The method is directly app- licable to seeds ... certification prior to marketing. They would .... fluorescent labels and automated scanning of the mi- crosatellite ...
Q Springer-Verlag 1999
Eur Food Res Technol (1999) 210 : 144–147
ORIGINAL PAPER
Antonella Pasqualone 7 Concetta Lotti Antonio Blanco
Identification of durum wheat cultivars and monovarietal semolinas by analysis of DNA microsatellites
Received: 22 February 1999 / Revised version: 19 April 1999
Abstract Microsatellites isolated in bread wheat were used to identify 20 Italian durum wheat (Triticum turgidum var. durum L.) cultivars. A total of 15 microsatellite primer pairs were tested against DNA extracted from leaf tissue, single seeds and semolina. Twelve markers showed allelic polymorphism among the 20 cultivars, providing a total of 41 different alleles. Firstly, an analysis of microsatellite informativeness in the chosen set of durum wheat cultivars was made and a set of highly polymorphic microsatellites was established. Secondly, among the most polymorphic cultivars, the minimum number of microsatellites able to distinguish all cultivars was determined. A set of five microsatellites was found sufficient to differentiate the durum wheat cultivars examined. The method is directly applicable to seeds and semolina, and is suitable for detecting seed mixtures in the same seed lot. Key words Polymerase chain reaction 7 DNA microsatellites 7 Durum wheat 7 Cultivar identification 7 Monovarietal semolina
control systems for durum wheat cultivar identification are needed. They are also necessary to assess the distinctness, uniformity and stability of new cultivars for certification prior to marketing. They would also be useful, a posteriori, to settle legal conflicts over the recognition of a seed stock. Microsatellites are simple sequence repeats ubiquitously interspersed in eukaryotic genomes [1] and show a high frequency of variation in the number of repeats in different individuals or accessions. This kind of polymorphism is easily detectable via the polymerase chain reaction (PCR) using specific primers on the flanking regions of the repeated sequence [2]. Preliminary studies indicated that the amplification of bread wheat microsatellites (WMS) in durum wheat [3] for cultivar identification purposes [4] was possible. In the present report, we demonstrate the application of bread WMS to the differentiation and identification of seeds and semolina from 20 Italian durum wheat cultivars.
Materials and methods Introduction The pasta-making industry requires semolina of high technological value obtained from durum wheat cultivars with a high protein content and good gluten quality. To keep the value of a monovarietal semolina or seed stock unchanged, no mixture with poor quality cultivars should be made. To avoid possible seed mixture or substitution in commercial transactions and to ensure inspections during milling, new routine quality A. Pasqualone (Y) Institute of Agricultural Industries, University of Bari, Via Amendola 165/A, I-70126, Bari, Italy e-mail: antonella.pasqualone6agr.uniba.it C. Lotti 7 A. Blanco Institute of Plant Breeding, University of Bari, Via Amendola 165/A, I-70126, Bari, Italy
Plant material. Twenty durum wheat cultivars (Fig. 2) included in the Italian National Register of Seeds, kindly provided by the Experimental Institute for Cereal Research, Foggia, Italy, were used. They were representative of the most marketed and processed Italian durum wheat germplasm. Twenty plants for each cultivar were grown in pots in the greenhouse and leaves were bulk harvested from 2-week-old seedlings. DNA extraction. DNA extraction from frozen-dried leaves was performed as described by Sharp et al. [5]. DNA extraction from single and half seeds was performed using the proteinase K method as described by Chunwongse et al. [6], using 4.8 ml of proteinase K (Sigma-Aldrich) in 200 ml of extracting solution. The same method was applied to DNA extraction from monovarietal semolinae obtained after grinding seeds with a Cyclotec 1093 Sample mill (Tecator). Microsatellite primers. Fifteen primer pairs of bread WMS localized on chromosomes of the A and B genomes were used: WMS 2, WMS 5, WMS 11, WMS 18, WMS 46, WMS 47 [7]; WMS 82, WMS 88, WMS 95, WMS 120, WMS 131, WMS 154, WMS 155,
145 WMS 186 [8]; WMS 135 [9]. The designation WMS corresponds to the probe and the designation Xgwm [Gatersleben (Germany) wheat microsatellite] to the corresponding locus name. The synthesis of primers was done by Gibco BRL-Life Technologies. Polymerase chain reaction. PCR amplifications from seeds and from semolinae were conducted as described by Chunwongse et al. [6], starting from 25 ml of extract in a final volume of 100 ml. PCR amplifications from leaves were performed as described by Röder et al. [7]. Amplification product analysis. Amplification products were separated by electrophoresis in 5% denaturing polyacrylamide (acrylamide:bisacrylamide 19 : 1 w:w plus 7 M urea) sequencing gels 0.4 mm thick, 50 cm long, in 1! TBE buffer (100 mM Tris-borate pH 8.0, 2 mM EDTA). PCR products were mixed with an equal volume of loading buffer containing 98% formamide, 0.1% bromophenol blue and 0.1% xylene cyanol. The mixture was incubated at 100 7C for 5 min and cooled on ice before loading. A 5 ml volume of each sample was analysed. Gels were run at constant power (80 W) for 3 h using a Sequi-Gen Sequencing Cell (BioRad). Fragment detection. For fragment detection, silver staining of the gels was performed as described by Bassam et al. [10]. Allele lengths were determined by comparing the amplified fragments with the molecular weight markers MspI/HaeIII (MBI Fermentas), 20 bp (BioRad) and 100 bp (MBI Fermentas).
