Experimental

This study was part of the European Union co-funded RIBESCO project aimed at establishing a core collection for the Northern European Ribes germplasm (Karhu et al., Reference Karhu, Antonius, Kaldmäe, Pluta, Rumpunen, Ryliskis, Sasnauskas, Schulte, Strautina and Grout2007). The project partners (P0, P1, P2, P3, P4, P5, P7, P8 and P9, abbreviations as indicated in the author affiliations) selected their own core collection candidate accessions for microsatellite (simple sequence repeat (SSR)) marker analysis. The total number of analysed accessions was 846, including 400 black currants (Ribes nigrum), 202 red and white currants (Ribes rubrum group), 242 gooseberries (Ribes uva-crispa) and 2 jostaberries (R. nigrum × R. uva-crispa). Since some partners did not have DNA analysis facilities available, they outsourced these services. The Tallinn University of Technology ran the analysis for partners P1, P3 and P4, and the University of Reading ran the analyses for P9. All the other partners analysed their own samples.

Six previously published markers, RJL-2, RJL-5, RJL-6, RJL-7, RJL-10 and RJL-11, were chosen for this study on the basis of their reported polymorphism in black currants (Brennan et al., Reference Brennan, Jorgensen, Woodhead and Russell2002). To be feasible for PCR amplification in our study, the reverse primer sequences of these markers required a reverse antisense translation (Table 1). In red and white currants and gooseberries, the markers RJL-5, RJL-7 and RJL-10 did not amplify any products of the expected size, or the resulting alleles were monomorphic in the test cultivars. Four new microsatellite markers were developed by partner P0, in accordance with the protocol presented by Korpelainen et al. (Reference Korpelainen, Kostamo and Virtanen2007), and named MTT-5, MTT-7, MTT-9 and MTT-32 (Table 1). Sequences for the markers were obtained from two accessions in the Finnish national germplasm collection from gooseberry ‘Ri532 – unknown cultivar’ for MTT-5, MTT-7 and MTT-9, and from white currant ‘Hele’ for MTT-32.

Table 1

Primer sequences and observed size ranges

There was no attempt to standardize the laboratory protocols. Each laboratory received leaf samples of the nine standard cultivars (five black currants, three red and white currants, and one gooseberry) from partner P1. Two gooseberry cultivars were added as optional standards for the laboratories focusing on gooseberry genotyping. Every laboratory optimized their own PCR amplification protocols for each primer pair in order to produce, for the standard genotypes, amplicons close to the size indicated by the work of partner P0. The results of the analyses of the standard cultivars were fairly comparable between those laboratories that used capillary electrophoresis (MegaBACETM-1000 DNA sequencer or ABI 3130 Genetic Analyzer) for allele detection. Deviations in the detected sizes in a particular allele of the same standard genotype ranged from 0 to 7 bp, depending on the locus. When partner P5 used Agilent 2100 Bioanalyzer for genotyping, their analysis results turned out to be considerably different when compared with the others, mainly due to the detection of additional amplification products. Later on, P5 re-analysed all their samples using capillary electrophoresis (CEQ 8000, Beckman-Coulter sequencer), which improved data compatibility.

After optimizing the protocols, each laboratory ran their set of candidate accessions. The resulting data were first analysed separately for each partner, and the diversity results were used as one of the inclusion criteria for the core collection selection. The data from different laboratories were subsequently combined by P0. To calibrate the results, conversion factors were calculated for each locus, using raw data (with the accuracy of two decimals) from the standard genotype analyses. Even after the calibration, the variation in the produced alleles was still slightly different across the separate analysis locations, which was likely to add some artificial diversity. However, such artificial variation probably exists in identical amounts in both candidate and core accessions, and therefore these two groups can be compared with each other. Allele frequencies, number of alleles per locus, genetic distances between accessions and principal coordinates analysis (PCoA) via distance matrix with data standardization were calculated using GENALEX 6 software (Peakall and Smouse, Reference Peakall and Smouse2006). Due to technical difficulties, partner P9 succeeded in analysing only two of the shared microsatellite loci, and their dataset had to be excluded from the analysis presented here.

The results showed that the selected core collection covered adequately the overall diversity of the analysed accessions (Fig. 1). A PCoA analysis of the selected core collection and all the non-selected candidate accessions for black currants is presented in Fig. 1. In the selected black currant core collection, 109 of the total of 171 detected alleles, i.e. 64%, were present. The corresponding result for red and white currants was 57 of the 71 alleles (80%), and for gooseberries 68 of the 111 alleles (61%).

Fig. 1

PCoA of the RIBESCO black currant core collection (Core) and the non-selected candidate cultivars (Candidate).

In addition to the results of genetic diversity analysis, each partner selected accessions to the core collection on account of other criteria, such as agronomical and culture historical value, yield quality, as well as pest and disease resistances. Altogether, 79% of the black currants, 82% of the red and white currants, and 79% of the gooseberries in the core collection were genotyped with the microsatellite markers. The accessions included in this collection are varieties or breeding lines of commercially important species of the genus. The final list of the RIBESCO core collection accessions can be found in the ECPGR Ribes and Rubus Database (http://www.ribes-rubus.gf.vu.lt/), and the detailed results of the SSR analyses are given in Supplementary Tables S1–S3 (available online only at http://journals.cambridge.org).