Isozyme analysis of genetic diversity in wild Sicilian populations of Brassica sect. Brassica in view of genetic resources management 1, 2 2 2 * ` ´´ Anna Geraci , Anne-Marie Chevre , Isabelle Divaret , Frederique Eber and Francesco 1 M. Raimondo 1 2 ` Dipartimento di Scienze Botaniche, Universita di Palermo, via Archirafi 38 90123, Palermo, Italy; INRA * ` Station d’ Amelioration, des Plantes Domaine de la Motte BP35327, 35653 Le Rheu, cedex, France; Author for correspondence;(e-mail: anna.geraci@infinito.it ) Received 13 March 2001; accepted in revised form 31 May 2002 Key words: Brassica sect. Brassica wild sicilian populations, Genetic resources, Genetic structure, Isozyme diversity Abstract In Sicily and in the small surrounding islands the section Brassica of the genus Brassica comprises five species, B. insularis Moris, B. incana Ten., B. macrocarpa Guss., B. rupestris Raf. and B. villosa Biv. These taxa represent a genetic resource as relatives of kale crops but some populations are endangered or threatened, thus isozyme analyses were performed to assess the genetic diversity degree at population and species levels in order to assist the design of conservation management programs. Eleven loci from five enzyme systems (aconitase, leucine aminopeptidase, 6-phosphogluconate dehydrogenase, phosphoglucoisomerase phosphoglucomutase) were analyzed in sixteen natural population (fifteen from Sicily, one from Calabria). Mean within-population genetic diversity was moderate (P 5 41%, A 5 1.54, H 5 0.16). In some cases a great number of heterozygous individuals were detected, in other cases fixation index (F) deviated significantly from Hardy-Weinberg genotypic expectations. A total of 37 alleles was recognized, six of which resulted exclusive to single populations. The among- population component of the total genetic diversity (Gst mean values) for each species was 0.30–0.37, indicating genetic differentiation among populations. Among B. villosa and B. rupestris populations genetic distance values resulted rather low and they resulted high with B. incana and B. macrocarpa populations. The results are discussed with regard to the distribution of the genetic diversity level and the genetic resources management. Introduction villosa, subsp. bivoniana (Mazzola et Raimondo) Raimondo et Mazzola, subsp. drepanensis (Caruel) Brassica sect. Brassica comprises ten species distrib- Raimondo et Mazzola, subsp. tinei (Lojac.) Raimondo uted in Mediterranean area and eastern Atlantic coast- et Mazzola) and to B. rupestris (subsp. rupestris, al areas ( Brassica oleracea L., B. cretica Lam., B. subsp. hispida Raimondo et Mazzola, subsp. bre- hilarionis Post, B. bourgeoui (Webb) O. Kuntze, B. visiliqua Raimondo et Mazzola) (Snogerup et al. montana Pourret, B. incana Ten., B. insularis Moris, 1990; Raimondo and Mazzola 1997). B. macrocarpa Guss., B. villosa Biv., B. rupestris This section, characterized by 2n 5 18, represents a Raf.) (Diederichsen 2001). Different subspecies are cytodeme (Harberd 1972) in which wild species and described as related to B. cretica (susbp. cretica, cultivated forms of B. oleracea are interfertile in some subsp. aegaea (Heldr. et Hal.) Snog., Gust. et Both., degrees (Kianian and Quiros 1992; Von Bothmer et al. subsp. laconica Gust. et Snog.), to B. villosa (subsp. 1995). The study and the conservation of wild rela- 2004 Kluwer Academic Publishers. Printed in the Netherlands. Genetic Resources and Crop Evolution 51: 137–146, 2004. 137
Transcript
Isozyme analysis of genetic diversity in wild Sicilian populations ofBrassica sect. Brassica in view of genetic resources management
1 2`Dipartimento di Scienze Botaniche, Universita di Palermo, via Archirafi 38 90123, Palermo, Italy; INRA*`Station d’Amelioration, des Plantes Domaine de la Motte BP35327, 35653 Le Rheu, cedex, France; Author
In Sicily and in the small surrounding islands the section Brassica of the genus Brassica comprises five species, B.insularis Moris, B. incana Ten., B. macrocarpa Guss., B. rupestris Raf. and B. villosa Biv. These taxa represent agenetic resource as relatives of kale crops but some populations are endangered or threatened, thus isozymeanalyses were performed to assess the genetic diversity degree at population and species levels in order to assistthe design of conservation management programs.