cal electrophoretic patterns were obtained from seed and semolina DNA, and from the DNA of the corresponding leaves (Fig. 1). PCR amplifications of different seeds from the same seed lot gave identical electrophoretic patterns, emphasizing the ability of this procedure to detect seed mixtures. All primer pairs produced microsatellite fragments in the durum wheat cultivars examined. No “null” loci were found. Polymorphism of the electrophoretic patterns (Fig. 1) was observed for 12 primer pairs. All the different patterns amplified with the same primer pair in the 20 cultivars were considered as alleles of the same microsatellite genetic locus. The number of detected alleles and their frequency over the cultivars were estimated and some indices of diversity were calculated for each microsatellite as a measure of its informativeness. Diversity indices (DI) [11], probabilities of identity (I) [12] and polymorphic information contents (PIC) [13] were calculated as follows: DIp1PA p 2i ipnP1
IpA p 4i c A
Results A total of 15 bread WMS were analysed. Leaf DNA was initially used to assess optimal PCR conditions as well as to screen the effectiveness of the 15 pairs of primers. Subsequently, a procedure of DNA extraction from single seeds and semolina, both able to supply smaller quantities of DNA than leaves, was carried out in order to develop a quick procedure of cultivar identification. Adequate levels of amplification and identi-
Fig. 1 Denaturing polyacrylamide silver-stained gel showing electrophoretic identical patterns of amplified fragments from DNA of cultivar Simeto obtained from leaves, seeds and semolina, respectively, with primers of microsatellite locus Xgwm5 (lanes 1–3); PCR fragments resulting from multiplex amplifications of Simeto DNA with primers of Xgwm5 and Xgwm46 (lane 4), and with primers of Xgwm46 and Xgwm135 (lane 5); polymorphism of microsatellite locus Xgwm5 in the 20 durum wheat cultivars examined (cultivar names are indicated at each lane)
1
n
A (2 pi pj) 2
ip1
jpic1
n
2
nP1
PICp1P A p 2i P A ip1
ip1
n
A 2 p 2i p 2j jpic1
where pi and pj are the frequencies of the ith and jth alleles in the 20 cultivars and n is the total number of such alleles. These indices provide an estimate of the discrimination ability of each microsatellite by taking into account not only the number of revealed alleles but also their relative frequencies. Thus, different values of DI, I and PIC for microsatellites detecting the same number of alleles are due to their different allelic distribution. For example, the microsatellite locus Xgwm46, which revealed five alleles, one of which had a very high frequency, had less discriminatory capability (Table 1) than Xgwm131, which also revealed five alleles but with similar frequencies. Usually, very highly discriminative loci show many alleles in similar frequencies. The calculated values for the 12 polymorphic microsatellites examined (Table 1) indicated that microsatellite loci Xgwm5, Xgwm11, Xgwm18, Xgwm46, Xgwm95, Xgwm131 and Xgwm135 were much more informative than the others since they had higher values of DI and PIC and lower values of I. The most polymorphic microsatellite locus was Xgwm131, which had the highest values of DI (0.76) and PIC (0.72), thus confirming preliminary data previously obtained over a smaller sample of seven durum wheat cultivars [4].
