Consequences of ex situ Conservation on the Genetic Integrity of ...

13 downloads 0 Views 546KB Size Report
The ex situ collection of germplasm is an institutional inno- vation to preserve germplasm outside of the native range of a species (Koo et al., 2004). Gene banks ...
RESEARCH

Consequences of ex situ Conservation on the Genetic Integrity of Germplasm Held at Different Gene Banks: A Case Study of Bread Wheat Collected in Pakistan Ryoko Hirano, Shakeel Ahmad Jatoi, Makoto Kawase, Akira Kikuchi, and Kazuo N. Watanabe*

ABSTRACT Genetic diversity and genetic integrity were tested for wheat (Triticum aestivum L.) landraces conserved in two gene banks with considerably different germplasm management systems. We identified two sets of 17 wheat accessions derived from identical sources collected in Pakistan, which were later deposited at the National Institute of Agrobiological Sciences, Japan (NIAS) and the Plant Genetic Resources Program, Pakistan (PGRP). Regeneration of the conserved germplasm was infrequent at both gene banks. Amplified fragment length polymorphism (AFLP) profiles did not indicate prominent changes in the banding patterns of the wheat landraces conserved at the two gene banks. No significant differences were observed in allele frequency of the detected loci in comparison to the original collections (materials preserved without regeneration and multiplication), except in one accession at NIAS. We conclude that genetic diversity was conserved at these two gene banks with different management systems. However, among the 161 scored loci, 26 AFLP bands disappeared at the same loci in NIAS and PGRP accessions, four disappeared only in NIAS accessions, and 28 only in PGRP accessions. At one particular locus, the band disappeared in 12 PGRP accessions but not in NIAS accessions. The results indicate that unique selection might occur at PGRP and that regeneration strategies to reduce it will be needed in the future.

R. Hirano, A. Kikuchi, K.N. Watanabe, Graduate School of Life and Environmental Sciences, Univ. of Tsukuba, Gene Research Center, 1-11, Tennodai, Tsukuba, Ibaraki 305-8572, Japan. S.A. Jatoi, Plant Genetic Resources Institute, National Agricultural Research Center, Islamabad, Pakistan. M. Kawase, National Institute of Agrobiological Sciences, 2-12, Kannondai, Tsukuba, Ibaraki 305-8602, Japan. Received 24 Nov. 2008. *Corresponding author ([email protected]). Abbreviations: AFLP, amplified fragment length polymorphism; AMOVA, analysis of molecular variance; NIAS, National Institute of Agrobiological Sciences, Japan; PGRP, Plant Genetic Resources Program, Pakistan.

T

he ex situ collection of germplasm is an institutional innovation to preserve germplasm outside of the native range of a species (Koo et al., 2004). Gene banks are playing an important role in the preservation, investigation, and utilization of germplasm of various crop species. Germplasm preserved in gene banks is intended for medium- to long-term storage under controlled conditions for future use (del Rio et al., 1997; Hawkes et al., 2000). Despite the considerable efforts of gene bank curators, conditions at gene banks are sometimes vulnerable because of economic constraints, insufficient infrastructure (e.g., power failure), natural disasters, and local confl icts. Therefore, a duplication approach has been undertaken to avoid the loss of conserved germplasm by placing material in other gene banks, both domestic and overseas (Plucknett et al., 1987; van Hintum and Visser, 1995). Duplicated entities allow for availability and access to the materials as well as equitable sharing among institutions (van Hintum, 2000). The original genetic attributes of accessions should be retained as much as possible after regeneration. Breese (1989) summarized Published in Crop Sci. 49:2160–2166 (2009). doi: 10.2135/cropsci2008.11.0675 © Crop Science Society of America 677 S. Segoe Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

