IMPORTANT LEGAL NOTICE - The information on this site is subject to a disclaimer and a copyright notice.
European Flag    Europa The European Commission Research Agriculture FAIR
index by sub-area Reformed CAP
index by country United Kingdom

The assessment of genetic purity in hybrid varieties of crops

Contract nr: FAIR-CT96-2069
Project nr: 2069
Project type: SC
Starting date: 01/02/1997
Duration: 39 months
Total cost: 1,925,697 EUR
EC Contribution: 1,065,000 EUR
Scientific Officer: Armin MÜNZINGER
Research topic: Optimisation of methods, systems and primary production chains
Acronym: Hybrid Purity Assessment

The prime objective of plant breeding is to produce new varieties (cultivars) of crops of agricultural and horticultural importance that represent a genuine improvement in yield and/or quality over existing varieties. An increasingly important trend is towards the production of hybrid varieties. Hybrids offer farmers increased yields, wider adaptability and reliability of both performance and quality by the exploitation of heterosis (`hybrid vigour'). There are also benefits to the plant breeding and seed industries in the commercialisation of hybrids, both in terms of better protection of intellectual property rights and of guaranteed annual seed sales to farmers.

But there is also a need to be able to assess the hybrid (genetic) purity of seed lots, as part of the seed certification process and as a means of general quality control. Only then can the genetic gains introduced by the breeding process be fully available to farmers.
Such purity testing can be carried out in the field by careful observation of morphological characters to recognise non-hybrid plants. However, these procedures are slow, labour-intensive and require large areas of land. A rapid laboratory-based test is required, which allows decisions to be made on seed lots prior to sowing. Despite some success with biochemical methods, they are unlikely to offer a long-term solution in most crops, especially those being developed as hybrid varieties. Methods also need to be rapid, highly automated and analyse multiple gene loci simultaneously. Second generation DNA profiling methods that are being developed and evaluated offer the greatest potential for such a system, particularly the use of sequence-tagged site microsatellites and AFLP (amplified fragment length polymorphism).

Newly bred varieties of crop plants must undergo statutory testing before they can be marketed within the EU to determine their eligibility for inclusion on a National List of Varieties. This testing requires that varieties must be distinct (D) from others known to exist, as well as uniform (U) and stable (S) in the characteristics used to distinguish and describe them. Plant breeders may protect their intellectual property in new varieties of crops through the statutory Plant Breeders Rights' (PBR) schemes, based on the same DUS principles. Current DUS testing relies mainly on morphological comparison of new (candidate) and existing varieties in the same trial. Guidelines for the testing of different crops are laid down by UPOV, consisting of lists of such characters and protocols. Many of the morphological characters can be affected by environmental interactions, being multigenic and continuously expressed, and this necessitates replication of observation. Some species also have insufficient descriptors to allow identification or discrimination of all varieties. There are thus compelling reasons to find more rapid and cost-effective procedures that could augment the morphologically-based approach.

(i) To develop where necessary, optimise and evaluate the use of DNA profiling methods (microsatellites and AFLP) for plant variety identification and registration and for hybrid purity assessment in crops of interest to European agriculture (maize, sunflower, and Brassica crops, primarily oilseed rape and the vegetable white kale)
(ii) To use maize as a model species for examining hybrids of various types and for comparison with the recognised and widely used biochemical methods for hybrid purity assessments
(iii) To use oilseed rape and kale as models for other Brassica crops where hybrid varieties of different types are commonly produced (e.g. Brussels sprouts, cabbage, cauliflower), to compare the results obtained with those from existing biochemical/molecular methods and to examine the transportability of techniques between species within the genus.
(iv) To develop not only the necessary profiling technology, but also the overall methodology required for the use of the techniques in the seed testing and certification context and to devise an operational system. This would involve consideration of both technical and statistical issues.
(v) To organise 'ring tests' to compare data obtained from the use of the optimised method(s) with a particular seed lot with those obtained from (any) existing biochemical analysis, and observed in field plots of the same lot. Whilst this comparison would apply primarily to the overall level of hybrid purity determined by the different approaches, it would also include the examination of any 'off-types'. This would form part of the assessment of an optimised operational system for seed certification.
(vi) To develop databases of varietal DNA profiles and potential applications of such techniques and databases to varietal registration.
This research thus promotes collaborative research and the development of harmonised and efficient systems for plant variety testing throughout the EU, as well as the availability of high quality varieties. It will provide tools for production of higher quality raw materials and increase the efficiency of the agri-industries within Europe. The DNA profiling systems can be used readily to trace varieties (and the products derived from them) throughout 'total quality management' systems, from propagative material to the end product, and contribute to the improvement of the quality of produce.

