Brief Communications

Brief Communications

Inheritance and Subcellular Localization of TriosePhosphate Isomerase in Dwarf Mountain Pine (Pinus mugo) I. J. Odrzykoski Several trees with expected heterozygous phenotype for triose-phosphate isomerase (TPI) were discovered in a population of dwarf mountain pine (Pinus mugo Turra) from southern Poland. As the inheritance of this enzyme in pines has not been reported, segregation of allelic variants was tested in eight trees with putative heterozygous phenotypes for two loci, TpiA and TpiB. Linkage between these and some other isozyme loci were studied and evidence for linkage has been found between TpiA and PgdA (r ⫽ 0.10) and between TpiB and DiaD (r ⫽ 0.36), but in single trees only. The subcellular localization of TPI isozymes was determined by comparing isoenzymes from the total extract with those found in fraction enriched in plastids, prepared by differential gradient centrifugation of cellular organelles. The more slowly migrating TPI-B isozyme is located in plastids. Inheritance of electrophoretic variants of many enzymes has been studied in pines and other gymnosperms since the beginning of the 1970s and especially during the 1980s ( El-Kassaby and White 1985; Paule 1990 for a bibliography). Allelic variants of different enzyme loci can be easily observed in these plants in a sample of haploid megagametophytes of heterozygous trees and they usually segregate in proportions expected in a test cross. Many reports have been published since the first studies of this kind (e.g., Bartels 1971; Rudin 1975) and some provided the first evidence for linkage groups and showed evidence for conservation of gene blocks

between different conifer genera (Conkle 1981; Guries et al. 1978; Rudin and Ekberg 1978). Polymorphic isozyme loci still provide a set of useful genetic markers in many population studies due to their technical simplicity of detection and the relatively low cost of data collection in comparison with DNA markers. Inheritance data for about 40 enzymes have been published to date and as electrophoretic phenotype is usually made up of the products of one to five separate genes, inheritance of roughly 80 loci has been studied in the genus as a whole ( Ledig 1998). Despite the large number of enzymatic systems available, not all enzymes frequently studied in other plant groups have a documented mode of inheritance in this genus. During the isozyme studies on populations of dwarf mountain pine (Pinus mugo Turra), I found several apparently heterozygous trees based on sporophytic phenotypes expected for the dimeric enzyme triose-phosphate isomerase ( TPI, EC 5.3.1.1). Because a formal inheritance study of this enzyme has never been published for pines, seeds from several heterozygous trees were collected and used to study allelic segregation and linkage relationships. Triose-phosphate isomerase is a glycolytic enzyme usually present in diploid homozygous plants as two isozymes with different subcellular localization (Wendel and Weeden 1989). The product of one nuclear gene is active in the cytosol and the other one, after translation, is actively transported into plastids (Pichersky and Gottlieb 1984). These two isozymes can be distinguished by comparison of the total cellular extract and the extract from a purified plastid preparation. The purpose of this study was to test the inheritance and linkage relationships of the two TPI isozymes found in buds and megagametophytes of P. mugo. Also, the subcellular localization of the two isozymes was inferred by comparing the phe-

notype from crude extracts with those enriched for plastids.

Materials and Methods Plant Material Winter buds and seeds from eight trees were taken from the ‘‘Bo´r na Czerwonem’’ peat bog population ( located near Nowy Targ, southern Poland) in late autumn. Small apical portions of the stem with winter buds and cones were collected from a single branch of each polycormic plant, and the buds were stored at ⫺20⬚C prior to enzyme extraction. Seeds were extracted from cones at room temperature and after desiccation stored at ⫺20⬚C. Isozyme Analyses Enzyme extraction. Three kinds of tissue were used for enzyme extraction: (1) a fragment of sporophytic tissue dissected from a winter bud, (2) seedlings, and (3) a sample of megagametophytes extracted from seeds germinated for about 10 days (to the moment when the radical of the seedling was 5 mm long). The tissue was homogenized in 0.1 M Tris-HCl buffer (pH 7.5), with the addition of 1 mM EDTA Na2, 10 mM KCl, 10 mM MgCl2 and 0.1% Triton X-100. Just before the extraction, 10 ␮l of 2-mercaptoethanol were added per 10 ml of extraction buffer. The extraction buffer for the bud tissue was modified by addition of 3% (w/v) PVP K-25. The homogenate was filtrated through a small piece of Miracloth tissue and absorbed into 2–2.5 mm wide Whatmann 3MM strips. Electrophoretic separation and isozyme detection. Isozymes were separated using horizontal 10% starch-gel electrophoresis and a modification of the discontinuous buffer system A (Conkle et al. 1982) using 190 mM boric acid and 40 mM lithium hydroxide (pH 8.3) as the electrode buffer and 50 mM Tris, 6 mM citric acid with the addition of 10% of the electrode buffer (pH 8.2) as a gel buffer. Electrophoresis was

271

performed in a refrigerated chamber under constant voltage (280 V) for about 4.5 h, to the moment when the ‘‘borate front’’ moved 8 cm from the origin. TPI isozymes were localized on top of the gel slices using the agar overlay technique (Manchenko 1994). The staining mixture contained 7.5 ml 50 mM Tris-HCl buffer (pH 7.5) with 1 mg of lithium salt of dihydroxyacetone phosphate (Sigma D7137) or a substrate prepared by hydrolysis of dihydroxyacetone phosphate dimethyl ketal (Sigma D7878), according to the supplier’s instructions, 2 mg NAD, 45 mg sodium arsenate, 2 mg MTT, 0.5 mg PMS, and 7.5 ml 1.8% of warm-water agar solution. The volume of staining assay was occasionally reduced to about 10 ml, when the location of the isozymes could be deduced from migration of pigment markers. For detection of the products of other enzyme loci (PGD, 6-phosphoglucose dehydrogenase; GOT, glutamic-oxaloacetic transaminase; GDH, NAD glutamate dehydrogenase; ADH, alcohol dehydrogenase; SDH, shikimate dehydrogenase; FLE, ‘‘fluorescent’’ esterase; ACO, aconitase; DIA, NADH-diaphorase; LAP, leucine aminopeptidase; MDH, NAD malate dehydrogenase; PGM, phosphoglucomutase; and ACP, acid phosphatase), the standard technique was used (Conkle et al. 1982) and designation of enzyme loci follows standards for Scots pine ( Niebling et al. 1987; Szmidt and Muona 1989). Chloroplast preparation. A small amount of purified plastids were obtained from cotyledons of 2-week-old seedlings by a density gradient centrifugation in buffered Percoll (Pharmacia) using a modified method described by Odrzykoski and Gottlieb (1984). The procedure involves a few simple steps. First, a pulp obtained by gentle maceration of cotyledons in the extraction buffer was centrifuged for 1 h in a Percoll gradient buffered with 25 mM HEPES-KOH buffer (pH 7.5) with the addition of 1 mM EDTA, 330 mM sorbitol, 1% ficol, 0.2 % BSA, 10 mM KCl, and 48 mM 2-mercaptoethanol using SW41Ti rotor and a Beckman ultracentrifuge at 15,000 rpm (36,000g). After centrifugation the layer of intact chloroplasts was transferred into a 1.5 ml Eppendorf tube, washed twice in a washing buffer, and the resulting pellet was homogenized in the same extraction buffer as used for seedlings. Isozymes extracted from seedlings and those from the

272 The Journal of Heredity 2001:92(3)

Figure 1. (A) Electrophoretic TPI phenotypes for buds of several individuals of P. mugo. TPI-A is invariable in all trees. The most common phenotype is TPIA11/TPI-B11 (e.g., line 1). The other phenotypes are TPI-B22 ( line 2), TPI-B12 ( line 7), TPI-B13 ( line 9). (B) TPI phenotype from seedlings with the most common phenotype ( lines 1, 2, 5, and 6), and from the chloroplast-enriched fraction obtained from the same seedlings ( lines 4 and 5 marked with ‘‘p’’).

plastid-enriched fraction were compared side by side in the same gel. Statistical Analysis For a statistical evaluation of segregation data and linkage relationships, the chisquared goodness-of-fit test was used, as described by Mather (1951). The data for the same locus pair from individual trees were pooled, if the tests results were homogeneous. Recombination values (r ⫽ R/ n) and their standard error SEr ⫽ (r(1 ⫺r)/ r)1/2 were estimated for pairs of loci with significant linkage only (R ⫽ the number of recombinant types observed in a sample of n megagametophytes).

Results Enzyme Phenotype for Winter Buds The most common phenotype of TPI is composed of two isozymes: TPI-A and TPIB ( Figure 1A, line 1). In the studied population some other phenotypes were detected, one for TPI-A ( labeled TPI-A12; Figure 2A, line 1) and three for TPI-B ( labeled TPI-B22, TPI-B12, and TPI-B13 ( Figure 1A, lines 2, 7, and 9, respectively). Subcellular Localization of Isozymes The seedlings of trees with the sporophytic phenotype TPI-A11/TPI-B11 had the same phenotype as the mother tree. In extracts from three independent preparations enriched in chloroplasts, only TPI-B isozyme was found ( Figure 1B, lines 3 and 4 marked with ‘‘p’’). Segregation in Heterozygous Trees Segregation was studied in three trees with the sporophytic phenotype TPI-A12 ( Figure 2A) and the results were homogeneous, showing the 1:1 ratio in all trees (⌺n ⫽ 166, ␹2(1:1) ⫽ 0.60, ␹2( het) ⫽ 0.299, df ⫽ 2). Similarly five trees with the phenotype TPI-B12 ( Figure 2B) showed a seg-

Figure 2. (A) Segregation of TpiA allozymes in megagametophytes of a tree with the sporophytic phenotype TPI-A12/TPI-B11 ( line 1). (B) Segregation of TpiB allozymes in megagametophytes of a tree with the sporophytic phenotype TPI-A11/TPI-B12 ( line 1).

regation ratio expected from the allelic variants (⌺n ⫽ 479, ␹2(1:1) ⫽ 0.012, ␹2( het) ⫽ 3.829, df ⫽ 4). These results suggest that the TPI phenotype in dwarf mountain pine is composed of the product of two genes: TpiA and TpiB. Linkage Studies Trees heterozygous for TpiA were also heterozygous for nine other enzyme loci, therefore studies of joint segregation were possible ( Table 1). The evidence for linkage was detected between TpiA and PgdA [6-phosphoglucose dehydrogenase ( EC 1.1.1.44), locus A] in one tree tested for this combination (r ⫽ 0.10, SEr ⫽ 0.04). A weak linkage was also detected between TpiA and GotC [glutamate oxaloacetate transaminase ( EC 2.6.1.1), locus C] also in one tree (r ⫽ 0.34, SEr ⫽ 0.05), and between TpiA and Fle [‘‘fluorescent’’ esterase ( EC. 3.1.1)] in one of two trees tested for these combinations (r ⫽ 0.33, SEr ⫽ 0.07). One or more trees were tested for a joint segregation of TpiB with 17 other markers, and the linkage was detected only for TpiB and DiaD [diaphorase ( EC 1.6.4.3), locus D] in one from two trees tested for this combination (r ⫽ 0.36, SEr ⫽ 0.04).

Discussion Little is known about the genetic control and variation of the important glycolytic enzyme TPI in pines and other gymnosperms. Recently the enzyme was surveyed for variation in population genetics studies of Pinus echinata, P. virginiana, and P. palustris ( Edwards and Hamrick 1995; Parker et al. 1997; Schmidtling and Hipkins 1998), however, no formal validation by segregation analysis was conducted. Nearly 550 plants from a peat bog population

Table 1. Two locus data and chi-squared analysis of allelic variants segregation at locus A (␹2A), locus B (␹2B), and joint segregation for the detection of linkage (␹ 2AB) for trees heterozygous for TpiA or TpiB Segregation Locus A

Locus B

TpiA: TpiA: TpiA:

PgdA GotC Fle

TpiA: TpiA: TpiA: TpiA: TpiA: TpiA: TpiB:

pooled Gdh AdhA AdhB SdhA PgdB MdhB DiaD

TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB: TpiB:

pooled Fle Gdh AdhA AdhB SdhA PgdA PgdB Aco MdhC GotA GotB DiaC LapA LapB Pgm Acp

Macrogametophyte class No. of trees A2B2 A2B1 A1B2 A1B1

Heterogeneity ␹2 (df )

Locus A ␹2(A)

Locus B ␹2( B)

Joint ␹2(AB)

1 1 1 1 2 3 1 1 1 1 2 1 1 2 2 3 3 1 4 5 5 2 2 1 2 1 1 1 2 2

— — — — 2.426 4.048 — — — — 0.014 — — 4.041 0.418 2.145 0.164 — 0.937 3.744 0.577 0.102 0.768 — 0.761 — — — 0.446 0.889

0.000 0.690 0.333 0.690 1.017 0.708 0.209 0.209 0.000 0.000 0.398 1.929 1.140 0.573 0.186 0.123 0.485 1.929 0.190 0.002 0.002 0.030 0.576 1.542 0.192 0.095 2.778 0.602 0.671 1.243

0.692 0.014 0.333 0.014 0.210 0.988 0.581 0.581 0.077 0.692 0.398 0.095 0.209 0.005 3.767 0.123 0.006 0.095 0.000 5.192 0.098 0.032 11.379** 0.602 3.461 3.429 0.309 0.386 0.055 0.006

33.923** 7.451* 5.333* 0.127 3.034 0.146 0.023 0.023 0.692 0.077 0.984 12.595** 0.209 8.763* 3.953 1.370 2.162 0.024 0.190 0.721 2.174 0.032 1.711 0.386 0.192 1.524 0.012 1.542 1.110 0.006

25 14 16 20 36 41 13 13 11 12 37 25 14 39 18 68 39 37 84 108 132 34 33 42 88 5 18 46 73 49

4 22 6 15 21 38 11 11 14 11 28 57 9 66 16 81 44 45 84 117 115 27 48 33 96 10 25 33 71 42

1 25 10 19 29 50 10 10 15 14 28 50 11 61 27 75 49 38 80 143 119 34 67 46 75 15 15 42 66 49

22 10 16 17 33 42 9 9 12 15 30 36 9 45 25 68 35 48 88 133 135 29 63 45 75 12 23 45 82 41

(1) (2)

(1)

(1)* (1) (2) (2) (1) (1) (4) (1) (1) (1)

(1) (1)

Data were pooled, when heterogeneity was not significant, but are shown separately (and pooled) for some cases (TpiA:Fle and TpiB:DiaD). *P ⬍ .05, **P ⬍ .01.

of P. mugo were tested for variability of this enzyme and some expressed an apparently heterozygous phenotype in the sporophytic tissue. The presumably heterozygous phenotype for fast-migrating TPI-A (a diffuse broad band of enzyme activity) was detected in 12 trees. Segregation studies showed that trees with this phenotype are indeed heterozygous, and in a sample of haploid megagametophytes the variants segregate in the 1:1 ratio. A three-banded phenotype was found in 85 trees for the more slowly migrating TPI-B, and segregation studies confirmed the heterozygosity of five of these trees. The enzyme phenotype of TPI in dwarf mountain pine is therefore composed of the products of two genes, a result commonly found also in diploid angiosperms (Gottlieb 1982; Weeden and Wendel 1989), excluding rare cases of duplication (or sometimes triplications) found in some genera (e.g. Pichersky and Gottlieb 1984; Wendel et al. 1989; and reports discussed by Weeden and Wendel 1989). The results of linkage studies showed that TpiA is probably linked to PgdA. This may locate TpiA in the well-established ‘‘linkage group E’’ (Conkle 1981), known in numerous pine species including the

closely related P. sylvestris (Goncharenko et al. 1994; Niebling et al. 1987; Szmidt and Muona 1989). The other enzyme loci from this group are GotC, SdhB, and MdhA. A weak linkage of TpiA with GotC was also detected in P. mugo, therefore the location of TpiA between PgdA and GotC is likely, but this should be confirmed with more detailed analysis. The evidence for linkage between TpiB and DiaD was also found in one tree. The second locus has been mapped in ‘‘linkage group A’’ (Conkle 1981), also in P. mugo, where DiaD and AdhA are tightly linked, with a recombination frequency of less than 0.10 (Odrzykoski IJ, unpublished data). The linkage of TpiB with loci from this group should be further tested, because at the same time no linkage was detected between TpiB and two other loci (GotA, AdhA) located close to DiaD. The subcellular localization experiment showed that the more slowly migrating TPI-B isoenzyme is the only one present in the plastid-enriched fraction. The fast-migrating TPI-A is likely to be a cytosolic form of this enzyme. This is a similar situation to that found in diploid flowering plants, where TPI is usually present as two isozymes encoded by separate nuclear

genes (Pichersky and Gottlieb 1984; Wendel et al. 1989). Only a little is known about the subcellular localization of enzymes used as genetic markers in population genetics studies in gymnosperms. Cytosolic and plastid isozymes of CuZn-superoxide dismutases from needles of Scots pine ( Karpinski et al. 1992; Wingsle et al. 1991) and Norway spruce ( Kroniger et al. 1992) are well characterized and the mitochondrial Mn-superoxide dismutase isozyme is known from the second species (Sehmer and Dizengremel 1998). Another example is provided by two NAD-dependent malate dehydrogenases (MDH2 and MDH3) resistant to ascorbic acid treatment, which suggests their mitochondrial localization, similar to the situation found in maize ( Breitenbach-Dorfer and Geburek 1995). The procedure of subcellular localization tested for pines in this report may be useful for better characterization of isozyme markers. It should help to identify products of homologous genes, which may be important especially in comparative studies (Conkle 1991). From the Department of Genetics, A. Mickiewicz University, 60–371 Poznan, Miedzychodzka 5, Poland. I appreciate the comments and suggestions of Dr. D. B. Wagner and by the anonymous reviewers. Address correspondence to I. J. Odrzykoski at the address above or e-mail: [email protected]. 䉷 2001 The American Genetic Association

References Bartels H, 1971. Genetic control of multiple esterases from needles and macro-gametophytes of Picea abies. Planta 99:283–289. Breitenbach-Dorfer M and Geburek T, 1995. Gene modifies electrophoretic properties of malate dehydrogenase in Norway spruce (Picea abies ( L.) Karst.). Hereditas 122:103–108. Conkle MT, 1981. Isozyme variation and linkage in six conifer species. In: Proceedings of the Symposium on Isozymes of North American Forest Trees and Forest Insects, Berkeley, CA, July 27, 1979. General Technical Report PSW-48. Washington, DC: USDA Forest Service, 11–17. Conkle MT, Hodgskiss PD, Nunnally LB, and Hunter SC, 1982. Starch gel electrophoresis of conifer seeds: a laboratory manual. General Technical Report PSW-64. Washington, DC: USDA Forest Service. Concle MT, 1991. Genetic diversity—seeing the forest through the trees. New For 6:5–22. Edwards MA and Hamrick JL, 1995. Genetic variation in shortleaf pine, Pinus echinata Mill. (Pinaceae). For Genet 2:21–28. El-Kassaby YA and White EE, 1985. Isozymes and forest trees: an annotated bibliography. Information Report BC-X-267. Canadian Forest Service, Pacific Forest Research Centre. Goncharenko GG, Padutov VE, and Silin AE, 1994. Construction of genetic maps for some Eurasian coniferous species using allozyme genes. Biochem Genet 32:223– 236. Gottlieb LD, 1982. Conservation and duplication of isozymes in plants. Science 216:373–380.