2160

WWW.CROPS.ORG

CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

factors to consider during the regeneration and multiplication of germplasm to minimize the effect of ex situ conservation on the genetic integrity of the germplasm. Mechanical contamination and differential genotypic survival in each stage during the regeneration process were among the factors. Unintentional selection, fertility differences, genetic drift, and genetic shift also contribute to the reduction of genetic diversity (Cross and Wallace, 1994; van Hintum et al., 2002). The management of the regeneration process, such as mating design and the number of parents, may affect the genetic component of resultant progeny as well. It is difficult to maintain genetic integrity due to genetic drift and random effects in the cases of outcrossing species and mixtures of mainly self-fertilizing species (van Hintum et al., 2002), such as wheat (Triticum aestivum L.). Despite being preserved at different locations, duplicate germplasm should be genetically equivalent. Although gene banks share the same objectives of securing longterm storage for maintaining the diversity and viability of germplasm, conditions and management systems vary among institutions. International institutes for the conservation of plant germplasm have provided guidance for implementing conservation management with adaptations for local operating conditions and target species (Breese, 1989; Hong and Ellis, 1996; Karp et al., 1997). Absolute standardization of gene bank management to maintain genetic integrity would be a challenging task. Researchers have studied the genetic integrity of wheat, potato (Solanum tuberosum L.), and rye (Secale cereale L.) materials deposited at gene banks (Börner et al., 2000; Chebotar et al., 2003; del Rio et al., 2006). Wheat landraces are often a mixture of different genotypes. Because of this heterogeneous nature, the effect of gene bank preservation on the genetic integrity of wheat landraces is different than that of registered varieties or breeders’ lines. The National Institute of Agrobiological Sciences (NIAS), Tsukuba, Japan, organized the collection expedition in collaboration with the Plant Genetic Resources Program (PGRP), National Agricultural Research Centre, Islamabad, Pakistan in 1989, with the aim of exploring different agroecological regions in Pakistan. Collection was targeted mainly at grain crops for conservation and utilization in crop improvement programs. A wide variety of crops, including wheat, were collected {56 species consisting mainly of rice (Oryza sativa L.), pearl millet [Pennisetum glaucum (L.) R. Br.], sorghum [Sorghum bicolor (L.) Moench], and minor millets}. The materials collected during the expedition were divided at the PGRP laboratory and subsequently deposited in NIAS and PGRP (Nakagahra et al., 1990). This provided a good opportunity to test the factors that affect duplicated germplasm conserved at different gene banks. This study tested whether conservation at two gene banks, one in Japan and the other in Pakistan, would alter the genetic integrity of wheat landraces from identical CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

sources. The main objectives were to track the changes in genetic diversity of wheat landraces through gene bank conservation and to compare the genetic integrities of the originally identical germplasms deposited in the two gene banks. Based on our findings, we discuss the accession management in a gene bank and importance of study of duplicate materials from different gene banks.

MATERIALS AND METHODS Plant Materials The bread wheat landraces used in this study were collected during the PARC/NIAR cereal collecting expedition in Pakistan in 1989. Thirty-three accessions of wheat germplasm were recorded in the collection expedition report (Nakagahra et al., 1990). After the expedition, germplasms were divided and subsequently deposited in NIAS ( Japan) and PGRP (Pakistan). The wheat germplasms were conserved for 17 yr in each gene bank under different institutional management policies. Each gene bank uses a different system for assigning accession numbers to the collected germplasm. NIAS uses collection numbers as the name of gene bank entries, whereas PGRP assigns different accession numbers upon deposition into the gene bank. Therefore, identities of the sources of NIAS and PGRP accessions were searched using the collection expedition report (Nakagahra et al., 1990) and passport information in the gene banks’ catalogs. The collection numbers, dates, sites, and altitudes for accessions were obtained from the gene bank catalog of PGRP (Afzal et al., 1995). After checking the correspondence of the collection year and country of origin, we identified 35 accessions from the collection expedition registered at PGRP. At NIAS, however, 135 accessions from the same expedition were registered because heterogeneous accessions were split based on morphological differences. The NIAS database is available online (www.gene.affrc. go.jp/databases-plant_search_en.php; verified 21 Sept. 2009). A set of germplasm accessions that were conserved without regeneration after the field collection in 1989 (Original) was used as the standard. Due to a lack of Original seed, amplified fragment length polymorphism (AFLP) profi les were compared between 17 accessions of Original and the derivatives registered at NIAS and PGRP. The list of gene bank materials used in this study is provided in the Supplementary Table S1.