Eleven loci from five enzyme systems (aconitase, leucine aminopeptidase, 6-phosphogluconate dehydrogenase,phosphoglucoisomerase phosphoglucomutase) were analyzed in sixteen natural population (fifteen from Sicily,one from Calabria). Mean within-population genetic diversity was moderate (P 5 41%, A 5 1.54, H 5 0.16). Insome cases a great number of heterozygous individuals were detected, in other cases fixation index (F) deviatedsignificantly from Hardy-Weinberg genotypic expectations.
A total of 37 alleles was recognized, six of which resulted exclusive to single populations. The among-population component of the total genetic diversity (Gst mean values) for each species was 0.30–0.37, indicatinggenetic differentiation among populations.
Among B. villosa and B. rupestris populations genetic distance values resulted rather low and they resulted highwith B. incana and B. macrocarpa populations.
The results are discussed with regard to the distribution of the genetic diversity level and the genetic resourcesmanagement.
Introduction villosa, subsp. bivoniana (Mazzola et Raimondo)Raimondo et Mazzola, subsp. drepanensis (Caruel)
Brassica sect. Brassica comprises ten species distrib- Raimondo et Mazzola, subsp. tinei (Lojac.) Raimondouted in Mediterranean area and eastern Atlantic coast- et Mazzola) and to B. rupestris (subsp. rupestris,al areas (Brassica oleracea L., B. cretica Lam., B. subsp. hispida Raimondo et Mazzola, subsp. bre-hilarionis Post, B. bourgeoui (Webb) O. Kuntze, B. visiliqua Raimondo et Mazzola) (Snogerup et al.montana Pourret, B. incana Ten., B. insularis Moris, 1990; Raimondo and Mazzola 1997).B. macrocarpa Guss., B. villosa Biv., B. rupestris This section, characterized by 2n 5 18, represents aRaf.) (Diederichsen 2001). Different subspecies are cytodeme (Harberd 1972) in which wild species anddescribed as related to B. cretica (susbp. cretica, cultivated forms of B. oleracea are interfertile in somesubsp. aegaea (Heldr. et Hal.) Snog., Gust. et Both., degrees (Kianian and Quiros 1992;Von Bothmer et al.subsp. laconica Gust. et Snog.), to B. villosa (subsp. 1995). The study and the conservation of wild rela-
2004 Kluwer Academic Publishers. Printed in the Netherlands.Genetic Resources and Crop Evolution 51: 137–146, 2004. 137
tives could thus be very interesting for the breeding ofagronomical varieties.
Gladis and Hammer (2001) propose to unite in onespecies (Brassica oleracea) all these wild and culti-vated forms and to treat the Atlantic and the wildBrassicas at subspecies and variety level.
Sicily represents one of the two centres of differen-tiation of this section, since the other one is in theeastern Mediterranaean area (Raimondo 1997a). Infact, three species, B. rupestris, B. villosa and B.macrocarpa, are endemic to the Sicilian area and twoother species, B. incana and B. insularis, also occurthere.
Their habitats consist of limestone cliffs at sea leveluntil 1000–1200 m. The populations are often re-stricted in size and distribution, basically because ofthe limited areas of cliffs, the competition with other
Figure 1. Map of distribution of 16 Brassica populations investi-species and the human disturbance i.e. grazing, fire,gated.
quarries, etc. Consequently, some of these popula-tions are endangered or threatened and need to bepreserved by genetic resources conservation measures(Raimondo et al. 1994 According to the enzyme system, the compositions).