Brief Communications 273

Guries RP, Friedman ST, and Ledig FT, 1978. A megagametophyte analysis of genetic linkage in pitch pine (Pinus rigida Mill.) Heredity 40:309–314. Karpinski S, Wingsle G, Olsson O, and Halgren JE, 1992. Characterization of cDNAs encoding CuZn-superoxide dismutases in Scots pine. Plant Mol Biol 18:545–555. Kroniger W, Rennenberg H, and Polle A, 1992. Purification of two superoxide dismutase isozymes and their subcellular localization in needles and roots of Norway spruce (Picea abies L.) trees. Plant Physiol 100:334–340. Ledig FT, 1998. Genetic variation in Pinus. In: Ecology and biogeography of Pinus (Richardson DM, ed). Cambridge: Cambridge University Press; 251–280. Manchenko GP, 1994. Handbook of detection of enzymes on electrophoretic gels. London: CRC Press. Mather K, 1951. The measurement of linkage in heredity. London: Methuen & Co. Niebling CR, Johnson K, and Gerholdt HD, 1987. Electrophoretic analysis of genetic linkage in Scots pine (Pinus sylvestris L.). Biochem Genet 25:803–814. Odrzykoski IJ and Gottlieb LD, 1984. Duplications of genes coding 6-phosphogluconate dehydrogenase in Clarkia (Onagraceae) and their phylogenetic implications. Syst Bot 9:479–489. Parker KC, Hamrick JL, Parker AJ, and Stacy EA, 1997. Allozyme diversity in Pinus virginiana (Pinaceae): intraspecific and interspecific comparisons. Am J Bot 84: 1372–1382. Paule L, 1990. Bibliography: isozymes and forest trees (1968–1989). Report 9. Swedish University of Agricultural Sciences, Department of Forest Genetics and Plant Physiology. Pichersky E and Gottlieb LD, 1983. Evidence for duplication of the structural genes coding plastid and cytosolic isozymes of triose phosphate isomerase in diploid species of Clarkia. Genetics 105:421–436. Pichersky E and Gottlieb LD, 1984. Plant triose phosphate isomerase isoenzymes. Purification, immunological and structural characterization, and partial amino acid sequence. Plant Physiol 74:340–347. Rudin D, 1975. Inheritance of glutamate-oxaloacetatetransaminases (GOT ) from needles and endosperms of Pinus sylvestris. Hereditas 80:296–300. Rudin D and Ekberg I, 1978. Linkage studies in Pinus sylvestris L. using macrogametophyte allozymes. Silvae Genet 27:1–12. Sehmer L and Dizengremel P, 1998. Contribution to subcellular localization of superoxide dismutase isoforms of spruce needles and oak leaves. J Plant Physiol 153: 545–551. Szmidt AE and Muona O, 1989. Linkage relationships of allozyme loci in Pinus sylvestris. Hereditas 111:91–97. Schmidtling RC and Hipkins V, 1998. Genetic diversity in longleaf pine (Pinus palustris): influence of historical and prehistorical events. Can J For Res 28:1135–1145. Weeden NF and Wendel JF, 1989. Genetics of plant isozymes In: Isozymes in plant biology (Soltis DE and Soltis PS, eds). Portland, OR: Dioscorides Press; 46–72. Wendel JF, Stuber CW, Goodman MM, and Becket JB, 1989. Duplicated plastid and triplicated cytosolic isoenzymes of triosephosphate isomerase in maize (Zea mays L.). J Hered 80:218–228. Wendel JF and Weeden NF, 1989. Visualization and interpretation of plant isozymes. In: Isozymes in plant biology (Soltis DE and Soltis PS, eds). Portland, OR: Dioscorides Press; 5–45. Wingsle G, Gardestrom P, Halgren JE, and Karpinski S, 1991. Isolation, purification, and subcellular localization of isozymes of superoxide dismutase in Scots pine (P. sylvestris L.) needles. Plant Physiol 95:21–28. Received February 16, 2000 Accepted January 15, 2001 Corresponding Editor: David B. Wagner

274 The Journal of Heredity 2001:92(3)

Extension of the Castle– Wright Effective Factor Estimator to Sex Linkage and Haplodiploidy C. D. Jones The Castle–Wright effective factor estimator gives a minimum estimate of the number of genes underlying complex traits. Because the Castle–Wright estimator does not rely on genetic markers, it is especially useful in genetically undeveloped species. In this article I describe two extensions of this estimator. The first corrects the estimator in heterogametic (XY) species with a partially sex-linked trait. In this case the traditional estimator underestimates gene number in F2 males and overestimates it in F2 females and backcross females and males. The second extension adapts the Castle–Wright equation to haplodiploid species. Since its creation by Wright (see Castle 1921), the Castle–Wright equation has become a widely used tool for estimating the number of genes affecting complex traits. Knowledge of the number of genes is important for two reasons. First, this information is important to quantitative genetics theory. As Lynch and Walsh (1998:231) point out, much of quantitative genetics theory assumes that many genes underlie a phenotypic trait. Therefore it is important to know how well and often this assumption is met. Second, accurate estimates of gene number may answer several important evolutionary questions. For example, current debate over the genetics of adaptation centers on the number of genes typically involved in adaptation ( Bradshaw et al. 1998; Orr and Coyne 1992; Tanksley 1993) Not surprisingly, several recent studies of the genetics of adaptation have employed Wright’s equation ( Hatfield 1997; Sezer and Butlin 1998). The Castle–Wright equation is especially useful in genetically undeveloped species because it does not require genetic markers. The data required are entirely phenotypic and derive from either an F2 cross or a backcross between two phenotypically divergent populations or species. For an F2 cross, the segregation variance of the F2(␴S2)—which is estimated from the phenotypic variance—and the mean trait values of the parents (P1,P2) are used to estimate the number of genes involved: nef ⫽

(P1 ⫺ P2 ) 2 . 8␴S 2

(1)

This equation rests upon several simplifying assumptions. First, it assumes that all alleles behave additively (h ⫽ 1/2), all loci are unliked, and all alleles have equal effects. It also assumes the two parental strains are homozygous for alternative alleles at all loci affecting the trait and all chromosomes are diploid. If these assumptions are not met, the equation usually underestimates the true number of genes. For example, tightly linked loci are treated as a single effective factor by the Castle–Wright equation. Therefore the Castle–Wright estimate is considered a minimum estimate of gene number. Despite these restrictive assumptions, a couple of recent genetic analyses of quantitative traits have suggested that the Castle–Wright estimator (and its improved descendants) can be fairly accurate. In a study of Populus, Wu et al. (1997) showed that estimates of gene number from the Castle–Wright equations were usually close to the number of factors found in quantitative trait loci (QTL) analysis. Likewise, Gurganus et al. (1999) found that the Castle–Wright equation accurately estimated of the number of factors underlying a difference in bristle number in Drosophila. Many improvements have been made to the Castle–Wright estimator, relaxing most of its key assumptions (reviewed in Lynch and Walsh 1998). Wright (1968) made several improvements. He modified the estimator for backcross data, showed how to estimate the effect of the largest factor involved, modified the equation for special cases of dominant and unequal allele effects, and presented methods for improving the estimation of the segregation variance, such as eliminating environmental variation. Later, Lande (1981) proved that the Castle–Wright equation could be applied to natural populations—that is, to populations not homozygous for all relevant factors. He also developed an equation for the variance of the estimator, var(nef ) ⫽

[

冢 冣冢 R2 8␴S 2

2

冣 冢

]

冣

4 var(R) var(␴S 2 ) ⫹ , R2 (␴S 2 ) 2 (2)

where R ⫽ (P¯1 ⫺ P¯2). Later, Cockerham (1986) presented a correction for the sampling variance of parental means. More recently, Zeng et al. (1990; Zeng 1992) studied the impact of linkage and unequal allelic effects on the estimator. Ollivier and Janss (1993) improved Wright’s corrections for dominance. Finally, Wu (1996)

corrected the estimator for parents with factors of both increasing and decreasing effects (gene dispersion). Despite these efforts, the effects of deviations from diploidy on the Castle– Wright equation have been ignored ( but see Chovnick and Fox 1953). In this article I consider the effects of violating this assumption in heterogametic species (i.e., in taxa in which the X chromosome is hemizygous in one sex) and in haplodiploid species. Because the case of haplodiploidy is simpler, I consider it first. Note that for simplicity, I illustrate these extensions using the traditional Castle– Wright equation. These extensions can be combined with other improvements to the Castle–Wright estimator detailed in Lynch and Walsh (1998).

Haplodiploid Species

Figure 1. Performance of the traditional Castle–Wright estimator in heterogametic species with a partially Xlinked trait. Actual number of genes is 20.

In haplodiploid organisms (such as bees and wasps), females are diploid and males are haploid. Although F2 and backcross females can be treated as normal diploids, males cannot; F2 males inherit their chromosomes solely from their mother and therefore carry a random assortment of grandparental chromosomes. Thus male F2 progeny are hemizygous at all loci. This lack of heterozygotes inflates the F2 segregation variance and leads to an underestimate of the number of effective factors (nef). Correcting for this effect in F2 males is analogous to correcting for F1 sexual haploids (see Chovnick and Fox 1953). Because the F2 variance of haplodiploids is twice that of diploids, it follows that (P ⫺ P2 ) 2 nef ⫽ 1 . 4␴S 2

(3)

Furthermore, it can be proved that Wright’s equation for estimating the effect of the largest factor is the same for diploids and haplodiploids (maximum effect, ⌽max ⫽ 兹1/nef). However, Lande’s equation for the variance needs modification. By the delta method, it can be shown that an approximate equation for the variance is var(nef ) ⫽

[

冢 冣冢 R2 4␴S 2

2

冣 冢

]

冣

4 var(R) var(␴S 2 ) ⫹ . R2 (␴S 2 ) 2 (4)

Computer simulations were used to test the accuracy of Equations 4 and 5. Both equations performed as well as the analogous diploid equations (data not shown). I provided Equation 3 to Weston et al. (1999), who applied it to data on the ge-

netics of wing size differences between two species of parasitic wasps (Nasonia). Using the traditional estimator, they would have found only 0.8 (SD ⫽ 0.38) effective factors contributing to the difference between these species. The corrected estimator suggests that the actual number is 1.6 (SD ⫽ 0.78), a number supported by their introgression data.

Sex Linkage Heterogametic species, such as humans and Drosophila, are effectively haplodiploid for their sex chromosomes (for simplicity, I assume males are the heterogametic sex). If some of the factors affecting a trait of interest are X linked, the Castle– Wright estimator is not appropriate. Because the X is never heterozygous in males, the variance of heterogametic F2 males will be inflated. This causes the Castle–Wright equation to underestimate nef. Figure 1 shows this underestimation is large when data from F2 males are considered, such as genetic studies of male secondary sexual characteristics (see True et al. 1997). In contrast, the traditional estimator overestimates nef when data derive from F2 females. This is because F2 females always get the same grand-maternal X chromosome from their father. The variance of the F2 will therefore be smaller than expected, leading to an overestimation of nef ( Figure 1). For similar reasons, estimates of gene number from backcross progeny of cross-

es using F1 males are also biased. In this case the Castle–Wright equation overestimates nef in both sexes ( Figure 1). However, employing F1 females in backcrosses avoids these problems. Correcting the estimator for sex linkage requires estimating the contribution of the X chromosome to the phenotype. This can be done by comparing the means of F1 males to those of the F1 females. Under an additive model, F1 males resemble their mothers more than F1 females, as males are hemizygous for the maternal X. Therefore the contribution of the X can be estimated as 円F¯ ⫺ F¯ 円 PX ⫽ ¯ 1(female) ¯ 1(male) . 円F1(female) ⫺ P(female) 円

(5)

From Figure 1, it is clear that this bias is relatively minor when Px, 0.25. As in haplodiploids, the corrected estimator is given by (P1 ⫺ P2 ) 2 . C␴S 2

nef ⫽

(6)

When data derive from F2 males, C⫽

8 . 1 ⫹ (PX )

(7)

When data derive from F2 females, C⫽

8

冢

冣

.

(8)

1 1 ⫺ PX 2

Or, when data derive from backcross progeny,

Brief Communications 275

C⫽

8

冢2 ⫺ 2 P 冣 1

.

(9)

1

X

As an example of the use of these estimators consider Val’s (1977) analysis of head shape differences in Hawaiian Drosophila. Male head width differs considerably between D. heteroneura and D. silvestris and a fraction of this difference was sex linked (⬃27%). Curiously Val found that in backcross data, nef depended on which sex was used in the backcross. Estimates based on backcrosses using F1 males were always greater than estimates based on crosses using F1 females. For example, nef ⫽ 4 in backcrosses to D. silvestris using females versus nef ⫽ 15 in backcrosses using males. Applying the above corrections to Val’s backcross data makes nef almost the same regardless of the sex used in the backcross (after correction, 4 versus 5.8). Similarly, in an F2 analysis of pteridine content in the heads of tsetse flies, McIntyre and Goodling (1996) underestimated the number of genes underlying this trait by 45–50% (cross 1, 2.3 versus 3.4; cross 2, 1.8 versus 2.6). In this article I have pointed out that the widely used Castle–Wright equation estimates of the number of effective factors may be biased in cases of haplodiploidy and heterogamety. I then presented two simple extensions to the estimator to correct for these cases. From the Department of Biology, University of Rochester, Rochester, NY, 14627. C. Jones is now at the Section of Evolution and Ecology, #1080, University of California, Davis, CA, 95616. I thank H. A. Orr and D. C. Presgraves for helpful discussion and commments on the manuscript. Also, I thank J. H. Werren for sharing his data. This work was supported by the David and Lucile Packard Foundation ( H. A. Orr), National Institutes of Health grant GM-51932 ( H. A. Orr), and a Caspari Fellowship from the University of Rochester (to C.D.J.). Address correspondence to the author at the current address above or E-mail: [email protected]. 䉷 2001 The American Genetic Association

Hatfield T, 1997. Genetic divergence in adaptive characters between sympatric species of stickleback. Am Nat 149:1009–1029. Lande R, 1981. The minimum number of genes contributing to quantitative variation between and within populations. Genetics 99:541–553. Lynch M and Walsh B, 1998. Genetics and analysis of quantitative traits. Sunderland, MA: Sinauer. McIntyre GS and Gooding RH, 1996. Variation in the pteridine content in the heads of tsetse flies ( Diptera: Glossinidae:Glossina Wiedmann): evidence for genetic control. Can J Zool 74:621–626. Ollivier L and Janss LL, 1993. A note on the estimation of the effective numbers of additive and dominant loci contributing to quantitative variation. Genetics 135: 907–909. Orr HA and Coyne JA, 1992. The genetics of adaptation: a reassessment. Am Nat 140:725–742. Sezer M and Butlin RK, 1998. The genetic basis of host plant adaptation in the brown planthopper (Nilaparvata lugens). Heredity 80:499–508. Tanksley SD, 1993. Mapping polygenes. Annu Rev Genet 27:205–233. True JR, Liu JJ, Stam LF, Zeng Z-B, and Laurie CC, 1997. Quantitative genetic analysis of divergence in male secondary sexual traits between Drosophila simulans and Drosophila mauritiana. Evolution 51:816–832. Val FC, 1977. Genetic analysis of the morphological differences between two interfertile species of Hawaiian Drosophila. Evolution 31:611–629. Weston RF, Quershi I, and Werren JH, 1999. Genetics of a morphological difference between two species of Nasonia. J Evol Biol 12:585–595. Wright S, 1968. Evolution and genetics of populations. Vol. 1. Chicago: University of Chicago Press. Wu RL, 1996. Quantitative genetic dissection of complex traits in a QTL-mapping pedigree. Theor Appl Genet 93:447–457. Wu R, Bradshaw HD, and Stettler RF, 1997. Molecular genetics of growth and development in Populus (Salicaceae). 5. Mapping quantitative trait loci affecting leaf variation. Am J Bot 84:143–153. Zeng Z-B, Houle D, and Cockerham CC, 1990. How informative is Wright’s estimator of the number of genes affecting a quantitative character? Genetics 126:235– 247. Zeng Z-B, 1992. Correcting the bias of Wright’s estimates of the number of genes affecting a quantitative character: a further improved method. Genetics 131: 987–1001. Received February 28, 2000 Accepted October 3, 2000 Corresponding Editor: Bruce S. Weir

References Bradshaw HD, Otto KG, Frewen BE, McKay JK, and Schemske DW, 1998. Quantitative trait loci affecting differences in floral morphology between two species of monkeyflower (Mimulus). Genetics 149:367–382. Castle WE, 1921 An improved method of estimating the number of genetic factors concerned in cases of blending inheritance. Proc Natl Acad Sci USA 81:6904–6907. Chovnick A and Fox AS, 1953. The problem of estimating the number of loci determining quantitative variation in haploid organisms. Am Nat 87:263–267. Cockerham CC, 1986. Modifications in estimating the number of genes for a quantitative character. Genetics 114:659–664. Gurganus, MC, Nuzhdin SV, Leips JW, and Mackay TFC. 1999. High-resolution mapping of quantitative trait loci for sternopleural bristle number in Drosophila melanogaster. Genetics 152:1585–1604.

276 The Journal of Heredity 2001:92(3)

Microsatellite Polymorphism in Closely Related Dogs L. Altet, O. Francino, and A. Sa´ nchez The effectiveness of microsatellites in parentage testing and individual identification has been proven in many species, including dogs. However, the use of these markers has not been extended to control for pedigrees in large populations of closely related animals. We have analyzed poly-

morphism in a set of 10 microsatellites over three generations of 360 pedigree rottweilers. Results were compared with two pure-bred populations of unrelated animals and with one population constituted by unrelated dogs of mixed breeds to measure polymorphism variation. We optimized this set of microsatellites to be analyzed by a semiautomated capillary electrophoresis method after amplification in two multiplex polymerase chain reactions (PCRs). The mean polymorphism information content (PIC) value in the rottweiler pedigree is 0.401 and the combined paternity exclusion probability (CPE) is 95.6%. These values are similar to those obtained in pure-bred populations of unrelated animals, and although polymorphism is reduced in relation to the pool population, we solved all paternity exclusions. In only a few cases did we have to use two additional microsatellites to solve individual identification of full-sib dogs. Microsatellites have been proven as a useful tool in parentage testing and individual identification in many species, owing to their high levels of polymorphism ( Hammond et al. 1994; Tautz 1989). However, these levels of polymorphism have an intrinsic limitation when analyzing dogs. Although there are more than 300 different breeds described all over the world, there is high genetic homogeneity within each breed because they have been inbred to select for characteristic traits. Therefore different lines of the same breed often share common ancestors and this results in decreased genetic variability within those breeds. Large numbers of microsatellites have been described in dogs to date ( Francisco et al. 1996; Holmes et al. 1993, 1995; Mellersh et al. 1994; Ostrander et al. 1993, 1995; Primmer and Mathews 1993; Thomas et al. 1997), however, genetic data has been pooled and analyzed mostly from mixed populations. In pure-bred populations, data about allele frequencies and polymorphism indexes are scarce ( Fredholm and Wintero 1995; Koskinen and Bredbacka 1999; Sutton et al. 1998; Zajc et al. 1997; Zajc and Sampson 1999). In these pure breeds, where only unrelated animals have been analyzed, intrabreed variation is reduced in relation to pooled populations. Moreover, breeders who rely on dog parentage testing and individual identification commonly use a few selected animals to produce salable pups. This creates difficulties, since microsatellite polymorphism analysis has not been utilized to

test for the paternity of closely related dogs. We have addressed this issue and present microsatellite polymorphism and allele frequencies in 360 pedigree rottweilers over the course of three generations. This population reached a maximum inbreeding coefficient of 16%. Our main purpose was to compare this microsatellite polymorphism with two different types of dog populations: one constituted of unrelated pure-bred animals with no common grandparents (golden retrievers and Labrador retrievers) and the other one constituted of 95 unrelated dogs of mixed breeds (pool population). In this way we are able to detect if microsatellite variability is substantially reduced in an inbred population where most of the animals share recent ancestors. These data are particularly relevant to dog breeders, which usually mate closely related individuals and consequently need to use accurate and sensitive parentage tests. Two multiplex PCR reactions were optimized for this set of microsatellites using a semiautomated fluorescent genotyping protocol.

Materials and Methods Animal Material The three generations of pedigree rottweilers originate from 47 crosses among 49 dogs used as breeding animals. Twentyseven of these breeding animals share recent common ancestors (at least grandparents) and 18 of the 47 crosses are inbred. The whole pedigree belongs to a single breeder and we consider it as a related pure-bred population. The pure-bred populations of unrelated dogs are composed of animals that have been collected from different breeders and have no common grandparents: 33 golden retrievers and 23 Labrador retrievers. The pool of 95 unrelated dogs includes animals from 24 different breeds: Newfoundland, Spanish greyhound, Belgian tervueren, German shepherd, Belgian groenendael shepherd, dachshund, Siberian husky, poodle, Yorkshire terrier, giant schnauzer, West Highland white terrier, Spanish mastiff, Neapolitan mastiff, boxer, basset hound, English cocker spaniel, Dalmatian, fila brasileiro, bullmastiff, Lhasa apso, Irish wolfhound, beagle, rottweiler, and American Staffordshire terrier. Dog genomic DNA was isolated as described elsewhere ( Francino et al. 1997). Microsatellite Markers A total of 10 unlinked markers have been studied (Mellersh et al. 1997, 2000), seven

dinucleotide markers—CPH5 and CPH9 ( Fredholm and Wintero 1995) and CXX 366, CXX410, CXX442, CXX459, and CXX474 (Ostrander et al. 1995)—and three tetranucleotide markers—CXX2001, CXX2010, and CXX2054 (Francisco et al. 1996). We have used two more 4 bp microsatellites— CXX2130 and CXX2158 (Francisco et al. 1996)—to solve the genetic identity in some cases of full sibs. These 10 microsatellites were amplified using two multiplex PCR reactions with five markers in each: multiplex-1 (CPH5, CXX366, CPH9, CXX474, and CXX459) and multiplex-2 (CXX2001, CXX2010, CXX2054, CXX410, and CXX442). The two multiplex reactions were carried out in 10 ␮l final reaction mixture containing PCR buffer (1⫻), 1.5 mM MgCl2, 0.2 mM of each dNTP (PE Biosystems), and 30–40 ng of dog genomic DNA. Primer concentration was optimized for each marker: 0.2 ␮M for 4 bp markers, 0.3 ␮M for CPH5, CPH9, and CXX366 markers, and 0.4 ␮M for the other 2 bp markers. One primer from each pair was fluorescently labeled with 6-FAM, TET, or HEX. Taq polymerase ( Life Technologies Inc.) was used at a final concentration of 0.075 U/␮l and 0.1 U/␮l in multiplex-1 and multiplex-2, respectively. Thermocycling conditions were 3 min at 94⬚C followed by 25 cycles of 94⬚C (30 s), 58⬚C for multiplex-1 and 55⬚C for multiplex-2 (30 s) and 72⬚C (30 s), followed by a final extension of 15 min at 72⬚C in an MJ Research Hot-Bonnet. The two additional microsatellites were used either together or were added to multiplex-1, using 0.4 ␮M of each primer. PCR reactions were analyzed by capillary electrophoresis in an ABI 310 Genetic Analyzer (Applied Biosystems, PE) and labeled PCR products were automatically sized relative to the internal standard (PRISM GENESCAN-350娂 TAMRA) with the GeneScan娂 Analysis 2.0 software. Computation and Analysis Allele frequencies and heterozygosity values were calculated with the Biosys-1 version 1.7 software package (Swofford and Selander 1989). The exclusion probability (PE) was calculated on the basis of the estimated allele frequencies (Jamieson 1994). Polymorphism information content (PIC) and PE values were calculated assuming that the genotypes of both parents were known ( Botstein et al. 1980; Jamieson 1994). We compared the mean PIC values per population using the Student’s t test with a significance level of 99.9% using the SAS package (SAS Institute 1995).