The work will develop, optimise and evaluate the use of two DNA profiling methods. Microsatellite and AFLP `Variety Test Sets' will be constructed, comprising a number of pairs of oligonucleotide primers revealing polymorphisms within a collection of reference varieties of maize, sunflower, oilseed rape and white cabbage. These microsatellite and AFLP Variety Test Sets will then be screened against a wide range of varieties, inbred lines and hybrids of different types for each of the four species. Comparative ring tests of two DNA profiling methods (sequence tagged site microsatellites and AFLP) for determining levels of hybrid purity in maize, sunflower, oilseed rape and white cabbage will then be organised.
Hybrid purity results obtained from the use of microsatellites and AFLP will be compared with those obtained from the use of existing biochemical (maize) and molecular (oilseed rape) methods of hybrid purity assessment and morphological assessment of field trials (maize, sunflower, oilseed rape).
Transportability of microsatellite and AFLP primers developed will be investigated for oilseed rape and white cabbage to other Brassica species. Consideration will be given to the development of operational systems for the use of microsatellites and AFLP for hybrid purity assessment in the context of seed testing and certification, in a range of different types of hybrid of each of the species. Databases of DNA profiling information will be compiled for varieties and hybrids for each of the species being considered.

Current situation/results:
1 Microsatellite Development
In oilseed rape, NIAB has identified some microsatellites using a LiCor automated sequencer. Others have been obtained from the literature, database searches and enriched library cloning. These have been evaluated for polymorphism and reliability. A total of 64 primer pairs were evaluated in total. From these, 11 were found to amplify successfully and were polymorphic in oilseed rape varieties.
For Sunflower, 25 putative microsatellite clones have been sequenced by NIAB; 14 of which show microsatellites or SSRs, but only four were suitable for primer design. Of these, two have been tested with a small number of varieties but no polymorphism was found. A further 2 microsatellites were obtained from the literature (Brunel, 1994) but found to be non-polymorphic. Of 14 microsatellites published on the WWW (Huestis, 1997) and tested, 12 amplify satisfactorily and 5 reveal some polymorphism.
For White cabbage, PRI (formerly CPRO) has tested 19 primer sets from the literature, the 10 sets designed from EMBL database sequences and the 36 sets designed on sequences obtained from PRI enriched. The 12 most informative primer pairs, which gave good quality products with clear and readily scored profiles, were taken forward for further testing and evaluation.
Finally, for Maize, UHOH has developed maize microsatellites by thorough examination of the publicly available primer sequences. SSR1.html
Techniques (PCR, MetaPhor gel electrophoresis) for analysing 19 selected microsatellites were also established; 10 fluorescence-labelled maize primer pairs have been employed with the ALF.

2 AFLP Development.
Oilseed rape: PGS tested a large set of AFLP primer combinations on two Brassica napus lines. 79 combinations yielding high-quality patterns were then evaluated on the set of ten representative lines from different countries. Two additional primer combinations (E32-M48 and E32-M62) were tested on the ten lines. The combination of MseI primers with EcoRI primer E32 appear to yield the highest average level of polymorphisms. PGS decided to use this primer as one common EcoRI primer in combination with six MseI primers for further development. The six combinations can be multiplexed cost-effectively in two sets of three combinations in the ABI377-assisted non-radioactive assay.
Sunflower: the AFLP methods have been adapted by NIAB to utilise the LiCor for analysis. Evaluation of potential primer combinations was initially by assessment of band patterns produced using DNA pooled from 10 plants each of two parent lines and their hybrid variety. Other criteria for selecting primer combinations for further analysis were applied at this stage, including uniform intensity and distribution of bands across the gel. Sixty four primer combinations have been tested on this basis - Eco RI primers with the following selective bases; AAC, AAG, ACA, ACC, ACG, ACT, AGC & AGG, in all combinations with the following Mse I primers; CAA, CAC, CAG, CAT, CTA, CTC, CTG & CTT. From this initial investigation, 32 combinations were selected for polymorphism analysis.
White cabbage: two primer combinations were extensively tested by PRI and a third primer combination chosen from the previously selected fourteen optimal ones, based on band quality, degree of polymorphism and reproducibility in a test set of OPVs as well as hybrid varieties with their parent lines. The three AFLP primer combinations chosen are E39 + M36, E39 + M38 and E37 + M33.
Maize: AFLP development has been completed by UHOH using P33-labelled primers. In total, 10 maize inbred lines were evaluated with 10 AFLP primer combinations.