Discussion By comparing data for all microsatellites, the minimum combination able to discriminate among the 20 culti-
146 Table 1 Informativeness of the polymorphic microsatellites over the 20 durum wheat cultivars examined Microsatellite locus
Allele number
Xgwm2 Xgwm5 Xgwm11 Xgwm18 Xgwm46 Xgwm47 Xgwm95 Xgwm 120 Xgwm131 Xgwm135 Xgwm154 Xgwm155
2 5 4 4 5 2 5 2 5 3 2 2
Mean value
3.4
Allele sizes (in base pairs) and allele type (a–e) a
b
228 168 190 176 175 175 118 100 152 158 120 135
230 170 224 177 177 190 123 109 153 161 138 136
c
d
e
173 225 178 179
175 226 180 181
177
124
125
127
154 163
155
158
vars, including the most closely related such as Duilio and its parents, Grazia and Giemme, was determined. The five highly polymorphic microsatellite loci Xgwm5, Xgwm11, Xgwm46, Xgwm131 and Xgwm135, according to their informativeness data, were able, in combination, to distinguish all cultivars with at most five amplifications. In addition, an identification key (Fig. 2) was worked out for cultivar identification by subsequent steps avoiding unnecessary amplifications. It should be noted that once the five differentiating microsatellites were identified, the extract from a single seed was sufficient for five subsequent different amplifications, leading to complete identification. As far as time demand is concerned, once the polymorphic microsatellites set has been established and the working conditions have been set up, the analysis of a single microsatellite can be made in about 9 h. Since the analysis of at most five microsatellites is needed to achieve complete identification and as the annealing temperatures of the five primer pairs are identical, it is possible to execute all amplifications simultaneously in the same block of the thermal cycler and to control all the amplification products on the same gel. Multiple loading of the gel can also increase the efficiency, thus minimizing the wastage of time and reagents. Furthermore, multiplex amplifications with primers containing fluorescent labels and automated scanning of the microsatellite alleles could improve the potential of this method for automation. In conclusion, it has been possible to establish a set of highly polymorphic microsatellites that in combination give an identification profile of the 20 durum wheat cultivars tested. The length of primers, medially 20 bp, their high annealing temperature and above all their high specificity give the technique the potential for performing reliable selective amplifications, therefore microsatellite assays have the robustness that is required for practical varietal identification in an industrial setting. Like all molecular technologies, this method represents a powerful tool for recognizing and dif-
182
Diversity index (DI)
Probability of identity (I)
Polymorphic information content (PIC)
0.255 0.740 0.570 0.585 0.485 0.480 0.540 0.480 0.760 0.540 0.480 0.420
0.588 0.119 0.263 0.258 0.290 0.385 0.243 0.385 0.188 0.286 0.385 0.424
0.222 0.696 0.490 0.501 0.461 0.365 0.509 0.365 0.720 0.466 0.365 0.332
0.528
0.318
0.458
Fig. 2 Identification key of the examined durum wheat cultivars based on microsatellite loci Xgwm11, Xgwm131, Xgwm46, Xgwm135 and Xgwm5. Each branch corresponds to a different allele type
ferentiating varieties, supplementing and refining the classical methods. It would be particularly useful in solving doubtful cases involving the identification of closely related cultivars.
147 Acknowledgements This research was supported financially by the Italian MURST 60%, Research Project “Identificazione del DNA in foglie e cariossidi di frumento duro e in semole e paste alimentari mediante analisi di microsatelliti.”
References 1. Tautz D, Renz M (1984) Nucleic Acids Res 12 : 4127–4137 2. Weber JL, May PE (1989) Am J Hum Genet 44 : 388–396 3. Korzun V, Röder MS, Wendehake K, Pasqualone A, Lotti C, Ganal MW, Blanco A (1999) Theor Appl Genet 98 : 1202–1207 4. Pasqualone A, Lotti C, Blanco A (1998) In: Troccoli A, De Vita P, Di Fonzo A (eds) Proceedings of the VI International Meeting on Durum Wheat. Experimental Institute for Cereal Research, Foggia, Italy, 30 April–2 May, 1998
5. Sharp PJ, Kreis M, Shewry PR, Gale MD (1988) Theor Appl Genet 75 : 286–290 6. Chunwongse J, Martin GB, Tanksley SD (1993) Theor Appl Genet 86 : 694–698 7. Röder MS, Plaschke J, Konig SU, Börner A, Sorrels ME, Tanksley SD, Ganal MW (1995) Mol Gen Genet 246 : 327–333 8. Plaschke J, Ganal MW, Röder MS (1995) Theor Appl Genet 91 : 1001–1007 9. Plaschke J, Börner A, Wendehake K, Ganal MW, Röder MS (1996) Euphytica 89 : 33–40 10. Bassam BJ, Caetano-Anolles G, Gresshoff PM (1991) Anal Biochem 196 : 80–83 11. Weir BS (1990) Genetic data analysis. Sinauer, Sunderland, Mass 12. Paetkau D, Calvert W, Stirling I, Strobeck C (1995) Mol Ecol 4 : 347–354 13. Weber JL (1990) Genomics 7 : 524–530