of the gels and the buffers were as follows:This paper evaluates the isozyme diversity of 161) aconitase (ACO), leucine aminopeptidasewild populations from Sicily and Calabria in order to
(LAP), phosphoglucoisomerase (PGI), phospho-obtain genetic resource information for meaningfulglucomutase (PGM)implementation of the conservation programs.• gel buffer: histidine pH 7 (DL-histidine, 5 mM
titrated with NaOH);• electrode buffer: Tris pH 7, 0.13 M titrated with
Materials and methodscitric acid;
A total of 16 populations was collected from natural2) 6-phosphogluconate dehydrogenase (6-PGD)habitats of Sicily, Egadi islands and Calabria in 1997
• gel buffer: morpholine-citrate pH 6.1; 2 mM(Table 1, Figure 1). Seeds were collected from seven(1 /20 dilution of electrode buffer)to ten different individuals and pooled to obtain a
• electrode buffer: citric acid 40 mM with pH ad-good representative sample of each population. Seedsjusted to 6.1 with N-3-aminopropylmorpholine.were stored under appropriate conditions (15 8C, low
humidity).Thirtyfive plants per population, grown in indi-
The enzymes migrated towards the anode. 35 mAvidual pots in a heated greenhouse (25 8C), were
were applied in pH 6.1 gels and 100 V in pH 7 gelsstudied.
during the first ten minutes. After the wicks wereremoved, gels were run at 40 mA and 150 V respec-
Isozyme analysis tively.Gel slices were incubated in a staining substrate
Fresh young leaves were crushed in 100 mL of solution, specific to each enzyme tested. Stainingextraction buffer containing tris HCl pH 7.5 and 1% solutions for ACO and LAP were described by Wen-reduced glutathion. Soaked wicks of each sample del and Stuber (1984) and Arus and Orton (1983),
`were inserted into a slit made across the gel (Chevre respectively. Solutions for PGI, PGM and 6-PGDet al. 1995). The technique used for horizontal elec- were as reported by Vallejos (1983).trophoresis in 11% starch gels is described by Kephart Zymograms and alleles nomenclature referred to
`(1990). Chevre et al. (1995) (Figure 2).
138
Table 1. Identification, geographical origin and size of Brassica populations investigated.
Identification Taxa Origin Size (number of individuals)
pop1 B. villosa subsp. villosa Monte Calcerame (Sagana) – Palermo 101–500pop2 B. villosa subsp. bivoniana M. Inici – Trapani 501–1000pop3 B. villosa subsp. bivoniana Caltabellotta – Agrigento .1000pop4 B. villosa subsp. drepanensis M. S. Giuliano - Erice - Trapani .1000pop5 B. villosa subsp. bivoniana C.da Serbatoio (Fontanarossa) - Trapani 11–50pop6 B. villosa subsp. tinei P.lla Scalazza (Marianopoli)- Caltanissetta 501–1000pop7 B. rupestris subsp. rupestris M. Pellegrino - Palermo .1000pop8 B. rupestris subsp. rupestris Rocca Busambra - Palermo .1000pop9 B. rupestris subsp. rupestris Stilo – Cosenza (Calabria) 51–100pop10 B. rupestris subsp. rupestris M. Sferrovecchio (S. Ciro) – Palermo 51–100pop11 B. rupestris subsp. brevisiliqua Capo San Vito (Isolidda) – Trapani 101–500pop12 B. rupestris subsp. hispida M. Pizzuta - Palermo .1000pop13 B. rupestris subsp. hispida Cozzo Cicero (Borgetto) - Palermo 51–100pop14 B. incana Gonato - Madonie - Palermo 101–500pop15 B. incana Passo della Zita (Longi) - Messina 101–500pop16 B. macrocarpa M. S. Caterina (Favignana, Isole Egadi) - Trapani 501–1000
Data analysis
For each locus the resulting zymograms were inter-pretated as allelic frequencies. For all populationslevels of allozyme diversity were estimated. A locuswas considered as polymorphic if the most commonallele was present at the frequency , 0.95 (P 95) or ,0.99 (P 99).
Wright (1951) fixation (F) index was calculated asF 5 12Ho/He with Ho, the observed heterozygosity(percentage of heterozygous for each population andlocus) and He, the expected heterozygosity calculatedfrom allelic frequencies according to the Hardy-Wein-berg law. Chi-square test was used to evaluate thesignificativity of the deviation from the Hardy-Wein-berg law.
For each population gene diversity, H, as describedby Nei (1973), was calculated for each polymorphiclocus.
For each polymorphic locus genetic diversity in thetotal populations (Ht), mean genetic diversity withinpopulations (Hs), the average genetic diversity amongpopulations (Dst) and the relative magnitude of genedifferentiation occurring among populations (Gst)were calculated (Nei 1973).