AFLP Analysis Seedlings of 11 to 16 individuals per accession were mixed and used for extraction of total DNA by the cetyl trimethyl ammonium bromide method (Doyle and Doyle, 1987). The AFLP analysis was performed following Vos et al. (1995). Approximately 2 μg of genomic DNA was digested with the restriction enzymes EcoRI and MseI in 12.5-μL reactions following ligation of adaptor nucleotides for both restriction ends. Pre-amplification of restriction fragments was conducted with AFLP Pre-amp Primer Mix I (Invitrogen, Carlsbad, CA) in 25-μL reactions. Selective amplifications were performed in 8 μL containing 0.625 μM WellRED (Beckman-Coulter, Fullerton, CA) fluorescently labeled forward primer (EcoRI primer), 0.625 μM MseI primer, 2.0 mM MgCl 2, 0.2 mM dNTPs, and 0.2 unit of platinum Taq polymerase (Invitrogen). Three selective

WWW.CROPS.ORG

2161

nucleotides were added to the 3' ends of MseI primers. Six primer combinations (EcoRI-ACT with MseI-CTA and-CTT; EcoRI-AGC with MseI-CAC, -CTA, -CTG, and -CTT) were tested. All amplifications were performed on a GeneAmp PCR System 9700 (Applied Biosystems, Inc., Foster City, CA). Amplified fragment length polymorphism fragments were separated and detected using a CEQ 8000 Genetic Analysis System (Beckman-Coulter) following the manufacturer’s protocol. One microliter of the amplification product was diluted with 10 μL of distilled water. The diluted amplification product was added to 23 μL of sample loading solution with 0.12 μL of internal size standard. Sample plates were loaded onto the CEQ 8000 system and then denatured at 90°C for 120 s before being separated through linear acrylamide containing 0.7 M urea in a 33 cm × 75 μm capillary array at 6.0 kV for 60 min. The reproducibility of AFLP fragments was determined by conducting eight repeat runs for a bulked sample surveyed, and yielded the same AFLP pattern for all set of primers.

Data Analysis Amplified fragment length polymorphism fragments were analyzed using the fragment analysis function of the CEQ 8000 software, version 8 (Beckman-Coulter). After scoring, the results were adjusted manually. Amplified fragment length polymorphism fragments that were not separated by 1.5 nucleotide differences were grouped. A binary matrix table was created according to the presence (1) or absence (0) of alleles. AFLP-SURV version 1.0 (Vekemans et al., 2002) was used to estimate pairwise relatedness coefficients, pairwise distances, and genetic variation within and between groups of accessions according to the methods of Lynch and Milligan (1994). Presence or absence of the fragments was compared between Original vs. NIAS and Original vs. PGRP for each accession that was derived from an identical source. Analysis of Molecular Variance (AMOVA; Excoffier et al., 1992) was performed using the software GeneAlEX 6.1 (Peakall and Smouse, 2006) on groups of accessions, the Original set, NIAS, and PGRP. The differentiations among three groups were estimated in terms of PhiPT, which is the proportion of variance among groups relative to total valiance. Tests of significance were based on 999 permutations.

RESULTS Germplasm Management in Different Gene Banks All wheat landraces collected in 1989 in Pakistan have experienced conservation in different gene banks, NIAS and PGRP, for 17 yr. Before registration in the gene bank, the collected material was regenerated by NIAS; however, information about whether this occurred at PGRP was not available. Due to the decline of seed viabilities, the material at the two gene banks was regenerated once over the entire 17 yr. The information pertaining to growing conditions and the seed regeneration system at the two gene banks displayed remarkable differences (Table 1). Both gene banks grew the wheat as a winter crop; however, the range of minimum temperatures during the growing season varied considerably. At NIAS, the regeneration was performed at four localities with different climatic conditions. The most prominent difference among the localities was minimum temperature in the wheat-growing season, which ranged from –6.7°C in Sapporo to 0.5°C in Fukuyama. At PGRP, regeneration was performed only in Islamabad, where the minimum temperature was 2.9°C. Nearly all the accessions at NIAS were separated into subaccessions based on the morphological variability of the progenies observed during regeneration. For example, accession 2426(3) was subdivided into six subaccessions: 2426(3)-(1), -(2), -(3), -(5), -(6), and –(7). The number of subdivisions ranged from two to eight depending on the heterogeneity in the original material. Because PGRP did not split or lump the accessions, the AFLP score of subaccessions was merged into one for the representative accession. The merging steps were as follows: (i) align and match the AFLP scores of all subaccessions sharing the same original accession number; (ii) identify the loci that do not match among the subaccessions; and (iii) if a band is present in at least one of the subaccessions, score the locus as present. This procedure made the NIAS accessions comparable with the Original dataset.

