Genetica (2010) 138:985–998 DOI 10.1007/s10709-010-9485-5
ORIGINAL RESEARCH
Effectiveness of in situ and ex situ conservation of crop diversity. What a Phaseolus vulgaris L. landrace case study can tell us Valeria Negri • Barbara Tiranti
Received: 10 April 2009 / Accepted: 9 August 2010 / Published online: 11 September 2010 Ó Springer Science+Business Media B.V. 2010
Abstract The effectiveness of in situ (on-farm) and ex situ conservation strategies to maintain total genetic diversity was assessed in a threatened Phaseolus vulgaris L. landrace. Farmer seed lots (subpopulations) were sampled initially and then after in situ and ex situ multiplication (two locations). The number of plants used in the ex situ multiplications (120) was much larger than that normally used in germplasm bank procedures and the farmer seed lots were kept separate. In situ, the landrace was multiplied by each farmer with the usual population size. Eighty plants from the initial population, the in situ and the two ex situ multiplications were individually tested using 26 microsatellite markers. Most of the genetic parameters showed a consistent decline in the ex situ populations compared with the in situ population, with a notable loss of less frequent alleles. The differentiation among the farmer subpopulations increased when the multiplication took place outside of the adaptation area. Although 120 plants were multiplied in each ex situ cycle, a bottleneck effect was present. In addition, tests for neutrality detected three loci that are involved in pathogen response and are potentially under selective effects. The diversity conservation and the management practices of autogamous landrace crops are discussed. Keywords Diversity management In situ versus ex situ Phaseolus vulgaris L. Landrace Microsatellite markers Population size and selection
V. Negri (&) B. Tiranti Dipartimento di Biologia Applicata, Universita` degli Studi di Perugia, Borgo XX Giugno 74, 06121 Perugia, Italy e-mail:
[email protected]
Introduction While many studies have focused on defining various aspects of conserving genetic resources (Marshall and Brown 1975; Guarino et al. 1995) much less work has been done on evaluating the effectiveness (i.e. the ability in maintaining total genetic diversity) of different conservation strategies (Li et al. 2002). It is widely acknowledged that both in situ and ex situ conservation have valuable and complementary roles to play. In situ (on-farm) conservation allows the population to evolve in relationship to the farmer’s needs and to sociocultural, economic and environmental changes (Brush 2000). In ex situ conservation, genetic resources are maintained outside their natural habitats (Convention on Biological Diversity, Art. 2, 1992) in facilities such as seed, in vitro and field genebanks, or in botanical gardens. This strategy allows for the study, distribution and use of plant genetic resources, but it is designed to maintain the genetic material in the state in which it was collected (Brush 2000). In ex situ conservation procedures, accessions often need to be multiplied before being stored in germplasm banks and periodically regenerated in order to increase the amount of seed and rejuvenate it. During multiplication or regeneration, a certain level of genetic erosion can be observed and the genetic integrity and structure of an accession can be modified, due to occasional crossing with other accessions, genetic drift and/or selective effects related to the multiplication environment (Spagnoletti Zeuli et al. 1995; Parzies et al. 2000; Bo¨rner et al. 2000; Chebotar et al. 2003; Rice et al. 2006). Models have been developed which establish guidelines and estimate the minimum and/or optimal sample size required to conserve the maximum allelic, genotypic and
123
986
Genetica (2010) 138:985–998
phenotypic diversity during the regeneration of germplasm (Crossa 1989; Crossa and Vencovsky 1994; Spagnoletti Zeuli et al. 1995; Vencovsky and Crossa 1999; Parzies et al. 2000; Lawrence 2002). However, few studies have documented the effectiveness or have compared the effects of in situ and ex situ conservation on crop and wild plant populations. Soleri and Smith (1995), working with maize varieties that evolved under arid conditions in the Hopi area of Arizona, reported marked differences between materials that had been maintained in situ and those that had been kept in a gene bank. These results were attributed to genetic drift, selection and gene flow. Comparing agronomic, morphological and isozyme variation between genebank conserved and farmer managed populations of the same farmers’ varieties of rice in Vietnam, Tin et al. (2001) reported that phenotypic differences were always observed while changes in the isozyme patterns were occasionally observed. Using Simple Sequence Repeat (SSR) markers, Gomez et al. (2005) found that genetic diversity decreased when bean landraces were maintained ex situ rather than in situ. In contrast, using the same markers, Rice et al. (2006) found that the genetic diversity of Jala maize populations maintained in situ and ex situ was substantially equal. Li et al. (2002) used Random Amplified Polymorphism DNA markers to check the effectiveness of ex situ and in situ conservation in Vatica guangxiensis, an endemic species in southwestern China. It was estimated that the ex situ populations contained about 88% of the total genetic diversity present in the remaining natural populations. Li et al. (2005) also evaluated the efficacy of in situ and ex situ conservation of Parashorea chinensis in southwestern China and found that in situ conservation maintains genetic diversity more efficiently than the ex situ strategy. In addition, in the above mentioned studies little information is given about the number of plants multiplied ex situ or the number of cycles and techniques adopted in the multiplications; therefore, little can be deduced about germplasm management. Among genetic resources, landraces (i.e. traditional crop varieties, also called ‘farmer varieties’, ‘local varieties’ or ‘primitive varieties’) have key importance and, since they are highly threatened, they deserve to be safeguarded with the highest priority (Negri et al. 2009). A landrace of an autogamous crop, Phaseolus vulgaris L., was taken as a case study in order to: 1.
2.
Compare the genetic outcome of ex situ multiplications in two different environments (one similar to and the other very different from the adaptation area) with in situ multiplication. Provide guidelines for germplasm management, especially concerning the number of individuals that should
123
3.
4.
be multiplied/regenerated in order to avoid genetic erosion. Gain some insights about the roles that selection and population size play in changing patterns of genetic variation and differentiation in multiplication/regeneration cycles. Obtain specific information about the possibility of maintaining the diversity of the P. vulgaris L. landrace under study outside the adaptation area. The on farm conservation of this landrace has little hope of success due to the disadvantaged area where it is grown, the advanced age of the farmers and the difficult management of the crop (see below).
Molecular markers that reveal polymorphisms at the DNA level are a powerful tool for assessing genetic diversity and structure in beans (Beebe et al. 2000, 2001; Masi et al. 2003; Papa and Gepts 2003; Sicard et al. 2005; Zizumbo-Villarreal et al. 2005). Among the various molecular markers available, SSRs were chosen because they are hypervariable and highly reproducible (Powell et al. 1996). Their presence, abundance and variation have been characterized in the bean genome (Yu et al. 2000; Blair et al. 2003).