Genetic distances among populations were esti-mated from allelic frequencies (Nei 1972) usingBIOSYS-1 program (Swofford and Selander 1989).
Dendrogram was then computed from the distancematrices using UPGMA option of the NEIGHBORprogram of Phylogeny Inference Package (PHYLIPFigure 2. Zymograms of: (a) Pgi-2 (pop8), (b) Pgm-2 (pop9), (c)
Lap-1 (pop7). Allelic interpretation of bands is shown. 3.5c., Felsenstein 1993).
139
Results population (pop7) and for B. incana populations(pop14 and 15).
Rare alleles (frequency , 5 %) were mainly foundAllele frequencies in B. rupestris (pop7 to 10), B. villosa subsp. villosa
(pop1), B. incana (pop15) and B. macrocarpaEleven polymorphic loci were observed from the five (pop16). Six alleles were specifically found in oneenzymes investigated (Table 2). In particular Lap-1 population (* in Table 2).locus showed the most variability with five differentalleles and only one population of B. villosa subsp. Genetic variabilitybivoniana (pop5) being monomorphic at this locus.The four loci belonging to the ACO system were also A total of 37 alleles was observed. For each popula-very variable with three alleles detected for each tion the mean proportion of polymorphism levels P95locus. Some loci were monomorphic for most popula- and P99, the mean number of alleles per locus acrosstions as Pgm-3 which was polymorphic only for one all loci (A) and per polymorphic locus (Ap) and thepopulation of B. rupestris subsp. rupestris (pop7). Wright’s fixation index (F) are shown (Table 3).The 6PGD system was also only polymorphic for this Mean number of alleles per locus (A) varied from
Table 2. Frequencies of the allozymes detected in 16 wild Brassica populations analyzed with 5 enzymatic systems.
System Locus Allele pop1 pop2 pop3 pop4 pop5 pop6 pop7 pop8 pop9 pop10 pop11 pop12 pop13 pop14 pop15 pop16
Table 3. Number of alleles (N), Average number of alleles per locus (A) and per polymorphic locus (Ap); average polymorphism P95 and P99,Wright index (F) and unbalanced loci for 16 wild Brassica populations.
1.19 to 1.72 with a mean value for all the populations found at Pgm-1 in pop6 (0.592), Pgm-2 in pop11of 1.54. Considering only the polymorphic loci, this (0.580), Aco-2 and Aco-3 in pop13 (0.530 andmean value (Ap) increased and ranged between 2 and 0.522), Lap-1 in pop1, pop2 (0.625) and pop102.6, with a general mean of 2.22. According to the (0.656). Mean values, considering all loci rangedpopulation, from 18% (pop5) to 63% (pop2, pop7, between 0.24 (pop2) and 0.067 (pop12). Highestpop14) of the loci studied were polymorphic. F val- values were found for B. incana (0.23, pop14), B.ues, Wrigth’s index, resulted negative for different rupestris subsp. hispida (0.21, pop13), B. villosapopulations indicating an excess of heterozygous subsp. bivoniana (0.21, pop3) and B. rupestris subsp.individuals with respect to the expected rate. Chi- rupestris (0.20, pop7). The mean value for all thesquare test showed that seven populations were not in populations was 0.16.equilibrium for some loci (Table 3). The Ht, Hs, Dst, Gst values calculated for each
Nei’s gene diversity index (H) calculated for each species for the polymorphic loci are shown in Table 5.polymorphic locus and for each population (Table 4) The total diversity (Ht) was slightly greater in B.varied from 0.056 (pop8 and pop9) at Pgm-1 locus to incana and B. villosa populations (0.338 and 0.3910.65 (pop10) at Lap-1 locus. Greater variability was respectively) than in B. rupestris populations (0.287).
Table 4. Nei (1973) diversity for each population at each locus.