3.1 Variety Test Set Development, microsatellites.
In oilseed rape, the most useful microsatellites identified above by NIAB were taken and used with 50 varieties to assess polymorphism levels. Varieties included winter, spring and forage types. 64 primer pairs were evaluated for size, quality and polymorphism and 11 were found to be polymorphic and reliable, but four showed only two alleles and a fifth was incompatible for multiplexing. Bn9, 12, 26 and 59 can be multiplexed and amplified successfully with a 600 anneal. Bn 9 and 26 overlap in size but are labelled with IR 700 and 800 dyes respectively. Clone 33 and MB5 can be multiplexed at 550 and overlap in size, but the use of different dyes avoids detection problems. The latter markers cannot be run on the same gel because they are similar in size to Bn9 and 26.
In sunflower, the available microsatellite primers that were found by NIAB to show polymorphism between and within varieties were tested using the LiCor.
In white cabbage PRI took the twelve most informative primer sets from Task 1, that is the sets that amplified unambiguously scorable products with the highest degree of polymorphism. These were combined into three multiplexes of four microsatellites each. One primer set was redesigned to obtain a PCR product size that fits into the multiplex C.
UHOH have developed a variety test set (VTS) for maize consisting of 5 primer pairs. A major criterion for selecting the primer pairs was the size range of alleles detected within 57 maize inbred lines: phi064 (Chromosome 1) covers the range of 75 to 113 bp, phi073 (Chromosome 3) 90 to 102 bp, phi077 (Chromosome 6) 123 to 149 bp, phi057 (Chromosome 7) 153 to 157 bp, and phi071 (Chromosome 10) 207 to 214 bp. Non-overlapping allele ranges are important for the future establishment of multiplex reactions. In order to cover most of the genome, microsatellites were selected from different chromosomes and with high PIC values to increase the chance of detecting impurities.

3.2 Variety Test Set Development, AFLP.
For oilseed rape, the AFLP variety test set for oilseed rape has been defined by PGS. Assessing genome coverage obtained with the VTS was also considered. Both polymorphism and genome coverage were considered as a criterion for VTS selection. To assess the genome coverage provided by the informative markers generated by the six primer combinations these markers were integrated in a published linkage map of B.napus. It was developed using a doubled-haploid mapping population derived from a cross between the spring variety Stellar and the winter variety Major and contains RFLP markers. This map has been aligned (unpublished information) with two other maps. The maps show a high degree of co-linearity. They are considered as reference maps for B.napus. AgrEvo-PGS has accessed the Stellar x Major mapping population. The six AFLP primer combinations were run on this population and segregating AFLP markers scored, added to the existing marker data and included in a new linkage analysis. Nineteen linkage groups were obtained. The markers generated by the proposed VTS covered 82% of the B.napus genetic map. The following selective AFLP primer combinations have been selected as the variety test set for B. napus: E32 in combination with M47, M48, M50, M51, M61, M62.
NIAB analysed 32 primer combinations for sunflower which fulfilled the criteria for selection with a wide range of varieties to determine polymorphism levels and reproducibility. 8 hybrid cultivars were chosen, including varieties from within one breeding programme and from a number of unrelated programmes plus specialist material. The number of polymorphic bands detected was taken as a measure of informativeness, and the most informative primer combinations taken for further analysis. Among these eight varieties, good levels of polymorphism were produced by 12 primer combinations. There were a minimum of six polymorphic bands with up to 19 polymorphisms. However, there were some discrepancies between these results and the earlier screening, suggesting that further development work is required.
For cabbage, PRI selected the following AFLP primer combinations from Task 2: E39/M36, E39/M38 and E37/M33. Finally, in maize, UHOH investigated 51 maize inbreds with EcoRI/MseI and PstI/MseI AFLP primer combinations. Generally the number of informative bands was higher with the EcoRI/MseI primer combinations, so exclusively EcoRI/MseI AFLP primer combinations were employed. Statistical analyses of this data will facilitate selection of the most suitable primer combinations as a variety test set.

