti cbasis with dentat edista lmargins, the number ofpairs of their difference sFu.rthermore, it was impor- of pinnae usually more than 10, and pinnae less tant to clarify the relationship between morpho- than 4 cm wide. Four varieties ofC fortun eiin-, logica vlariation and geneti cclones, cluding var,,fortunei, vat; clivicoia, var. interme- In the present study, fresh leaf samples were dium, and var. atrqpunctattim have been recog- co]lected from mature individua losf Cvrtomium nized based on differefi ciens (1 )the number of fortu ni'en ifour distant llyocate dpopulation sin pair sofpinnae, (2 )the coler ofthe indusium ,and Japan. In all population si,ndividua ltshat could (3 )the presence or absence of an auricle on the be identifi etod the differe nvtarieties of C, fortu- TABT.E 1. Distinctive niorphological charactei'istics of four varieties of Cyrtomrt{m.rZ )sernstuu nleaiif io" Japan. Number of paii 'osf Color of center of Presence or abscnce ef auricle ptnnae indusium at antcTior base ofpinnae q,rtontiunlfortunei var,.fi)rtunei 15-305-2010-!510-20 grayish whitc prcsent/'absent var. ctivicota grayish whitc presentlabsent var, intermedium dark brown present var. atropunctat'um dark brown present NII-Electronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics June2011 OoTsuKI & AL.-Gcnetic Variation in (lyrtomittmfortunei 3 nei grew together. We firs dtetermined the gello- varicties correlatcd with morphological charac- types of a]1 samples, then distinguish ecdlones teristic so,r do the four varieties correspond to using the rbcL sequence and allozyme variation differe gnetneti cclones? as geneti cmarkers, In asexually reproducing or- ganisms ,the minimum biologic aunlit is the ge- Materials and Methods netic clone. We therefore concluded that the most Study eracient method of aAalyzing apogamous speci ¢s sites andsaiuple cotleetions such as C. fortu nwaesi use genetic markers to Samples of Clyrtomiu mfortu nuseedi in this identif cylenes growing within loca lpopulations, study were col]ected from four population sin Ja- Such an intensiv geeneti canatysis ofapopul ation pan: Moroyama, Saitama Pref; Kawazu, Shizuo- of apogarnous fern shas never been performed. ka Prefi K;obe, Hyogo Prcf; and Fuchu, Hiroshi- Allozyme variation is a powerful codominant ma Prcf (Fi g1,) .In cach population m,any moF marker- for populatio ngeneti canalyses of wild phological variants, which could be assigned to plants .Nucleotide sequencc variation in rbeL, scveral differe nvatrieties of C. fortun egrie,w to- which is encoded in chloroplast DNA and is ma- gethcr. ternally inherite idn ferns ,is a usefu1 tool fbr de- In general, whole fres hleaf samples were col- termining maternal races in loca ]fern popula- lecte dfrom mature individua lisn each popula- tions (Yata bete al. 1998, Murakami et aL1999). tion. In the Moroyama, Kobe, and Fuchu popula- Lu et al, (2005 s)equenced the chloroplast DNA tions, quadrats were establishcd and the location rhcL and trnL-F rcgions of 19 species of the of cach individua wlas rccorded befbre the sam- dyrtomiu min,cludin gC fortun eanid, cight spe- ples were collccted. In the Moroyama population, cies of related genera to establish the molecular 51 mature individua lsof Clyrtomiu mfortunei phylogeny, They did not, however, analyze the were examined in an area of 60 x 20 m. In the varietjes of C fortun eTihe, rhcLs firo mthe four Kobe popu]atio n,mature leave swere collected C.fortunehaive been from83individua]lsn 100 20 varieties of not analyzed. an area of × m and After identifyi ntghc genetic clones, we com- from an additional 24 nearby individual sI.n the pared morphological variation between the same Fuchu populatio nl,eaf samplcs from 55 individu- and differe ngtenetic types within loca lpopula- als