But for all the species, the diversity within popula- villosa showed a greater diversity within populationstions (Hs) was greater than the diversity among for Lap-1 and B. incana for 6Pgd-1. The loci in-populations (Dst) in the same proportion (Figure 3). volved in the populations differentiation (Gst) were
Although, the differentiation in Hs and Dst was also different: Pgm-2 and Aco-4 for B. villosa andsimilar according to the species, the structure of the Lap-1 and Pgm-1 for B. incana.genetic diversity at the studied loci was different. ForB. rupestris (Table 5a), the total diversity (Ht) and thediversity within populations (Hs) was greater at the Genetic distancesAco-3 and Lap-1 loci. For this species, Pgm-2 and6Pgd-1 were important for the differentiation among Matrix of genetic distance (Table 6), shows thatthe populations (Gst). The total diversity (Ht) was distance values among B. villosa populations (pop 1greater at Lap-1 and Pgm-2 loci for both B. villosa to 6) were comprised between 0.06 and 0.26 and(Table 5b) and B. incana (Table 5c) species. But B. ranged from 0.05 to 0.20 among B. rupestris popula-
Table 5. Diversity index: total genetic diversity (Ht), genetic diversity within populations (Hs), genetic dufferentiation (Gst) for each species.
rocarpa from the villoso-rupestris group in which aclear separation of populations at specific and sub-specific rank was not well detectable.
Discussion
From the analysis of allelic presence in our popula-tions, 37 alleles were detected. B. rupestris and B.villosa show a more similar composition whereas B.incana and B. macrocarpa resulted more distinct andmore differentiated. The presence of six exclusivealleles was a very important remark because it char-acterizes the populations.
Figure 3. Differentiation of the total diversity in Hs (within The mean value of polymorphic loci ranged be-populations) and Dst (among populations).
tween 18% in B. villosa subsp. bivoniana (pop5) and63% in B. villosa subsp. bivoniana (pop2), B. rupes-tris subsp. rupestris (pop7) and B. incana (pop14).
tions (pop7 to 13). A distance of 0.22 was found The lowest value of polymorphism was found in popbetween B. incana populations. 5 which is the smallest population studied having less
The minimum distance (0.06) between B. rupestris than 50 individuals in its natural habitat. The 36%and B. villosa was found between pop13 and pop3 value obtained for B. macrocarpa (pop16) is similarwhereas the maximum value (0.29) was found be- to the result obtained by Lazaro and Aguinagaldetween pop2 and pop12. (1998a) for a population of B. macrocarpa (36.36%).
The distances between B. incana and B. villosa and The average of gene diversity, computed as Nei’sbetween B. incana and B. rupestris were similar (0.29 (H) statistic, is comprised (mean values) betweento 0.51). 0.067 and 0.24 although great fluctuations were de-
High values of genetic distance were found also tected for different loci. The mean values of H arebetween B. macrocarpa and the other species, ranging lower than those obtained by Lanner-Herrera et al.from 0.20 to 0.46. (1996) for wild populations of B. oleracea from
The dendrogram (Figure 4), obtained with Spain, France, and Great Britain for which the aver-UPGMA algorithm, on the basis of these distances age diversity was 0.40. However, as these authorsshowed a great separation of B. incana and B. mac- report, this index gives information on the variation in
Figure 4. Dendrogram of 16 wild Brassica populations from the distance matrix described in Table 1 and the UPGMA algorithm.
a population but does not describe the organization of presented an excess of heterozygotes.This could bethe variation in the population. For example two explained by assuming the presence of subpopulationspopulations of B. incana are very different for H or competition phenomena in a population as well asvalues (0.23 pop14, 0.09 pop15) but their allelic extensive gene exchanges (Brown 1979) and alsocomposition is very similar; on the contrary B. incana hypothesizing the self-incompatibility systems, often(pop14) B. villosa subsp. bivoniana (pop2), B. rupes- found in many Brassica populations (Watanabe andtrus subsp. hispida (pop13) have similar values (0.23, Hinata 1999). The presence of rare alleles could also0.24 and 0.21 respectively) but a very different allelic have an influence on the number of heterozygotes,composition. because the probability to find them as heterozygotes
Several factors have been associed with differences is higher than to find them as homozygotes. For twoin values of polymorphism (P), allelic number (A), populations, pop4 (B. villosa subsp. drepanensis) andgene diversity (H) such as life form, regional dis- pop14 (B. incana), an excess of homozygous indi-tribution, geographical range, breeding system (Ham- viduals was found. For these populations, the generick et al. 1979; Hamrick 1990). According with these exchange seems restricted either by geographicalparameters, high values are generally associed with isolation or by biological process.mixed system of reproduction between autogamy and Gst values could be related with breeding systemallogamy; low values are found in autogamous and in and seed dispersal mechanism (Loveless and Hamrickendemic plants. In our case, we must keep in mind 1984). In Brassica populations investigated, meanthat three species (B. macrocarpa, B. rupestris, B. values of Gst were comprised between 0.30–0.37.villosa) are endemic, some populations (as pop5) are Therefore a breeding system mixed between au-constitued by very few individuals and other popula- togamy and allogamy (probably prevailing because oftions as B. rupestris subsp. hispida (pop12) and B. self-incompatibility) and a seed dispersal-mechanismmacrocarpa (pop16) that are larger in size, are iso- for gravity or provoked by animals could be hypoth-lated from the geographic point of view. esized.