Results and Discussion Microsatellite allele frequencies and PIC values tend generally to be described in populations of unrelated animals. We were interested in measuring how these PIC and PE values are reduced when analyzing inbred populations. In such populations where most of the animals share recent ancestors, 10 microsatellites might be considered insufficient in order to distinguish closely related animals. Indeed, Sutton et al. (1998) found it necessary to use two typing systems ( DNA fingerprinting and microsatellites, or DNA fingerprinting alone) in order to elucidate such problems. In this study we have analyzed 10 microsatellite polymorphism in 360 pedigree rottweilers over three generations (R). Results were compared with two pure-bred populations of unrelated animals [golden retriever (G) and Labrador retriever ( L)] and also with a pool of 95 dogs from 24 different breeds (P). Allele frequencies for each microsatellite are shown in Table 1. It is interesting to note that the mean PIC value obtained for these 10 microsatellites in R show no significant differences with the pure-bred populations of unrelated dogs we have analyzed (G and L) (shown in Table 2). These PIC values are similar to the ones previously described analyzing 19 microsatellites in three populations of unrelated pure-bred dogs ( Zajc et al. 1997, 1999). The sustained polymorphism in R may be explained in two ways. First, the breeding animals possessed a relatively high level of heterozygosity. Second, despite an inbreeding coefficient of 16% in some crosses and a number of backcrosses among the R pedigree progeny, the breeder used a sufficient number of breeding animals to maintain variability. The mean PIC value for P shows significant differences in relation to the pure breeds (G, L, and R). This population contains a higher level of genetic heterogeneity since it comprises 24 different breeds with some characteristic alleles within each marker that result in a mean PIC value of 0.70. Otherwise this value does not guarantee sustained polymorphism levels when a certain pure breed is analyzed. For example, in our pure-bred populations (G, L and R), the PIC values range from 0.03 to 0.69, with only 40% of the PIC values being higher than 0.5 for the same microsatellites. Moreover, although there are some breed-specific alleles, the major differences between purebred populations are the relative allele

Brief Communications 277

Table 1. Breed allele frequencies and polymorphism information content (PIC) for each locus Locus

Alleles P

G

L

R

Locus

Alleles P

G

L

R

CXX366

160 164 166 238 242 PIC

.284 .519 .198 .231 .038 .540

.371 .581 .048 .419 — .427

.955 .023 .023 .457 — .085

.014 — .986 .229 — .027

CXX2010

226 230 234

.086 .441 .204

— .048 .532

— .478 .065

.568 — .203

PIC

.655

.438

.459

.519

CPH5

105 107 109 111 113 115 169 PIC

.011 .368 .142 .400 .068 .011 .028 .621

— .406 — .594 — — — .366

— .239 — .043 .717 — — .366

— .332 .297 .226 — .144 — .680

CXX442

157 159 161 163 165 167

.167 — .139 .483 .100 .083

.385 .019 — .558 .038 —

.133 — .233 .633 — —

— — — .985 .015 —

PIC

.669

.445

.468

.029

CPH9

133 137 139 141 145 147 149 151 PIC

.011 .394 .005 .005 .441 .059 .074 .011 .574

— .484 — — .516 — — — .375

.217 .348 — — .435 — — — .567

— .240 — — .760 — — — .298

CXX459

149 151 153 155 157 159 161 167 PIC

.022 .250 .114 .136 .147 .163 .163 .005 .808

— .355 .323 .226 .032 — .065 — .661

— .227 .023 .023 .682 — .045 — .430

— — .569 — .424 — .007 — .380

CXX474

107 109 111 113 115 117 119 121 170 174 178 PIC

.314 .016 .197 .080 .245 .059 — .090 .038 .054 — .753

.015 — .652 — — — .167 .167 .182 .030 — .470

— — .174 .696 .065 — .043 .022 — — — .443

.052 — .120 .391 — — — .437 .030 .060 .001 .569

CXX2054

140 144 148 150 154 158 162 166

.005 .102 .027 .161 .274 .161 .065 .113

— .076 — — .136 .045 .076 .455

— .109 — .587 .152 .087 .022 .043

— .003 — .065 .835 .006 — —

PIC

.822

.697

.578

.282

CXX2001

120 124 128 132 136 140 144 148 150 152 154 156 160 PIC

.005 .016 .229 .186 .027 .229 .154 .090 — — .016 .048 — .802

.030 — .045 .167 — .424 .121 .212 — — — — — .692

— — .152 — — .065 .609 .087 .022 .022 — .022 .022 .564

— .001 .389 — — .200 .099 .018 — — — .293 — .662

CXX410

93 97 102 105 108 110 112 114 116 118 122

.057 .172 — .046 .109 .040 .190 .034 .259 .075 .017

— .078 — .078 — .406 .344 .094 — — —

— .043 — .043 — .174 .652 — .087 — —

— — .003 .261 — .486 — — .004 .246 —

PIC

.823

.645

.497

.566

P, pool; G, golden retriever; L, Labrador retriever; R, rottweiler.

frequencies at each individual loci, as has previously been reported ( Fredholm and Wintero 1995; Zajc et al. 1997). CXX366 has the same number of alleles in G and in R, but shows an enormous difference in the PIC values (0.34 and 0.05, respectively) due to the different distribution of allele frequencies. Therefore it is important to

describe allele frequencies of microsatellites in a certain pure breed of dogs in order to select a useful set of markers for parentage control in that particular breed. Microsatellite sequence is also an important consideration when choosing markers. Although greater instability has been described for tetranucleotide motifs,

Table 2. Mean values for the polymorphism information content (PIC) and heterozygosities (Ho and He) and combined paternity exclusion probability (CPE) for each population He

Ho Pool Golden Retriever Labrador Retriever Rottweiler

.550 .592 .517 .452

(.052) (.028) (.061) (.083)

* P ⬍ .01.

278 The Journal of Heredity 2001:92(3)

.750 .607 .505 .452

PIC (.028) (.034) (.051) (.083)

.707 .522 .446 .401

making them more polymorphic than dinucleotide repeats, we have not found any correlation between the repeat motif and the polymorphism level of these microsatellites. The most informative markers for each breed were CPH5 (2 bp) in the R population, CXX2054 (4 bp) in the G and L populations, and CXX410 (2 bp) in the P population. It has also been reported that longer repeats can generate alternatively sized alleles more frequently than shorter ones ( Francisco et al. 1996). However, the longest microsatellite marker we analyzed (CXX2010) was not the most polymorphic one in any of our breeds. Furthermore, our results suggest that polymorphism levels in dogs depend on the specific pure breed in which a microsatellite has been studied. Markers CXX2001, CXX2054 and CXX410 do not always follow the above 2 bp or 4 bp repeat motifs, perhaps because of some variation contained within the repeat, as has been previously described for sequenced tetranucleotide markers ( Francisco et al. 1996). Alternatively it may be due to variability in the flanking regions of the repeats (Grimaldi and Crouau-Roy 1997). Similar results have been reported by Sutton et al. (1998) for 4 bp repeat microsatellites, with alleles separated by 2 bp or less. In conclusion, this work has addressed and substantiated the use of microsatellites for parentage testing and individual identification in a large population of closely related dogs. Although microsatellite polymorphism is reduced in the rottweiler pedigree if compared to the mixedbreed population, it is similar to the populations composed of unrelated purebred dogs. Therefore 10 microsatellites, multiplexed in two different PCR reactions, were enough to solve all paternity exclusions. Only a limited number of cases required the use of two additional microsatellites to allow individual identification of full-sib dogs. We also demonstrate the existence of allele-specific patterns in the G, L, and R breeds. Taken together, our results suggest that microsatellite polymorphism data obtained from heterogeneous populations cannot always be extrapolated to specific breeds. This fact must be taken into account when implementing microsatellite-based assay for parentage cases, especially in closely related dogs.

CPE (.103)* (.128) (.136) (.228)

.999 .986 .964 .956

From the Unitat de Gene`tica i Millora, Departament de Cie`ncia Animal i dels Aliments, Facultat de Veterina`ria, Edifici V, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain. We are thankful to the breeders Pere Pujals and Didier Schaer and to the Veterinary Clinic Hospital from the Universitat Auto`noma de Barcelona for providing the samples used in this

study. We are also thankful to Atilio Aranguren for SAS computation analysis and to Simon Boa and Marcel Amills for their critical reviews. Address correspondence to Laura Altet at the address above or e-mail: [email protected].

Swofford D and Selander RB, 1989. BIOSYS-1: a computer program for the analysis of allele variation in population genetics and biochemical systematics, release 1.7. Urbana: Illinois Natural History Survey.

䉷 2001 The American Genetic Association

Tautz D, 1989. Hypervariability of simple sequences as a general source for polymorphic DNA markers. Nucleic Acids Res 17:6463–6471.

References

Thomas R, Holmes NG, Fischer PE, Dickens HF, Breen M, Sampson J, and Binns MM, 1997. Eight canine microsatellites. Anim Genet 28:153–154.

Botstein D, White RL, Skolnick M, and Davis RW, 1980. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet 32:314–331. Francino O, Amills M, and Sa´nchez A, 1997. Canine Mhc DRB1 phenotyping by PCR-RFLP analysis. Anim Genet 28:41–45. Francisco LV, Langston AA, Mellersh CS, Neal CL, and Ostrander EA, 1996. A class of highly polymorphic tetra-nucleotide repeats for the canine genetic mapping. Mammal Genome 7:359–362. Fredholm M and Wintero AK, 1995. Variation of short tandem repeats within and between species belonging to the Canidae family. Mammal Genome 6:11–18. Grimaldi M-C and Crouau-Roy B, 1997. Microsatellite allelic homoplasy due to variable flanking sequences. J Mol Evol 44:336–340. Hammond H, Jin L, Zhong Y, Caskey CT, and Chakraborty R, 1994. Evaluation of 13 short tandem repeat loci use in personal identification applications. Am J Hum Genet 55:175–189. Holmes NG, Dickens HF, Parker HL, Binns MM, Mellersh CS, and Sampson J, 1995. Eighteen canine microsatellites. Anim Genet 26:132–133. Holmes NG, Mellersh CS, Humphreys SJ, Binns MM, Holliman A, Curtis R, and Sampson J, 1993. Isolation and characterization of microsatellites from the canine genome. Anim Genet 24:289–292. Jamieson A, 1994. The effectiveness of using co-dominant polymorphic allelic series for (1) checking pedigrees and (2) distinguishing full-sib pair members. Anim Genet 25(suppl 1):37–44. Koskinen MT and Bredbacka P, 1999. A convenient and efficient microsatellite-based assay for resolving parentages in dogs. Anim Genet 30:148–149. Mellersh C, Holmes N, Binns M, and Sampson J, 1994. Dinucleotide repeat polymorphisms at four canine loci ( LEI 003, LEI 007, LEI 008 and LEI 015). Anim Genet 25: 125. Mellersh CS, Hitte C, Richman M, Vignaux F, Priat C, Jouquand S, Werner P, Andre´ C, DeRose S, Patterson DF, Ostrander EA and Galibert F, 2000. An integrated linkage-radiation hybrid map of the canine genome. Mammal Genome 11:120–130. Mellersh CS, Langston AA, Acland GM, Fleming MA, Ray K, Wiegand NA, Francisco LV, Gibbs M, Aguirre G, and Ostrander EA, 1997. A linkage map of the canine genome. Genomics 46:326–336. Ostrander EA, Sprague GF Jr, and Rine J, 1993. Identification and characterization of dinucleotide repeat (CA)n markers for genetic mapping in dog. Genomics 16:207–213. Ostrander EA, Mapa FA, Yee M, and Rine J, 1995 One hundred and one simple sequence repeat-based markers for the canine genome. Mammal Genome 6:192–195. Primmer CR and Mathews ME, 1993. Canine tetra-nucleotide repeat polymorphism at VIAS-D10 locus. Anim Genet 24:332. SAS Institute, 1995. SAS user’s guide: statistics, version 6.12. Cary, NC: SAS Institute. Sutton MD, Holmes NG, Brennan FB, Binns MM, Kelly EP, and Duke EJ, 1998. A comparative genetic analysis of the Irish greyhound population using multilocus DNA fingerprinting, canine single locus minisatellites and canine microsatellites. Anim Genet 29:168–172.

Zajc I, Mellersh CS, and Sampson J, 1997. Variability of canine microsatellites within and between different dog breeds. Mammal Genome 8:182–185. Zajc I and Sampson J, 1999. Utility of canine microsatellites in revealing the relationships of pure bred dogs. J Hered 90:104–107. Received March 27, 2000 Accepted October 31, 2000 Corresponding Editor: Stephen J. O’Brien

Inheritance Pattern of RAPD Markers in Melipona quadrifasciata (Hymenoptera: Apidae, Meliponinae) M. G. Tavares, E. H. Ribeiro, L. A. O. Campos, E. G. Barros, and M. T. V. A. Oliveira Melipona quadrifasciata is an important pollinator agent in several regions of Brazil. Data concerning the genetics of this species are scarce in the literature. In this work we used the random amplified polymorphic DNA (RAPD) technique to determine the degree of polymorphism and the inheritance pattern of these molecular markers in this species. Our ultimate goal is to establish tools to be used in the study of the genomic organization of M. quadrifasciata. Genomic DNA from progenies F1 and BC1 were assayed with 79 different primers, yielding an average of 6.67 bands and 1.68 polymorphisms per primer. Three types of polymorphisms were detected: band presence/absence, band intensity, and fragment-length polymorphisms. Most of the observed polymorphisms were band presence/absence, typical of RAPD-dominant markers. The number of observed polymorphisms and their segregation in accordance with a Mendelian proportion confirm the importance of this technique for genome analysis of species like M. quadrifasciata that are poorly studied at the genetic level. Melipona quadrifasciata, a species of stingless bee popularly known as ‘‘mandac¸aia,’’ is an important pollinator agent in several Brazilian ecosystems. Many scientific studies have characterized the biology

and the behavior of these bees, however, very little effort has been devoted to the development and use of molecular genetic markers for this species. In spite of this, M. quadrifasciata has a great potential to become a model organism for genetic studies because of its lack of sting and the possibility of making controlled crosses in laboratory, and so its genomic organization must be known. In order to have markers to construct a linkage map for this species, this work aimed at the evaluation of the frequency of random amplified polymorphic DNA (RAPD) polymorphisms and the determination of their inheritance pattern within M. quadrifasciata. This map will be very useful for further genome analysis in Melipona and will also facilitate the mapping of quantitative trait loci (QTLs) and the characterization of the system for sex determination in this species, allowing comparisons with maps already known for other Hymenoptera species. The polymerase chain reaction (PCR)RAPD technique is based on the amplification of genomic DNA using decamer primers of random sequence in a PCR (Welsh and McClelland 1990; Williams et al. 1990). The amplification products are separated electrophoretically and can be directly visualized without using specific radioactive probes. Polymorphisms due to base changes at the primer annealing sites or due to deletions or insertions in the sequence flanked by the primers are frequent among genotypes and can be used as molecular markers. Many articles described this technique as a new tool to analyze the genome of insect species. Most of them dealt with the identification of species and subspecies, including mosquitoes ( Flavia et al. 1994; Wilkerson et al. 1993), aphids ( Black et al. 1992; Puterka et al. 1993), white fly (Gawel and Bartlett 1993), grasshoppers (Chapco et al. 1992), fruit fly ( Baruffi et al. 1995; Haymer and McInnis 1994), and several Hymenoptera parasites ( Edwards and Hoy 1993; Landry et al. 1993). The simplicity of the RAPD analysis, and the small amount of DNA required for the reactions, makes this technique a powerful and efficient tool for genetic analysis of various species. However, most RAPD markers are inherited in a dominant fashion. For this reason, RAPDs provide less information than codominant markers, such as microsatellites (Williams et al. 1990). Nevertheless, here we demonstrate that M. quadrifasciata is quite suitable for analysis with RAPD markers because it is

Brief Communications 279

a haplodiploid insect. The dominance of RAPD markers, in this case, can be overcome by performing the analyses in haploid drones, which supply information about the loci present in the parental female, distinguishing the homozygous loci from the heterozygous.

Materials and Methods Genetic Material A heterozygous virgin queen for the enzymatic marker hydroxybutyrate dehydrogenase (Hbdh, E.C. 1.1.1.30) was crossed with a drone that had been previously sterilized by treatment with gamma-ray irradiation (60,000r) emitted by a cobalt-60 pump. Subsequently the queen was introduced in a colony from which the original queen had been removed. Sixty haploid adult drones originated from this cross ( F1) were used to determine the type and frequency of RAPD polymorphisms. One of the haploid drones was backcrossed to the parental queen, resulting in the BC1 progeny. These progeny consisted of diploid males and females (workers and queens) at a frequency of 1:1 and a small amount of haploid males originated from nonfertilized eggs. The haploid males were detected by cytogenetic analysis of brighteyed pupae (data not shown). Enzymatic analyses of Hbdh (Alfenas et al. 1991) were performed in the BC1 progeny to select heterozygous diploid drones and workers for this locus which were used for the DNA analysis. In this way the inheritance pattern of RAPD markers in the haploid and diploid progeny could be accessed. DNA Extraction and PCR Amplification Each individual of the F1 and BC1 progenies was frozen in liquid nitrogen and immediately separated in two parts: head and mesosome, and metasome. The metasome of the BC1 individuals was used to determine the Hbdh phenotype, and the head and mesosome were used for DNA extraction. In both techniques a different pestle and mortar were used for each sample to eliminate cross-contamination. The genomic DNA was extracted as described by Waldschmidt et al. (1997). The amplification reaction mixture (25 ␮l) contained 3.5 mM MgCl2, 10 mM/50 mM TrisKCl (pH 8.3), 0.1 mM of each dNTP (dATP, dTTP, dGTP, dCTP), 0.4 ␮M of a decamer primer (Operon Technologies, Alameda, CA), 1 U of Taq DNA polymerase, and 12.5 ng of genomic DNA. The mixture was placed in a thermocycler model PTC-100

280 The Journal of Heredity 2001:92(3)

(MJ Research) programmed for 40 cycles. Each cycle consisted of 15 s at 94⬚C, 30 s at 35⬚C, and 1 min at 72⬚C. After the 40th cycle a final extension step of 7 min at 72⬚C was performed. The amplification products were resolved in 1.2% agarose gels immersed in TBE (90 mM Tris-borate, 10 mM EDTA), stained with ethidium bromide (10 ␮g/ml), visualized, and documented under ultraviolet ( UV) light. Preliminary tests (not shown) defined 79 primers to be used in the RAPD analyses. These primers produced consistent bands that could be easily scored. The polymorphic bands were scored as 1 (presence) and 0 (absence). The scoring process was performed twice by two different people. The two readings were compared and data considered ambiguous were removed. The bands were then analyzed for 1:1 segregation using the chisquared test (P ⬎ .05).

Results and Discussion Amplification of the genomic DNA of M. quadrifasciata with 79 random primers yielded 527 bands, with an average of 6.67 bands per primer. Three hundred ninetyfour bands were monomorphic and 133 were polymorphic, an average polymorphism of 1.68 bands per primer. Most RAPD markers are inherited in a dominant way. In general, it is not possible to distinguish heterozygous- from homozygous-dominant individuals at such loci; both have the ‘‘band present’’ phenotype. In this work, three types of polymorphisms were observed: band presence/absence, band intensity, and fragment length polymorphisms ( Figure 1). These types of polymorphisms occurred 99, 12, and 22 times, respectively. The majority of the markers (95.5%) for which the queen was heterozygous segregated 1:1 in the F1 drones (P ⬎ .05). These types of polymor-

Figure 1. RAPD products generated from haploid drones of M. quadrifasciata showing (a) band presence/absence (primer OPR-6), (b) band intensity (primer OPH-17), and (c) fragment length polymorphisms (primer OPC-01). The arrows point out the polymorphic bands.

Figure 2. RAPD products generated from F1 and BC1 progenies, showing the Mendelian inheritance of the polymorphisms for band presence/absence (arrows), generated with primer OPH-19 (A) and OPR-6 ( B). Lanes are as follows: 1, BC1 parental drone; 2–9, haploid drones ( F1); 10–17, diploid drones ( BC1); 18–25, workers ( BC1).

Figure 3. RAPD products generated from F1 and BC1 progenies, showing the Mendelian inheritance of the band intensity polymorphisms (arrows), generated with primer OPH-11. Lanes are as follows: 1, BC1 parental drone; 2– 9, haploid drones ( F1); 10–17, diploid drones ( BC1); 18–25, workers ( BC1).

phisms have also been observed during the construction of linkage maps for Apis mellifera ( Hunt and Page 1992), Neurospora crassa (Williams et al. 1990), Helianthus (Rieseberg et al. 1993), and Eucalyptus (Grattapaglia and Sederoff 1994). Considering the three possible types of polymorphisms, some primers were selected to verify the inheritance pattern of these markers in the BC1 progeny. For this purpose, we analyzed gels containing the amplification products of the parental drone, haploid drones, diploid drones, and workers ( Figures 2 and 3). As observed in Figure 2A, the presence/ absence polymorphism generated by primer OPH-19 segregated 1:1 in the F1 individuals, revealing that the queen was heterozygous for this locus. As the drone used to generate the BC1 progeny possessed the marker, each of the progeny also presented it. On the other hand, DNA analysis of the parental drone with primer OPR-6 ( Figure 2B, lane 1), which also generated presence/absence polymorphism in the F1, revealed that this individual did not show the band. Consequently this marker showed a Mendelian segregation within the diploid BC1 progeny, as expected for a cross involving a heterozygous queen and a hemizygous recessive male (not showing the corresponding band). Twelve markers (9.02%) segregated in the haploid drones of the F1 as band intensity polymorphisms, showing that the queen was heterozygous for these loci. It was also observed that these markers segregated in the BC1 progeny independent of the parental drone phenotype ( Figure 3). The segregation of these polymorphisms in the BC1 progeny may be due to intrinsic characteristics of this type of marker which may lead to differential amplifications. The differences could be the result of a different number of sequences in tandem or the degree of mismatches between the primer and its binding site. One of the amplification products generated with primer OPC-1 in the F1 segregated for band size ( Figure 4). Codominant alleles must be amplified with the same primer and be present simultaneously in heterozygous individuals. As expected, both alleles for this locus were present in half of the workers and half of the diploid drones in the BC1 progeny. The other half of the progeny and the parental drone showed only the small-size band ( Figure 4). These results clearly show that the queen was heterozygous for this locus. Additional hybridization experiments

Brief Communications 281

Tingey SV, 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531–6535. Received July 12, 1999 Accepted October 31, 2000 Corresponding Editor: Ross MacIntyre

Chromosomal Mapping of 18S-28S rRNA Genes and 10 cDNA Clones of Human Chromosome 1 in the Musk Shrew (Suncus murinus) Figure 4. RAPD products generated from F1 and BC1 progenies, showing the Mendelian inheritance of the fragment length polymorphisms (arrows), generated with the primer OPC-01. Lanes are as follows: 1, BC1 parental drone; 2–8, haploid drones ( F1); 9–17, diploid drones ( BC1); 18–25, workers ( BC1). Note the occurrence of only one of the bands for each F1 drone and the presence of heterozygous individuals in the BC1 progeny ( lanes 13, 16, 19, and 22).

should be performed to confirm the homology between the two ‘‘alleles’’ detected in this locus. The clear segregation pattern present in the F1 and BC1 progenies demonstrate the usefulness of RAPD markers in genetic analysis of M. quadrifasciata. Genetic linkage analyses among the markers detected in this work are presently under way to establish a linkage map for this species to be used in future genetic studies.