were collected in an area of50 × 50 m. Quad- tions and between diffbre pnotpulations. rat size differe damong populati'on sbecause Each geneti ctype that could be distinguishedquadrat swere set so that the total nurnber ofsam- by rbcL and/or allozyrne markers was tentativety ples in each quadrat was approximately 50, de- recognized as a separate clone. It should be noted spite diffbrin dgensiti eosfindividuals, The den- that geneti ctypes differin ign any geneti cmarker sity of plant sof C. foJ"tu in'n etihe Kawazu popu- belong to differe nctlones, but types with identi- latio nwas too low to establish a quadrat ,IR all cal genotypes do not nccessarily belong to the populations, the environment, such as the amount same clone. of. di're csutnlight (amoun otf shade), moisture in In this study, the fbl]owin gfour questions the soil, and topographical feature swere ob- were specifically considered: (1 )Do several dir served and recorded for each sample. feren tgeneti cclones exist in each population We identif itehed four varieties of C);rtomium where several morphological varieties grow to- fortu none ithe basi sof the distineti vmoerpho- gether? (2 )If several genetic clones do exist in logica clharacters shown in Table 1. We tentative- each local populatio nh,ow many clones are pres- ly identifi eidndividua iwsith more than 15 pairs ent ln each population? .(3 )Are the same clones of pinnae as var. fortun eanid those with 15 or (individ nsaholwisng the same genetic type) fewer pairs ofpjnnae as var. clivicola, Clyrtomium shared, ¢ven ameng remote population so,r are fortu nvaer,i clivicola and var. intermedizam were differen tclones restricted to a local population? found in the Moroyama and Kobe population sal,- (4 )Are genetic difference osbserved among thc though those varieties were not always distin- NNIII-IE-leEcltreoncitcronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics 4 ActaPhytotax.Gcobot. Vbl.62 ma "';o- lilo 200 400 600 800 1OOOkm 'fbur FIG. 1. Location ofthe study sitcs and nurllbcr ofindividuals examined in each population. guishabi edue to centinuity in their morphologi- leaf samples using 2x hexadecyltrimethylammo- cal variatiofi. Only var. fortun eanid var, cgivi- nium bromide (CTAB )solution according to the colum were in the Fuclt upopulation; vaT. inter- method of Doyle and Doyle (l99 0wi)th some medium was not found. Var. atropunetatum was modifications, In brieL 100 mg of dried leaves also in the Kawazupopulation, in addition to var. were ground into a fin epowder using a TissueLy- fortun aendi var. c:livicota. ser (Rets cQhIAGEN, Germany), Next, 500 uL of 'Iiwo pinnae were removed from each leaf 2x CTAB solution was added, mixed, and heated sample. One pinna was maintained as a fresh leaf at 550C for 20 min. Afier chloroform-isoamyl al- sample at 4eC in the refrigerator until it was used cohol (24: e1xt)raetion, isopropano plrecipitation ibr allozyme and ploidy analyses, The other pin- was performed .The DNA pellet obtaincd was na was dried with silica gel and kept in a sma]1 washed with 70% ethanol. air dried ,and redis- plasti bcag at room temperature until it was used solved in 50 ptL TE buffer P,CR amplification of for DNA extraction. Fresh root samples were col- the rbcL gene was performed usingNova Taq }lot lecte dfrom representative individua lsin the Star tDNA Polymerase (Novage Mna,dison, WI), Kobe populatio nT.he remaining portion of the 1× Ampdirect Plus Buffer (Shimaz uKy,eto} Ja- leaves was dried in newspapcrs to serve as a pan) ,and the af and cr prim.crs of Hasebe et al. voucher specimen. All voucher specimens have (1994 I)n .additi.on, four original interna lprimers been deposite din the Makino Herbarium (MAK) dF (5LGGTGTTGGATTCAAGCTGGIL3' )d,R of Tokyo Metropelitan University. (5LGAGCCTG[EACGCAAGCTTCT-3'), e2F (5'-GCGGTGGACTTGATTTCACA-3 'an)d , e2R DIVI telxtraction andsequenc:ing ofrbc gLene (5LGACAATTGGTGCACCCAACT-3' w)ere dc- [[bta DlNA was extracted from si1ica gel dried veloped in this study and used for sequencing, NII-Electronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics Junc 2011 OOTSUKI & AL.