Materials and methods The landrace in situ and the farmers The landrace, called ‘‘A pisello’’, is grown by a few elderly farmers in the area of Colle di Tora (Rieti Province, central Italy) located in the Apennine mountains on the steep slopes of Lake Turano. It is genetically and morphologically distinct from other landraces grown in central Italy (Negri and Tosti 2002) and is characterized by an aggressive climbing ability, a late flowering (August) and harvesting date (end of October), a white flower and a small whitish seed. Because of its very particular thermal, humidity and edaphic requirements it cannot be grown with success in other areas (Tiranti 2005). In particular the farmers say that, when it is cultivated at lower than usual altitudes, it is highly infested by bruchids (Bruchidae). It has a rich local niche market due to its smooth, doughy taste and very limited production. While the market for ‘A pisello’ could be expanded, the landrace appears to be in danger of extinction because it is only cultivated by five elderly farmers (designated PL, GF, DL, PM and PG, average age is 64 years old) who are not interested in the potential business. Given the potential for integrating landrace conservation with profitable economic activities in disadvantaged areas that exists in Italy, ‘A pisello’ was chosen as a case study to model conservation initiatives at the national level (see Acknowledgments).
Genetica (2010) 138:985–998
The landrace is cultivated in small, separate fields for a total hectarage of 2–3 ha. The fields are located at different altitudes, ranging from 748 (PG) to 890–910 m asl (PL and GF neighbouring fields, respectively) with an intermediate altitude for PM and DL (both 816 m asl, also neighbouring) and have different exposures (PL and GF have a northern exposure while all the others have an eastern exposure). Each farmer inherited this landrace from his parents and has always multiplied his own seed. Another landrace of P. vulgaris and one landrace of P. coccineus are also grown in the area (the two bean species are cross fertile when P. vulgaris is the mother plant, Escalante et al. 1994). The farmers do not apply pesticides, but use mechanical tools and chemical fertilisers in addition to manure (on average 30 N, 18 P2O5, 30 K2O, 4 MgO kg ha-1, respectively, are usually distributed in a postemergence application). Type and quantity of fertilisers applied may vary from one farm to another in different years (i.e. manure may not be applied in a certain year; chemical nitrogen may be applied as NH4NO3 or (NH2)2CO). There are no other differences in the way the farmers manage this crop. The average number of plants multiplied annually by each farmer is equal to 700, 800, 1,000, 1,300, and 7,000 plants for the farmers GF, DL, PG, PM and PL, respectively. At harvest, each farmer selects the seeds that will be used for the next season. One farmer (DL) selects for a small seed size, while the others select for a large seed size. The intensity of selective pressure applied is variable. For example, PM applies a rigorous selection and chooses both the legumes and seeds that most conform to his own ideotype, while PG only applies a gross selection. In addition, selection intensity applied by farmers may vary from one year to another depending on the amount of seed produced (i.e. it is likely relaxed when the yield is low). Farmers do not usually exchange seeds. According to the farmers, this landrace originated from a seed sample brought to Colle di Tora at the end of the 19th century by a woman who returned from South America. Phaseolin analysis confirmed that the landrace belongs to the Andean gene pool (phaseoline ‘‘C’’ type; unpublished data). Seed collection and multiplication Molecular markers showed that this landrace is a structured population (Tiranti and Negri 2007). Consequently, seeds were collected from the respectively storage facility of four farmers (PL, GF, DL, and PM) in the autumn of 1999 (CT1999, initial population). The subpopulation PG was not included in this study because of possible extensive introgression from neighbouring different varieties and lack of a rigorous selection by the farmer (Tiranti and Negri 2007).
987
The initial population was subsequently multiplied in the adaptation area (Colle di Tora) and in two different environments, Perugia (central Italy, 320 m asl) and Potenza (southern Italy, 620 m asl; Fig. 1). Each farmer seed lot was kept distinct during multiplication. In the adaptation area, the initial population was multiplied by each farmer for 2 years (2000–2001) under the traditional agronomic practices and with the usual average population size. A sample of each farmer’s harvest was collected from their seed storage facility in the autumn of 2001 (population CT2001, in situ multiplication). In Perugia the landrace was multiplied for 2 years (2000–2001, population PG2001), while in Potenza the landrace was only multiplied for 1 year (population PZ2000). For both ex situ multiplications, 30 plants from each of the four farmer samples (total 120 plants) were grown each year. Single plants were grown in jiffy pots and then transplanted 90 9 90 cm apart in the experimental fields in a randomised block design with three replicates of 40 plants each (10 plants for each farmer subpopulation per plot). Crop management was similar to that of the farmers including fertilization (60 N, 36 P2O5, 60 K2O, 8 MgO kg ha-1, considered that manure was not applied). Outcrossing with different landraces/cultivars possibly present in neighbouring farms was prevented by the spatial isolation of the experimental field (500 m), although this practice is unusual for common bean multiplication in germplasm banks. Eventual intercrossing among farmer subpopulations was not controlled. A bulk sample of each farmer subpopulation was harvested each year by pooling the harvests from the three replicates. No selection was applied by us in preparing the following year seedlings. Data relative to each farmer subpopulation in the different multiplication years will be referred to by the farmer acronym and the location and harvest year written as a subscript (e.g. PLCT1999). Genomic DNA isolation The examined sample size was 80 plants for each population (CT1999, CT2001, PG2001 and PZ2000) with the exception of CT1999 where 78 plants were sampled. DNA was extracted from 20 plants from each farmer subpopulation in each population (with the exception of the PMCT1999 where 18 plants were sampled), 318 individuals were studied in total. DNA extraction was performed according to a slightly modified procedure of Doyle and Doyle (1990). The quality of the DNA was checked by electrophoresis in a 1% agarose gel containing ethidium bromide (0.5 lg/ml) in 19 TBE (89 mM Tris–HCl, 89 mM boric acid, plus 2 mM EDTA). DNA was quantified using a DU650 Spectrophotometer (Beckman).
123
988
Genetica (2010) 138:985–998
Fig. 1 The populations and subpopulations under study and their relationships. CT (Colle di Tora), PG (Perugia), PZ (Potenza) are the multiplication sites, further explanation in the text
DNA fingerprinting Table 1 reports the details of the twenty-six SSR primers used. Twenty-two SSRs were chosen from those developed from gene-coding sequences, of which nineteen were assigned to seven different linkage groups (Yu et al. 2000; Masi et al. 2003). Four additional non-coding microsatellites were chosen from those developed by Blair et al. (2003), three of which were assigned to different linkage groups. Overall, ten of the eleven P. vulgaris linkage groups were examined. Polymerase chain reaction (PCR) amplifications were performed in 20 ll of reaction volume with 25 ng of genomic DNA, 10 lM each of the fluorescence-labelled forward primer and of the reverse primer, 1.25 mM of each dNTP and 5 U of the Taq DNA polymerase (Invitrogen). For the first set of 22 primers, the PCR conditions consisted of an initial denaturation step at 95°C for 2 min, followed by 30 cycles at 94°C for 2 s (denaturation), 47–52°C (depending on the primer) for 10 s (annealing), 72°C for 2 min (elongation) and a final elongation step at 72°C for 30 min. For the second set of four primers, the PCR conditions were the same except for the denaturation and annealing which were carried out for 1 min. The 26 SSR loci were fluorescently labelled in different colours (FAM, NED, VIC). DNA amplification was performed by using a GeneAmp 9700 thermocycler (Applied Biosystems).