Edwards O and Hoy M, 1993. Polymorphisms in two parasitoids detected using random amplified polymorphic DNA (RAPD) PCR. Contr Theor Appl Pest Manage 3:243–257.

From the Departamento de Biologia Geral, Universidade Federal de Vic¸osa. Av. PH Rolfs, s/n, 36.571-000, Vic¸osa, MG, Brazil ( Tavares, Ribeiro, Campos, Barros) and Departamento de Biologia, Instituto de Biocie ˆncias, Letras e Cie ˆncias Exatas de Sa˜o Jose´ do Rio Preto ( IBILCE/UNESP), Sa˜o Jose´ do Rio Preto, SP, Brazil (Oliveira). We are grateful to M. A. Del Lama for suggestions and assistance. This research was supported by CNPq (Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico) and FAPEMIG ( Fundac¸a˜o de Amparo a` Pesquisa de Minas Gerais). Address correspondence to Mara Garcia Tavares at the address above.

Haymer D and McInnis D, 1994. Resolution of populations of the Mediterranean fruit fly at the DNA level using random primers for the polymerase chain reaction. Genome 37:244–248.

䉷 2001 The American Genetic Association

References Alfenas AC, Peters I, Brune W, and Passador GC, 1991. Eletroforese de proteı´nas e isoenzimas de fungos e esseˆncias florestais. Vic¸osa, Brazil: Imprensa Universita´ria. Baruffi L, Damiani G, Guglielmino CR, Bandi C, Malacrida AR, and Gasperi G, 1995. Polymorphism within and between populations of Ceratitis capitata: comparison between RAPD and multiloci enzyme electrophoresis data. Heredity 74:425–437. Black WC IV, DuTeau N, Putrka G, Nechols J, and Pettorini J, 1992. Use of the random amplified polymerase chain reaction (RAPD-PCR) to detect DNA polymorfisms in aphids. Bull Entomol Res 82:151–159. Chapco W, Ashton N, Martel R, and Antonishyn N, 1992. A feasibility study of the use of random amplified polymorphic DNA in the population genetics and systematics of grasshoppers. Genome 35:569–574.

282 The Journal of Heredity 2001:92(3)

Flavia G, Dimopoulos G, and Louis C, 1994. Analysis of the Anopheles gambiae genome using RAPD markers. Insect Mol Biol 3:149–157. Gawel N and Bartlett A, 1993. Characterization of differences between whiteflies using RAPD-PCR. Insect Mol Biol 2:33–38. Grattapaglia D and Sederoff R, 1994. Genetic linkage maps of Eucalyptus grandis and E. urophylla using a pseudo-testcross: mapping strategy and RAPD markers. Genetics 137:1121–1137.

Hunt GJ and Page RE, 1992. Patterns of inheritance with RAPD molecular markers reveal novel types of polymorphism in honey bee. Theor Appl Genetics 85:15–20. Landry B, Dextraze L, and Boivin G, 1993. Random amplified polymorphic DNA markers for DNA fingerprinting and genetic variability assessment of min parasitic wasp species used in biological control programs of phytophagous insects. Genome 36:580–587. Puterka GJ, Black WC, Steiner WM, and Burton RL, 1993. Genetic variation and phylogenetic relationships among worldwide collections of the Russian wheat aphid Diuraphis noxia, inferred from allozyme and RAPD-PCR markers. Heredity 70:604–618. Rieseberg LH, Choi H, Chan R, and Spore C, 1993. Genomic map of a diploid hybrid species. Heredity 70: 285–293. Waldschmidt AM, Saloma˜o TMF, Barros EG, and Campos LAO, 1997. Extraction of genomic DNA from Melipona quadrifasciata ( Hymenoptera:Apidae, Meliponinae). Braz J Genet 20:421–423. Welsh J and McClelland M, 1990. Fingerprinting genomes using PCR with consensus tRNA gene primers. Nucleic Acids Res 19:861–866. Wilkerson RC, Parsons TJ, Albright DG, Klein TA, and Braun MJ, 1993. Random amplified polymorphic DNA (RAPD) markers distinguish cryptic mosquito species ( Diptera:Culicidae:Anoplheles). Insect Mol Biol 1:205– 211. Williams JGK, Kubelik AR, Livak KJ, Rafalski JA´, and

A. Kuroiwa, K. Matsubara, T. Nagase, N. Nomura, J. K. Seong, A. Ishikawa, R. V. P. Anunciado, K. Tanaka, T. Yamagata, J. S. Masangkay, V.-B. Dang, T. Namikawa, and Y. Matsuda The direct R-banding fluorescence in situ hybridization (FISH) method was used to map 18S-28S ribosomal RNA genes and 10 human cDNA clones on the chromosomes of the musk shrew (Suncus murinus). The chromosomal locations of 18S28S ribosomal RNA genes were examined in the five laboratory lines and wild animals captured in the Philippines and Vietnam, and the genes were found on chromosomes 5, 6, 9, and 13 with geographic variation. The comparative mapping of 10 cDNA clones of human chromosome 1 demonstrated that human chromosome 1 consisted of at least three segments homologous to Suncus chromosomes (chromosomes 7, 10, and 14). This approach with the direct R-banding FISH method is useful for constructing comparative maps between human and insectivore species and for explicating the process of chromosomal rearrangements during the evolution of mammals. The musk shrew [Suncus murinus (Soricidae, Insectivora)] is distributed widely from eastern Africa to East and Southeast Asia, and the basic chromosome number of this species is 2n ⫽ 40. The chromosome number ranges from 30 to 40 in wild populations of several localities (Aswathanarayana and Prakash 1976; Ishikawa et al. 1989; Sam et al. 1979; Yosida 1982), and variation has been reported in the size and morphology of both autosomes and sex chromosomes (Manna and Talukdar 1967; Obara and Miyai 1981; Sharma et al. 1970; Yosida 1982). Laboratory lines of this species have been established and

several spontaneous mutations in coat hair, behavior, diabetes, morphology, and others have been reported ( Ishikawa et al. 1998; Matsuura et al. 1999; Ohno et al. 1994, 1998). Effects of various emetic and antiemetic drugs have been well studied using Suncus for its potential use as an experimental animal model of mammals in emetic research ( Ueno et al. 1987), because Suncus is one of the few mammalian species (e.g., dog, cat, monkey) that vomit in response to emetic drugs. Therefore Suncus is very useful as a small experimental mammal for screening drugs that prevent motion sickness (Matsuki et al. 1997; Okada et al. 1995). The importance of Suncus is increasing in the field of biomedical sciences as an experimental animal, however, genetic analysis is almost impossible in this species because no chromosome maps have been established. Direct R-banding fluorescence in situ hybridization ( FISH) is a powerful technique for constructing high-resolution cytogenetic maps rapidly and efficiently by localizing cloned DNA sequences precisely onto banded metaphase chromosomes (Matsuda et al. 1992; Takahashi et al. 1990). Furthermore, this technique makes it possible to detect chromosomal homology between different species by localizing cDNA clones isolated from map-rich species to chromosomes of map-poor species ( Kuroiwa et al. 1998). In this study, to apply the direct R-banding FISH method for genome mapping in Suncus, first we demonstrated replication R-banding patterns of Suncus chromosomes and compared the karyotypes of the five laboratory lines derived from different localities. Using this method we mapped 18S-28S ribosomal RNA (rRNA) genes in the five laboratory lines and wild animals captured in Southeast Asia. Furthermore, 10 cDNA clones of human chromosome 1 were localized directly to Suncus chromosomes to detect conserved homology between human chromosome 1 and Suncus chromosomes.

Table 1. Chromosomal locations of 18S-28S ribosomal RNA genes in five laboratory lines and wild animals

DNA Probes A 6.6 kb mouse genomic DNA fragment was used for chromosomal localization of

Localities

No. of animals

Chromosome no.

TKU OKI BAN NAG KAT Wild shrews Wild shrews

Tokunoshima (Japan) Okinawa (Japan) Bangladesh Nagasaki (Japan) Katmandu ( Nepal) Tabon (the Philippines) Hanoi ( Vietnam)

么2 么1 么1 么2 么3 么2乆3 乆5

5, 6, 6, 5, 5, 5, 5,

18S-28S rRNA genes ( Kominami et al. 1982). Human cDNA clones of HSA1 (Homo sapiens chromosome 1) listed in Table 2 were used for comparative mapping between human chromosome 1 and Suncus chromosomes. The cDNA clones were isolated from the size-fractionated cDNA libraries of the human brain (Seki et al. 1997) and these clones have been mapped by GeneBridge 4 radiation hybrid panel ( Ishikawa et al. 1997). Their cytogenetic locations were estimated by the DNA markers close to the cDNA clones, of which precise locations on chromosomal bands have been determined [see the genome directory in Nature 1995; 377(suppl)]. The information on these clones is available to the public in the KDRI database ( http://www.kazusa.or.jp). KIAA444 and 463 clones showed 89.2% and 99.3% identities to human 218kD Mi-2 and OCT (plexin A2) genes, respectively ( Ishikawa et al., 1997). Other clones showed less than 65% identities. Chromosome Preparation and in situ Hybridization Cell culture for R-banded chromosome preparations and FISH were performed as described by Matsuda and Chapman

9, 9, 9, 6, 6, 6, 6,

13 13 13 9, 13 9, 13 9, 13 9, 13

(1995) with slight modifications. Lymphocytes were isolated from the spleen of adult animals, washed twice with serumfree TC199 medium, and transferred to 25 cm2 culture flasks containing 10 ml TC199 supplemented with 20% fetal calf serum, 3 ␮g/ml concanavarin A, 10 ␮g/ml lipopolysaccharide, 50 ␮g/ml HA15 (Murex), and 5 ⫻ 10⫺5 M mercaptoethanol. The splenocytes were cultured for 46 h, and thymidine (300 ␮g/ml) was added in culture medium. After 14 h the cells were washed twice with serum-free TC199 medium, transferred to the culture medium, and then cultured with BrdU (30 ␮g/ml) for an additional 3.5 h. Colcemid (0.02 ␮g/ml) was added 30 min before harvesting the cells (total culture time with BrdU is 4 h). The chromosome slides were exposed to ultraviolet ( UV) light after staining with Hoechst 33258 and stained with Giemsa for analyzing R-banded patterns. For Gband analysis, the cells were cultured without thymidine and BrdU treatments. The G-banded chromosomes were obtained by trypsin-treatment as described by Seabright (1971). The DNA probes were labeled by nick translation with biotinylated 16-dUTP (Roche Diagnostics). The hybridized 18S-

Table 2. List of human cDNA clones used for FISH mapping and their chromosomal locations

Materials and Methods Animals Five laboratory lines and wild animals listed in Table 1 were used in this study. The laboratory lines are maintained at the Laboratory of Animal Genetics, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan.

Lines

Gene Clone no. ( KIAA)a no.b

Accession no.c

Size (kb)

444 445 448 449 450 452 454 456 458 463

AB007913 AB007914 AB007917 AB007918 AB007919 AB007921 AB007923 AB007925 AB007927 AB007932

6.6 6.5 6.6 6.5 7.0 6.3 6.7 6.3 6.6 6.3

HG0035 HG0102 HG0135 HG0177 HG0217 HG0323 HG0502 HG0633 HG0715 HG0942

Chromosomal locations

Number of cells with specific signalsf

HSAd

SMUe

0g

1g

2g

3g

4g

1p36.2 1p36.1 1p22.1-p22.3 1q32.1-p32.3 1p36.2-p36.3 1p34.1-p34.3 1p13.3-p21.1 1q31 1p36.2 1q25.3-q31.3

14q4.4-q4.5 10q5.1 10q5.1 10q4.1 10q5.2 14q3.5 7q1.8-q1.9 10q3.3-q3.4 14q4.4-q4.5 10q4.3

77 80 73 77 84 87 84 77 84 57

0 0 10 0 3 7 0 0 10 3

13 13 10 20 13 3 13 10 3 23

7 0 7 3 0 3 0 10 0 10

3 7 0 0 0 0 3 3 3 7

Accession numbers in KDRI database ( http://www.kazusa.or.jp). Clone numbers in KDRI database. c Accession numbers in DDBJ, EMBL, and GenBank databases. d Locations identified by using GeneBridge 4 radiation hybrid panel ( Ishikawa et al. 1997). e Data in the present study. f Number of cells examined is 100. g Number of chromatid with signals per cell.

a

b

Brief Communications 283

28S rRNA gene probes were stained with fluoresceinated avidin ( Vector Laboratories). The hybridized cDNA probes were reacted with goat antibiotin antibodies ( Vector Laboratories) and then stained with fluoresceinated donkey anti-goat IgG ( Nordic Immunology). The slides were stained with 0.75 ␮g/ml propidium iodide (PI) for observation. FISH images were observed under a Nikon fluorescence microscope using Nikon filter sets B-2A and UV2A. Kodak Ektachrome ASA100 films were used for microphotography.

Results and Discussion

Figure 1. The G- ( left) and replication R-banded (right) patterns in NAG line of Suncus murinus.

Figure 2. The morphological variation of chromosomes 16 and Y in the five laboratory lines. M, SM, A, and SA are metacentric, submetacentric, acrocentric, and subacrocentric chromosomes, respectively. Arrowheads on the left side of the chromosomes indicate the centromeric regions.

284 The Journal of Heredity 2001:92(3)

Variations of Chromosomal Morphology The G- and R-banded karyotypes of the NAG line are shown in Figure 1. The G- and R-banded patterns were analyzed in detail using 13 and 10 metaphase spreads of the NAG line, respectively. Ambiguous or unusual banded patterns were observed for only a small number of chromosomes, 1.9% (10/520 chromosomes) and 1.5% (6/400 chromosomes) in G- and R-banded metaphases, respectively. These included unclear bands, dark-stained minor bands, and light-stained major bands. The replication R-banded pattern was the complete reverse of the G-banded pattern obtained by trypsin treatment as shown in Figure 1. Next we examined 4, 5, 5, 5, and 23 metaphase spreads of the OKI, TKU, BAN, KAT, and NAG lines, respectively, and found morphologic variations in chromosomes 16 and Y ( Figure 2). We measured the length of the long arm (q) and short arm (p) of chromosomes 16 and Y in these five lines, and calculated the ratio of the long arm to the short arm (r ⫽ q/p). The r values of chromosomes 16 were 3.4, 4.1, 16.8, 23.3, and 39.6 in the OKI, TKU, NAG, BAN, and KAT lines, respectively. The r values of Y chromosomes were 1.2, 1.3, 1.3, 1.4, and 2.2 in the OKI, TKU, NAG, BAN, and KAT lines, respectively. Following the nomenclature for the centromeric position of chromosomes ( Levan et al. 1964), chromosome 16 was subacrocentric in the OKI and TKU lines and acrocentric in other three lines. Y chromosomes were submetacentric in the KAT line and metacentric in the other four lines ( Figure 2). The morphology of chromosomes 16 and Y was consistent with that reported by Rogatcheva et al. (1996). Geographical variation in the size and morphology of the sex chromosomes has been reported for both the X chromosome (Obara and Miyai 1981) and the Y chromosome (Sharma et al. 1970; Yosida 1982). Y chromosomes were not different in the

Figure 3. Chromosomal localization of 18S-28S ribosomal RNA genes to chromosomes in (a,b) TKU, (c) OKI, and (d) KAT lines. A 6.6 kb genomic fragment of the 18S-28S ribosomal RNA gene was used as a biotinylated probe. Numbers indicate the chromosomes to which ribosomal RNA genes are localized. PI-stained R-banded patterns and Hoechst 33258 stained G-banded pattern are demonstrated in (a,c,d) and ( b), respectively.

size among the laboratory lines used in this study, hence the variation found in this study appears to be caused by pericentric inversion. Robertsonian fusion of acrocentric chromosomes has been reported by many researchers (Aswathanarayana and Prakash 1976; Ishikawa et al. 1989; Rogatcheva et al. 1997; Sam et al. 1979; Yosida 1982). The variation in the morphology of chromosome 16 was first found by chromosome banding analysis in this study. Chromosomal Locations of 18S-28S rRNA Genes The chromosomal locations of 18S-28S rRNA genes in the five laboratory lines

and wild animals are shown in Table 1, and the hybridization patterns in the TKU, OKI, and KAT lines are shown in Figure 3. The hybridization signals were observed in the telomeric regions of chromosomes 9 and 13 in all the lines and wild animals, while variations in signal distributions were found in the short arms of chromosomes 5 and 6. The signals were detected in chromosome 5 in the TKU line, chromosome 6 in the OKI and BAN lines, and both chromosomes 5 and 6 in the NAG and KAT lines and the wild animals. No variation was observed between individuals in the wild populations of the Philippines and Vietnam. We observed 130, 123,

Figure 4. Chromosomal localization of the (a,b) KIAA444 and (c,d) 445 clones to Suncus chromosomes. Arrows indicate the hybridization signals. KIAA444 and 445 were localized subregionally to SMU14q4.4-q4.5 and SMU10q5.1, respectively. R- and G-banded patterns are demonstrated in (a,c) and ( b,d), respectively.

115, 135, 112, 144, and 108 metaphase spreads in the OKI, KAT, TKU, BAN, and NAG lines, and the Vietnam and Philippines wild animals, respectively. The frequencies of metaphases, in which the probes completely hybridized to the chromosomes with low copy number of the genes, were 43.8, 70.3, 87.0, 61.4, 84.3, 87.5, and 70.4% in the OKI, KAT, TKU, BAN, and NAG lines, and the Vietnam and Philippines wild animals, respectively. The mtDNA haplotype analysis by Yamagata et al. (1995) indicated that the wild populations of Suncus were classified into three groups: the continental group ( Bangladesh and Nepal), the island’s group (insular countries and Vietnam), and the Malay group. Following this classification the BAN and KAT lines should be classified into the continental group, and the OKI, NAG, and TKU lines and the wild animals of the Philippines and Vietnam into the island’s group. However, the five laboratory lines and the wild populations were not classified into the two groups by the distribution patterns of the 18S-28S rRNA genes. The present results suggest that the chromosomal locations of the 18S-28S rRNA genes were conserved in chromosomes 9 and 13, and that either of chromosomes 5 and 6 with signals in the wild population was fixed in the TKU, OKI, and BAN lines in the process of domestication. Rogatcheva et al. (1997) reported that the 18S-28S rRNA genes were localized to the telomeric region of chromosomes 9 and 13 and the short arms of chromosome 5 in the four males of the KAT line by FISH methods. In this study, the hybridization signals of the rRNA genes in the short arms of chromosome 6 were observed in addition to chromosomes 5, 9, and 13 in three males of the KAT line maintained in our laboratory ( Table 1). These results suggest a possibility that this variation arose in the process of domestication from the wild population with polymorphism in the chromosomal distribution of the rRNA genes. Mapping of Human cDNA Clones We localized 10 cDNA clones of HSA1, which were isolated as long-sized fulllength cDNA clones, to Suncus murinus (SMU) chromosomes. The hybridization patterns of two clones, KIAA444 and 445, on Suncus chromosomes are demonstrated in Figure 4. The chromosomal locations of the 10 cDNA clones and the hybridization efficiency for each clone to Suncus chromosomes are shown in Table 2. No consistent fluorescence signals were ob-

Brief Communications 285

Figure 5. A comparative map between HSA1 and Suncus chromosomes. The chromosomal segments of HSA1 homologous to Suncus chromosomes are represented on the right side of the ideogram of HSA1. The clone numbers are represented by accession numbers in the KDRI database. The ideograms of HSA1 and SMU7, 10, and 14 were demonstrated on referring to ISCN (1995) and Rogatcheva et al. (1996), respectively.

served on other chromosomes. The comparative mapping in this study revealed the presence of homology between HSA1 and Suncus chromosomes 7, 10, and 14 (SMU7, 10, and 14) ( Figure 5). The homologous regions of SMU10 identified by six genes were divided into four segments in HSA1. The order of KIAA456, 449, and 463 clones in HSA1 was conserved in Suncus chromosomes, and the other three genes, KIAA450, 445, and 448, were mapped in the small q5.1-q5.2 region of SMU10. Three genes, KIAA444, 458, and 452, were mapped in the distal region of SMU14, however, the homologous block was interrupted by KIAA445 in HSA1. These results suggest the presence of multiple inversion events in HSA1 that occurred after the divergence of primates and insectivores. In view of the fact that the insectivore appeared in the Cretaceous period around 150 million years ago (Ohno 1993), comparing insectivore chromosomes with human chromosomes is one of important approaches for explaining the process of chromosomal evolution in mammals. Di-

286 The Journal of Heredity 2001:92(3)

xkens et al. (1998) suggested by the ZOOFISH analysis that only 10 breakages were necessary to transform the human karyotype into the karyotype of the common shrew (Sorex araneus, Insectivora). Although the small segments and intrachromosomal inversions were not detectable in their study, they revealed that the present-day human karyotype was very similar to the ancestral mammalian founder karyotype. On the contrary, our present study indicates a possibility that more chromosomal rearrangements occurred between human and S. murinus, although the information of comparative mapping is limited to HSA1-linked genes. The search of chromosomal homology by comparative mapping of functional genes provides a clue for clarifying the phylogenetic relationship between human and insectivore chromosomes and the ancestral genome structure prior to the separation of the primate and insectivore lineage. A considerable amount of information on comparative mapping to human chromosomes is being accumulated for several

other species than mouse. The comparative maps of human chromosome 1 and chromosomes of several species, including rat, zebrafish, and goat, were reported recently (White et al. 1999). Our research is the first report of comparative mapping between Suncus and human, and the direct R-banding FISH used in this study will contribute to construction of the comparative maps between Suncus and other mammalian species. It is necessary to increase mapping data, and making a substantial map between human and S. murinus is our essential subject in the future. From the Laboratory of Animal Genetics, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan ( Kuroiwa, Ishikawa, Anunciado, Tanaka, Yamagata, and Namikawa), Laboratory of Cytogenetics, Division of Bioscience, Graduate School of Environmental Earth Science, Hokkaido University, Sapporo, Japan ( Kuroiwa and Matsubara), Laboratory of Gene Structure I, Kazusa DNA Research Institute, Chiba, Japan ( Nagase and Nomura), Division of Laboratory Animal Science, Medical Research Center, College of Medicine, Yonsei University, Seoul, Korea (Seong), College of Veterinary Medicine, University of the Philippines at Los Ban˜os, Laguna, the Philippines (Anunciado and Masangkay), Faculty of Animal Science, Hanoi Agricultural University, Gialam, Hanoi, Vietnam ( Dang), and Chro-

mosome Research Unit, Faculty of Science, Hokkaido University, North 10, West 8, Kita-ku, Sapporo 060-0810, Japan (Matsuda). This work was supported by a research grant (no. 09680829) from the Ministry of Education, Science, Sports, and Culture of Japan (to Y.M.). Address correspondence to Yoichi Matsuda at the address above or e-mail: [email protected].

and segregating the sucrase deficients (suc/suc). Exp Anim 43:111–113.