-Genetic Variati oinn C.vriomiumforiwnei 5 The conditions fbr PCR amplificatiens were as Shira sihi (1988 W)b. examined phosphoglucom u- fo11ows: initia ldenaturation at 950C for 10 min, tase (PGM) 6,-phosphogluconat deehydrogenase 45 cycles at 940C for30 s, at 580C fbr 30 s, and at (6PG) h,exokinas e(HK) ,and leucin eaminopep- 720C fo r90 s. The PCR product swere incubated tidase (LAP) f,o11owin gthe procedures of Shirai- at 370C for 30 min and at 800C fbr 20 min with shi (1988 )Lo,ci were numbercd, with the most 5% ExoSAP-IT (USB ,Cleveland, OhiQ) to re- anodal locus as 1 and progressively more cath- rnove single-stranded DNA. For cycle sequencing odal loc iwith higher numbers. Allele swere simi- reactions, a BigDyc Terminator kit version 3,1 larl yindicate dat each locus ,with the most an- CAppli eBidosystems ,Foster City ,CA) was used. odal fbrm designate d"d' and progressively slow- The reactlon mixtures were analyzed on an auto- er forms as "b," "c," and so on. mated DNA sequencer model 3100 (Appli Beido- systems), The nucleotide sequences obtained Estimation qfreproductiv meode were aligned using ClustalX2 (Lark ietn aL The reproductive mode (apogamo uors sexu- 2007). al) in most homosporous ferns can be estimated simply by counting the number of spores in a spo- Mblecularphylogenetic analysis rangium. Sexual ferns usually have 64 spores per To compare the rhcL sequences obtained from sporangium, whereas apogamous ferns have 32 Japanese Cyrtomiumfortzan ethiowsei otbthaincd (Manto n1950) .We counted the number of spores by Lu et aL (2005 f)rom Chinese pla'nt sm,axi- per sporangium in the voucher specimens from mum parsimony (MP) analysis was performed the above four localitie tso determine the repro- using MEGA 5 software (http w:wwl.mlegasoft- ductiv emode. ware.net!). The MP tree NNras obtained using the Close-Neighbor-Interchange algorithm with Clytotogica lobservation andploicly analysis us- search leve l1 in which the ini't tiraeels were ob- ing.17b cwytometry tained with randem addition of sequences (10 Mitoric chromosomes were observed in mate- replicates). Analysis involved the 2 nucleotide se- rials collected from six plants w,hich showed die quences of rbcL firo mJapanese plant sas weli as feren tgenetic types, in the Kobe population, In the 16 rbcL sequences reportedby Lu et ai, (2005)this population, the larges tnumber of different from C. macrqpdyllum, C, omeiense, C, uropIryl- genetic types was identifi ebdy rbcL sequencing lum,C. C. C,fortunei, Therefbre, cap:yotideum, aeguibasis, and allozyme analysis. cytological ob- C devexiscapul Ca.e f,alcatu Cm., nephrolepioi- servations and ploidy analyses of (lyrtomiumfor- dbs, C. shingianum, C. hemionitis ,C. grossum, tunei were performed en this population. Root C. chingianum, C. tonehitoicl easnd, C. gui- tips were pretreated with O,O02 M 8-hydroxy- zhouense from China. A total of 1237 positions quinoline fbr 6 h at approximately 200C, After were in the final dataset of the TbcL sequences, fixation in 45% acetic acid for 15-30 min, the root Polystichum gonchit iwsas used as an outgroup. tips were hydrolyze din 1 N HCI at 600C and then Bootstrap analysis with 1000 replicates was per- squashed in 2% aceto-orcein. Chromosomes formed to evaluate internal support for the trees were observed with an Olympus BX-41 micro- obtained. scope and photographed with an Olympus DP-50 digital camera. Alloayrn eanalysis Furthermore, to