123
For electrophoresis, 2 ll of the PCR product were mixed with 10 ll of a 75:1 solution of formamide and GENESCAN-500 (ROX) size standard (Applied Biosystems) solution. Microsatellite polymorphisms were electrophoresed and analyzed with a fluorescent detection method using an ABI 307 sequencer (Perkin-Elmer). Visualization and sizing of DNA fragments were performed using the GENESCAN 3.1 software (Applied Biosystems). Data analysis Genetic diversity among and within populations Data sets relative to each population and the total were obtained by coding each detected allele in base pairs (bp). For each population (CT1999, CT2001, PG2001 and PZ2000) and for each subpopulation (PL, GF, DL, and PM) within a population, the following were calculated: total number of alleles (NA), number of alleles with frequency greater than 0.05 (common alleles, Na [ 0.05), number of alleles with a frequency lower than 0.05 (rare alleles, Na \ 0.05) and number of private alleles (No). The observed heterozygosity (Hobs) and the Nei’s unbiased estimate of gene diversity (He, expected heterozygosis, Nei 1987) were also separately worked out for each population and for each subpopulation as implemented in the GenAlEx software (Peakall and Smouse 2005). To test the significance of the pairwise
Genetica (2010) 138:985–998
989
Table 1 Genebank entry, locus name, linkage group, allele size range and number of alleles detected throughout the samples examined relative to the 26 microsatellites used Genbank entry
Locus name (symbol)
Linkage Allele Group size range
Allele Genbank number entry
Locus name (symbol)
Linkage Allele group size range
X79722a
Sn-glycerol-3-phosphate acyltransferase (PVPLB)
B2
149
1
M75856a Phathogenesis-related protein 3 (PHVPVPR3A)
B11
X04660a
Phytohemagglutinin pseudogene (PVPDLEC1)
X13329a
5’ flanking sequence of glutamine synthetase beta subunit gene (PVGLNB)
X59469a
Chalcone synthase
B2
165–167 2
150–158 3
J01263a
Beta-phaseolin (PHVBCSP)
B7
155–171 2
B4
201–220 3
X52626a,c Alpha phaseolin (PVAPHASE)
B7
170–176 2
B1
138–140 3
M18094a
Hydroxyproline richglucoprotein (PHVHRGPB)
179
X60000a
Small subunit of ribulose 1,5 B4 biphosphate carbosylase/oxygenase (PVBLB)
128–144 5
M13968a
Chitinase (PHVCHM)
178–182 2
X74919a
Endochitinase (PVGEC9)
B5
132–182 4
U10419a
Nitrite reductase (PVU10419)
203
1
X80051a
B9
192–208 8
X63525a
Lipoxygenase
305
1
U77935a
NADP-dependent malic enzyme (PVME1G) DnaJ-like protein
K03288a
Phytohemagglutinin (PHA-E)
B4
U28645a
Embryo-specific acidic transcriptional activator Phytohemagglutinin (PHA-L)
K03289a
1
X02980a,c Alpha-phaseolin (PVVPHASAR)
B7
192
1
125–128 2
M68913a,c Arcelina (PHVARC1A)
B4
193
1
115
1
Bmd–12b
B6
163–166 3
B4
144–154 4
Bmd-41b
B10
232–247 2
B2
85–95
2
Bng225/R common bean genomic clone Bng91/R common bean genomic clone
X57022a
Small subunit of ribulose 1,5biphosfatecarboxylase/oxigenase (PVSS15BCO)
B4
150–155 2
Bmd-43b
Bng112/R common bean genomic clone
X04001a
Glutamine synthetase (PVGSR1)
B1
164–165 2
Bmd-44b
Bng125/R common bean genomic clone
a
Allele number
176
B8
1
135–136 2
Yu et al. (2000)
b
Blair et al. (2003)
c
Linkage group assigned later (Masi et al. 2003)
genetic diversity estimates, a Wilcoxon Two Sample test was performed using the Institute of Phonetic Sciences (IFA) on line service (www.fon.hum.uva.nl/Service/Statistics/). This is a non parametric statistical method used to compare independent groups of sampled data; it tests the hypothesis that the different samples in the comparison were drawn from the same distribution or from distributions with the same median. Like most non parametric tests, it is performed on ranked data, so the measurement observations are converted to their ranks in the overall data set: the smallest value is given a rank of 1, the next smallest is given a rank of 2, and so on. Genetic differentiation among and within populations Differentiation was initially studied on the entire data set considering differences in allele frequencies and FST
values. To determine if significant differences in allele frequencies at each locus exist among populations and among subpopulations within a population, the exact test described by Raymond and Rousset (1995) was applied using TFPGA 1.3 software (Miller 1997). This procedure analyzes each locus in a data set to determine if differences in allele frequencies exist. In addition, the Fisher’s Combined Probability test (Sokal and Rohlf 1995) was used as a global test over loci to determine the overall significance. FST values were then computed for each pairwise comparison between the four distinct populations (CT1999 vs. CT2001; CT1999 vs. PG2001; CT1999 vs. PZ2000; CT2001 vs. PG2001; CT2001 vs. PZ2000 and PZ2000 vs. PG2001) by using the MSA software (Dieringer and Schlo¨tterer 2002). The significance of the pairwise comparisons was tested by the Wilcoxon Two Sample test using the already mentioned on-line service (www.fon.hum.uva.nl/Service/Statistics/).
123
990
FST values for the among subpopulations within each single population differentiation, were then computed using the above mentioned software. P-values were obtained by permutating genotypes 10 000 times. Spatial differentiation Spatial differentiation of populations was studied through an Analysis of Molecular Variance (AMOVA) approach with the three multiplications (CT2001, PG2001 and PZ2000) as the group level and the 12 subpopulations as the population level. AMOVA was carried out as implemented in the GenAlEx software (Peakall and Smouse 2005). The significance of sources of variation was tested on the basis of 999 permutations. Estimating temporal changes in effective population size The effective size of a population (Ne) influences the amount of genetic drift and, hence, the rate of genetic diversity loss, the rate of fixation of unfavourable alleles and the efficiency of the selection at maintaining beneficial alleles. It is a critical factor in regenerating germplasm accessions (see for example http://www.ecpgr.cgiar.org/Workgroups/grain_ legumes/grain_legumes.htm) and can be taken as a measure of genetic representativeness of a sample. Monitoring Ne is then important for estimating the population evolution under different conditions and in different environments. The investigation was limited to the population level because monitoring population Ne was of major concerns in this study since a landrace is usually multiplied/regenerated as a single accession in genebanks and pollen flow, that blurs population structure, could not be excluded in the locations where the landrace was multiplied. There is considerable genetic and environmental variation for outcrossing in P. vulgaris L. and the outcrossing level can reach 67% (Wells et al. 1988). In addition, in Colle di Tora subpopulations are grown near each other or, sometimes, near other landraces (Tiranti and Negri 2007) and in the ex situ multiplications subpopulations were grown in the same experimental field which could have facilitated intercrossing. Temporal changes in effective size (Ne) were estimated according to Waples (1989). For a selfing species the expression becomes (Goldringer et al. 2001): Ne ¼ t=2½Fc 1=Si 1=Sf ; where t is the number of generations between the populations studied, Fc is the estimator of the standardised variance of allelic frequencies, Si and Sf are the individuals sampled at generation i and f, respectively. Confidence intervals (CI) for Ne were estimated as suggested by Waples (1989) who noted that the appropriate formula is that for the CI of a variance.