䉷 2001 The American Genetic Association

Okada F, Saito H, and Matsuki N, 1995. Blockade of motion- and cisplatin-induced emesis by a 5-HT2 receptor agonist in Suncus murinus. Br J Pharmacol 114:931–934.

References Aswathanarayana NV and Prakash KLS, 1976. A new chromosome number for the Asian house shrew, Suncus murinus. Chromosomes Today 5:447. Dixkens C, Klett C, Bruch J, Kollak A, Serov OL, Zhdanova N, Vogel W, and Hameister H, 1998. ZOO-FISH analysis in insectivores: ‘‘evolution extols the virtue of the status quo.’’ Cytogenet Cell Genet 80:61–67. ISCN, 1995. An international system for human cytogenetic nomenclature (Mitelman F, ed). Basel: S. Karger. Ishikawa A, Akadama I, Namikawa T, and Oda S, 1989. Development of a laboratory line (SRI line) derived from the wild house musk shrew, Suncus murinus, indigenous to Sri Lanka. Exp Anim 38:231–237. Ishikawa A, Naito J, and Namikawa T, 1998. Open eyelids at birth, a new mutation in the musk shrew, Suncus murinus. Exp Anim 47:105–110. Ishikawa K, Nagase T, Nakajima D, Seki N, Ohira M, Miyajima N, Tanaka A, Kotani H, Nomura N, and Ohara O, 1997. Prediction of the coding sequences of unidentified human genes. VIII. 78 new cDNA clones from brain which code for large proteins in vitro. DNA Res 4:307– 313. Kominami R, Mishima Y, Urano Y, Sakai M, and Muramatsu M, 1982. Cloning and determination of the transcription termination site of ribosomal RNA gene of the mouse. Nucleic Acids Res 10:1963–1979. Kuroiwa A, Watanabe T, Hishigaki H, Takahashi E, Namikawa T, and Matsuda Y, 1998. Comparative FISH mapping of mouse and rat homologues of twenty-five human X-linked genes. Cytogenet Cell Genet 81:208–212. Levan A, Fredga K, and Sandberg AA, 1964. Nomenclature for centromeric position on chromosomes. Hereditas 52:201–220. Manna GJ and Talukdar M, 1967. Chromosomes of bone marrow cells of the Indian house shrew, Suncus murinus. Mammalia 31:288–294. Matsuda Y and Chapman VM, 1995. Application of fluorescence in situ hybridization in genome analysis of the mouse. Electrophoresis 16:261–272. Matsuda Y, Harada Y-H, Natsuume-Sakai S, Lee K, Shiomi T, and Chapman VM, 1992. Location of the mouse complement factor H gene (cfh) by FISH analysis and replication R-banding. Cytogenet Cell Genet 61: 282–285. Matsuki N, Wang CH, Okada F, Tamura M, Ikegaya Y, Lin SC, Hsu YN, Chaung LJ, Chen SJ, and Saito H, 1997. Male/female differences in drug-induced emesis and motion sickness in Suncus murinus. Pharmacol Biochem Behav 57:721–725. Matsuura A, Ohno T, Matsushima T, Namikawa T, and Ishikawa A, 1999. Delayed development of reflexes and hyperactive locomotion in the spontaneous mutant ‘‘waltzing’’ of the musk shrew, Suncus murinus. Exp Anim 48:191–197. Obara Y and Miyai T, 1981. A preliminary study on the sex chromosome variation of the Ryukyu house shrew, Suncus murinus riukiuanus. Jpn J Genet 56:365–371.

Ohno T, Yoshida F, Ichikawa Y, Matsuo S, Hotta N, Terada M, Tanaka S, Yamashita K, Namikawa T, and Kitoh J, 1998. A new spontaneous animal model of NIDDM without obesity in the musk shrew. Life Sci 62:995– 1006.

Rogatcheva MB, Serdyukova NA, Biltueva LS, Perelman PL, Borodin PM, Oda S, and Graphodatsky AS, 1997. Localization of the genes for major ribosomal RNA on chromosomes of the house musk shrew, Suncus murinus, at meiotic and mitotic cells by fluorescence in situ hybridization and silver staining. Genes Genet Syst 72: 215–218. Rogatcheva MB, Borodin PM, Matsuda Y, and Oda S, 1996. Standard karyotype of the house musk shrew, Suncus murinus ( Insectivora, Soricidae). Cytologia 61: 197–208. Sam C-K, Young H-S, and Dhaliwal SS, 1979. The G- and C-bands in relation to Robertsonian polymorphism in the Malayan house shrew, Suncus murinus (Mammalia, Insectivora). Caryologia 32:355–363. Seabright M, 1971. A rapid banding technique for human chromosomes. Lancet 7731:971–972 Seki N, Ohira M, Nagase T, Ishikawa K, Miyajima N, Nakajima D, Nomura N, and Ohara O, 1997. Characterization of cDNA clones in size-fractionated cDNA libraries from human brain. DNA Res 4:345–349. Sharma T, Pathak S, and Ray-Chaudhuri SP, 1970. Large sex chromosomes of Indian house shrew, Suncus murinus ( L.). Nucleus 13:62–69. Takahashi E, Hori T, O’Connell P, Leppert M, and White R, 1990. R-banding and nonisotopic in situ hybridization: precise localization of the human type II collagen gene (COL2A1). Hum Genet 86:14–16. Ueno S, Matsuki N, and Saito H, 1987. Suncus murinus: a new experimental model in emesis research. Life Sci 41:513–518. White PS, Forus A, Matise TC, Schutte BC, Spieker N, Stanier P, Vance JM, and Gregory SG, 1999. Report of the fifth international workshop on human chromosome 1 mapping 1999. Cytogenet Cell Genet 87:143–171. Yamagata T, Ohishi K, Faruque MO, Masangkay JS, BaLoc C, Vu-Binh D, Mansjoer SS, Ikeda H, and Namikawa T, 1995. Genetic variation and geographic distribution on the mitochondrial DNA in local populations of the musk shrew, Suncus murinus. Jpn J Genet 70:321–337. Yosida TH, 1982. Cytogenetical studies on Insectivora, II. Geographical variation of chromosomes in the house shrew, Suncus murinus (Sorisidae), in East, Southeast and Southwest Asia, with a note on the karyotype evolution and distribution. Jpn J Genet 57:101–111. Received April 21, 2000 Accepted November 11, 2000 Corresponding Editor: Williams S. Modi

A Cryptic RRY(i) Microsatellite From Atlantic Salmon (Salmo salar): Characterization and Chromosomal Location

Ohno S, 1993. Patterns in genome evolution. Curr Opin Genet Dev 3:911–914.

J. L. Martinez, P. Moran, and E. Garcia-Vazquez

Ohno T, Oda S, and Namikawa T, 1994. TESS line: a laboratory line of the musk shrew (Suncus murinus, Insectivora), triple-homozygous for the curly hair (ch), cream coat-color (cr) and red-eyed dilution (rd) genes

In this article we describe the isolation and characterization of a cryptic RRY(i) micro-

satellite from an Atlantic salmon genomic cosmid library. The chromosomal location of the microsatellite-containing cosmid was performed by fluorescent in situ hybridization (FISH) showing a single-locus signal located on an interstitial position of an acrocentric pair. The suitability of this type of microsatellite marker for population genetic analysis and for the development of a genetic map in this species is discussed. In addition, the usefulness of cosmid libraries for physical mapping of microsatellite markers and therefore for the integration of physical and genetic maps is pointed out. Most eukaryotic genomes contain a considerable number of repetitive noncoding sequences that exist as both dispersed copies and tandem arrays. Microsatellites (tandemly repeated motifs of 1–5 bp) belong to this second category. Microsatellite loci can be defined by their specific flanking sequences showing a high degree of length polymorphism (Weber 1990), which can be analyzed by the polymerase chain reaction (PCR) followed by sizing on polyacrylamide gels (Weber and May 1989). This polymorphism, joined with their even and apparently random distribution in the genome, makes microsatellite loci very useful as markers for genetic mapping and identity control and they have been used for the development of high-resolution genetic maps of species such as human and mouse (Chapman and Nadeau 1992; Weissenbach et al. 1992). Low-resolution genetic marker maps, also based on microsatellite markers, are being developed in a wide variety of commercially important species, such as pig, chicken, cattle, rainbow trout, tilapia, and flat oyster ( Buchanan et al. 1993; Kocher et al. 1998; Moran 1993; Naciri et al. 1995; Rohrer et al. 1994; Young et al. 1998). Genetic linkage maps are complemented with physical mapping, which enables the assignment of linkage groups to specific chromosomes ( Ellegren et al. 1994; Toldo et al. 1993). The development of fluorescent in situ hybridization ( FISH) using microsatellite-containing cosmids as probes has been an important advance and has been used in different map projects ( Dickens et al. 1999; Fischer et al. 1996; Toldo et al. 1993). This method is of particular importance in species, like fishes, whose karyotypes are not standardized, since FISH can simultaneously allow chromosome identification and genetic data integration.

Brief Communications 287

nealing at 57⬚C for 20 s, and extension at 72⬚C for 20 s. The final extension was at 72⬚C for 5 min. PCR products were run on 5.6% acrylamide, 5.6 M urea denaturing gels, and detected by silver staining using the DNA Silver Staining System (Promega). The sizes of allele products were estimated by comparison with pUC sequence reactions. Mendelian inheritance of this microsatellite locus was tested in two half-sib families. To obtain a first estimate of its variability, 30 wild adults caught in the Esva River (Spain) were analyzed. Figure 1. Segregation of microsatellite locus SS10 in an Atlantic salmon half-sib family. M, male parent; F, female parent.

In this work we describe the isolation and characterization of a single trinucleotide locus microsatellite from an Atlantic salmon (Salmo salar) cosmid library. We also report the chromosomal location of the microsatellite-containing cosmid clone on the Atlantic salmon chromosome complement.

Materials and Methods Isolation and Characterization of the Locus Microsatellite A cosmid genomic library has been constructed in superCosI according to manufacturer’s instructions (Stratagene, La Jolla, CA). A (GAC)6 oligonucleotide was kinased with (␥ 32P). Positive clones were isolated and DNA extracted by the standard alkali lysis. Cosmid DNA was digested with several restriction enzymes and analyzed by Southern blotting. Positive restriction fragments smaller than 1.3 kb were subcloned into pUC and sequenced with the Sequenase 2.0 sequencing kit (Amersham, Sweden). Clone SS10 was chosen for microsatellite analysis after being mapped by FISH. Two primers flanking the trinucleotide repeat were designed for PCR amplification of this microsatellite (submitted to the EMBL, accession number AJ012206). FISH Chromosome obtention. Metaphase chromosomes were obtained from lymphocyte cultures. Two to 3 ml of venous blood was extracted from the dorsal vein of several Atlantic salmon adults and stored in heparinized tubes. Lymphocytes were purified and cultured according to standard procedures. Cultures were incubated at 19⬚C for 5 days. Six hours before harvesting,

288 The Journal of Heredity 2001:92(3)

colchicine was added to a final concentration of 0.01 ␮g/ml. Cells were treated with 0.5% KCl and fixed in methanol:acetic acid (3:1). Slides were prepared according to standard procedures. Probes The chromosomal location of locus SS10, characterized in this study, was established using as probe the whole microsatellite-containing cosmid clone labeled with biotin 16-dUTP by nick translation according to the manufacturer’s recommendations (Roche Diagnostics). Chromosome slides were pretreated with RNase and pepsin as described by Wiegant et al. (1991). Repetitive sequences were suppressed by prehybridizing 100 ng of the labeled probe with 100 ␮g of sonicated salmon testes DNA. After overnight hybridization at 37⬚C, the slides were washed for 10 min at 42⬚C in 50% formamide 2⫻ SSC and then washed twice for 5 min in 0.1⫻ SSC at 50⬚C. Detection of signals was performed according to Penda´s et al. (1993). Images were obtained using a Zeiss axioscope epifluorescent microscope equipped with a CCD camera (Photometrics). Microsatellite Analyses PCR amplifications were carried out using the GeneAmp PCR System 2400 from Perkin-Elmer Cetus, with samples containing approximately 50 ng Atlantic salmon DNA, 10 mM Tris-HCl pH 8.8, 1.5 mM MgCl2, 50 mM KCl, 0.1% Triton X-100, 20 pmol of each primer, 1 U Dynazyme II DNA polymerase ( Finnzymes Oy), and 250 ␮M dNTP in a final volume of 20 ␮l. PCRs were performed with an initial denaturing step (5 min at 95⬚C) followed by 35 cycles consisting of denaturation at 95⬚C for 20 s, an-

Results and Discussion The detailed sequence analysis of the positive clone SS10 showed that the repeat motif present in this clone was composed of two different interspersed trinucleotide sequences (AAC and GAC), the longest single triple array being seven GAC repeats. According to Jacobson et al. (1993), this sequence can be considered a ‘‘cryptic repeat’’ because the nature of the long tandem repeat is only appreciated when the sequence is categorized into purines and pyrimidines. When this transformation was done to our sequence, we obtained a cryptic repeat consisting of 82 triplets (cRRY(82)). In human, mouse, and yeast sequences, cryptic RRY are abundant and, like simple RRY, are nonrandomly distributed with respect to both sequence and location, being the trinucleotides GGC or AGC predominant within human cRRY(i) (Gostout et al. 1993) and preferentially located in coding and 5⬘ untranslated regions (Ricke et al. 1995). Whether these cryptically simple regions within genes are important for the function of the gene product or represent relatively weakly selected parts of the gene remains unclear. When the two half-sib Atlantic salmon were analyzed using this cRRY microsatellite, we observed a perfect codominant single-locus Mendelian inheritance ( Figure 1). In the sample from the Esva River, seven alleles were detected, ranging from 380 to 456 bp in length. The 434 bp allele showed the highest frequency (0.55), far from the others which showed frequencies of 0.21 (425 bp allele), 0.1 (380 bp allele), 0.06 (413 bp allele), 0.03 (391 and 456 bp alleles), and 0.02 (409 bp allele). The heterozygosity observed for this population was 0.63. In all cases only one or two alleles per individual were observed. Most of the Atlantic salmon microsatellites characterized in other works are

comprised of two base pair repeat units, usually (GT )n or (GA)n motives. A disadvantage of dinucleotide repeat polymorphisms is that in acrylamide gels, each allele is revealed as several shadow bands that sometimes obscure the position of other allelic fragments, which makes genotyping difficult or impossible: for example, it is difficult to differentiate heterozygotes from homozygotes for alleles differing in length by only two nucleotides. In this work, and despite the size of the repeat segment in this polymorphic locus ( longer than 380 bp), all the alleles at the cRRY microsatellite locus could be identified unambiguously and no stutter bands were observed ( Figure 1). This could be due to the complexity of the sequence, which can prevent the substantial polymerase stuttering that is commonly seen when more monotonous tandem repeats [e.g., (GT )n] are amplified by PCR (Gostout et al. 1993). Similar results have been observed by other authors in different species ( Edwards et al. 1992; Francisco et al. 1996; Naish 1998; O’Reilly et al. 1996). The unambiguous allele sizing of trinucleotide and tetranucleotide core motives in comparison with dinucleotide core sequences leads us to consider these types of microsatellite loci to be more suitable genetic markers for population analyses. As expected from the Mendelian monogenic inheritance detected in the two studied families, the FISH of the whole microsatellite-containing cosmid clone shows a single-locus signal in most of the cells analyzed. The signals were located on an interstitial position of an acrocentric chromosome pair ( Figure 2). Accordingly we believe that this microsatellite can be used for anchoring the developing genetic and physical map in Atlantic salmon. As previously reported by Lundin et al. (1999) and Martinez et al. (1999), the use of cosmid libraries in Atlantic salmon for isolation and characterization of molecular markers allows the integration of physical and genetic maps and also the identification of the different chromosome pairs. Looking at the results obtained in this work, we conclude that the isolation of microsatellite markers from cosmid clones is a useful tool for the development of the genetic and physical map in species like fish, with poorly standardized karyotypes. From the Departmento de Biologia Funcional, Facultad Medicina, Universidad de Oviedo, Julian Claveria s/n, 33006 Oviedo, Spain. We are grateful to Dr. A. M. Pendas for technical assistance and to Dr. J.

Figure 2. Chromosomal location of the Atlantic salmon SS10 microsatellite locus (arrows).

B. Taggart for the half-sib families. Jose Luis Martinez received a fellowship from the Regional Government (Consejeria de Agricultura) of the Principado de Asturias. This work was supported by the Spanish DGICYT (PB98-1570). Address correspondence to Jose Luis Martinez at the address above or e-mail: [email protected]. 䉷 2001 The American Genetic Association

Jacobson DP, Schmeling P, and Sommer SS, 1993. Characterization of the patterns of polymorphism in a ‘‘cryptic repeat’’ reveal a novel type of hypervariable sequence. Am J Hum Genet 53:443–450. Kocher TD, Lee W, Sobolewska H, Penman D, and McAndrew B, 1998. A genetic linkage map of a cichlid fish, the tilapia (Oreochromis niloticus). Genetics 148: 1225–1232. Lundin M, Mikkelsen B, Moran P, Martinez JL, and Syed M, 1999. Cosmid clones from Atlantic salmon: physical genome mapping. Aquaculture 173:59–64.

References Buchanan FC, Littlejohn RP, Galloway SM, and Crawford AM, 1993. Microsatellites and associated repetitive elements in the sheep genome. Mamm Genome 4:258– 264. Chapman VM and Nadeau JH, 1992. The mouse genome: an overview. Curr Opin Genet Dev 2:406–411. Dickens HF, Holmes NG, Ryder E, Breen M, Thomas R, Suter N, Sampson J, Langford CF, Ross M, Carter NP, and Binns MM, 1999. Use of cosmid-derived and chromosome-specific canine microsatellites. J Hered 90:52– 54. Edwards A, Hammond HA, Jin L, Caskey CT, and Chakraborty R, 1992. Genetic variation at five trimeric and tetrameric tandem repeat loci in four human population groups. Genomics 12:241–253. Ellegren H, Chowdhary B, Johansson M, and Andersson L, 1994. Integrating the porcine physical and linkage map using cosmid-derived markers. Anim Genet 25: 155–164. Fisher PE, Holmes NG, Dickens HF, Thomas R, Binns MM, and Nacheva EP, 1996. The application of FISH techniques for physical mapping in the dog (Canis familiaris). Mamm Genome 7:37–41. Francisco LV, Langston AA, Mellersh CS, Neal CL, and Ostrander EA, 1996. A class of highly polymorphic tetranucleotide repeats for canine genetic mapping. Mamm Genome 7:359–362. Gostout B, Liu Q, and Sommer SS, 1993. ‘‘Cryptic’’ repeating triplets of purines and pyrimidines (cRRY(i)) are frequent and polymorphic: analysis of coding cRRY(i) in the proopiomelanocortin (POMC) and TATAbinding protein ( TBP) genes. Am J Hum Genet 52:1182– 1190.