4 Evaluation of Variety Test Sets.
NIAB has evaluated six primer pairs in oilseed rape, comprising the microsatellite VTS using 50 varieties of various types. For each primer pair, the separation rate (defined as s = number of pairs of varieties separated / total number of pairs of varieties, as a percentage) was determined, both alone and in combination with others. Those primer pairs, which amplified multiple polymorphic bands (alleles), tended to give much higher levels of separation than those which have only a limited number of allelic states.
For example, the primer pair Bn12, which amplifies 3-6 differently sized bands per variety, gave a separation rate of 83.8%, whilst SLA2G showed only two alleles with a ratio of 47:3, giving a separation rate of 11.5%. Combining the data from more than one microsatellite increased the separation rates. When used in combination the two most polymorphic primers pairs, Bn12 and clone 33, provided a separation rate of 95.1%. This could be increased further still by the addition of data from other primer pairs, such that for instance Bn12, clone 33, Bn26 and Bn27 in combination achieved a 97.8% separation.
PGS has started the AFLP variety test set screening for oilseed rape on 33 varieties representing a wide range of germplasm. A final evaluation of the variety test set will be done by calculating the marker index for each primer combination based on the results of the screening of 33 lines. The marker index not only considers both the level of polymorphism and the frequency of the marker alleles in germplasm.
NIAB has tested the five most useful and polymorphic microsatellite primer pairs in sunflower in a range of varieties. These are now being evaluated with a series of parent and hybrid lines. In white cabbage the multiplexes of microsatellite primer sets from Task 3 have been tested by PRI on a series of 70 hybrid varieties of diverse types using bulks of four plants from each variety. The multiplexes could distinguish all 70 varieties from each other. This level of discrimination was accomplished using only two of the multiplexes, and in several of the variety bulks more than two alleles for some of the primer sets were found. B. oleracea being essentially a diploid species, this implies variability at these loci between the individuals of the bulk sample, in turn suggesting variability of the parent lines at these particular markers.
The AFLP variety test set evaluation in white cabbage was performed by PRI on the same 70 hybrid varieties as have been used in the microsatellite analysis above, using the three primer combinations from Task 3. All varieties could be distinguished from each other. In this case, this could already be achieved using only either primer combination E39/M36, which produces 53 polymorphic bands, or E39/M38, which produces 47 polymorphic bands. Primer combination E37/M33, which produces the lowest number of polymorphic bands (40) was by itself not able to distinguish all varieties.
UHOH has completed variety test set evaluation for maize microsatellites using the five primer pairs selected above, plus an additional 14, to investigate 55 maize inbred lines. This information has been used to confirm selection of the VTS. AFLP data obtained in Task 2 using 10 AFLP primer combinations for 10 maize inbreds, as well as those obtained using 51 maize inbreds and 8 AFLP primer combinations, will be analysed within the next months to select, confirm and evaluate the most appropriate AFLP VTS. Microsatellites appear to be the preferable marker type compared to AFLPs for the second half of this project in maize. Especially for highly polymorphic species like maize, where many markers exist and multiplexing is possible, microsatellites should detect impurities or heterogeneities in seed lots quicker and with lower costs than AFLPs.