123
Genetica (2010) 138:985–998
The estimated Ne was then compared with the demographic effective size worked out as in Caballero (1994) using census values corrected for the mean inbreeding values. One hundred and twenty, 120 and 9,800 (estimate) individuals bred in PG2001, PZ2000 and CT2001, respectively. Their mean inbreeding values, which were worked out as Wright Fis (Weir and Cockeram 1984) by using the GenAlEx software (Peakall and Smouse 2005), were equal to 0.999, 0.960 and 0.843, respectively. Testing the presence of selective effects To address aim 3 of the paper, it was important to check which loci could be subject to selective effects when multiplication/regeneration of a bean landrace is carried out in situ or ex situ and in relationship to the initial population. In order to take into account the existence of a population structure, pairwise selective tests were initially performed for each possible subpopulation combination. If the tests did not give any positive result, they were then performed for each possible population combination. The Beaumont and Nichols (1996) approach, implemented in Fdist2 software, was the first test applied. It uses computer simulation to detect loci where the genetic diversity within (heterozygosity) and between populations (FST) does not conform to the prediction of a simple infinite or finite-island model obtained by coalescent simulations. Loci with unusually high FST values conditional on heterozygosity are considered to be potentially under divergent selection. The method calculates heterozygosity and the Cockerham and Weir (1993) FST for each locus in the sample. Coalescent simulations based on an infinite allele mutational model are used to generate data sets with the mean FST similar to the empirical distribution. The observed FST estimates are compared with the simulation output (the simulated distribution of FST) to identify potential outliers. The distribution of FST as a function of heterozygosity is characterized by estimating the 0.5 (1 Pvalue), median and the 0.5 (1 ? Pvalue) quantiles of the distribution. In a first step, the neutral expectation was based on the overall mean value of FST calculated from all the markers and setting a probability value equal to 0.95. Markers with FST values outside the 0.95 limits were then removed and a new analysis was performed with a recalculated mean value of FST. This procedure was repeated until no FST marker value was found outside the limits. Markers with FST values outside the 0.95 limits in the last analysis were considered as outlier loci; 20 000 runs were used at each step. The DetSel (Vitalis et al. 2001) test was used to detect outlier loci in spatial comparisons between subpopulations. The approach relies on a model in which a common ancestor population splits into two populations that
Genetica (2010) 138:985–998
subsequently diverge only by random drift (Vitalis et al. 2001; Vitalis et al. 2003). Population-specific parameters as functions of probabilities of identity for pairs of genes taken within or between populations are defined. For each population i (i = 1 or 2), the divergence parameter (Fi) is simply related to the ratio of divergence time t over the population size Ni, that is Fi * 1 - exp (-t/Ni). Single locus estimates of these parameters (conditioned on the number of alleles in the two populations) can be calculated for each locus by using the parameters of the model: mutation rate (l), ancestral population size before the bottleneck (Ne), ancestral population size during the bottleneck (N0) and the number of generations during the bottleneck (s0). A joint distribution of F1 and F2 under neutral expectations can be generated by using coalescent simulations and every locus that falls outside the resulting confidence envelope is considered to be potentially under selection. A null distribution was obtained based on a range of parameters (l varying from 10-3 to 10-5, Ne from 103 to 104) that generated a number of alleles similar to that in the observed sample (N0 = 30 and t = s0 = 2 in our study). This procedure provided a maximum number of realistic scenarios that gave a more robust null confidence envelope. 50,000 runs were performed for each possible pair of subpopulations. The significance level was set at 0.95. While both Fdist and DetSel methods are based on the principle that genetic differentiation between samples is expected to be higher for loci under divergent selection than for the rest of the genome, they rely on different assumptions and parameters. Therefore, they can detect different loci putatively under selection. Only loci detected as outliers in multiple comparisons are most probably under selective effects, while loci with atypical behaviour in a single comparison (i.e. which do not exhibit parallel trends in several comparisons) should not be considered to be under selection, but rather as outliers due to chance alone (Campbell and Bernatchez 2004; Storz 2005; Vasema¨gi et al. 2005). Loci detected as outliers in multiple comparisons that involve the same location are putatively under selection in that location and reveal local selective effects. Outlier SSRs from spanning coding sequences could be considered candidates for genes under selection. Nonetheless, this should be validated since there is a risk of false positives in all methods that try to detect a signature of selection (Storz 2005). Since independent confirmatory evidence can come from co-localisation with loci definitely under selection, the known map localization of the outlier loci was compared with that of genes and quantitative trait loci (QTLs; Blair et al. 2003, 2006; Koinange et al. 1996).
991
Results Genetic diversity among and within populations Eighteen loci out of the 26 tested were polymorphic and a total of 61 alleles (ranging from 85 to 305 bp size) were detected overall (Table 1). A total of 44, 49, 33 and 37 alleles were found in CT1999, CT2001, PG2001 and PZ2000, respectively (Table 2) of which 12, 12, 2 and 3, respectively, were rare alleles. Each population was characterized by exclusive (private) alleles. Six, 11, 2 and 3 exclusive alleles were found in CT1999, CT2001, PG2001 and PZ2000, respectively (Table 2). Exclusive alleles that were found in the multiplied populations should be considered new alleles that arose by mutation or introgressed from neighbouring bean crops. One private allele in CT1999 and one in CT2001 were detected with frequencies higher than 0.05 (e.g. the 144 bp allele of locus K03289 in CT1999, frequency of 0.069, and the 220 bp allele of locus X04660 in CT2001, frequency of 0.063). Some private alleles characterized a certain subpopulation in each population. Of the 13 polymorphic loci detected in CT1999, just one, with a high initial frequency (0.94), became fixed in CT2001, while 7 (54%) and 5 (38%) loci became fixed in PG2001 and in PZ2000, respectively (not shown). Both PG2001 and PZ2000 had fewer alleles (33 and 37) with respect to the initial population (44). Considering the new alleles that arose, this corresponds to a proportional reduction equal to -30 and -23%, respectively, (Table 2) which was mostly due to the loss of rare alleles (-85 and 77%, respectively). Consequently, the proportion of common (Na [ 0.05) alleles, which was 71% in the initial population CT1999, increased markedly in the populations multiplied outside the adaptation area (94 and 92%, PG2001 and in PZ2000, respectively). The proportions of common, rare and private alleles in CT2001 were similar to those detected in the initial population. The observed heterozygosity was generally low, and in agreement with the reproductive system of the species studied (Table 2), both for populations and for subpopulations within a population. The population multiplied in situ (CT2001) had the highest observed (0.020) and expected (0.130) heterozygosity values that were twenty and 1.8–1.4 times higher than those recorded in the populations multiplied ex situ (Table 2). CT2001 He values were always significantly different from those of the other populations (Table 3). The difference in heterozygosity between CT2001 and CT1999 was mostly due to a greater number of high frequency alleles in the PLCT2001 and DLCT2001 subpopulations. It is
123
992
Genetica (2010) 138:985–998
Table 2 Number of alleles (NA), number of alleles with frequency greater than 0.05 (Na [ 0.05), number of private alleles (No), observed heterozygosity (Hobs) and gene diversity (He) detected in each subpopulation and population PL
GF
DL
PM
POP
Table 3 Above the diagonal: significance of He values in pairwise comparisons. Below the diagonal: FST values and their significance in pairwise comparisons between populations. On the diagonal (in italics): FST values among subpopulations within each single population He
CT1999
CT2001
PG2001
PZ2000
FST
In situ
CT1999
0.315**
**
NS
NS
NA
32
27
36
33
44
CT2001
0.082 NS
0.286**
**
**
Na [ 0.05
30
27
35
33
32
PG2001
0.253 **
0.144 **
0.712**
NS
No
1
0
2
2
6a
PZ2000
0.125 NS
0.080 NS
0.200 **
0.441**
Hobs
0.002
0
0.004
0.006
0.003
He
0.04
0.02
0.06
0.04
0.056
NA
39
29
43
34
49
Na [ 0.05
35
29
43
33
37
CT1999
* Significant at P B 0.05; ** significant at P B 0.01
CT2001
No
0
0
0
2
2
Hobs
0
0
0
0.006
0.001
He
0.00
0.01
0.03
0.05
0.071
Fisher’s tests carried out over the entire data set and the four population hierarchical data sets were always significant at P B 0.0001 (not shown). The pairwise estimates of genetic differentiation between populations were only significant for the comparisons involving PG2001 (Table 3, FST values below the diagonal). When single populations were analysed, the genetic differentiation among subpopulations was always significant, but differentiation was more evident in the ex situ populations than in the in situ ones (Table 3, FST values on the diagonal).