determin ethe ploidy leve lof Fresh pinnae were ground in 1 ml of cold ex- individua lgsrowing in the Kobe population in traction buffe r(pH = 7.5 )containing O,1 mM detai lt,he DNA content ofeach nucleus extracted Tris-HCI ,1 mM EDTA-4Na, 10 rnM KCI, 10 mM from fresh pinnae samples was measured by flow MgC12, O.4% 2-mercaptoethanol, and 10% poly- cytometry using a CyFlow Ploidy Analyzer vinylpyrrolidone. Enzymes were resolved on 69t6 (Part eMcun,ster, Germany) as well as a Cystain polyacrylamide gels fo11owin gthe procedures of UV Precis eP kit (Part Meucn,ster, Germany), A NNIII-IE-leEcltreoncitcronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics 6 ActaPhytoiax,Geobot. Vbl.62 sample lea fsegment approximately 5 × 5 mm Database of Japan under acccssion numbers was finel ychoppcd with a razor blade in O,5 ml ef AB598689 and AB598690, respectively. Seven nucleus extraction buffe rfrom the kit ,then fi1- base pair substitutions were observed between tered thrQugh a 50 psm mesh and stained with 1 ,5 the two sequences. Leaf samptes collected from ml ofstaining bu ffe rcontaining 4Z 6'-diamidino- the Moroyama and Kobe population sshewed 2-pherrylind ohlyedrochlorid ferom the kit .Fresh both a and P rbcL sequences (Tab l2e). In con- leaf tissues of a diploi dindividua olf C. fortuneitrast ,samp]es collected from the Kawazu and Fu- collected in China (Nakat eot al. 1995) and an- chu population sshowed only sequence tt, othcr triploi dindiv{dual from the Kobe popula- tion, whose ploidy leve lwas determined from .4{tblecuiaTpbylogeneticanalysis chromosome counts during this study, were used Molecular phylogeneti canalysis of rbcL se- as eontrols, First ,the DNA content per nucleus quences obtained from (;),rtomitt mwas per- was measured for the diploi dand triploid con- forme dby the MP method to infer the phyloge- trols. We then analyzed 80 samples collected netic positio onf the sequences a and B .One of the from the Kobe populatio nunder the same condi- 30 most parsimoniou strees <leng t"=h 78) ob- tlons. tained in this study is shown in Fig .2 with boot- strap probabiliti Ietss .consistency index was O.72 Adorphologicalobservation and retention index was O.gO, The rbcL sequence After identifyi ntghe allozyme types, mor- S from Japanese C. fortun weasi the same as in C phologica clharacters among the differe ngtenetic aequibasis, Cyrtomi uaemquibasis is more close- types were analyzed and compared, The charac- ly related to C. macropbyltum. rbcL sequence a ters used were (1 )number of pair sof pinnae ,(2) differ eidn two base pair sfrom the Chines eC. color of indusium ,and (3 p)resenc eor al]senc oef fortu nreepoirted by Lu et al, (2005 b)u,t still an auricle on the anteri'or base of the pinna .Dii fbrmed a clade with i t(Fig,2). ferencc sin the characters (1 a)rnong the allozyme types were statistically tested using the Steel- Atloayme variation Dwass' test. We performed electrophoretic analysis of four enzyme systems and determined the genotype of euantptcat oifognenet viarication 224 individua losf Cyrt'omiumfor tWtet nweerei, To quantif ythe clonal diversity detected in able to resolvc six loci in total. FoT PGM and 6PG each population as well as within (lyrtomiumfor- enzyme systems, two loc ift.m- a1nd Ilgm-2, tunei, Simpson's diversi tiyndex (D) was calcu- and opg- 1and 6Pg-2) wcre detected ,although late dfor every populatio nas well as for the entire 6Pg- s1howed no allelic variation and was exclud- specEes: ed from furthe ranalysis. For HK andi LAP en- D - 1-£l{[ni(ni-1)]! )[ }]N,(N-1 zyme systems, one locus each was detected. where ni is the number of indivick]al osf vari- In total, eight electrophoretically distinguish- ant i and N is the nu mber of individua lcsollected, able types were recognized (Fi g3. and