Table 1. Basic information for the National Institute of Agrobiological Sciences, Japan (NIAS) and the Plant Genetic Resources Program, Pakistan (PGRP). Gene bank location

Regeneration Regeneration Geographic position frequency field location of field

Growing season

Temp. of growing No. of individuals used season (min./max.)† for regeneration °C

NIAS

Tsukuba, Japan

0 or 1 time since gene bank entry

Sapporo

43°00' N, 141°24' E

Sept.–June

–6.7/22.3

Morioka

39°44' N, 141°08' E

Sept.–June

–5.3/23.8

Tsukuba

36°01' N, 140°06' E

Oct.–June

–2.7/24.9

Fukuyama

34°30' N, 133°23' E

Oct.–June

0.5/26.9

Islamabad

33°42' N, 73°08' E

Nov.–June

2.9/29.7

about 15

PGRP Islamabad, Pakistan

Once in 1998

Few



Temperature of NIAS based on the monthly mean daily maximum and minimum temperatures from 1989 to 2007 reported on the Japan Meteorological Agency web page (www.data.jma.go.jp/obd/stats/data/en/smp/index.html; accessed 1 June 2008, verified 3 Aug. 2009), and data for the PGRP site was based on Afzal et al. (1995).

2162

WWW.CROPS.ORG

CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

Diversity Indices and Genetic Variation A total of 161 AFLP markers were scored. More polymorphic bands per locus were detected in Original samples (60) as compared to NIAS (49) and PGRP (48), although chisquare tests showed no significant differences in Original vs. NIAS (P = 0.256) and Original vs. PGRP (P = 0.121). Percentages of polymorphic loci were 37.3, 29.8, and 30.4% in Original, NIAS, and PGRP, respectively. The expected heterozygosities under Hardy–Weinberg genotypic proportions were 0.069, 0.062, and 0.064 for the Original, NIAS, and PGRP, respectively, and the chi-square tests showed no significant differences in Original vs. NIAS (P = 0.986) and Original vs. PGRP (P = 0.978). Table 2 shows AMOVA results on AFLP variation of the wheat landraces demonstrating that the most part (94%) of the total variation was caused by allelic variation within groups of accessions (within gene banks) and rest (6%) was explained as variation among the groups. PhiPT value, which summarizes the degree of differentiation among population divisions, was 0.063 (P < 0.001) among three groups. NIAS vs. PGRP exhibited the largest value of PhiPT (0.090, P = 0.001), followed by Original vs. PGRP (0.056, P = 0.001) and Original vs. NIAS (0.044, P = 0.004).

Changes in the Allele Numbers between NIAS and PGRP

Table 2. Results of Analysis of Molecular Variance (AMOVA) conducted on amplified fragment length polymorphism (AFLP) variation among and within groups of Original, the National Institute of Agrobiological Sciences, Japan (NIAS), and the Plant Genetic Resources Program, Pakistan (PGRP). Variation source Among groups

df

Sum of squares

Variance component

% of variation

2

27.59

0.36

6

Within group

48

251.88

5.25

94

Total

50

274.47

5.61

100

AFLP bands disappeared in either NIAS or PGRP accessions (Fig. 1). Of the 58 loci, 26 displayed a loss of bands in NIAS and PGRP accessions, whereas four loci exhibited a loss of bands only in NIAS accessions and 28 loci only in PGRP accessions. The highest number of disappeared bands was observed at locus 40_61_141, for which 12 accessions in PGRP showed a loss of bands. However, no changes occurred at this locus in NIAS accessions.

DISCUSSION Maintenance of Genetic Diversity of Wheat Landraces through Gene Bank Conservation Wheat landraces collected in Pakistan maintained overall genetic diversity through 17 yr of preservation at NIAS and PGRP gene banks. The loss of alleles was observed in both NIAS and PGRP accessions but the genetic diversity and allele frequency was not significantly changed by chi-