Martinez JL, Moran P, Pendas AM, Taggart JB, and Garcia-Vazquez E, 1999. Genetic characterization and chromosomal location of a single locus GT microsatellite from Atlantic salmon. Anim Genet 30:399–400. Moran C, 1993. Microsatellite repeats in pig (Sus domestica) and chicken (Gallus domesticus) genomes. J Hered 84:274–280 Naciri Y, Vigouroux Y, Dallas J, Desmarais E, Delsert C, and Bonhomme F, 1995. Identification and inheritance of (GA/TC)n and (AC/GT )n repeats in the European flat oyster Ostrea edulis. Mol Mar Biol Biotechnol 4:83–89. Naish KA and Skibinski DOF, 1998. Tetranucleotide microsatellite loci for the Indian major carp. J Hered 53: 886–889. O’Reilly PT, Hamilton LC, McConnell SK, and Wright JM, 1996. Rapid analysis of genetic variation in Atlantic salmon (Salmo salar) by PCR multiplexing of dinucleotide and tetranucleotide microsatellites. Can J Fish Aquat Sci 53:2262–2298. Pendas AM, Moran P, and Garcia-Vazquez E, 1993. Ribosomal RNA genes are interspersed throughout a heterochromatic chromosome arm in Atlantic salmon. Cytogenet Cell Genet 63:128–130. Ricke DO, Liu Q, Gostout B, and Sommer SS, 1995. Nonrandom patterns of simple and cryptic triplet repeats in coding and noncoding sequences. Genomics 26:510– 520. Rohrer GA, Alexander LJ, Keele JW, Smith TP, and Beattie CW, 1994. A microsatellite map of the porcine genome. Genetics 136:231–245. Toldo SS, Fries R, Steffen P, Neibergs HL, Barendse W, Womack JE, Hetzel DJS, and Strazinger G, 1993. Physically mapped, cosmid-derived microsatellite markers

Brief Communications 289

as anchor loci on bovine chromosomes. Mamm Genome 4:720–727. Weber JL, 1990. Informativeness of human (dCdA)n.(dG-dT )n polymorphism. Genomics 7:524–530. Weber JL and May PE, 1989. Abundant class of human DNA polymorphism which can be typed using the polymerase chain reaction. Am J Hum Genet 44:388–396. Weissenbach J, Gyapay G, Dib C, Vignal A, Morissette J, Millasseau P, Vaysseix G, and Lathrop M, 1992. A second-generation linkage map of the human genome. Nature 359:794–801. Wiegant J, Ried T, Nederlof PM, van der Ploeg M, Tanke HJ, and Raap AK, 1991. In situ hybridization with fluoresceinated DNA. Nucleic Acids Res 19:3237–3241. Young WP, Wheeler PA, Coryell VH, Keim P, and Thorgaard GH, 1998. A detailed linkage map of rainbow trout produced using doubled haploids. Genetics 148: 839–850. Received January 18, 2000 Accepted January 15, 2001 Corresponding Editor: Bernie May

A QTL Study of Cattle Behavioral Traits in Embryo Transfer Families S. M. Schmutz, J. M. Stookey, D. C. Winkelman-Sim, C. S. Waltz, Y. Plante, and F. C. Buchanan Two behavioral traits, temperament and habituation, were measured in 130 calves from 17 full-sib families which comprise the Canadian Beef Cattle Reference Herd. Using variance components, heritability was calculated as 0.36 for temperament and 0.46 for habituation. Genotyping of 162 microsatellites at approximately 20 cM intervals allowed the detection of six quantitative trait loci (QTL) for behavior traits on cattle chromosomes 1, 5, 9, 11, 14, 15. The inheritance of behaviors in domestic animals is of considerable interest to livestock producers, but it has been the focus of relatively few studies. In part, this may be because of the difficulty of assessing or quantifying a behavior for statistical analyses. For example, most studies with cattle rely on a subjective scoring scale to assess temperament during some handling procedure ( Dickson et al. 1970; Hearnshaw and Morris 1984; Tulloh 1961; Voisinet et al. 1997). Temperament of an animal can be defined as ‘‘an animal’s behavioral responses to handling by humans’’ ( Burrow et al. 1997), including its excitatory or inhibitory reactions, level of motor activity, persistent habits, emotionality, alertness, etc. ( Hurnik et al. 1995), and as such is not easily quantified. However, certain aspects of temperament such

290 The Journal of Heredity 2001:92(3)

as excitability and the level of motor activity during handling have been quantified ( Burrow and Dillon 1997; Stookey et al. 1994) and proven to be persistent over time (Grandin 1993). In addition, these objective measurements have been correlated to at least one physiological response—heart rate (Piller et al. 1999; Waynert et al. 1999). Because selection for certain behaviors is considered to be useful to humans and/or to the animal (Schmutz and Schmutz 1998) it would be beneficial to establish that such behaviors are inherited and therefore could potentially be mapped. Some studies in humans have been conducted to evaluate relationships between behavior and specific candidate genes, often chosen on the basis of neurochemical properties. Polymorphisms in type 4 dopamine receptor ( DRD4) ( Ekelund et al. 1999) and in dopamine receptor type 2 ( Noble et al. 1998) have been associated with novelty seeking as assessed by the temperament and character inventory ( Ekelund et al. 1999). This same assessment was used by Kumarkiri et al. (1999) to conclude that alleles in the serotonin transporter transcriptional control region were associated with cooperativeness. An intronic polymorphism in tryptophan hydroxylase was used as a marker to show it appears to be associated with antagonistic behavior (Manuck et al. 1999). Quantitative trait loci (QTL) mapping is being used in humans and mice for complex traits, and some of these include behaviors. We report here on a study that attempts to map two behavioral traits using a QTL approach in beef cattle: the response to isolation during handling (which we believe to be a reflection of temperament) and habituation to the handling procedure.

Methods Animals A herd of 17 families composed of 130 embryo transfer calves was used in this study. The calves were raised by surrogate mothers or recipient dams, as opposed to their biological mothers. As is typical of cow herds, the sires were not in contact with their calves either. The calves were weaned from their recipient dams at 6 months of age and trucked for approximately 2 h from the ranch where they were born to the University of Saskatchewan Beef Research Station. The newly weaned calves arrived in six groups of a few families each over a period of 4

months. The calves were unloaded from the trucks and group penned until each was individually weighed and measured within the next hour. Behavior Measurements Each group was assessed upon arrival at the feedlot and again on a single day when they ranged in age from 8 to 12 months. The difference between the initial score at weaning and this later measurement, we call ‘‘habituation,’’ since the animals had been weighed in this same building, and therefore held briefly in this same device, every other week since weaning. During the behavioral assessment, cattle were moved single file through an indoor handling facility and held individually on an electronic platform scale for 1 min. Solid sliding doors and sides prevented the animal from seeing other cattle. The amount of movement made by the animal during the 1-min test was quantified by an electronic movement measuring device (MMD) attached to the load cells of the weight scale (Stookey et al. 1994). The MMD samples the analogue voltage signal at 122 discrete time intervals per second. Any movement by the animal on the scale causes the signal to fluctuate. A peak is recorded whenever a trend of increasing or decreasing voltage is reversed. The number of peaks recorded is correlated to the amount of movement that can be detected by video analysis (Stookey et al. 1994). We call this response to isolation ‘‘temperament,’’ in the sense that agitation and movement during handling can be thought of as a reflection of an animal’s temperament. Heritability Calculation Heritability was calculated from variance components obtained from analysis of variance (ANOVA) using a nested design with biological dams nested within sires, since each sire was mated to more than one dam and hence had more than one full-sib family. Genotyping One hundred and sixty-two microsatellites were selected at approximately 20 cM intervals throughout the genome. We chose microsatellites that had six or more alleles whenever these were available, at approximately 20 cM intervals. Polymerase chain reaction (PCR) was used to genotype all parents and calves from DNA extracted from blood ( Buchanan et al. 1994). Genotypes were scored twice independently

and discrepancies were resolved or the samples were retyped. QTL Analysis It was necessary to use an analysis that could summarize information about each marker-QTL linkage ( Knott and Haley 1992) across several small full-sib families. We used a modified identical-by-descent analysis procedure. Each sib pair within a family (i.e., a family of 5 sibs would constitute 10 sib pairs) was designated as being like or unlike in their genotype at a given marker. An unpaired t test was used to test the absolute differences in the phenotypes (i.e., MMD score) of like versus unlike groups within each family. The ttest values were squared and summed across all the families and the probability was determined using a chi-squared distribution ( Xu and Atchley 1995) with degrees of freedom equal to the number of families (Weller et al. 1990). A P ⱕ .00156 was considered significant ( Knott et al. 1996).

Results The movement score, as a response to isolation, had a mean of 89 ⫾ 5.9 at weaning, with a standard deviation of 56. The movement scores ranged from 21 to 284. The difference between this score and the score several months later, or habituation, had a mean decrease of 17 ⫾ 6 (SD ⫽ 70). The range was 267 more to 232 less, indicating that some cattle were more agitated in the first measurement while others were more agitated during the second. In his review, Burrow (1997) states that most studies show temperament improved with age and increased handling. The heritability of temperament at weaning was calculated as 0.36 and of habituation as 0.46. Several QTL were detected for one and/or both of these traits. Chromosomal locations for QTL detected for both behaviors are shown in Table 1. Of anecdotal interest, one night a pen of heifers was scared by something and two animals became so agitated they ran into a post in their corral and each broke their jaw. Both were members of a single family and had high scores (119 and 149) detected during their behavioral measurements. We suggest this exemplifies the problem such a temperament can cause an animal.

Discussion We found a slightly higher heritability for temperament than most other studies.

Table 1. QTL for behavioral traits Significance Chromosome cM

TemperaMicrosatellite ment

Habituation

1 5 9 11 14a

BMS574 RM103 ILSTS013 ILSTS036 RM180 ILSTS008 ADCY2

0.00001 0.00009 0.00006 0.00001

15 a

14 29 44 57 19 35 12

0.00001 0.00001 0.00001 0.00001 0.00002 0.00001 0.00001

0.00001

Adjacent markers are reported for this chromosome. The P value used to detect significance was .00156. cM are estimates from the centromere.

However, most studies, including ours, agree that temperament is of moderate heritability in beef cattle. Fordyce et al. (1996) studied Shorthorn cross beef cattle in Australia and found a heritability of 0.08–0.014 for their temperament score but 0.32–0.70 for flight distance. Le Neindre et al. (1995) reported a heritability of 0.22 for a docility score based on a Limousin heifer’s reactions to being limited to a corner of a pen by a handler. Morris et al. (1994) used a subjective scale of 1–8 for assessing cows during weighing and 1–5 to rate temperament in terms of ease of handling during herding into pens by stockmen in Angus and Hereford beef cattle. They report low to moderate heritabilities with wide standard errors. Mourao et al. (1998) estimate the heritability of temperament as 0.27 using dam-daughter regression analysis of subjective scores between 1 and 5 based on aggressiveness. Using a restraint test in a chute of weaned calves, Stricklin et al. (1980) reported a higher heritability, similar to that found in this study, of 0.44–0.48 which was calculated using paternal half-sib correlations. This measurement of temperament, although subjectively scored, was obtained from the test situation most similar to that used in our study. In the present study, the embryo transfer protocol used to produce the families created a unique group of individuals deprived of the opportunity to imitate their biological dams’ behavior during rearing. Each calf was carried to term and reared by a surrogate mother making it possible to study innate behavioral tendencies with the confounding effects of maternal influence on early experience somewhat diluted. The high heritabilities of behavioral traits that we report may be a reflection of having equalized some of the effects of early environmental (maternal) influences. The modification in our analysis from

Figure 1. A histogram showing an example chromosome, cattle chromosome 1, with the number of families in which both parents had genotypes that were alike, in which neither parent was heterozygous, in which only one parent was heterozygous, and in which both parents were heterozygous. The number of alleles found for each microsatellite is shown below to indicate that a higher number of alleles found did not guarantee a higher proportion in which both parents were heterozygous.

the more typical scoring of zero, one, or two alleles shared was done because most families had only one or the other parent heterozygous at a marker instead of both heterozygous, as shown in Figure 1. Some families had both parents heterozygous, but they had the same genotype and the offspring were thus uninformative. A few had neither parent heterozygous. Therefore we did consider the more typical IBD approach, but we did not feel it suited our dataset as well as one might expect. Although we might miss some QTL using only ‘‘like or unlike,’’ we believed that this approach was conservative. We report six QTL localizations where both of these related behavioral traits were detected. Relatively few coding genes are yet mapped in cattle, but we attempted to look for candidate genes in the regions near these QTL. Cannabinoid receptor (CNR1) was previously mapped by in situ hybridization to the region of chromosome 9 (Pfister-Genskow et al. 1997), where one QTL was found. This gene was mapped in humans to chromosome 6q1415 ( Hoehe et al. 1991). Type 2 and 4 dopamine receptors ( DRD2 and DRD4) are localized to human chromosome 11p15.5. DRD2 is mapped to cattle chromosome 15 (Amarente et al. 1999). Although few studies have attempted QTL mapping of behavioral traits in animals, Turri et al. (1999) reported three localizations for ‘‘emotionality’’ in mice, chromosomes 1, 12, and 15. Their behavior measurements for this trait were the total distance a mouse traversed in an open field arena in 5 mm and the number

Brief Communications 291

of fecal boli during this 5-min time period. The region on mouse 1, from 92 to 105 cM, is homologous to the upper third of cattle chromosome 16. The area on mouse chromosome 15, at approximately 43 cM, in which they found a QTL, is homologous to cattle chromosome 14 in the region where a QTL was found in this study. From the Department of Animal and Poultry Science (Schmutz, Winkelman-Sim, and Buchanan) and the Department of Herd Medicine and Theriogenology (Stookey and Waltz), University of Saskatchewan, Saskatoon, Canada, and Bova Can Labs, Saskatchewan Research Council, Saskatoon, Canada (Plante). The authors are grateful for the financial support provided by the Canadian Cattlemen’s Association, Alberta Cattle Commission, and Natural Science and Engineering Research Council ( NSERC). The Saskatchewan Agricultural Development Fund provided support for the development and use of the device (M.M.D.) for quantifying animal movement. The help provided by the staff of the Termuende Farm and the University of Saskatchewan Beef Research Facility was also greatly appreciated. Address correspondence to Sheila Schmutz, Department of Animal and Poultry Science, 51 Campus Dr., University of Saskatchewan, Saskatoon S7N 58A, Canada, or e-mail: [email protected]. 䉷 2001 The American Genetic Association

References

of farm animal behavior, 2nd ed. Ames: Iowa State University Press. Knott SA, Elsen JM, and Haley CS, 1996. Methods for multiple-marker mapping of quantitative trait loci in half-sib populations. Theor Appl Genet 93:71–80.

SSR Markers for Quercus suber Tree Identification and Embryo Analysis

Knott SA and Haley CS, 1992. Maximum likelihood mapping of quantitative trait loci using full-sib families. Genetics 132:1211–1222.

A. Go´mez, B. Pintos, E. Aguiriano, J. A. Manzanera, M. A. Bueno

Kumakiri C, Kodama K, Shimizu E, Yamanouchi N, Okada S, Noda S, Okamoto H, Sato T, and Shirasawa H, 1999. Study of the association between serotonin transporter gene regulatory region polymorphism and personality traits in a Japanese population. Neurosci Lett 263:205–207. Le Neindre P, Trillat G, Sapa J, Menissier F, Bonnet J, Chupin J, and Le Neindre P, 1995. Individual differences in docility in Limousin cattle. J Anim Sci 73:2249–2253. Manuck S, Flory J, Ferrell R, Dent K, Mann J, and Muldoon M, 1999. Aggression and anger-related traits associated with a polymorphism of the tryptophan hydroxylase gene. Biol Psychol 45:603–614. Morris C, Cullen N, Kilgour R, and Bremner K, 1994. Some genetic factors affecting temperament in Bos taurus cattle. N Z J Agric Res 37:167–175. Mourao GB, Bergmann JAG, and Fereira MBD, 1998. Differencas geneticas e estimacao de coeficientes de herdabilidade para temperamento em femeas zebus e F1 Holandes X zebu. Rev Brasil Zootec 27:722–729. Noble EP, Ozkaragoz TZ, Ritchie TL, Zhang X, Belin TR, and Sparkes RS, 1998. D2 and D4 dopamine receptor polymorphisms and personality. Am J Med Genet 81: 257–267.

Amarante M, Lopez C, and Womack JE, 1999. Dopamine receptor D2 maps to bovine chromosome 15. Anim Genet 30:398.

Pfister-Genskow M, Weesner G, Hayes H, Eggen A, and Bishop M, 1997. Physical and genetic localization of the bovine cannabinoid receptor (CNR1) gene to bovine Chromosome 9. Mamm Genome 8:301–302.

Buchanan FC, Adams LJ, Littlejohn RP, Maddox JF, and Crawford AM, 1994. Determination of evolutionary relationships among sheep breeds using microsatellites. Genomics 22:397–403.

Piller CAK, Stookey JM, and Watts JM, 1999. Effects of mirror-image exposure on heart rate and movement of isolated heifers. Appl Anim Behav Sci 63:93–102.

Burrow HM, 1997. Measurements of temperament and their relationships with performance traits of beef cattle. Anim Breed Abstr 65:477–495.

Schmutz SM and Schmutz JK, 1998. Heritability estimates of behaviors associated with hunting in dogs. J Hered 89:233–237.

Burrow HM and Dillon RD, 1997. Relationships between temperament and growth in a feedlot and commercial carcass traits of Bos indicus crossbreds. Aust J Exp Agric 37:407–411.

Stookey JM, Nickel T, Hanson J, and Vandenbosch S, 1994. A movement measuring device for objectively measuring temperament in beef cattle and for use in determining factors that influence handling. J Anim Sci 74(suppl 1):133.

Burrow HM, Seifert GW, and Corbet NJ, 1988. A new technique for measuring temperament in cattle. Proc Aust Soc Anim Prod 17:54–157.

Stricklin WR, Heisler CE, and Wilson LL, 1980. Heritability of temperament in beef cattle. J Anim Sci 5(suppl 1):109–110.

Dickson DP, Barr GR, Johnson LP, and Wieckert DA, 1970. Social dominance and temperament of Holstein cows. J Dairy Sci 53:904–907.

Tulloh NM, 1961. Behaviour of cattle in yards. II. A study in temperament. Anim Behav 9:25–30.

Ekelund J, Lichtermann D, Jarvelin M, and Peltonen L, 1999. Association between novelty seeking and the type 4 dopamine receptor gene in a large Finnish cohort sample. Am J Psychiatry 156:1453–1455. Fordyce G, Howitt C, Holroyd R, O’Rourke P, and Entwistle K, 1996. The performance of Brahman-Shorthorn and Sahiwal-Shorthorn beef cattle in the dry tropics of northern Queensland. 5. Scrotal circumference, temperament, ectoparasite resistance, and the genetics of growth and other traits in bulls. Aust J Exp Agric 36: 9–17. Grandin T, 1993. Behavioral agitation during handling of cattle is persistent over time. Appl Anim Behav Sci 36:1–9. Hearnshaw H and Morris CA, 1984. Genetic and environmental effects on a temperament score in beef cattle. Aust J Agric Res 35:723–733. Hoehe MR, Caenazzo L, Martinez M, Hsieh WT, Modi WS, Gershon ES, and Bonner TI, 1991. Genetic and physical mapping of the human cannabinoid receptor gene to chromosome 6q14-q15. New Biol 3:880–885. Hurnik JF, Webster AB, and Siegel PB, 1995. Dictionary

292 The Journal of Heredity 2001:92(3)

Turri M, Talbot C, Radcliffe R, Wehner J, and Flint J, 1999. High-resolution mapping of quantitative trait loci for emotionality in selected strains in mice. Mamm Genome 10:1098–1101. Voisinet BD, Grandin T, Tatum JD, O’Connor SF, and Struthers JJ, 1997. Feedlot cattle with calm temperaments have higher average daily gains than cattle with excitable temperaments. J Anim Sci 75:892–896. Waynert D, Stookey JM, Schwartzkopf-Genswein K, Watts J, and Waltz C, 1999. The response of beef cattle to noise during handling. Appl Anim Behav Sci 62:27– 42. Weller JI, Kashi Y, and Soller M, 1990. Power of daughter and granddaughter designs for determining linkage between marker loci and quantitative trait loci in cattle. J Dairy Sci 73:2524–2537. Xu S and Atchley WR, 1995. A random approach to interval mapping of quantitative trait loci. Genetics 141: 1189–1197. Received May 5, 2000 Accepted January 16, 2001 Corresponding Editor: Bruce S. Weir

Three Quercus simple sequence repeat (SSR) markers were amplified by polymerase chain reaction (PCR) from nuclear DNA extracts of trees and in vitro-induced haploid embryos from anther cultures of Quercus suber L. These markers were sufficiently polymorphic to identify 10 of 12 trees located in two Spanish natural areas. The same loci have been analyzed in anther-derived haploid embryos showing the parental tree allele segregation. All the alleles were present in the haploid progeny. The presence of diverse alleles in embryos derived from the same anther demonstrated that they were induced on multiple microspores or pollen grains and they were not clonally propagated. Also, diploid cultures and mixtures of haploid-diploid tissues were obtained. The origin of such cultures, either somatic or gametic, was elucidated by SSR markers. All the embryos showed only one allele, corroborating a haploid origin. Allelic composition of the haploid progeny permitted parental identification among all analyzed trees. The embryogenic process originated during anther culture may have different origins, for example, haploid cells such as microspores or pollen grains, or somatic cells from anther tissues. Therefore embryo genetic composition may be of either gametic or sporophytic, depending on which cells are induced. Anthers subjected to stress conditions can become a target for embryo induction ( Bueno et al. 1997). When embryogenesis appears as a result of stress, the probability of obtaining haploid embryos from microspores or pollen grains is greater, rendering this method more interesting than embryogenesis induction by plant growth regulators. Data for isolated microspore cultures clearly indicate that hormones are not required for embryogenic induction. In fact, at least benzylaminopurine has been shown to inhibit pollen embryogenesis in tobacco ( Kyo and Harada 1990). Due to the importance of doubled haploid plants for plant breeding, their production through anther culture would be especially useful in species with long generation times and strong inbreeding de-

Table 1. Comparison between Q. suber and other Quercus species (Steinkellner et al. 1997b) microsatellite loci Species

SsrQpZAG15

SsrQpZAG46

SsrQpZAG110

Q. Q. Q. Q. Q. Q. Q.

108–152 (36/11) 120–128 (5/4) 108–128 (15/9) 124–126 (13/2) 117 (8/1) 104–135 (9/6) 118–124 (12/4)

190–222 (14/9) 216–228 (5/4) 194–222 (14/11) 183–205 (13/5) 182–202 (8/6) 187–209 (9/5) 188–192 (12/3)

206–262 (32/7) 200–208 (4/3) 203–255 (14/10) 215–261 (13/8) 221–225 (8/3) 213–232 (6/7) 220–238 (12/6)

petraea robur pubescens cerris palustris rubra suber

Sizes are given in bp. Number of individuals tested/number of alleles found are in parentheses.

pression, which make traditional breeding methods impractical. Only a few isolated reports have appeared describing the formation of embryos in anther cultures of forest trees such as Quercus suber ( Bueno et al. 1997). The type of origin, either haploid, dihaploid, or diploid, of anther-derived embryos has been rarely studied in forest tree species or even in other plant species ( Hayward et al. 1990; Mu¨ller-Starck and Jo¨rgensen 1991). Nevertheless, flow cytometry has permitted the distinction of haploid embryos, diploid embryos, and explants composed of cells of both types, all of them coming from the same anther in cork oak (Q. suber). The proportion of haploid embryos formed in anther cultures has been higher than that of diploids or haploid/diploid mixtures. This has led to the conclusion that the embryos with double DNA content were not heterozygotic diploids with the parental genome but dihaploids. The necessity of testing this assumption led us to use molecular markers, such as simple sequence repeat (SSR). Codominant molecular markers are more suitable for this study because homozygotic and heterozygotic individuals can be distinguished. For this purpose, isozyme markers were used initially (Mu¨ller-Starck and Jo¨rgensen 1991), but presently DNA markers with a higher number of alleles are available, such as microsatellites, that facilitate sample analysis through selective amplification of repeated sequence fragments. The use of microsatellite markers is generally restricted to species in which they are designed, due to the high degree of homology necessary between primers and sample DNA. Sometimes there are amplifications available for one species derived from closely related species during evolution for which those primers were designed ( Fields and Scribner 1997; Primmer et al. 1996; Sun and Kirkpatrick 1996). This is the case for (GA)n microsatellites in the genus Quercus. Most SSRs localized by

Steinkellner et al. (1997a,b) in Q. petraea can be polymerase chain reaction (PCR) amplified using the same primers in other oaks and even some SSRs have been found in other species of the Fagaceae. In the present work, microsatellite markers are used for the first time in cork oak. Those markers can be used for tree identification, genotypic characterization, heterozygosity evaluation, and determination of the ploidy level in anther embryos induced by stress treatments. Furthermore, in the case of highly heterozygotic trees, a test of parental exclusion and genotype identification might be used for identification of the paternal tree of the gametic embryos, based on the analysis of its haploid progeny.