NA
29
31
35
29
37
Spatial differentiation
Na [ 0.05
29
31
35
28
34
No
0
0
2
0
3c
Hobs
0
0.002
0.004
0
0.001
He
0.05
0.03
0.10
0.05
0.092
b
No
2
0
3
1
Hobs
0.01
0.008
0.006
0.056
0.020
He
0.11
0.03
0.18
0.07
0.130
NA
26
26
27
30
33
Na [ 0.05
26
26
27
30
31
11
Ex situ PG2001
PZ2000
The number of rare alleles (Na \ 0.05) can be calculated by subtracting the Na [ 0.05 value from the NA value a
One of the exclusive alleles of CT1999 was found in 3 subpopulations (PL, DL and PM)
b
Five of the exclusive alleles of CT2001, were found in PL and DL (4 alleles) and in PM and DL (1 allele) subpopulations c One of the exclusive alleles of PZ2000 was amplified in both DL and PM subpopulations
also important to recall that 18 plants were sampled in PMCT1999. Differentiation among and within populations Overall populations were significantly differentiated at 16 of the 18 polymorphic loci, the exceptions were the U77935 and X57022 loci (not shown). Subpopulations within populations were significantly differentiated at 9 and 13 loci in CT1999, and CT2001, respectively, but only at 5 and 8 loci in PG2001 and PZ2000, respectively (not shown). Two loci (X74919 and X80051) were significantly differentiated in all the tests carried out. The overall loci
123
AMOVA showed that among population, subpopulation and individual differences are significant (Phi values equal to 0.116, 0.407 and 0.476, respectively, P B 0.001). Most of the variation is due to differences among individuals (52%) and among subpopulations (36%), while a rather limited amount of genetic variation (12%) is due to populations. The high level of differentiation among subpopulations appears to be due to differential changes in allele frequencies in different subpopulations. In the PG2001 population, for example, the 174 bp allele (locus X74919), reached fixation in PLPG2001, but had a frequency of 0.25 in DLPG2001. Estimates worked out without including the new alleles gave similar results. This shows that the new alleles that arose in the populations multiplied in situ and ex situ did not influence differentiation. Temporal evolution of effective population size The effective size estimate between the initial population (CT1999) and the populations multiplied outside the adaptation area (PZ2000 and PG2001) was very low (Ne = 11.4 and 15.6, respectively) and was significantly different from the Ne between CT1999 and CT2001 (117.9, Table 4) which
Genetica (2010) 138:985–998
993
Table 4 Effective population size (Ne) computed on the basis of temporal variation of allelic frequencies from those of initial population at 18 SSR marker loci, its confidence intervals, demographic effect size (N) and Ne/N ratio t
n
PZ2000
1
21
PG2001
2
CT2001
2
Ne
CI
11.4
4.9–24.1
21
15.6
0.4–22.9
30
117.9
28.5–infinity
N
Ne/N 61.2
0.19
60.0
0.26
5,317.4
0.02
t is the number of generations, n is the number of independent alleles used to work out the estimator of allelic frequency variance
was ten times higher. This is clearly due to the fact that only 120 individuals were multiplied ex situ, while thousands of individuals were maintained in situ. All of the Ne estimates were lower than the expected demographic values (N) and the Ne/N value ranged from 0.26 to 0.02. Signals of selective effects The Beaumont and Nichols’ tests for neutrality (1996) applied to the 18 polymorphic markers did not reveal any locus putatively under selective effects at the subpopulation level. Selective effects at the population level were then evaluated. Two loci were shown to be outliers: the M13968 locus in CT1999 vs. PZ2000 and CT1999 vs. PG2001 and the X59469 locus in all the comparisons involving PG2001 (CT1999 vs. PG2001, PG2001 vs. CT2001, PG2001 vs. PZ2000; Fig. 2). It should be noted that no locus was found to be putatively under selective effects in the CT1999 vs. CT2001 comparison. The DetSel spatial pairwise comparisons at the subpopulation level showed that locus X74919 was putatively under selective effects in the PMCT2001 vs. PMPG2001, DLCT2001 vs.
DLPZ2000 and PMPG2001 vs. PMPZ2000, comparisons (Fig. 3). To avoid bias due to the population structure, no comparisons at the population level were carried out. Other loci were detected as outliers by both tests, but, since they only occurred in one pairwise comparison, they were not considered to be reliable outliers, but rather due to chance alone (Campbell and Bernatchez 2004; Storz 2005; Vasema¨gi et al. 2005). Two of the outlier SSRs detected in this study are colocalized with important QTLs mapped by Blair et al. (2006). The X59469 locus (linkage group B2) and the X74919 locus (linkage group B5) are flanking markers to a ‘seed weight’ (sw), and a ‘days to flowering’ (df 5.1) QTL, respectively.