5 Hybrid Purity Assessment - Ring Tests.
In oilseed rape, two hybrid varieties were analysed by NIAB and their parental lines with four of the VTS microsatellites. Both hybrids showed a range of off-types, the origin of which appeared to be due to heterogeneity among the parental lines, contamination with stray pollen or seed lot contamination. Maternal selfing could be ruled out with the majority of cases. Parental lines showed a range of off-type contamination, from uniform to 20% off-types. This is presumably a result of the breeding process, which does not focus on homozygosity at non-morphological loci.
NIAB also tested sunflower varieties and their parental lines, using the microsatellite VTS. These analyses are currently being undertaken.
In white cabbage PRI have chosen two varieties from the 70 analysed in Task 4 to screen more extensively using two of the microsatellite multiplexes. From >100 individuals of each variety, variety 1, nine selfings could be shown. In both varieties, a few unexpected alleles were also found, indicating some variability in the parent lines used for the production of these hybrids. In one case for variety 1, a possible "outside" pollination could be inferred. Thus it is already possible to discern all relevant aspects of genetic impurities in hybrid varieties using a relatively small set of variable microsatellite markers, without having to refer to the parent lines. The selfings inferred by the use of microsatellites were also completely corroborated by the AFLP data. The possible "foreign" pollination inferred by microsatellite analysis was also corroborated.
LUFA used maize supplied by UHOH comprising six hybrids and parent. In parallel, the respective hybrids and parent lines have been grown at UHOH to produce leaf material from 50 individuals per hybrid and 20 individuals per parent line. The results will be compared with those obtained using the microsatellite and AFLP VTS by UHOH.

6 Comparison of Hybrid Testing Data - maize/oilseed rape.
Maize samples supplied by UHOH are also being used to compare results from biochemical (zein and isozyme analysis) and molecular analysis (LUFA).

Standardisation of the method and its simplification enables a direct comparison of different hybrids or inbred lines. The optimised method is being applied to the samples supplied by UHOH. Method development indicates the possibilities for computer-based identification of samples.

7 Comparison of Hybrid Testing Data - field trials.
The oilseed rape field trial has been planted at NIAB, comprising 5 hybrid varieties and 5 parental lines (one line is the female parent of three of the hybrids and not all parent lines of all hybrids are represented). Additional plots were drilled consisting of hybrid varieties deliberately contaminated with known percentages of other varieties or parental lines. DNA sampling of the plots is now complete and gel analysis is underway. Problems in purity have been noted for by INSPV for sunflower in five commercial hybrids and their inbred lines. Androsterile plants have been observed in some commercial varieties. This will facilitate provision of mixtures in the trial at INSPV next year. There are five commercial hybrids of sunflower being tested, together with the male and female lines. Maize samples (mixtures of six varieties and varying proportions of their female lines) have been prepared and supplied by UHOH and sown at two trial sites in Spain. The results from the two sites will be compared.

8 Transportability of cabbage and oilseed rape microsatellite and AFLP primers.
Three microsatellite multiplexes comprising the white cabbage VTS have been tested by PRI against four Savoy cabbage, two red cabbage, nine cauliflower and four Brussels sprouts hybrid varieties. The microsatellites worked well in these crop types, with good amplification and some new alleles being noted. All varieties could be distinguished from each other, using only two of the multiplexes. The crop types also formed groups that could clearly be distinguished from each other. Transportability of the microsatellites was not limited to within Brassica oleracea alone, for two of the multiplexes tested have also worked in twelve oilseed rape varieties that were obtained from partner 1.


National Institute of Agricultural Botany (NIAB)
Huntingdon Road
UK-CB3 0LE Cambridge
Tel: +44 1223 34 22 72
Fax: +44 1223 27 76 02


  • Albrecht MELCHINGER
    Universität Hohenheim (UHOH)
    Fruwirthstr. 21
    D-70599 Stuttgart
    Tel: +49 711 459 23 34
    Fax: +49 711 459 23 43

  • Bart LAMBERT
    Plant Genetic Systems nv (PGS)
    Jozef Plateaustraat 22
    B-9000 Gent
    Tel: +32 9 235 84 85
    Fax: +32 9 224 06 94

  • Lioba Friese
    Staatliche Landwirtschaftliche Untersuchungs- und Forschungsanstalt Augustenberg (LUFA)
    Nesslerstr. 23
    D-76227 Karlsruhe-Durlach
    Tel: +49 721 946 81 00
    Fax: +49 721 946 82 09

  • Ben VOSMAN
    PRI - Agrotechnological Research Institute
    Droevendaalsesteeg 1
    P.O. Box 16
    NL-6700 AA Wageningen
    Tel: +31 317 47 69 80
    Fax: +31 317 41 59 83

    Jose Abascal 4
    E-28003 Madrid
    Tel: +34 91 347 41 62
    Fax: +34 91 347 41 68
  Search Top