Considering the fixation index, a great number of The same structure of the diversity was observedpopulations, in particular large sized populations, for each species with the intra-populational diversity
144
(Hs) representing the higher part of total diversity tion (Gemmill 1998). Thus, a knowledge of the(Ht) (60% for B. incana, 62% for B. rupestris and
levelsand distribution of genetic diversity of threat
58% for B. villosa). The coefficient of genic differen-ened or
endangered species is an important element intiation (Gst) is comprised between 0.30 and 0.37.
arrang- ing conservation programs.
These mean values were comparable with those re- These genetic data have a great importance inported by Lanner-Herrera et al. (1996) for B. oleracea leading conservation actions for these species, rela-populations which ranged between 0.11 and 0.50 and tives to kale crops, that are in some case endangeredwith those reported by Lazaro and Aguinagalde or threatened and to provide information necessary for(1998a) for B. rupestris group (0.11–0.31). In the maximization of genetic diversity which is an im-Sicilian taxa the Gst values show high variability for portant consideration at the intraspecific and specificeach polymorphic locus varying from 0.03–0.06 to level (Loo et al. 1999). Concerning conservation0.70–0.79. High values as 0.72 (Pgm-2 ), 0.56 (6Pgd- methods, Gillies et al. (1997) suggested that in situ1 ) in B. rupestris, 0.52 (Pgm-2 ), 0.49 (Aco-4 ), 0.47 conservation strategies should seek to conserve popu-(Aco-2 ) in B. villosa, 0.79 (Lap-1 ), 0.70 (Pgm-2 ) in lations that show interpopulational genetic variation,B. incana, show that these loci play an important role to prevent the loss of genetic diversity within a taxon.of differentiation among the populations. On the other hand, ex situ conservation measures
Concerning the genetic distances, B. incana popu- should include the collection of germplasm fromlations (in the same group) show a clear separation separated populations with significant genetic differ-
´from the villoso-rupestris populations. B. mac- ences (Martın et al. 1997). Our observations suggestrocarpa, too, forming a separate clade, is very dif- that is necessary to preserve in situ the populations asferentiated from the other Sicilian species. This evi- dynamic conservation. That should be accompanieddence is confirmed by RAPD analyses (Geraci et al. by the threat prevention and by periodic monitoring2001) because in these taxa specific genomic zones (ecologically, genetically) of the status of these popu-were found. Also Lazaro and Aguinagalde (1998a, lations. On the other hand these data could be very1998b) realized by isozymes and molecular markers, useful in ex situ conservation programmes in order tofound B. incana more strictly related to B. montana- quantify and to document the diversity that can beB. oleracea group even if taxonomically it belongs to present in a collection and at the same time to use
´B. rupestris group (Gomez-Campo 1980, 1999). B. these information to allow to maximize per accessionvillosa and B. rupestris populations, showing distance of genetic diversity (Lamboy et al. 1994). In this case,values very heterogenous, are not separated as well as considering that the level of diversity inter- and intra-by morphological characters and under geographical populational were comparable could be useful toand ecological data (Raimondo et al. 1991). It shows conserve a large sample of a large number of popula-the close relationship between these species that tions as well as a large number of individuals withinpartly overlap and then can hybridize. However, also each population. Moreover populations showing ex-Lanner-Herrera et al. (1996) studying North-Euro- clusive alleles deserve particular attention as well aspean wild populations relatives of B. oleracea, ob- those made up by few individuals that are threatenedserved that with isozyme analysis populations of the in their natural habitat.same country were not always related.