The genetic integrity assessment of each duplicated germplasm was made based on the presence or absence of bands. We classified the observations into three Table 3. Summary of amplified fragment length polymorphism (AFLP) categories: disappeared (present to absent, 1 to 0), loci comparisons between Original and duplicates at the National Instiappeared (absent to present, 0 to 1), and unchanged tute of Agrobiological Sciences, Japan (NIAS) and the Plant Genetic (1 to 1 and 0 to 0). The average numbers of alleles Resources Program, Pakistan (PGRP).† that disappeared and appeared were 2.29 and 7.65 Accession Original NIAS PGRP at NIAS and 5.18 and 4.82 at PGRP, respectively name No. of compared loci Dis. Unch. App. Dis. Unch. App. (Table 3). Thus, fewer bands disappeared at NIAS 2303(5) 161 2 151 8 2 152 7 and new bands were also observed at both gene 2426(3) 161 3 150 8 14 146 1 banks. There was no correlation between the num- 2428(2) 161 6 150 5 4 149 8 ber of bands that appeared and the number of sub- 2432(6) 161 0 157 4 4 157 0 accessions at NIAS. 2435(1) 161 1 153 7 2 155 4 The average numbers of unchanged bands at 2436(2) 161 0 159 2 2 157 2 NIAS (153.06) and PGRP (153.00) were almost 2439(2) 161 4 140 17 4 148 9 the same. We analyzed the numbers of unchanged 2440(2) 161 1 140 20 1 157 3 alleles for each accession at NIAS and PGRP in 2445(1) 161 0 154 7 13 141 7 comparison with the Original (Tables 4, 5). Chi- 2505(1) 161 0 158 3 2 157 2 square tests showed that there were no significant 2513(2) 161 4 155 2 4 153 4 differences, except for one accession preserved at 2588(2) 161 7 138 16 11 139 11 NIAS [2440(2), P = 0.02]. 2590(1) 161 1 158 2 1 157 3

Genetic Drift during Gene Bank Conservation To analyze the effect of genetic drift during the gene bank conservation, we compared the number of disappeared bands at each locus. Among the 161 AFLP loci scored, we identified 58 loci for which CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

2592(1)

161

0

161

0

3

157

1

2593(1)

161

0

153

8

5

153

3

2593(3)

161

2

151

8

4

150

7

2594(1)

161

8

140

13

12

139

10

Average

161

2.29 151.06



7.65

5.18

151.00

4.82

Dis., number of AFLP bands that disappeared during 17 yr of gene bank conservation; Unch., number of bands that did not change during gene bank conservation; App., number of bands that appeared during gene bank conservation.

WWW.CROPS.ORG

2163

Table 4. Comparison of unchanged amplified fragment length polymorphism (AFLP) loci between Original and the National Institute of Agrobiological Sciences, Japan (NIAS).†