Materials and Methods Plant Material DNA from leaves was obtained from 12 Q. suber trees. Eight trees, located in Ca´ceres, were numbered 1H–8H, and the other four trees, located in Madrid, were labeled J, 1P, 2P, and 3M. Embryonic tissue for DNA extraction was sampled from Q. suber haploid embryo cultures obtained from tree 3M, following Bueno et al. 1997: (1) Twelve embryos from two anthers, six from anther 164 and six from anther 169, the haploid origin of which has been assessed by flow cytometry analysis. (2) Six embryos from anther 104, which have shown a haploid-diploid composition by flow cytometry analysis. (3) Six embryos from anther 212. In these embryos diploid DNA levels have been observed by flow cytometry analysis. DNA Isolation, SSR Loci, and Amplification Conditions DNA from leaves was extracted as described by Ziegenhagen et al. (1993) with a posterior purification with the GENECLEAN威 kit ( BIO 101). DNA from embryos

was extracted as described by Doyle and Doyle (1990). Three Q. petraea microsatellite loci, (GA)n repeats, were amplified with the primers SsrQpZAG15, SsrQpZAG46, and SsrQpZAG110 designed by Steinkellner et al. (1997a) with the PCR amplification profile as described by Barreneche et al. (1998). These loci were assayed on the basis of their observed heterozygosity and conservation between Quercus species (Steinkellner et al. 1997a,b). Each 25 ␮l amplification reaction contained 20 ng of genomic DNA, 0.2 ␮M of fluorescently labeled forward primer and unlabeled reverse primer (Progenetic), 200 ␮M each dNTPs, 50 mM KCl, 10 mM Tris-HCl (pH 9), 2.5 mM MgCl2 and 0.5 U of Taq-DNA polymerase ( Ecogen). Fluorescent labeled PCR products were separated and analyzed on a semiautomated sequencer (ABI-Prism, Perkin-Elmer). Standards were used for length determination of alleles. Heterozygosity and Power of Discrimination Indexes As a measure of the information provided by each locus, heterozygosity was calculated using Nei’s (1973) formula: H ⫽ n/(n ⫺ 1)*1 ⫺ ⌺pi2, where pi is the frequency of allele i in the n analyzed trees. Power of discrimination for each locus was calculated using the formula PD ⫽ 1 ⫺ ⌺Pi2, where Pi is the frequency of genotype i ( Kloosterman et al. 1993). Mendelian Allele Segregation Mendelian allele segregation was analyzed by chi-squared test.

Results DNA amplification in leaf extracts of the trees analyzed was successfully obtained in all three loci. Also, all amplification products had the expected size for all microsatellites ( Table 1). It is noteworthy that the extraction system designed for pine needles by Ziegenhagen et al. (1993) has also been applicable to DNA isolation from cork oak leaves, which contain characteristic substances that might interfere with Taq polymerase activity and hamper SSR fragment amplification. All loci were polymorphic in cork oak ( Table 1) with 3–6 alleles per locus, the average being 4.67 alleles per locus. Table 2 shows the Hardy–Weinberg expected heterozygosity (H), which range between 0.475 and 0.685 per locus. The average value of this index for all three loci studied

Brief Communications 293

Table 2. Allele sizes given in bp and Genetic indexes of the three ssrQpZAG markers used to generate DNA profiles of 12 cork oak trees Tree

SsrQpZAG15

SsrQpZAG46

SsrQpZAG110

1H 2H 3H 4H 5H 6H 7H 8H 1P 2P J 3M H PD

120–124 120 120–124 120–124 118–124 120 120–124 124–134 120 120 120–124 120–124 0.569 0.625

188 188 188–192 188–190 188 188 188 190–192 188 188–192 188 190–192 0.475 0.597

220–224 224 222–238 224–232 224–226 224–238 222–224 224 224 224 220 222–238 0.685 0.819

Mean 0.576

Total 0.681 0.896

H, heterozygosity; PD, power of discrimination.

Discussion

was 0.576. The power of discrimination between genotypes was 89.6%. All three microsatellite loci showed a similar level of polymorphism to that found in other species of the genus Quercus (Steinkellner et al. 1997b). Four alleles were found for locus ssQpZAG15 among the 12 cork oak trees analyzed, in a similar frequency to that found in Q. petraea. Locus ssQpZAG110 had more alleles, while locus ssQpZAG46 had fewer alleles in Q. suber than in Q. petraea. A total of 13 alleles have been found ( Table 2), four of which were unique.

Three primers designed by Steinkellner et al. (1997a) for microsatellite amplification in Q. petraea were used for DNA analysis of Q. suber. Tree Analysis All three loci tested were polymorphic and the average rate of heterozygotic loci found in a set of 12 trees was as high as 55.6%. This is not remarkable if we take into consideration that those markers were chosen for this study on the basis of high polymorphism. Semiautomated sequencer output revealed only a single peak for loci of homozygous trees and two peaks of different size in the case of heterozygotes. It was concluded that this system is adequate for size determination of microsatellites ( Bredemeijer et al. 1998). Microsatellite polymorphism has provided a new approach to the genetic analysis of cork oak and an efficient tree identification system, due to the high discrimination power obtained for genotypic differentiation. In our case, only three loci were necessary to identify 10 of 12 trees tested. Two fully homozygot-

Embryo Analysis Genotype identification of embryos induced in anthers from tree 3M was performed by means of all three microsatellites assayed ( Table 3). Six embryos from each anther and four anthers from the same tree were tested for each locus, thus a total of 24 embryos. All alleles of the parent tree were also found in the progeny, but only one allele per locus was amplified in each embryo. The pattern of allelic heredity was analyzed by a chi-squared test, obtaining a 1:1 segregation for locus ssQpZAG46 only ( Table 3). A total of five genotypes were differentiated.

Table 3. Number of anther -induced embryos bearing each microsatellite allele detected in cork oak tree 3M

Loci SsrQpZAG15 SsrQpZAG46 SsrQpZAG110

3M tree allele ( bp) 120 124 190 192 222 238

Anther 104

164

169

212

Total no. of embryos

— 6 — 6 6 —

6 — 4 2 — 6

— 6 — 6 6 —

1 5 6 — 6 —

7 17 10* 14* 18 6

* 1:1 segregation at the 0.05 significance level (chi-squared test).

294 The Journal of Heredity 2001:92(3)

ic trees for all three loci, with the same allelic composition, were found (namely 2H and 1P). On the other hand, three individuals were heterozygotic for all three loci. A high degree of heterozygosity has already been detected for oaks by means of isozyme analysis. In Q. petraea and Q. robur, for instance, a population of 1606 juvenile plants showed a 21.9 and 21.3% heterozygosity, respectively (Mu¨ller-Starck 1990). Embryo Analysis We obtained a haploid progeny of pollen embryos through anther culture from tree 3M, which has a high degree of heterozygosity. Previous studies on the ploidy level of cork oak anther embryos were performed by flow cytometry to determine their origin, either gametic or somatic. A high percentage of embryos were in fact haploid, confirming their origin from microspores or pollen grains. Nevertheless, some exceptions were found, revealing a diploid genome, such as embryos from anther 212 or a mixture of haploid and diploid genome, such as in anther 104. One of the aims of our analysis by microsatellite markers was to prove the hypothesis of a spontaneous duplication of the haploid genome. The results obtained in this work verify this hypothesis. Four anther cultures previously analyzed by flow cytometry for ploidy level determination were chosen. From these anthers, six embryos per anther were analyzed by microsatellite amplification, corroborating the allelic pattern of the parent tree ( Table 3). Both alleles from each locus were found in the progeny and different haplotype combinations have been obtained. As is shown in Table 3, diploid anther embryos had only one allele per locus, revealing a homozygotic genome for all loci tested. These results confirm the applicability of microsatellite markers as indicators of the ploidy level in embryo regeneration from anther cultures, as it was previously performed by isozymes in Q. petraea (Mu¨ller-Starck and Jo¨rgensen 1991). Again, in the case of haploid-diploid embryos induced on anther 104, a single allelic combination (124, 192, and 222 bp) was observed. Both alleles of locus ssQpZAG46 (190 and 192 bp) segregated 1:1 in the haploid embryo progeny induced on anther 164. Also for dihaploid embryos from anther 212, both alleles of ssQpZAG15 (120 and 124 bp) were present.

The use of microsatellite markers has permitted the verification of the hypothesis that anther embryos are induced from different microspores or pollen grains, as it has been proved by the diverse genetic composition of embryos from the same anther. This confirms previous results obtained by isozymes and RAPDs ( Bueno et al. 2000). Parental Tree Analysis by Descendant Embryo Analysis Allele segregation in the haploid descendants can be used for the heterozygosity analysis of an individual so that direct DNA analysis is not necessary for the parental tree. The high rate of polymorphism observed also permitted the identification of the parent tree by parental exclusion. In this case locus ssQpZAG15 has two alleles, of 120 and 124 bp each, which are present in trees 1H, 3H, 4H, 7H, J, and 3M, excluding other trees as possible fathers. The locus ssQpZAG46 has two alleles present in the embryos, of 190 and 192 bp, which are present in trees 8H and 3M only. The combination of both exclusion criteria reveals 3M as the parental tree. The same result can be obtained with locus ssQpZAG110, with alleles of 222 and 238 bp, only present in trees 3H and 3M. The principle of parental exclusion could be applied in our embryo cultures and only two loci were sufficient for parental identification.

Conclusions Nuclear DNA microsatellites are an adequate system for the tree identification in cork oak thanks to a high discrimination power among genotypes. Both alleles of each SSR locus were inherited by anther embryos, and Mendelian segregation (1:1) could be statistically proved by chisquared test in one case. The homozygosity of both haploid and diploid anther embryos has been proved by microsatellite markers, revealing a certain rate of spontaneous DNA duplication. Embryo origin from multiple microspores or pollen grains inside a cork oak anther has been found by the different genetic composition of those embryos. The parent tree genome can be deduced from the haploid embryo progeny. SSR markers were used for the first time in cork oak, corroborating their applicability for many genetic analyses. From the INIA-CIFOR, Ctra. de La Corun˜a Km 7, 28040 Madrid, Spain (Go´mez, Pintos, Aguiriano, and Bueno) and ETSI Montes, UPM, Madrid, Spain (Manzanera).

This study was supported by INIA project SC 98-081. Address correspondence to M. A. Bueno at the address above or e-mail [email protected]. 䉷 2001 The American Genetic Association

FISH Mapping of the 5S and 18S-28S rDNA Loci in Different Species of Glycine

References

P. Krishnan, V. T. Sapra, K. M. Soliman, and A. Zipf

Barreneche T, Bodenes C, Lexer C, Trontin J-F, Fluch S, Streiff R, Plomion C, Roussel G, Steinkellner H, Burg K, Favre J-M, Glo¨ssl J, and Kremer A, 1998. A genetic linkage map of Quercus robur L. (pedunculate oak) based on RAPD, SCAR, microsatellite, minisatellite, isozyme and 5S rDNA markers. Theor Appl Genet 97:1090–1103. Bredemeijer GMM, Arens P, Wouters D, Visser D, and Vosman B, 1998. The use of semi-automated fluorescent analysis for tomato cultivar identification. Theor Appl Genet 97:584–590. Bueno MA, Agundez D, Go´mez A, Carrascosa MJ, and Manzanera JA, 2000. Haploid origin of cork oak anther embryos detected by enzyme and RAPD gene markers. Int J Plant Sci 161:363–367. Bueno MA, Go´mez A, Boscaiu M, Manzanera JA, and Vicente O, 1997. Stress-induced formation of haploid plants through anther culture in cork-oak (Quercus suber). Physiol Plant 99:335–341. Doyle JJ and Doyle JL, 1990. Isolation of plant DNA from fresh tissue. Focus 12:13–15. Fields RL and Scribner KT, 1997. Isolation and characterisation of novel waterfowl microsatellite loci: crossspecies comparisons and research application. Mol Ecol 6:199–202. Hayward MD, Olesen A, Due IK, Jenkins R, and Morris P, 1990. Segregation of isozyme marker loci among androgenetic plants of Lolium perenne L. Plant Breed 104: 68–71. Kloosterman AD, Budowle B, and Daselaar P, 1993. PCRamplification and detection of the human D1S80 VNTR locus. Amplification conditions, population genetics and application in forensic analysis. Legal Med 105: 257–264. Kyo M and Harada H, 1990. Specific phosphoproteins in the initial period of tobacco pollen embryogenesis. J Plant Physiol 137:525–529. Mu¨ller-Starck G. 1990. Genetic studies in Quercus robur L. and Quercus petraea Liebl. in the Federal Republic of Germany. EC report MA 1B–0012-D (AM).1–32. Brussels: Commission of the European Communities. Mu¨ller-Starck G and Jo¨rgensen J, 1991. Enzyme gene markers as indicators of the initial ploidy in anther cultures of trees. Can J For Res 21:1141–1144. Nei M, 1973. Analysis of gene diversity in subdivided populations. Proc Natl Acad Sci USA 70:3321–3323. Primmer CR, Moller AP, and Ellegren H, 1996. Polymorphisms revealed by simple sequence repeats. Trends Plant Sci 1:215–222. Steinkellner H, Fluch S, Turetschek E, Lexer C, Streiff R, Kremer A, Burg K, and Glo¨ssl J, 1997a. Identification and characterization of (GA/CT )n-microsatellite loci from Quercus petraea. Plant Mol Biol 33:1093–1096. Steinkellner H, Lexer C, Turetschek E, and Glo¨ssl J, 1997b. Conservation of (GA)n microsatellite loci between Quercus species. Mol Ecol 6:1189–1194. Sun HS and Kirkpatrick BW, 1996. Exploiting dinucleotide microsatellites conserved among mammalian species. Mammal Genome 7:128–132. Ziegenhagen B, Guillemaut P, and Scholz F, 1993. A procedure for mini-preparations of genomic DNA from needles of silver fir (Abies alba Mill.). Plant Mol Biol Rep 11:117–121. Received October 2, 1999 Accepted November 11, 2000 Corresponding Editor: James L. Hamrick

Wild germplasms are often the only significant sources of useful traits for crops, such as soybean, that have limited genetic variability. Before these germplasms can be effectively manipulated they must be characterized at the cytological and molecular levels. Modern soybean probably arose through an ancient allotetraploid event and subsequent diploidization of the genome. However, wild Glycine species have not been intensively investigated for this ancient polyploidy. In this article we determined the number of both the 5S and 18S-28S rDNA sequences in various members of the genus Glycine using FISH. Our results distinctly establish the loss of a 5S rDNA locus from the ‘‘diploid’’ (2n ⫽ 40) species and the loss of two from the (2n ⫽ 80) polyploids of Glycine. A similar diploidization of the 18S-28S rDNA gene family has occurred in G. canescens, G. clandestina, G. soja, and G. max (L.) Merr. (2n ⫽ 40). Although of different genome types, G. tabacina and G. tomentella (2n ⫽ 80) both showed two major 18S-28S rDNA loci per haploid genome, in contrast to the four loci that would be expected in chromosomes that have undergone two doubling events in their evolutionary history. It is evident that the evolution of the subgenus Glycine is more complex than that represented in a simple diploid-doubled to tetraploid model. Progress in soybean [Glycine max ( L.) Merr.] improvement has been slow due to an overall lack of genetic variation in the germplasm, inherent difficulties in crossing, and a lack of cytogenetic and molecular markers ( Keim et al. 1990). In a crop species with limited genetic variability, such as soybean ( Delannay et al. 1983; Keim et al. 1989; Specht and Williams 1984), wild germplasms are often the predominant sources of genes for crop improvement. The genus Glycine, contained within the tribe Phaseoleae, has been divided into two subgenera, Glycine and Soja. The subgenus Glycine consists of 15 wild perennial species, mostly diploid (2n ⫽ 40) and some allopolyploids (2n ⫽ 80) (Singh 1993). The subgenus Soja (2n ⫽ 40) contains the cultigen Glycine max ( L.) Merr.

Brief Communications 295

and its wild annual progenitor G. soja Sieb. and Zucc. (Singh 1993). Nearly all the genera of tribe Phaseoleae have a chromosome number of 2n ⫽ 22. As no members of the genus Glycine have a confirmed diploid chromosome number of 20 or 22, soybeans are thought to have arisen through an ancient allotetraploid event involving both chromosome doubling and chromosome loss, followed by the subsequent diploidization of the genome ( Danna et al. 1996). However, the putative original progenitor species have not been identified ( Hymowitz and Singh 1987; Kumar and Hymowitz 1989; Lackey 1980), nor have the wild species been closely investigated for evidence of this ancient polyploid event. The next generation of evolutionary studies has moved beyond simple base addition/deletion frequency correlations and is focused on analysis of genome organization and synteny. However, despite the considerable attention and resources committed, the high-density, marker-saturated genetic maps and genomic DNA sequence data tell us relatively little about the large-scale physical organization of the chromosomes (Schmidt and Heslop-Harrison 1998). Probes for DNA repeats (e.g., ribosomal, microsatellite, telomeric, etc.) have become powerful tools for discerning chromosomal organization and have expanded our knowledge of evolutionary, genetic, and taxonomic relationships and have been used in practical applications such as agricultural forensics (individual identification) and cultivar tracking. The nuclear genes encoding both 5S and 18S-28S ribosomal RNA (rRNA) consist of highly conserved repeat units arranged in one or more tandem arrays up to 10,000 bp long and variable nontranscribed spacer regions. In plants, the 5S rRNA genes are arrayed independently, while the 18S, 5.8S, and 26S rRNAs are produced together from a 45S rRNA precursor gene. In addition to multiple genes within an array, there may be multiple arrays ( loci) on the same or different chromosomes. Localization of multiple repetitive sequences by fluorescence in situ hybridization ( FISH) provides a novel mechanism for viewing genomic organization and chromosome structure. These sequences can also act as landmarks for observing gene location, clustering, and orientation. Here we present results on the distribution, copy number, and location of both 18S-28S and 5S rDNA in species of wild perennial Glycine. The evolution of the agronomically important soybean, Glycine

296 The Journal of Heredity 2001:92(3)

max ( L.) Merr., turns out to be much more complex than a simple comparison of chromosome numbers would suggest.

digested fragment of the 5S ribosomal RNA repeat of Acacia melanoxylon in pUC 118, provided by Dr. R. Appels, CSIRO, Australia. The plasmid pGMR3, containing a 4.5 kb EcoRI-digested fragment of the 18S-28S ribosomal RNA repeat of G. max in pBR325, provided by Dr. E. Zimmer, Smithsonian Institution, Washington, DC, was used for the 18S-28S rRNA site localization. Both the plasmids were isolated by the alkaline lysis plasmid maxiprep method as described by Silhavy et al. (1984). Whole plasmid DNA was labeled with biotin-14-dATP ( BRL) using the Gibco BRL BioNick娂 Labeling System or with digoxigenin-11-dUTP using the Boehringer Mannheim Nick Translation Kit. Commercially purchased E. coli DNA, sheared to an average fragment size of 200–500 bp, was used as the carrier DNA.

Materials and Methods Plant Material and Metaphase Preparation Seeds of wild Glycine species—G. canescens (PI 440936 and 446937), G. clandestina (PI 339656 and 440958), G. soja (PI 81762), G. tabacina (PI 193232, 378704 and 440996), and G. tomentella (PI 441005), kindly provided by Dr. Theodore Hymowitz, Department of Agronomy, University of Illinois, Urbana, and of G. max ( L.) Merr., cultivar Bedford (from the Alabama A&M Seed Laboratory)—were used as the sources for metaphase chromosome spreads. The terminal 1 cm of the roots was excised from individual seedlings, pretreated in 2.5 mM 8-hydroxyquinoline for 4 h at room temperature and fixed overnight in freshly prepared, room temperature, 3:1 (v/v) ethanol : glacial acetic acid. The root tips were treated with 0.1 N HCl for 5 min before incubation in a cell wall digestive enzyme cocktail of 5% R-10 cellulase and 1% pectolyase Y-23, for a duration of 5–20 min based on the length and thickness of the root tips, in a 37⬚C water bath. Metaphase spreads were prepared from the terminal 1 mm of the enzymetreated root tips as described by Jewell and Islam-Faridi (1994).

In situ Hybridization (Islam-Faridi and Mujeeb-Kazi 1995) Slides were immersed in 30 ␮g/ml RNase/ 2⫻ SSC for 45 min at 37⬚C, denatured in 70% formamide/2⫻ SSC for 70 s at 70⬚C and then dehydrated in 70, 95, and 100% ethanol for 2 min each at ⫺20⬚C. Probe mix (deionized formamide, 50% dextran sulfate, 15 ␮g/slide E. coli carrier DNA and 30 ng/slide labeled probe DNA in 2⫻ SSC) was denatured at 80⬚C for 10 min, chilled on ice, applied to the slide, covered with a 20 mm ⫻ 40 mm coverslip, and sealed with rubber cement. Following overnight incubation at 37⬚C, the slides were rinsed at 40⬚C in 2⫻ SSC twice for 5 min each, 2⫻ SSC/50% formamide for 10 min, and 2⫻

Probe and Carrier DNA Probes for 5S rRNA were generated from pAM033 which contained a 470 bp BamHI-

Table 1. 18S-28S and 5S rDNA sites in species of genus Glycine Number of sites /metaphase spread 18S-28S rDNA Species G. canescens PI 440936 PI 446937 G. clandestina PI 339656 PI 440958 G. soja PI 81762 G. max Bedford G. tabacina PI 440996

2n

Genome symbol

5S rDNA

Observed

Expecteda

Observed Expecteda

2b 2

4 4

2 2

4 4

GG

2 2 2

4 4 4

2 2 2

4 4 4

GG

2

4

2

4

4 major 2 minor N.D. N.D.