In conclusion, isozyme analysis was very useful tocharacterize Brassica sect. Brassica populations as Acknowledgementsregards allelic structure and variability within-popula-tion. ´The authors are grateful to Prof. C. Gomez-Campo for
Enzymatic study was very interesting to assess the the interesting discussions and to Prof. M. Gustafssongenetic structure of the populations. In fact, genetic for the revision of the manuscript and the helpfulvariation within a taxon could be critical for the long suggestions.term survival and continuous evolution of a popula-tion or a species (Huenneke 1991). Population genetictheory predicts that a decrease of heterozygosity will
Referenceslead to reduced individual fitness and populationviability and that a decrease in allelic diversity will Arus P. and Orton T.J. 1983. Isozyme and linkage relationships oflimit the ability of a species to track changing selec- isozyme loci in Brassica oleracea. J. Hered. 74: 405–412.
145
Von Bothmer R., Gustafsson M. and Snogerup S. 1995. Brassica Kianian S.F. and Quiros C.F. 1992. Trait inheritance, fertility, andsect. Brassica (Brassicaceae) II. Inter- and intraspecific crosses genomic relationships of some n 5 9 Brassica species. Genet.with cultivars of B. oleracea. Genetic Resources and Crop Resour. Crop Evol. 39: 165–175.Evolution 42: 165–178. Lamboy W.F., Mcferson J.R., Westmann A.L. and Kresovich S.
Brown A.D.H. 1979. Enzyme polymorphism in plant populations. 1994. Application of isozyme data to the management of theTheor. Pop. Biol. 15: 1–42. United States national Brassica oleracea. L. genetic resources`Chevre A.M., Delourme R., Eber F., Margale E., Quiros C. and collection. Genet. Resour. Crop Evol. 41: 99–108.Arus P. 1995. Genetic analysis and nomenclature for seven Lanner-Herrera C., Gustafsson M., Falt A-S and Bryngelsson T.isozyme systems in Brassica nigra, B. oleracea, and B. campes- 1996. Diversity in natural populations of wild Brassica oleraceatris. Plant Breeding 114: 473–480. as estimated by isozyme and RAPD analysis. Genet. Resour.
Diederichsen A., 2001. Brassica L. In: Hanelt, P. and Institute of Crop Evol. 43: 13–23.Plant Genetics and Crop Plant Research (eds), Mansfelds En- Lazaro A. and Aguinagalde I. 1998a. Genetic diversity in Brassicacyclopedia of Agricultural and Horticultural Crops. Springer oleracea L. (Cruciferae) and Wild Relatives (2n 5 18) usingBerlin - Heidelberg. isozymes. Annals of Botany 82: 821–828.
Felsenstein J. 1993. PHYLIP (PHYlogenetic Inference Package) Lazaro A. and Aguinagalde I. 1998b. Genetic diversity in Brassicaversion 3.5 c distributed by the author. Departement of Genetics, oleracea L. (Cruciferae) and Wild Relatives (2n 5 18) usingUniversity of Washington, Seattle /WA. RAPD Markers. Annals of Botany 82: 829–833.
Gemmill C.E., Ranker T.A., Ragone D., Perlman S.P. and Wood Loo A.H.B., Tan H.T.W., Kumar P.P and Saw L.G. 1999. Popula-K.R. 1998. Conservation genetics of endangered endemic tion analysis of Licuala glabra Griff. var. glabra (Palmae) usingHawaiian genus Brighamia (Campanulaceae). Amer. J. Bot. 85: RAPD profiling. Annals of Botany 84: 421–427.528–539. Loveless M.D and Hamrick J.L. 1984. Ecological determinants of
`Geraci A., Divaret I., Raimondo F.M. and Chevre A.M. 2001. Use genetic structure in plant populations. Annu. Rev. Ecol. Syst. 15:of RAPD markers to investigate the genetic relationships be- 65–95.
´ ´tween Sicilian wild populations of Brassica. Plant Breeding 120: Martın C., Gonzales-Benito M.E. and Iriondo J.M. 1997. Genetic193–196. diversity within and among populations of a threatened species:
´Gillies A.C.M., Cornelius J.P., Newton A.C., Navaro C., Hernandez Erodium Paularense Fern. Gonz. and Izco. Molecular Ecology 6:M. and Wilson J. 1997. Genetic variation in Costa Rican 813–820.populationsof the tropical timber species Cedrela odorata L. Nei M. 1972. Genetic distance between populations. Amer. Natu-assessed using RAPDs. Molecular Ecology 6: 1133–1145. ralist. 106: 283–292.