square test as compared to the Original seed materials, except accession 2440(2) at NIAS. Phi statistics of AMOVA demonstrated a NIAS 1–1 matches 0–0 matches statistically significant level of genetic difNo. of Accession Exp. Obs. χ2‡ P Exp. Obs. χ2‡ P subaccessions ferentiation among groups of accessions. name However, these values were smaller than the 2303(5) 64 62 0.06 0.8 97 89 0.72 0.40 1 PhiST values among different market classes 2426(3) 69 66 0.13 0.72 92 84 0.76 0.38 6 of wheat cultivars in the United States, 2428(2) 67 61 0.54 0.46 94 89 0.28 0.60 3 which attribute very small total variance 2432(6) 68 68 0.00 1.00 93 89 0.18 0.67 4 compared with the value of the samples used 2435(1) 65 64 0.02 0.90 96 89 0.55 0.46 3 in this study (Barrett and Kidwell, 1998). 2436(2) 67 67 0.00 1.00 94 92 0.04 0.84 2 Since only a small part of the total AFLP 2439(2) 63 59 0.25 0.61 98 81 3.57 0.06 3 variation was explained with the between 2440(2) 66 65 0.02 0.90 95 75 5.33 0.02* 5 groups variation, we concluded that the 2445(1) 65 65 0.00 1.00 96 89 0.55 0.46 2 genetic differentiation was significant but 2505(1) 68 68 0.00 1.00 93 90 0.10 0.75 2 the effect was small. 2513(2) 68 64 0.24 0.63 93 91 0.04 0.83 2 The chi-square test examined whether 2588(2) 67 60 0.73 0.39 94 78 3.28 0.07 3 or not alleles of NIAS and PGRP accessions 2590(1) 66 65 0.02 0.90 95 93 0.04 0.84 4 differed from those of Original accessions; on 2592(1) 70 70 0.00 1.00 91 91 0.00 1.00 6 the other hand, AMOVA evaluated genetic 2593(1) 67 67 0.00 1.00 94 86 0.74 0.39 8 differences between groups of accessions 2593(3) 65 63 0.06 0.80 96 88 0.73 0.39 6 in relation to variation within group (gene 2594(1) 69 61 0.93 0.34 92 79 2.14 0.14 5 banks) using a nonparametric resampling *Significant at P < 0.05. approach (Excoffier et al., 1992). The main † Exp., expected number of AFLP bands in the accession (equivalent to number of bands observed in Original); Obs., observed number of bands in NIAS accession. reason for the inconsistency was probably that ‡ df = 1. AMOVA considers both the number of bands and the frequencies and thus is more sensitive than chi-square test. Phenotypic variation observed during the regeneration and subsequent subaccession procedure at NIAS indicated Table 5. Comparison of unchanged amplified fragment length that the original collections were mixtures of different genopolymorphism (AFLP) loci between Original and the Plant Genetic Resources Program, Pakistan (PGRP).† types. The overall genetic diversity and allelic integrity of the accessions were maintained despite the heterogeneous 0–0 matches PGRP 1–1 matches characteristics of the wheat landraces collected from the Accession P Exp. Obs. χ2‡ P Exp. Obs. χ2‡ name field. The genetic structures of groups of accessions in both 2303(5) 64 62 0.07 0.80 97 90 0.54 0.46 gene banks, however, were affected by the registration and 2426(3) 69 55 3.56 0.06 92 91 0.01 0.92 conservation. Cross and Wallace (1994) simulated the degree 2428(2) 67 63 0.25 0.61 94 86 0.74 0.39 of genetic diversity loss within heterogeneous self-pollinat2432(6) 68 64 0.25 0.62 93 93 0.00 1.00 ing plants through regeneration, assuming that mixture of 2435(1) 65 63 0.06 0.80 96 92 0.17 0.68 genetically distinct 1000 individuals with 10% selection pres2436(2) 67 65 0.06 0.80 94 92 0.04 0.84 sure per regeneration cycle. They found rapid truncation 2439(2) 63 59 0.27 0.60 98 89 0.91 0.34 of genetic variation after two to three regeneration bouts, 2440(2) 66 65 0.02 0.90 95 92 0.10 0.75 with only 50% of the original heterogeneity retained. Our 2445(1) 65 52 3.25 0.07 96 89 0.55 0.46 study showed that the different conditions of the two gene 2505(1) 68 66 0.06 0.81 93 91 0.04 0.83 banks had more impact on the genetic structure than on 2513(2) 68 64 0.25 0.62 93 89 0.18 0.67 allele conservation of the wheat landraces. The preservation 2588(2) 67 56 2.16 0.14 94 83 1.46 0.23 of alleles might have been due to the relative homogeneity of 2590(1) 66 65 0.02 0.90 95 92 0.10 0.75 accession compared with the simulation of Cross and Wal2592(1) 70 67 0.13 0.71 91 90 0.01 0.92 lace (1994), low frequency of regeneration, and management 2593(1) 67 62 0.40 0.53 94 91 0.10 0.75 practices taken in both gene banks. Our results indicated that 2593(3) 65 61 0.26 0.61 96 89 0.55 0.46 several regenerations may accelerate alteration of the origi2594(1) 69 57 2.53 0.11 92 82 1.22 0.27 nal genetic components as shown by the significant level of † Exp., expected number of AFLP bands in the accession (equivalent to number genetic differentiation revealed by phi statistics. We emphaobserved in Original); Obs., observed number of bands in PGRP accession. size that regeneration was conducted only once in the both ‡ df = 1.

2164

WWW.CROPS.ORG

CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

gene banks thus, continual monitoring of the genetic diversity conserved in gene bank accessions will be necessary.