Discussion At each locus different alleles and different frequencies characterized the populations studied. Large differences were observed especially between the population multiplied in situ and the populations multiplied ex situ (to be recalled that the same number of plants was analysed in each population). The former showed the same level of allelic diversity that was observed in the initial population, while both ex situ populations showed a lower allelic diversity. Indeed, ex situ, even though subpopulations were kept distinct from one another and many plants per population (120) were transplanted in each multiplication cycle and in each location, a loss of rare alleles, allele fixation and an increased homozygosity were observed, due to a reduction in population size. In situ, a loss of alleles was certainly prevented by maintaining the usual population dimension; in addition, a gain of alleles due to introgression from
Fig. 2 Distribution of FST values as a function of heterozygosity (HS) in the comparisons CT1999 vs. PG2001 (a), CT1999 vs. PZ2000 (b), PG2001 vs. CT2001 (c) and PG2001 vs. PZ2000 (d). Each square represents a SSR marker and the outer lines represent the 95% confidence envelope. The M13968 locus is detected as an outlier in all the comparisons including CT1999 (a, b); the X59469 locus is detected as an outlier in all the comparisons including PG2001 (a, c, d)
123
994
Genetica (2010) 138:985–998
Fig. 3 Spatial pairwise comparison between subpopulations per- c formed with DetSel: plot of PMCT2001 vs. PMPG2001 (a), DLCT2001 vs. DLPZ2000 (b), PMPG2001 vs. PMPZ2000 (c). Each square represents a SSR marker and the lines represent the 95% confidence envelope. The X74919 locus can be seen as being potentially under selection
neighbouring populations can not be excluded (outcrossing was only controlled ex situ). Ex situ, however, gene diversity was substantially maintained by multiplying the initial population for one or 2 years under the conditions set out in this study, but this was only due to the increased frequency of the most common alleles in these multiplications. Loss of low frequency alleles has been reported in ex situ collections compared to in situ conservation for both autogamous and allogamous populations (Gomez et al. 2005; Parzies et al. 2000; Li et al. 2005; Rice et al. 2006). Gene diversity was reduced in ex situ maintained barley and bean landraces (Parzies et al. 2000; Gomez et al. 2005). However, it should be noted that the results of the present study can not be directly compared with those obtained elsewhere because of the different type/number of markers used, different numbers of regeneration cycles and/or different propagation system of the species under study. The above reported findings may be due to unsatisfactory sampling procedures used when the ex situ repositories were renewed or to on-farm population changes that occurred more rapidly than generally supposed. In fact, surprisingly rapid changes in agronomic performances have been reported in rice landraces (Tin et al. 2001) and in experimental populations of wheat in response to climatic conditions (Goldringer et al. 2001, 2006; Rhone´ et al. 2008). The genetic structure of the ‘‘A pisello’’ landrace was also profoundly modified by ex situ multiplications which increased subpopulation differentiation, as a consequence of allele loss and fixation. Very strong effects were observed after two multiplication cycles in Perugia. The effect on subpopulation differentiation was less evident after only one cycle was carried out in an environment similar to that of adaptation (Potenza). A clearer clue to the role played by different environments could have been obtained if it had been possible to have two multiplication cycles in Potenza. As commonly found (Frankham 1995; Goldringer et al. 2001; Raquin et al. 2008), the estimated effective size was much smaller than the demographic effective size based on the true number of plants cultivated in each generation. Indeed the Ne values based on allelic frequency variation between populations were quite different considering the ex situ or the in situ populations. The Ne value for the in situ population was similar to those observed in experimental populations of another autogamous species (i.e. wheat) that were multiplied ex situ for 10–15 generations sowing a similar number of plants as sown by farmers for
123
this landrace (Goldringer et al. 2001; Raquin et al. 2008). In contrast, Ne values after one or two generations of ex situ multiplication were ten times lower than the Ne value
Genetica (2010) 138:985–998
after in situ multiplication. This is clearly due to a genetic drift effect caused by multiplying fewer individuals ex situ than used by farmers in situ (also allowing for year to year variation in the number of plants in situ). For autogamous crop regeneration, Vencovsky and Crossa (1999) showed that the effective size decreases markedly when the number of parents that actually produce seeds decreases and when random sampling of the seed is applied. In this study, which was mainly aimed at controlling the changes of the genetic make up at the population level, we avoided random sampling of the seed at the population level by taking the same number of plants from each subpopulation at each multiplication cycle, but we did not control the contribution to the next generation of single individuals within subpopulation (in other words: the bulk of subpopulation seed was not built up by taking the same amount of seed from each plant). Consequently, the low Ne values detected for ex situ multiplications were probably due, beside a bottleneck effect, to an unequal contribution of single genotypes to the next generation as a consequence of different levels of adaptation to the biotic and abiotic stresses in the new multiplication environment (see discussion on selective effects below). The Ne/N ratio values were in the range measured in natural animal and plant populations (Frankham 1995). This ratio depends on fluctuations in population size and variance in family size. The Ne/N ratio of in situ multiplication was tenfold lower than for ex situ multiplications due to a lower Ne compared to N. In situ both farmer and environmental selection does occur (see introduction and Tiranti and Negri 2007) so that each new plantation is likely to derive from a low number of plants. Ex situ we did not apply any conscious selection, only environmental selection would eventually operate, the number of individuals multiplied was the same year by year (i.e. demographic fluctuations did not occur), and the contribution of each subpopulation to the next generation was equalized. When the two samples from the adaptation area (i.e. Colle di Tora) were compared, the Fdist test did not show any locus as under selective effects, while they were found when the initial population and the populations multiplied ex situ were compared. Other selective effects were revealed when the Detsel test was applied to subpopulations multiplied in different environments. At least three loci (X59469, M13968 and X74919), among those that, according to the results of the exact test, differentiate multiplications, are probably under selective effects because they were detected as outliers in multiple comparisons involving different multiplication environments (Campbell and Bernatchez 2004; Storz 2005; Vasema¨gi et al. 2005).
995
The X59469 SSR is present in the pathogenesis-related chalcone synthase protein gene that is involved in defence against pathogen attack and other important metabolic processes such as symbiosis, biosynthetic pathway of pigments and exposure to ultraviolet light (Peters et al. 1986). M13968 is present in a chitinase encoding gene. Chitinase enzymes are induced in response to pathogen attacks and abiotic stress and are some of the major defence enzymes used by the plants against fungal and bacterial invaders (Serrano et al. 2007). The bean chitinase protein (CHI) degrades the structural fungal cell wall polysaccharides and acts synergistically with hydrolytic enzyme glucanase against fungi in vitro (Margis-Pinheiro et al. 1994; Lima et al. 2002). The X74919 locus is also present in a pathogenesis-related protein gene (an endochitinase). The non neutral behaviour of these SSRs could then be related to different selection pressures exerted by pathogens in the considered multiplication areas (PG, in particular, is located at a low altitude where pests are more likely to attack). Finally, the X59469 and X74919 outlier SSRs, are colocalized with important QTLs and genes that control seed weight and flowering date. Consequently, their non-neutral behaviour may alternatively indicate selective effects at other loci via hitchhiking. Overall, reduction of population size, as the main factor, and selection, played a role in changing the patterns of genetic variation and differentiation with multiplication cycles outside the adaptation area.
