Nuclear DNA Content Variation Within the Rosaceae
Total Page:16
File Type:pdf, Size:1020Kb
University of South Carolina Scholar Commons Faculty Publications Biological Sciences, Department of 9-1-1992 Nuclear DNA Content Variation Within the Rosaceae E. E. Dickson Cornell University K. Arumuganathan Cornell University Stephen Kresovich University of South Carolina - Columbia, [email protected] J. J. Doyle Cornell University Follow this and additional works at: https://scholarcommons.sc.edu/biol_facpub Part of the Biology Commons Publication Info Published in American Journal of Botany, ed. Judy Jernstedt, Volume 79, Issue 9, 1992, pages 1081-1086. © American Journal of Botany 1992, Botanical Society of America. This Article is brought to you by the Biological Sciences, Department of at Scholar Commons. It has been accepted for inclusion in Faculty Publications by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. AmericanJournal of Botany 79(9): 1081-1086. 1992. NUCLEAR DNA CONTENT VARIATION WITHIN THE ROSACEAE' E. E. DICKSON,2,5 K. ARUMUGANATHAN,3 S. KRESOVICH,4 AND J. J. DOYLE2 2L. H. BaileyHortorium and 3Departmentof Plant Breeding and Biometry, CornellUniversity, Ithaca, New York 14853-4301;and 4USDA-ARSPlant Genetic Resources Unit, Cornell University, Geneva, New York 14456-0462 NuclearDNA contenthas beenestimated using flow cytometry for 17 speciesand eightcultivars of Malus and for44 speciesof 29 othergenera within the Rosaceae. Compared to otherangiosperms, diploid genome sizes vary little within the familyRosaceae and withinthe genus Malus. C-valuesof generawithin the subfamilies Spiraeoideae and Rosoideaeare amongthe smallest of flowering plants thus far reported. In general,the Maloideae have the largest diploid genomes of the family,consistent with their higher chromosome numbers and presumedpolyploid origin. The Rosaceae, includingsuch economicallyimportant MATERIALS AND METHODS plants as almond, apple, strawberry,and rose, is consid- ered a naturalgroup held togetherby similaritiesin floral Leaf materialof rosaceous taxa was collectedfrom sev- structures.The foursubfamilies are definedby fruittype eral sources (Table 1). Voucher specimens are deposited (Robertson,1974), and each includespolyploid series with at theL. H. Bailey Hortorium(BH). Chromosome counts, fairlyconsistent chromosome base numbers:Spiraeoideae whereavailable, were gatheredfrom the literature(Table (x = 9); Amygdaloideae (x = 8); Rosoideae (x = 7, 8, and 2). 9); and Maloideae (x = 17) (Sax, 1931, 1932). The sub- For the determinationof nuclear DNA content,sus- familyMaloideae, with its relativelyhigh base chromo- pensions ofintact nuclei wereprepared from young leaves some number,has been hypothesizedto be eitherof au- by chopping(Galbraith et al., 1983) accordingto the pro- topolyploid (Darlington and Moffett, 1930) or cedure of Arumuganathanand Earle (1991la) and stained allopolyploid origin(Sax, 1931, 1932; Stebbins, 1950). withpropidium iodide. The isolation procedurewas fur- Althoughnumerous chromosome numbershave been ther modified for some members of the subfamilyRo- reportedfor Rosaceae, the amount of DNA per nucleus soideae bythe addition of 1% polyvinylpyrrolidone (PVP- (C-value) has been reportedfor only 14 species of the 40) to the initial isolation solution,without the addition family(Bennett and Smith, 1976, 1991; Bennett,Smith, of propidium iodide. The modifiedprocedure increased and Heslop-Harrison, 1982; Arumuganathanand Earle, the number of intact nuclei isolated. The mean fluores- 1991b). Apart from the utilityof genome size data in cence intensity,frequency, standard deviation, and coef- ongoingmolecular studies in thisimportant plant family, ficientof variation of the propidium iodide-stained nuclei the amount and distributionof nuclear DNA content at 488 nm were recordedwith an EPICS PROFILE flow variation among related taxa may give insightsinto ge- cytometer(Coulter Electronics, Hialeah, FL). Chicken nomic evolution that underlies or parallels speciation erythrocytes(2.33 pg DNA/2C; Galbraith et al., 1983) (Raina and Narayan, 1984; Ohri and Khoshoo, 1986; wereincluded with our samples as internalstandards (Fig. Price, 1988). 