Consequences of Gene Bank Conservation on Genetic Integrity Some single-allele changes were observed for each accession in both NIAS and PGRP in comparison with Original. Genetic drift and random effects might have caused most of these changes. Considering the fact that about half of all disappeared loci were observed at the same loci in both NIAS and PGRP samples, conditions of the collected seeds from the field may affect the genetic component of gene bank accessions. Poor germination in field conditions, failure in seed setting, and differential seed maturity may be possible factors causing genetic drift in heterogeneous populations. NIAS separates heterogeneous accessions into pure lines, a practice that succeeded in reducing the allele changes from the Original (four loci) as compared to the PGRP (28 loci). Cross and Wallace (1994) recommended the preservation of individual accessions as pure lines to prevent the loss of genetic variations. The AFLP band for locus 40_61_141 disappeared in 12 PGRP accessions, but no change was observed in NIAS accessions. Although AFLP markers are supposed to be neutral (Vos et al., 1995), locus 40_61_141 disappeared selectively. Therefore, it may be linked closely with an allele that is sensitive to a selective pressure existing at PGRP but not at NIAS. With the continued presence of the selection pressure, PGRP entries may deviate markedly from the original allelic composition in the future.

Conclusions and Implications for Gene Bank Conservation In the present study we made a comparative assessment of the genetic diversity of wheat landraces conserved at gene banks in Japan and Pakistan. Although the number of samples assayed and regeneration cycles were relatively small, our findings highlight the need for research on the genetic integrity of materials conserved in gene banks. The overall genetic diversity of wheat landraces collected in Pakistan was conserved at NIAS and PGRP despite considerable differences in germplasm management at the two gene banks. However, AMOVA revealed that a small but significant genetic differentiation between the set of accessions was conserved in both gene banks. Frequent loss of bands at a particular locus of PGRP accessions implied possible unintentional selection at that gene bank. Our findings indicated that splitting heterogeneous collections into subaccessions is an effective way of preserving germplasm, although it increases the size and cost of maintaining a collection (van Treuren and van Hintum, 2001). When comparing the genetic integrity of each NIAS subaccession with the Original, we did not detect significant differences in the number of unchanged alleles. The disappearance of AFLP bands was relatively CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009

Figure 1. Disappearance of bands in wheat landraces at different loci studied. Open bars indicate number of disappeared bands in the National Institute of Agrobiological Sciences, Japan (NIAS) compared with Original, and shaded bars indicate that of the Plant Genetic Resources Program, Pakistan (PGRP). Amplified fragment length polymorphism (AFLP) bands disappeared at 58 loci. Almost half (26/58 = 44.8%) of the changes were observed at the same loci for NIAS and PGRP. At locus 40_61_141, the band disappeared in 12 accessions of PGRP entries but no change was observed in NIAS entries.

high in PGRP accessions as compared to those of NIAS. Efficient gene bank management depends on capacity and resource availability but the splitting approach used at NIAS appears to succeed in preventing the loss of alleles during regeneration. Wheat landraces result from a long history of adaptation under local environments and selection by farmers. We previously observed that a particularly high genetic diversity of wheat landraces exists in northern Pakistan (Hirano et al., 2008). We had intended to explore the same area again to assess the status of genetic diversity in wheat landraces nearly two decades after the first exploration, but the geopolitical situation in the region did not permit us to initiate such a study. This situation highlights the fact that ex situ gene bank conservation can play an important role as a safeguard of in situ conservation and the importance of accession management for the maintenance of genetic integrity of gene bank accessions. Acknowledgments We thank the National Institute of Agrobiological Sciences, Japan, and Plant Genetic Resources Program, Pakistan, for providing materials and information on gene bank management. We are also grateful to recommendations from the two anonymous reviewers whose comments have improved the quality of this manuscript. This study was sponsored by Step-up Support Project 2007, University of Tsukuba, Japan, and Higher Education Commission Support 2007, Pakistan.

References Afzal, M., A. Zahoor, M.S. Bhatti, and A. Qayyum. 1995. Wheat germplasm catalog. Part 1. Plant Genetic Resources Institute, National Agricultural Research Institute, Islamabad, Pakistan. Barrett, B.A., and K.K. Kidwell. 1998. AFLP-based genetic diversity assessment among wheat cultivars from the Pacific Northwest. Crop Sci. 38:1261–1271. Börner, A., S. Chebotar, and V. Korzun. 2000. Molecular characterization of the genetic integrity of wheat (Triticum aestivum