Conclusions The extent and distribution of genetic variation of a population which is maintained in the adaptation area in situ (on-farm) and ex situ must be understood in order to devise strategies that will efficiently maintain diversity in conservation programs. Farmers generally manage their populations in situ by using the usual population size and agronomic techniques. Ex situ multiplication/regeneration procedures are generally less strict than the ones used in this study because genebank managers have to cope with many technical and financial constraints and are often forced to adopt suboptimal conditions. Altoveros and Rao (1998) reported that 24–35% of the curators grow fewer than the recommended minimum number of plants (40) when the pedigree method is used and 48–65% grow fewer than the minimum (60 plants) required when the bulk method is used. However, not even 120 plants, replication and separate harvests of subpopulations during multiplication/regeneration are sufficient conditions for maintaining the allelic diversity of an accession such as the one under study. It should be stressed
123
996
that since it is not known which alleles will prove to be the most useful in the future, it would be wise to conserve all of them. This is especially true in a rapidly changing environment such as that the current one (IPCC 2007). In contrast, the population that was multiplied in situ did not show signs of genetic erosion, remained dynamic and incorporated new alleles that arose by mutation or were introgressed from other neighbouring populations. The results of this study clearly show that in situ conservation is the most effective way to maintain the diversity present in a population such as the P. vulgaris landrace under study. This strategy also allows long term biological evolutionary processes to be maintained and local knowledge systems to change over time. The in situ-on farm conservation of landrace diversity should rely on encouraging farmers to continue to select and manage the local crop. The primary way to achieve this goal would be to increase the market value of the diverse local crop populations so as to give farmers the incentive to keep growing them. If it is not possible to maintain a landrace in situ, then ex situ storage have to be used. One or two multiplications/ regenerations, carried out under the conditions applied here, even outside of the adaptation area, will maintain gene diversity due to the most frequent alleles, at least. Genes that are involved in controlling disease resistance are likely to be under selection when multiplication is carried out outside the adaptation area. Therefore, diseases should be carefully controlled when multiplication or regeneration takes place outside the adaptation area. During bean accession regeneration X59469, M13968 and X74919 could be used to monitor changes arising from to the above-mentioned selective pressures. Having detected loci that are probably affected by selection pressure, this study has also opened up several prospects for analysing gene functionality and diversity conservation. Fitness related variation among genotypes bearing different alleles can be measured directly in ad hoc designed experiments which could give conclusive evidence about the role of these genes. The roles of X59469 and X74919 loci, which were shown to be under selective effect also in another study (Tiranti and Negri 2007), should be investigated further. Moreover, the approach used in this study can be easily extended to other multiplication case studies and marker types. Other landraces multiplied under different ecological conditions may show that the same or other loci are involved in adaptation to biotic and abiotic conditions. Acknowledgments Financial support was provided by the Italian Ministry of University and Scientific Research (PRIN project n. 9907384522_006 ‘Evaluation of ‘‘species-environment’’ systems for ‘‘in situ’’ conservation of genetic resources of cultivated species. In situ conservation of a self pollinating species. Evaluation of the
123
Genetica (2010) 138:985–998 system: landraces of Phaseolus vulgaris in a mountain area’). Thanks are due to the anonymous referees for useful suggestions and to Dr. I. Goldringer (INRA, Moulon) and Dr. P. E. Jorde (CEES, Oslo) for discussing the Ne estimating methods with us.
References Altoveros NC, Rao R (1998) Analysis of information on seed germplasm regeneration practices. In: Engels JMM, Rao RamanthaR (eds) Regeneration of seed crops and their relatives. Proceedings of a consultation meeting 4–7 December 1995. ICRISAT, Hyderabad, India Beaumont MA, Nichols RA (1996) Evaluating loci for use in the genetic analysis of population structure. Proc R Soc Lond Ser B Biol Sci 263:1619–1626 Beebe S, Skroch PW, Tohme J, Duque MC, Pedraza F, Nienhuis J (2000) Structure of genetic diversity among common bean landraces of Middle American origin based on correspondence analysis of RAPD. Crop Sci 40:264–273 Beebe S, Rengifo J, Gaitan E, Duque MC, Tohme J (2001) Diversity and origin of Andean landraces of common bean. Crop Sci 41:854–862 Blair MW, Pedraza F, Buendia HF, Gaita´n-Solı´s E, Beebe SE, Gepts P, Tohme J (2003) Development of a genome-wide anchored microsatellite map for common bean (Phaseolus vulgaris L.). Theor Appl Genet 107:1362–1374 Blair MW, Iriarte G, Beebe S (2006) QTL analysis of yield traits in an advanced backcross population derived from cultivated Andean X wild common bean (Phaseolus vulgaris L.) cross. Theor Appl Genet 112:1149–1163 Bo¨rner A, Chebotar S, Korzun V (2000) Molecular characterization of the genetic integrity of wheat (Triticum aestivum L.) germplasm after long-term maintenance. Theor Appl Genet 100:494–497 Brush SB (2000) The issues of in situ conservation of crop genetic resources. In: Brush SB (ed) Genes in the field. On farm conservation of crop diversity. IPGRI, Rome, IDRC, Ottawa, and Lewis Publishers, Boca Raton Caballero A (1994) Developments in the prediction of effective population size. Heredity 73:657–679 Campbell D, Bernatchez L (2004) Genomic scan using AFLP markers as a means to assess the role of directional selection in the divergence of sympatric whitefish ecotypes. Mol Biol Evol 21:945–956 Chebotar S, Ro¨der MS, Korzun V, Saal B, Weber WE, Bo¨rner A (2003) Molecular studies on genetic integrity of open-pollinating species rye (Secale cereale L.) after long-term genebank maintenance. Theor Appl Genet 107:469–1476 Cockerham CC, Weir BS (1993) Estimation of gene flow from F-statistics. Evolution 47:855–863 Crossa J (1989) Methodologies for estimating the sample size required for conservation of outbreeding crops. Theor Appl Genet 77:153–161 Crossa J, Vencovsky R (1994) Implications of the variance effective population size on the genetic conservation of monoecious species. Theor Appl Genet 89:936–942 Dieringer D, Schlo¨tterer C (2002) Microsatellite analyser (MSA): a platform independent analysis tool for large microsatellite data sets. Mol Ecol Notes 3:167–169 Doyle JJ, Doyle JL (1990) Isolation of plant DNA from fresh tissue. Focus 12:13–24 Escalante A, Coello G, Eguiarte LE, Pin˜ero D (1994) Genetic structure and mating systems in wild and cultivated populations of Phaseolus coccineus and P. vulgaris (Fabaceae). Amer J Bot 81:1096–1103