8

4

8

N.D. N.D.

4 4

8 8

8

4

8

AA 40 40 A1A1 40 40 40 40 (complex) 80

PI 193232 PI 378704 G. tomentella

80 80

PI 441005

80

DD (complex) 4 major 2 minor

The number of sites expected in a tetraploid (2n ⫽ 4x ⫽ 40) and octaploid (2n ⫽ 8x ⫽ 80) based on an ancestral polyploid event in the evolution of Glycine. b One pair of signals ⫽ two sites ⫽ one locus.

a

Figure 1. Fluorescence photomicrographs of metaphase chromosomes from various Glycine species hybridized with a digoxigenin-labeled 5S rDNA probe. Signals were detected using Cy3 and chromosomes were counterstained with 4,6-diamidino-2-phenylindole ( DAPI). Images were digitally captured in gray scale and the appropriate colors for the chromosomes and the signals were superimposed and contrast adjusted using the IPLab Spectrum P software. (A) G. canescens [(2n ⫽ 40), subgenus Glycine], showing one 5S rDNA locus. (B) G. clandestina [(2n ⫽ 40), subgenus Glycine], a single 5S rDNA locus. (C) G. soja [(2n ⫽ 40), subgenus Soja], one 5S rDNA locus. (D) G. max [(2n ⫽ 40), subgenus Soja], one 5S rDNA locus. (E) G. tabacina (2n ⫽ 80), two 5S rDNA loci. (F) G. tomentella (2n ⫽ 80), two 5S rDNA loci. Bar represents 5 ␮m.

Figure 2. Fluorescence photomicrographs of metaphase chromosomes from various Glycine species hybridized with a digoxigenin-labeled 18S-28S rDNA probe. Signals were detected using Cy3 and chromosomes were counterstained with 4,6-diamidino-2-phenylindole ( DAPI). (A) G. canescens [(2n ⫽ 40), subgenus Glycine], showing one 18S-28S rDNA locus. (B) G. clandestina [(2n ⫽ 40), subgenus Glycine], a single 18S-28S rDNA locus. (C) G. soja [(2n ⫽ 40), subgenus Soja], one 18S-28S rDNA locus. (D) G. max [(2n ⫽ 40), subgenus Soja], one 18S-28S rDNA locus. (E) G. tabacina (2n ⫽ 80), two major and one minor 18S-28S rDNA loci. (F) G. tomentella (2n ⫽ 80), two major and one minor 18S-28S rDNA loci. Both the major and minor loci in the polyploids are located on separate chromosomes. Bar represents 5 ␮m.

Brief Communications 297

SSC for 5 min. The slides were incubated at room temperature in solutions of 2⫻ SSC for 5 min, 1⫻ SSC twice for 5 min each, and 4⫻ SSC/0.2% Tween-20 for 5 min. Signal from biotin-labeled probes was amplified and detected with sequential applications of 5 ␮g/ml FITC-avidin DCS in 5% BSA-4⫻ SSC/0.2% Tween-20 for 30 min at 37⬚C, 5 ␮g/ml biotinylated-antiavidin D in 5% NGS-4⫻ SSC/0.2% Tween-20 for 30 min at 37⬚C and 5 ␮g/ml FITC-avidin DCS in 5% BSA-4⫻ SSC/0.2% Tween-20. Signal from digoxigenin-labeled probes was amplified with 2 ␮g/ml mouse antidigoxigenin (MAD) in 5% BSA-4⫻ SSC/0.2% Tween-20 for 30 min at 37⬚C and detected with 5 ␮g/ ml Cy3 anti-mouse in 5% NGS-4⫻ SSC/0.2% Tween-20 for 30 min at 37⬚C. Between steps, slides were washed four times in 4⫻ SSC/0.2% Tween-20 for 5 min each at 37⬚C. Chromosomes were stained with 3 ␮g/ml DAPI in McIlvaine’s buffer (9 mM citric acid, 80 mM Na2HPO4·H2O, 2.5 mM MgCl2, pH 7.0) for 45 min at room temperature and destained in 4⫻ SSC/0.2% Tween-20 for 20 s. Slides detected with FITC-avidinDCS were further stained with 20 ␮g/ml of propidium iodide-2⫻ SSC for 30 min at room temperature and destained for 20 s in 2⫻ SSC. Vectashield娂 antifade agent was applied to the slides before a 20 mm ⫻ 40 mm coverslip was placed over the slides. Metaphase Observation and Photography Images were digitally captured in gray scale using a Nikon cooled-CCD camera system and standard Olympus filter sets for ultraviolet ( DAPI), triple band pass ( DAPI/Cy3), and blue (PI/FITC) excitation. The appropriate colors for the chromosomes and the signals were superimposed and contrast adjusted using the IPLab Spectrum P software on an Apple Macintosh Power PC. The final images were printed using Adobe Photoshop, version 5.0.

Results Single-label FISH, rather than dual-label FISH, was used to detect 5S rDNA and 18S28S rDNA signals with the greatest possible sensitivity. Only sites at which two signals were visible, that is, one per chromatid, were scored. Major sites were defined as those giving very large pairs of signals observable in all interphase and metaphase cells. Smaller FISH signals, detectable in 30–40% of metaphase cells ob-

298 The Journal of Heredity 2001:92(3)

served, were described as minor signals. The description and the number of sites for both 18S-28S and 5S rDNA signals are given in Table 1. Representative photomicrographs of the results are shown in Figures 1A–F and 2A–F. Two high-stringency washes were effective in eliminating background hybridization and greatly enhanced the reliability of our results. The additional 1⫻ SSC washes increased the signal : noise ratio by minimizing nonspecific hybridization. Signal intensity was not compromised by the higher stringency washes for both the 18S28S and the 5S rDNA probes. One major pair of 18S-28S rDNA FISH signals were observed in all of the interphase and metaphase spreads examined in the AA genome types of G. canescens and G. clandestina (2n ⫽ 40) ( Figure 2A,B). G. soja and G. max (2n ⫽ 40), belonging to the GG genome type, also exhibited one major pair of 18S-28S rDNA FISH signals in all the spreads examined ( Figure 2C,D). The polyploids G. tabacina and G. tomentella (2n ⫽ 80) ( Figure 2E,F) exhibited a pattern of two major and one minor pair of FISH signals. All of the 18S-28S rDNA signals observed were telomerically or subtelomerically located. A single major pair of 5S rDNA FISH signals was observed in all of the interphase and metaphase spreads in the (2n ⫽ 40) members ( Figure 1A–D) of the genus Glycine that we examined. Only two major pairs of 5S rDNA FISH signals were seen in the polyploid G. tomentella and G. tabacina (2n ⫽ 80) ( Figure 1E,F). The 5S rDNA loci were located distally on the chromosome pairs of all the species examined.

Discussion The ‘‘diploid’’ (2n ⫽ 40) species examined in the subgenus Glycine and the subgenus Soja, irrespective of their genome type, have a single 5S rDNA locus per haploid genome. The polyploids (2n ⫽ 80) G. tomentella and G. tabacina clearly have two 5S rDNA loci per haploid genome. Previous FISH studies have shown G. max ( L.) Merr as having a single 5S rDNA locus (Shi et al. 1996) and a single 18S-28S rDNA locus (Skorupska et al. 1989). CHEF gel electrophoresis detected a single 5S rDNA locus in G. soja ( Danna et al. 1996) and Southern blot analysis showed two major 5S repeats in G. tomentella ( Doyle and Brown 1989). However, this is the first confirmation of these results by FISH. If the genus Glycine has an allotetraploid origin ( Hymowitz and Singh 1987; Kumar and Hy-

mowitz 1989; Lackey 1980; Zhu et al. 1994) as has been postulated, then our results clearly demonstrate the loss of a 5S rDNA locus from the ‘‘diploid’’ species and loss of two from the polyploids of Glycine. A similar diploidization of the 18S-28S rDNA array has apparently occurred in G. canescens, G. clandestina, G. soja, and G. max ( L.) Merr. (2n ⫽ 40) since only half the expected number of loci per haploid genome were found. Both G. tabacina and G. tomentella (2n ⫽ 80), although of different genome types, showed two major loci per haploid genome in the metaphase spreads observed, as opposed to the four loci that would be expected in chromosomes that have undergone two doubling events in their evolutionary history. More than 50% of the spreads examined also detected an additional minor locus on a different chromosome from that of the major locus. Evolutionary Implications Both 5S and 18S-28S rDNA loci have experienced a diploidization event for members of both subgenera Soja and Glycine through physical loss of the sequences and not just loss of function. As these loci are on different chromosomes, at least in G. max (Shi et al. 1996), there must have been a separate deletion of each locus in the evolutionary past. Loss or addition of rDNA loci during the evolution of a polyploid plant species has been documented in Triticum ( Kim et al. 1993; Mukai et al. 1991), Aegilops ( Badaeva et al. 1996), Gossypium (Crane et al. 1993; Hanson et al. 1996), and Avena (Jellen et al. 1988). Investigations of newly formed polyploids (Comai et al. 2000; Xu et al. 2000) show great genomic and phenotypic instability. Mechanisms responsible for the variation in the size and the number of the rDNA loci may include (1) translocation breakpoints near the locus that may have occurred and the sites may have been duplicated following polyploidization of the species ( Hanson et al. 1996), (2) minor sites may have been added or deleted through nonhomologous unequal crossing over within the locus (Arnheim et al. 1980; Seperack et al. 1988), and (3) telomeric or subtelomeric positions of the rDNA loci would possibly allow significant rearrangements to occur without deleterious effects to the cells ( Bennett 1982; Hanson et al. 1996). Additions, deletions, and rearrangements of genetic material, with concomitant phenotypic abnormalities, can occur but must eventually stabilize into genetically stable species (Grant

et al. 2000; Paterson et al. 2000; Wendel 2000). Such early instability could account for the presence of genome duplication (Shoemaker et al. 1996; Lee et al. 1999), satellite chromosomes ( Huiyu and Ruiyang 1984) as well as diploidization (Grant et al. 2000; Hadley and Hymowitz 1973) within the genus. Diploidization of both the rDNA loci could have occurred soon after the original polyploid event that resulted in the 2n ⫽ 40 (4 FISH-detectable sites) Glycine ancestor, which was itself the result of a polyploidization of 2n ⫽ 20 (2 FISH-detectable sites) species that then evolved into the present-day 2n ⫽ 40 and 2n ⫽ 80 Glycine spp. It can also be speculated, but less likely, that each Glycine member underwent deletion events independently after species radiation, indicating possible deletion hotspots in the chromosomes involved. The time frame of these diploidization events may be challenged if (1) more than one rDNA locus is found in the remaining (2n ⫽ 40) Glycine spp. or (2) if (2n ⫽ 20) members can be found ( Kumar and Hymowitz 1989). If the report of a G. max (2n ⫽ 20) (Pillai 1976) can be confirmed, the number of 5S and 18S-28S rDNA loci must be investigated. The number of 5S rDNA loci in the (2n ⫽ 80) Glycine members suggests an origin through polyploidy of two diploidized parents. However, the presence of major and minor 18S-28S rDNA loci complicate such a simple scenario. Confounding an explanation is the tremendous diversity found within both species, indicating possible multiple origins (G. tabacina) ( Doyle et al. 1999) or active radiation/speciation (G. tomentella) (Singh et al. 1998). Multiple accessions within each species will have to be investigated to evaluate homologous relationships, if any, among their rDNA loci. As FISH yields semiquantitative results, that a major and minor site are homologous or orthologous will depend on sequence information and the multiplicity of an array repeat unit within a locus. The 18S-28S rDNA minor sites that we have observed could either represent reduction of a major site through partial deletion, the addition of a smaller array from a larger rDNA array by unequal rearrangement, or partial array duplication. Rapid evolution of multigene families is likely to produce readily detected polymorphisms between related species or among members of a species ( Danna et al. 1996). Repeat units within a tandem array of a multigene family typically undergo concert-

ed evolution ( Dover 1986). Unequal crossing over can also change copy number of the array repeat unit as observed for the 5S rRNA genes in flax plants subjected to environmental stress (Schneeberger and Cullis 1991). Patterns of evolution in plants have been uncovered through analysis of such polymorphisms in the rDNA in species of the Triticeae ( Kim et al. 1993; Mukai et al. 1991), Arabidopsis (Maluszynska and Heslop-Harrison 1993a), Gossypium (Crane et al. 1993), and Brassica ( Delseny et al. 1990; Maluszynska and Heslop-Harrison 1993b). It is evident that the evolution of the subgenus Glycine is much more complex than represented by a simple diploid → tetraploid (via chromosome doubling) model. Additional FISH studies on other species within the subgenus Glycine are necessary to completely decipher the evolutionary history of this important genus. And lastly, the restrained multiplicity of rDNA loci and their easy detection by FISH facilitates their use as a powerful tool for studies on the evolutionary behavior of repetitive gene families in soybean and other species. From the Department of Plant and Soil Science, Alabama A&M University, 4900 Meridian St., Carver Complex South, Room 213, Normal, AL 35762. The authors thank Lori Morgenrath and Dr. Robert Zahorchak of Research Genetics for their CCD camera system, Harold Anthony and Rudy Pacumbaba Jr. for their assistance with the computer systems and two anonymous reviewers for their helpful comments. Contributed by the Agricultural Experiment Station, Alabama A&M University, journal no. 423. This research was supported by USDA Capacity Building Grant no. 94-38814-0556. Address correspondence to A. Zipf at the address above or e-mail: [email protected]. 䉷 2001 The American Genetic Association

References Arnheim N, Krystal M, Schmickel R, Wilson G, Ryder O, and Zimmer E, 1980. Molecular evidence for genetic exchanges among ribosomal genes on nonhomologous chromosomes in man and ape. Proc Natl Acad Sci USA 77:7323–7327. Badaeva ED, Friebe B, and Gill BS, 1996. Genome differentiation in Aegilops. 2. Physical mapping of 5S and 18S–26S ribosomal RNA gene families in diploid species. Genome 39:1150–1158. Bennett MD, 1982. Nucleotypic basis of the spatial ordering of chromosomes in eukaryotes and the implications of the order for genome evolution and phenotypic variation: a subtitle. In: Genome evolution ( Dover GA and Flavell RB, eds). London: Academic Press; 239– 262. Comai L, Tyagi AP, Winter K, Holmes-Davis R, Reynolds SH, Stevens Y, and Byers B, 2000. Phenotypic instability and rapid gene silencing in newly formed Arabidopsis allotetraploids. Plant Cell 12:1551–1567. Crane CF, Price HJ, Stelly DM, Czeschin DG Jr, and McKnight TD, 1993. Identification of homeologous chromosome pairs by in situ DNA hybridization to ribosomal RNA loci in meiotic chromosomes of cotton (Gossypium hirsutum). Genome 36: 1015–1022.

Danna KJ, Workman R, Coryell V, and Keim P, 1996. 5S rRNA genes in tribe Phaseoleae: array size, number, and dynamics. Genome 39:445–455. Delannay X, Rodgers DM, and Palmer RG, 1983. Relative genetic contribution among ancestral lines to North American soybean cultivars. Crop Sci 23:944–949. Delseny M, McGarth JM, This P, Chevre AM, and Quiros CF, 1990. Ribosomal RNA genes in diploid and amphiploid Brassica and related species: organization, polymorphism, and evolution. Genome 33:733–744. Dover GA, 1986. Molecular drive in multigene families: how biological novelties arise, spread and are assimilated. Trends Genet 2:159–165. Doyle JJ and Brown AHD, 1989. 5S nuclear ribosomal gene variation in the Glycine tomentella polyploid complex ( Leguminosae). Syst Bot 14:398–407. Doyle JJ, Doyle JL, and Brown AHD, 1999. Origins, colonization, and lineage recombination in a widespread perennial soybean polyploid complex. Proc Natl Acad Sci USA 96:10741–10745. Grant D, Cregan P, and Shoemaker RC, 2000. Genome organization in dicots: genome duplication in Arabidopsis and synteny between soybean and Arabidopsis. Proc Natl Acad Sci USA 97:4168–4173. Hadley HH and Hymowitz T, 1973. Speciation and cytogenetics. In: Soybeans: improvement, production and uses (Caldwell BE, ed). Madison, WI: American Society of Agronomy; 97–116. Hanson RE, Islam-Faridi MN, Percival EA, Crane CF, Ji Y, McKnight TD, Stelly DM, and Price HJ, 1996. Distribution of 5S and 18S–28S rDNA loci in a tetraploid cotton (Gossypium hirsutum L.) and its putative diploid ancestors. Chromosoma 105:55–61. Huiyu Z and Ruiyang C, 1984. A diploid strain of wild soybean (Glycine soja) with four-satellited chromosomes. Soybean Sci 3:81–83. Hymowitz T and Singh RJ, 1987. Taxonomy and speciation: a subtitle. In: Soybeans: improvement, production, and uses, 2nd ed. Agronomy Monograph no.16 (Wilcox JR, ed). Madison, WI: American Society of Agronomy; 23–48. Islam-Faridi MN and Mujeeb-Kazi A, 1995. Visualization of Secale cereale DNA in wheat germ plasm by fluorescent in situ hybridization. Theor Appl Genet 90:595– 600. Jellen EN, Phillips RL, and Rines HW, 1988. Molecular genetic characterization of oat ribosomal DNAs. In: Agronomy Abstracts. Anaheim, CA: American Society of Agronomy; 169. Jewell DC and Islam-Faridi N, 1994. A technique for somatic chromosome preparation and C-banding of maize: a subtitle. In: The maize handbook ( Freeling M and Walbot V, eds). New York: Springer-Verlag; 484–493. Keim P, Shoemaker RC, and Palmer RG, 1989. RFLP diversity in soybean. Theor Appl Genet 77:786–792. Keim P, Diers BW, Olson TC, and Shoemaker RC, 1990. RFLP mapping in soybean: association between marker loci and variation in quantitative traits. Genetics 126: 735–742. Kim NS, Kuspira J, Armstrong K, and Bhambhani R, 1993. Genetic and cytogenetic analyses of the A genome of Triticum monococcum. VII. Localization of rDNAs and characterization of the 5S rRNA genes. Genome 36:77–86. Kumar PS and Hymowitz T, 1989. Where are the diploid (2n ⫽ 2x ⫽ 20) genome donors of Glycine Willd. ( Leguminosae, Papilionoideae)? Euphytica 40:221–226. Lackey JA, 1980. Chromosome numbers in the Phaseoleae ( Fabaceae: Faboideae) and their relation to taxonomy. Am J Bot 67:595–602. Lee JM, Bush AL, Specht JE, and Shoemaker RC, 1999. Mapping of duplicate genes in soybean. Genome 42: 829–836. Maluszynska J and Heslop-Harrison JS, 1993a. Molecular cytogenetics of the genus Arabidopsis: in situ local-

Brief Communications 299

ization of rDNA sites, chromosome numbers and diversity in centromeric heterochromatin. Ann Bot 71:479– 484. Maluszynska J and Heslop-Harrison JS, 1993b. Physical mapping of rDNA loci in Brassica species. Genome 36: 774–781. Mukai Y, Endo TR, and Gill BS, 1991. Physical mapping of the 18S.26S rRNA multigene family in common wheat: identification of a new locus. Chromosoma 100: 71–78. Paterson AH, Bowers JE, Burow MD, Draye X, Elsik CG, Jiang C-X, Katsar CS, Lan T-H, Lin Y-R, Ming R, and Wright RJ, 2000. Comparative genomics of plant chromosomes. Plant Cell 12:1523–1539. Pillai RVR, 1976. Diploids among the cultivars of soybean (Glycine max Linn.) in Manipur. Sci Cult 42:519– 521. Schmidt T and Heslop-Harrison JS, 1998. Genomes, genes and junk: the large-scale organization of plant chromosomes. Trends Plant Sci 3:195–199. Schneeberger RG and Cullis CA, 1991. Specific DNA al-

300 The Journal of Heredity 2001:92(3)

terations associated with the environmental induction of heritable changes in flax. Genetics 128:619–630.

lationships in the genus Glycine Willd. Genome 41:669– 679.

Seperack P, Slatkin M, and Arhneim N, 1988. Linkage disequilibrium in human ribosomal genes: implications for multigene family evolution. Genetics 119:943–949.

Skorupska H, Albertsen MC, Langholz KD, and Palmer RG, 1989. Detection of ribosomal RNA genes in soybean, Glycine max ( L.) Merr., by in situ hybridization. Genome 32:1091–1095.

Shi L, Zhu T, and Keim P, 1996. Ribosomal RNA genes in soybean and common bean: chromosomal organization, expression, and evolution. Theor Appl Genet 93: 136–141. Shoemaker RC, Polzin K, Labate J, Specht J, Brummer EC, Olson T, Young N, Concibido V, Wilcox J, Tamulonis JP, Kochert G, and Boerma HR, 1996. Genome duplication in soybean (Glycine subgenus soja). Genetics 144: 329–338. Silhavy TJ, Berman ML, and Enquist LW, 1984. Experiments with gene fusion. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Singh RJ, 1993. Plant cytogenetics. Boca Raton, FL: CRC Press. Singh RJ, Kollipara KP, and Hymowitz T, 1998. The genomes of Glycine canescens F.J. Her., and G. tomentella Hayata of Western Australia and their phylogenetic re-

Specht JE and Williams JH, 1984. Contribution of genetic technology to soybean productivity-retrospect and prospects: a subtitle. In: Genetic contributions to yield gains of five major crop plants ( Fehr WR, ed). Madison, WI: American Society of Agronomy; 49–74. Wendel JF, 2000. Genome evolution in polyploids. Plant Mol Biol 42:225–249. Xu SJ, Singh RJ, Kollipara KP, and Hymowitz T, 2000. Hypertriploid in soybean: origin, identification, cytology, and breeding behavior. Crop Sci 40:72–77. Zhu T, Schupp JM, Oliphant A, and Keim P, 1994. Hypomethylated sequences: characterization of the duplicated soybean genome. Mol Gen Genet 244:638–645. Received May 9, 2000 Accepted November 11, 2000 Corresponding Editor: Reid G. Palmer