Gladis Th. and Hammer K. 2001. Nomenclatural notes on the Nei M. 1973. Analysis of gene diversity in subdivided populations.Brassica oleracea - group. Genet. Resour. Crop Evol. 48: 7–11. Proc. Nat. Acad. Sci. USA 70: 3321–3323.´Gomez-Campo C. 1980. Morphology and morphotaxonomy of the Raimondo F.M. 1997a. Les membres italiens du complex Brassica
´ ´tribe Brassiceae. In: Tsunoda S., Hinata K. and Gomez–Campo oleracea: Leur distribution et specifites ecologiques. BocconeaC. (eds), Brassica Crops and Wild Allies. Biology and Breeding. 7: 103–106.Japan Scientific Society Press, Tokyo, pp. 3–31. Raimondo F.M., Gianguzzi L. and Ilardi V. 1994. Inventario delle´Gomez-Campo C. and Gustafsson M. 1995. Prospect for in situ specie ‘‘a rischio’’ nella flora vascolare nativa della Sicilia.conservation of Brassica oleracea wild relatives In: Gass T., Quad. Bot. Ambientale Appl. 3: 65–132.Gustafsson M., Astley D. and Frison E. (compilers), Report of a Raimondo F.M., Mazzola P. and Ottonello D. 1991. On the tax-working group on Brassica (second meeting, 13–15 November onomy and distribution of Brassica sect. Brassica (Cruciferae)1994, Lisbon, Portugal). pp. 72–74. European Cooperative Pro- in Sicily. Fl. Medit. 1: 63–86.gramme for Crop Genetic Resources Networks (ECP/GR). Inter- Raimondo F.M. and Mazzola P. 1997. A new taxonomic arrange-national Plant Genetic Resources Institute, Rome, Italy. ment of the sicilian members of Brassica sect. Brassica. Lagas-´Gomez-Campo C. 1999. Biology of Brassica Coenospecies. calia 19: 831–838.Elsevier. Snogerup S., Gustafsson M. and Von Bothmer R. 1990. Brassica
Hamrick J.L. 1990. Isozymes and the analysis of genetic structure sect. Brassica (Brassicaceae) I Taxonomy and variation. Will-in plant populations. In: Soltis D.E. and Soltis P. (eds), Isozymes denowia 19: 271–365.in Plant Biology. Dioscorides Press, Portland, OR. Swofford D.L. and Selander R.B. 1989. BIOSYS-1. Release 1.7.
Hamrick J.L., Linhart Y.B. and Mitton J.B. 1979. Relationships Illinois Natural History Survey, Champagne, IL.between life history characteristics and electrophoretically–de- Vallejos C.E. 1983. Enzyme activity staining. In: Tanksley S.D. andtectable genetic variation in plant. Annu. Rev. Ecol. Syst. 10: Orton T.J. (eds), Isozymes in Plant Genetics and Breeding, Part173–200. A. Elsevier Science Publishers, Amsterdam, pp. 469–516.
´Harberd D.J. 1972. A contribution to the cyto-taxonomy of Bras- Watanabe M. and Hinata K. 1999. Self-Incompatibility. In: Gomez-sica (Cruciferae) and its allies. Bot. J. Linn. Soc. 65: 1–23. Campo C. (ed.), Biology of Brassica Coenospecies. Elsevier, pp.
Huenneke L.F. 1991. Ecological implications of genetic variation 149–183.in plant populations. In: Falk D.A. and Holsinger K.E. (eds), Wendel J.F. and Stuber C.F. 1984. Plant isozymes: EnzymesGenetics and conservation of rare plants. Oxford University studied and buffer systems for their electrophoretic resolution inPress, New York. starch gels. Isozyme Bull. 17: 4–11.
Kephart S.R. 1990. Starch gel electrophoresis of plant isozymes: a Wright S. 1951. The genetic structure of populations. Annals ofcomparative analysis of techniques. Amer. J. Bot. 77: 693–712. Eugenics 15: 323–354.