Genetica (2010) 138:985–998 Frankham R (1995) Effective population-size adult-population size ratios in wildlife: a review. Genet Res 66:95–107 Goldringer I, Enjalbert J, Raquin AL, Brabant P (2001) Strong selection in wheat population during ten generation of dynamic management. Genet Sel Evol 33(suppl1):S441–S463 Goldringer I, Prouin C, Rousset M, Galic N, Bonnin I (2006) Rapid differentiation of experimental populations of wheat for heading time in response to local climatic conditions. Ann Bot 98:805–817 Gomez OJ, Blair MW, Frankow-Lindberg BE, Gullberg U (2005) Comparative study of common bean (Phaseolus vulgaris L.) landraces conserved ex situ in genebanks and in situ by farmers. Gen Res Crop Evol 52:371–380 Guarino L, Rao RR, Reid R (1995) Collecting plant genetic diversity, technical guidelines. CAB International, Wallingford IPCC (2007) Fourth assessment report climate change 2007: synthesis report. Intergovernmental Panel on Climate Change, Geneva, Switzerland Koinange EMK, Singh SP, Gepts P (1996) Genetic control of the domestication syndrome in common bean. Crop Sci 36:1037–1045 Lawrence MJ (2002) A comprehensive collection and regeneration strategy for ex situ conservation. Gen Res Crop Evol 00:1–11 Li Q, Xu Z, He T (2002) Ex situ genetic conservation of endangered Vatica guangxiensis (Dipterocarpaceae) in China. Biol Conserv 106:156 Li Q, He T, Xu Z (2005) Genetic evaluation of the efficacy of in situ and ex situ conservation of Parashorea chinensis (Dipterocarpaceae) in southwestern China. Biochem Genet 43:387–406 Lima VM, Magioli C, de A Gerhardt LB, Tarre´ E, Menezes RMG, Sachetto-Martins G, Pinheiro MM (2002) Bean class IV chitinase promoter is modulated during plant development and under abiotic stress. Plant Physiol 116:512–521 Margis-Pinheiro M, Marivet J, Burkard G (1994) Bean class IV chitinase gene: structure, developmental expression and induction by heat stress. Plant Sci 98:163–173 Marshall DR, Brown AHD (1975) Optimum sampling strategies in genetic conservation. In: Frankel OH, Hawkes JG (eds) Genetic resources for today and tomorrow. Cambridge University Press, Cambridge, pp 53–80 Masi P, Zeuli PL, Donini P (2003) Development and analysis of multiplex microsatellite markers sets in common bean (Phaseolus vulgaris L.). Mol Breed 11:303–313 Miller MP (1997) Tools for population genetic analyses (TFPGA), version 1.3. A windows program for the analysis of allozyme and molecular population genetic data. Distributed by the author Negri V, Tosti N (2002) Phaseolus genetic diversity maintained on farm in Central Italy. Gen Res Crop Evol 49:511–520 Negri V, Maxted N, Vetelainen M (2009) European landrace conservation: an introduction. In: Vetelainen M, Negri V, Maxted N (eds) European landraces: on-farm conservation, management and use. Bioversity Technical Bulletin No. 15, Bioversity International, Bioversity International publication, Rome, Italy, pp 1–22, ISBN 978-92-9043-805-2 Nei M (1987) Molecular evolutionary genetics. Columbia University Press, New York Papa R, Gepts P (2003) Asymmetry of gene flow and differential geographical structure of molecular diversity in wild and domesticated common bean (Phaseolus vulgaris L.) from Mesoamerica. Theor Appl Genet 106:239–250 Parzies HK, Spoor W, Ennos RA (2000) Genetic diversity of barley landrace accessions (Hordeum vulgare subsp. vulgare) conserved for different lengths of time in ex situ gene banks. Heredity 84:476–486 Peakall R, Smouse PE (2005) Genalex 6: genetic analysis in excel. Population genetic software for teaching and research.
997 Australian National University, Camberra, http://www.anu. edu.au/BoZo/GenAlEx Peters NK, Frost JW, Long SR (1986) A plant flavone luteolin induces expression in Rhizobium meliloti nodulation genes. Science 233:977–980 Powell W, Morgante M, Doyle JJ, McNicol JW, Tingey SV, Rafalski AJ (1996) Genepool variation in genus Glycine subgenus soja revealed by polymorphic nuclear and chloroplast microsatellites. Genetics 144:793–803 Raquin AL, Depaulis F, Lambert A, Galic N, Brabant P, Goldringer I (2008) Experimental estimation of mutation rates in a wheat population with a gene genealogy approach. Genetics 179:2195–2211 Raymond M, Rousset F (1995) Genepop (version 1.2)—population genetics software for exact tests and ecumenicism. J Hered 86:248–249 Rhone´ B, Remoue´ C, Galic N, Goldringer I, Bonnin I (2008) Insight into the genetic bases of climatic adaptation in experimentally evolving wheat populations. Mol Ecol 17:930–943 Rice EB, Smith ME, Mitchell SE, Kresovich S (2006) Conservation and change: a comparison of in situ and ex situ conservation of Jala maize germplasm. Crop Sci 46:428–436 Serrano AR, Del Castillo JL, Novo JJ, Ocan˜a AF, Rodrı´guez MVG (2007) Chitinase and peroxidase activities in sunflower hypocotyls: effects of BTH and inoculation with Plasmopara halstedii. Plant Biol 51:149–152 Sicard D, Nanni L, Porfiri O, Bulfon D, Papa R (2005) Genetic diversity of Phaseolus vulgaris L. and Phaseolus coccineus L. landraces in central Italy. Plant Breed 124:464–472 Sokal RR, Rohlf FJ (1995) Biometry: the principles and practice of statistics in biological research, 3rd edn. Freeman WH and Co, San Francisco Soleri D, Smith SE (1995) Morphological and phenological comparisons of two Hopi maize varieties conserved in situ and ex situ. Econ Bot 49:56–77 Spagnoletti Zeuli L, Sergio L, Perrino P (1995) Changes in the genetic structure of wheat germplasm accessions during seed rejuvenation. Plant Breed 114:193–198 Storz JF (2005) Using genome scans of DNA polymorphism to infer adaptive population divergence. Mol Ecol 14:671–688 ˚ (2001) Diversity and adaptation in rice Tin HQ, Berg T, Bjørnstad A varieties under static (ex situ) and dynamic (in situ) management. A case study in the Mekong Delta, Vietnam. Euphytica 122:491–502 Tiranti B (2005) Varieta` locali italiane di Phaseolus vulgaris L.: livelli di diversita`, struttura genetica e strategie di conservazione [Italian landraces of P. vulgaris L.: diversity, genetic structure and conservation strategies]. PhD thesis, University of Perugia, p 324 Tiranti B, Negri V (2007) Selective micro-environmental effects play a role in shaping genetic diversity and structure in a Phaseolus vulgaris L. landrace: implications for on-farm conservation. Mol Ecol 16:4942–4955 Vasema¨gi A, Nilsson J, Primmer CR (2005) Expressed sequence taglinked microsatellites as a source of gene-associated polymorphisms for detecting signatures of divergent selection in Atlantic salmon (Salmo salar L.). Mol Biol Evol 22:1067–1076 Vencovsky R, Crossa J (1999) Variance effective population size under mixed self and random mating with applications to genetic conservation of species. Crop Sci 39:1282–1294 Vitalis R, Dawson K, Boursot P (2001) Interpretation of variation across marker loci as evidence of selection. Genetics 158:1811–1823 Vitalis R, Dawson K, Boursot P, Belkhir K (2003) DetSel 1.0: a computer program to detect markers responding to selection. J Hered 94:429–431
123
998 Waples RS (1989) A generalized approach for estimating effective population size from temporal changes in allele frequency. Genetics 121:379–391 Weir BS, Cockeram CC (1984) Estimating F-statistic for the analysis of population structure. Evolution 38:1358–1370 Wells WC, Isom WH, Waines JG (1988) Outcrossing rates of six common bean lines. Crop Sci 28:177–178
123
Genetica (2010) 138:985–998 Yu K, Park SJ, Poysa V, Gepts P (2000) Integration of simple sequence repeat (SSR) markers into a molecular linkage map of common bean (Phaseolus vulgaris L). J Hered 91:429–434 Zizumbo-Villarreal D, Colunga-Garcia Marin P, De la Cruz EP, Delgado PV, Gepts P (2005) Population structure and evolutionary dynamics of wild-weedy-domesticated complexes of common bean in a Mesoamerican region. Crop Sci 45:1073–1083