WWW.CROPS.ORG

2165

L.) germplasm after long-term maintenance. Theor. Appl. Genet. 100:494–497. Breese, E.L. 1989. Regeneration and multiplication of germplasm resources in seed genebanks: The scientific background. International Board for Plant Genetic Resources, Rome. Chebotar, S., M.S. Röder, V. Korzun, B. Saal, W.E. Weber, and A. Börner. 2003. Molecular studies on genetic integrity of open-pollinating species rye (Secale cereale L.) after long-term genebank maintenance. Theor. Appl. Genet. 107:1469–1476. Cross, R.J., and A.R. Wallace. 1994. Loss of genetic diversity from heterogeneous self-pollinating genebank accessions. Theor. Appl. Genet. 88:885–890. del Rio, A.H., J.B. Bamberg, and Z. Huaman. 1997. Assessing changes in the genetic diversity of potato gene banks. 1. Effects of seed increase. Theor. Appl. Genet. 95:191–198. del Rio, A.H., J.B. Bamberg, and Z. Huaman. 2006. Genetic equivalence of putative germplasm collections held at CIP and US potato genebanks. Am. J. Potato Res. 83:279–285. Doyle, J.J., and J.L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19:11–15. Excoffier, L., P.E. Smouse, and J.M. Quattro. 1992. Analysis of molecular variance inferred from metric distance among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 131:479–491. Hawkes, J.G., N. Maxted, and B.V. Ford-Lloyd. 2000. The ex situ conservation of plant genetic resources. Kluwer Academic, Dordrecht, the Netherlands. Hirano, R., A. Kikuchi, M. Kawase, and K.N. Watanabe. 2008. Evaluation of genetic diversity of bread wheat landraces from Pakistan by AFLP and implications for a future collection strategy. Genet. Res. Crop Evol. 55:1007–1015. Hong, T.D., and R.H. Ellis. 1996. A protocol to determine seed storage behaviour. IPGRI Tech. Bull. 1. Int. Plant Genet. Resour. Inst., Rome. Karp, A., S. Kresovich, K.V. Bhat, W.G. Ayad, and T. Hodgkin. 1997. Molecular tools in plant genetic resources conservation:

2166

A guide to the technologies. IPGRI Tech. Bull. 2. Int. Plant Genet. Resour. Inst., Rome. Koo, B., P.G. Pardey, and B.D. Wright. 2004. Introduction. p. 1–5. In B. Koo et al. (ed.) Saving seeds. CAB Int., Wallingford, UK. Lynch, M., and B.G. Milligan. 1994. Analysis of population genetic structure with RAPD markers. Mol. Ecol. 3:91–99. Nakagahra, M., M. Kawase, and T. Nagamine. 1990. A report of PARK/NIAR cereal collecting expedition in Pakistan (1989). Natl. Inst. of Agrobiological Resour., Tokyo. Peakall, R., and P.E. Smouse. 2006. GENALEX 6: Genetic analysis in Excel: Population genetic software for teaching and research. Mol. Ecol. Notes 6:288–295. Plucknett, D.L., N.J. Smith, J.T. Williams, and N.M. Anishetty. 1987. Gene banks. p. 72–96. In D.J. Plucknett et al. (ed.) Gene banks and the world’s food. Princeton Univ. Press, Princeton, NJ. van Hintum, T.J.L. 2000. Duplication within and between germplasm collections. III. A quantitative model. Genet. Res. Crop Evol. 47:507–513. van Hintum, T.J.L., N.R. Sackville Hamilton, J.M.M. Engles, and R. van Treuren. 2002. Accession management strategies: Splitting and lumping. p. 113–120. In J.M.M. Engels et al. (ed.) Managing plant genetic diversity. CABI, New York. van Hintum, T.J.L., and D.L. Visser. 1995. Duplication within and between germplasm collections: II. Duplication in four European barley collections. Genet. Res. Crop Evol. 42:135–145. van Treuren, R., and T.L.J. van Hintum. 2001. Identification of intra-accession genetic diversity in selfi ng crops using AFLP markers: Implications for collection management. Genet. Res. Crop Evol. 48:287–295. Vekemans, X., T. Beauwens, M. Lemaire, and I. Roldan-Ruiz. 2002. Data from amplified fragment length polymorphism (AFLP) markers show indication of size homoplasy and of a relationship between degree of homoplasy and fragment size. Mol. Ecol. 11:139–151. Vos, P., R. Hogers, and M. Bleeker. 1995. AFLP: A new technique for DNA fi ngerprinting. Nucleic Acids Res. 23:4407–4414.

WWW.CROPS.ORG

CROP SCIENCE, VOL. 49, NOVEMBER– DECEMBER 2009