Table 2) and named allozyme types A through H. Geno- Results typ¢s and the number of indMduals ef each allo- zyme type observed in the four populations ex- IVbeeleot isedgeuence variation ofrbc gLene amined are shown in Table 2. In all four popula- We determined the 1317-b pnucleotide se- tions examined in this study, individnal swith quence of the rbcL gene for 224 individual sof seyeral allozyme types were observcd, indicating C);rtomi fuomrtun efiiro mthe four populations. that several differc ncltones were growlng togeth- 'PsEro types of rbcL sequences were obtained, er in each population (Fi g4.) .In the Moroyama hereafter referred to as rbcL sequellces q and B, and Kobe population st,hree and fiv etypes were The sequences have been deposite din the DNA identified r,espectively. Three al]ozyme types (A, NII-Electronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnSyts teSmaytsictsematics June2011 OoTsuKI&AT..-Genetic VariationinOrtomiumfortwnei 7 TtxBLE 2. Distributio nof the two rbcL Lypes ( aand b) in four population sof C.vrtomiumfortuneis,lat. Total individuals rheL-type number of Population B ct examined Moroyama 28118555 23o22o 5111le75S KawazuKobeFuchu v FIG ,2. 0ne ef the most parsimonious trees based on rbcL sequence of Japanese Ortomium fortun aendi Chinese spccies of CFrtomiu rmeportcdby Lu et al. (2005 ()leng ==L 7h8 steps; consistency index = O.72 ;retention indcx == O.80) ,Percentage ofreplicatctrcesinwhichtheassociatedtaxaclusteredtogetherinthebootstraptest(1000repticates)andnumberoFbasc pai rsubstitution are shown above and bclow cach branch, rcspcctive]y. NNIII-IE-leEcltreoncitcronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics 8 ActaPtiytotax.Geobot. Vbl.62 E, and H) were in common between these two ligh tgray indusia ,a]though they varied in leaf population sa,lthough the populations are remote, morphology. separated by approximately 500 km (Tab l3e). The rel ationship bctwcen allozyme types and Only individua losfallozyme type H had rbcL se- number of pairs of pinnae is shown in Table 4. In quence B, whereas those of all other allozyme addition, variation in number ofpairs ofpinnae types had rbcL sequence a, Table 3 summarizes in each allozyme type is shown in Fig. 6, In Fjg. 6 the results ofSimpson's diversit iyndex (P). i tcan be seen that allozyine type D has a signifi- cantty greate rnumber of pinnae than other allo- rlYpe Reproductiv emode zyme types (A ,B, C, E, F and H). G, how- Reproductiv emode was estimated by eount- ever, was not significantly differe nftrom types A ing the number of spores per sporangium for 140 to E includin gtype D, [IYp eH tended to have a individua lgs6 from the Moroyama population; lower number ofpinnae, but not significantly less 11 from Kawazu; 46 from Kobe; and 43 from Fu- than in other types (A ,B, C, E and F). Large vari- chU), At leas ttwo individua lisn each ofthe eight ation in the number of pair sof pinnae was ob- clones recognized in this study were include din served in all allozyme types, although it is still the analyses. All individua lesxamined had 32 pessibl ethat several differe nctlones were con- spores per speraRgium, suggesting that plant sof tained in one alLozyme type, The relationship be- ail eight allozyme types were apogameus. tween the varieties of (lyrtomiumfortunei and al- lozyme types is shown in Table 5. Morphology From eur genetic analyses and morphological Clytotogicatanalysis observations we found that all individua losf al- The chromosomes of a few individua losf lozyme type H with rhcL sequence P had bicol- each ef the fiv eallozyme types from the Kobe ored indusi a(Fi 5g).. Plants with allozyme type B populatio nwere counted (Fi g7,). All allozyme in the Kawazu populatio nhad bicolore dindusia, types showed 2n = ca, 