1). In this study,flow cytometry was used to estimatenu- The mean 2C-value of each sample was determinedby clear DNA contentsof 28 genera fromeach of the four multiplyingthe ratio of the fluorescencemeans of the Rosaceae subfamilies.Compared to Feulgen densitome- sample nuclei and chickenerythrocytes by the amount of try or reassociation kinetics,flow cytometryis a rapid DNA per chicken erythrocyte(2.33 pg). The standard and reliable method for estimatingC-values in plants deviation forthe mean 2C-value of each sample was cal- (Galbraith et al., 1983; De Laat, Gohde, and Vogelzang, culated by takinginto account the variances of the fluo- 1987; Raybum et al., 1989; Raybum, 1990; Michaelson rescencefrom both the sample and the internalstandard et al., 199 la). We here reportlow levels of nuclear DNA (W. Lamboy, personal communication): variationwithin the Rosaceae and findthat Spiraeoideae C-values are among the smallest of angiosperms.Rela- a2(samp/crbc) = U2(samp)/M(crbc)2 tivelylarge C-values of Maloideae supportthe polyploid + [U2(crbc) x M(samp)2]/M(crbc)4 originof the subfamily. Sample C-value standard deviation = o(samp/crbc) x 2.33 where I Receivedfor publication 25 November1991; revision accepted 27 April1992. (a2 = fluorescencevariance The authorsthank Steven Spongberg of theArnold Arboretum for M = fluorescencemean helpingin thecollection of leaf samples; Martha Rangel-Lugo for pro- samp = sample nuclei vidingchicken blood; and Kathy Anderson and Jim Slattery for assisting crbc = chicken erythrocytes in theoperation of the flow cytometer. 5 Authorfor correspondence. Our standard deviations are up to twice as large as for 1081 1082 AMERICAN JOURNAL OF BOTANY [Vol. 79 thosein whichwe assumedno variancefor the standard. TABLE 1. Originalsources and vouchersof plants used in thisstudy. reportsof C-value standarddeviations: 1) as- AA = Arnold Arboretum; CU = Cornell University;NY- Previous SAES = New YorkState Agricultural Experiment Station, Geneva, sumethe standard is constantwithout variance (Rothfels New York et al., 1966;Narayan, 1982; Price et al., 1983;Goldblatt, Walbot,and Zimmer,1984; Laurie and Bennett,1985; Spiraeoideae: Simsand Price,1985); 2) use a standardassuming a con- Exochordagiraldii, AA 1503-52A,EED 872. Neilliasimensis, AA stantmean and a constantstandard deviation (Galbraith 144-81,EED 875. Physocarpusbracteatus, AA 1235-85,EED 856; et al., 1983); or 3) give standarddeviations around the P. opulifolius,AA 1279-82-A,EED 857. Spiraeachinensis, AA 223- mean fluorescenceor mean densityof thesample in ar- 83, EED 855; S. crenata,AA 1230-85,EED 859; S. pubescens,AA bitraryunits (Chooi, 1971; Priceand Bachmann,1975; 541-83,EED 853; S. nipponica,CU, EED 900; S. sargentiana,AA Raybumet al., 1989; Raybum,1990; McMurphyand 1233-85,EED 854; S. wilsonii,AA 953-85-A,EED 852. Stephanan- Raybum,1991). dra incisa,CU, EED 902. Amygdaloideae: Osmaroniacerasiformis, AA 274-85,EED 879. Prinsepiauniflora, RESULTS AND DISCUSSION AA 7188-A,EED 848. Prunuspersica "Red Haven,"NYSAES, no voucher;P. serotina,NYSAES, EED 879; P. subhirtella,CU, EED The genomicDNA contentsof 44 rosaceousplants (29 903. genera),estimated from the mean fluorescence values of 500-2,000nuclei per sample,are listedin Table 2. The Rosoideae: ofvariation were less than 5% forall reported Dryasoctopetala, CU, EED 892.Duchesnea indica, Geneva, NY, EED coefficients 898. Neviusiaalabamensis, AA 1809-71,EED 873. Potentillafruc- samples.Previously obtained 2C-values of Rosaceae are ticosa,CU, EED 899. Rhodotypusscandens, AA 680-79,EED 851. includedin Table 2: six estimatedby Feulgenmicroden- Rosa multiflora,Geneva, NY, EED 897. Rubusodoratus, CU, EED sitometry(Bennett and Smith, 1976, 1991; Bennett, Smith, 896. and Heslop-Harrison,1982) and eightothers obtained by Maloideae: flowcytometry recently reported by one of us (Arugu- Amelanchiersp., NYSAES, EED 876. Aroniaarbutifolia, AA 1905- manathanand Earle, 1991b). The values obtainedby 81,EED 870. Chaenomelesspeciosa, NYSAES, EED 881. Cotoneaster Feulgendensitometry agree closely with the flowcyto- melanocarpa,Geneva, NY, EED 882. Crataeguscrus-galli, Geneva, metryvalues of other species within the same subfamily. NY, EED 883. Cydoniaoblonga, NYSAES, EED 877. Eriobotrya japonica,CU, no voucher.Malus spp.,NYSAES. Mespilusgerman- ofMalus x domestica ica, NYSAES, EED 878. Photineaparvifolia, AA 577-76,EED 867. Polyploidseries -The 2C-values Pyracanthacoccinea, CU, EED 901; P. sp. "Royal,"AA 194-49-A, and Prunus species correlatewith theirchromosome EED 869. Pyruscalleryana, CU, EED 895. Sorbusalnifolia, CU, EED counts.Malus x domesticaaccessions with chromosome 880; S. americana,NYSAES, EED 894. numbersof 2x, 3x, and 4x have 2C-values (standard deviation)of 1.55 (0.21), 2.51 (0.20), and 2.86 (0.43) pg, respectively,deviations from a perfectnumerical series well withinthe variationfound within apple (Table 3). genomewas reportedfor Cardamine amara L. (0.11 pg/ The 2C value of thediploid sweet cherry Prunus avium 2C) (Bennettand Smith,1991), another member of the (0.67 pg) is approximatelyhalf that of the tetraploid hy- Brassicaceae. bridcherry Prunus x sp. '4X' (1.36 pg). Othersmall C-values occur within the Rosaceae. Rosa In contrast,the nuclear DNA contentsof Spiraeapu- wichuraianaCrepin (Rosoideae) has a genomesize of0.2 bescensand S. nipponica,reportedly diploids, are two and pg/2C(Bennett and Smith,199 1). Withinthe