123, All genetic types de- but their rbcL sequence was a. Individual osf a]] tected in the populatio nwere therefore deter- other allozyme types (A ,C-G) had unifbrmly mined to be triploid, Flow cytometry showed that the triploi dcon- trol had 1.5 times more DNA per nucleus than the ABCDEFGH diploid Thisdiploidindividual Cyrto- control, of Pgm- 1 a e miumfoxtunei from China had rbeL a. Thus, its b-e-ee-e- c]ose relationship to the Japanese triploi dC. .fbr- c ee tunei could be supposed. The DNA content per a-b-c nucleus in the 80 individua lfsrom the Kobe pop- Pgm-2 --e -- - e- -- ulation and in the triploid contTol was identical, and 1.5 times the nuclear content of the diploid Hk a----ee-- control. All 80 individua lwsere therefore detcr- b--e mined to be triploid. We concluded that all indi- a - viduals of C. fortu nien tihe Kobe population are 6 p g - 2 b - - ab e ee -e - apogamoustriploids, -ee- --D- c---- ..- Discussion L a p a-e -ee be -ee-ee In this study, two rbcL types (a and B) and FIG. 3. Zymograms of Japanese q,rlomiu fmortun es.iiat. eight allozyme types (A-H) werc detecte din Band pattern sof four enzyme systems (PGM, HK, 6PG, (llyi'tomiumfortu Onnei t,he basis ofmorphology, a<nAd- ILAlP)). are shown fbr each ofthc eight allozyrne types fbur varieties were identifie hdo,wever, a greater NII-Electronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorPrl anPtlSyastnetma tSicysstematics JLLn 2e011 OoTstJK I& AL.-Genetic Variati einn CI]?rtomiztmfortevnei 9 TABLE 3, Distributio nof eight allozyme types in four populations of (;)rrtomiuin]?)rtu sn.elait. and Simptien' sdiversit yindex (D )calculated on the basi sofallozyme variation in each population .Individua [numbers ofeacli allozyrne typc found in the four loca plopulations as well as va lue sofD are shown. -CJ'ALIozyme type Simpson's Population A B D E F G H Index Moroyama 1 27 23 O.53O.55O.77O,14O.82 KawazuKobeFuchu 6 5 37 26255 11 11213 22 5]5t Total 38 6 5 11 45 su k / pt ee ec,1'g'i ee ,g. ,{fi,k, es E,#gas .sff ll', igwFg*g, ew gewa eetw iggeg 'g"t',/E7f' pt･.'t･ e . p a e e tfi M o2 e r asostsecy rncaetmt4eN]a oo s" o mo tKht"wtes#beea ee tca = - ny 8 x A l #li" ee"eee D e E .g-.-/ME ,s e ee e F . e e G i L d i S tgk!Q!!!! ' se. H llltk'its 'ca e to nc sc ca pm'slenme(m}] Fuchu FIG.4. Maps showing distributi oond' Lhe eigh tallozyme types in the Moroyama (Saita Prrenfa.) K,obe (Hyog oPref. )and Fuchu (HiroshimaPretl)populations, NNIII-IE-leEcltreoncitcronic Library Service TThhee JJaapapnaesneSeosciee tySociety ffoorr PPllanatnt SSyystsetmaetmicastics 10 Acta Phytotax. Geobot. Vol. 62 ..･/.;s. ,,le.f. t't 2 4 F]G. 5. Indusia ofthe four vaTieties of evftomtum fortun e1:i v.ar, forttt n2:e si,a.r, clivicolum. 3: var. alropunciarttm. 4: var. in- r= termediuni (ba r 1 cm). number ofgenetic clones were observed in C.,for- peculia rto C. fort" nvaer.i intermedium, It was tunei .In the Kobe and Moroyama populations,therefore concluded that C. fortun vear.i interme- individua lsof allozyme type El had rbcL se- dium is gepeticall ydiffere nitn it snuclear and quence P ,whereas those ofall the other types had plasti dgenomes from other varieties of C. fortu- rbcL sequence a, thereby showing that rbcL P nei. Seven base pair swere differe nbtetween rbcL type was geneticall ydlfferen ftrom the rbcL or u and B .The differenc ewsere as large as those type even in the nuclear genome (allozy mieoci). between distinc stpecies (Yata bete al. 2009). Zy- Furtherinore, only individua lwsith rbcL P and mograms of the H type are similar to zymograms FIb the H allozyme patter nhad bicolored indusia, ofthe other types (Fi g3.). ctarify the origin of which is one of the morphQ!ogical charaeteristics var. intermedium, careful comparison with mem- NII-Electronic Library Service