Over the years, the CIMMYT/ICARDA durum project has made steady progress in breeding durum wheat for increased productivity, yield stability and grain quality. In 2002, major advances were made in identifying and genetically mapping both the quantitative trait loci (QTLs) which control two traits influencing the milling quality of durum wheat, and the QTLs that control protein content. In total, 13 QTLs were identified, and strong QTL x environment interactions were detected for protein content in particular. These QTLs will be used in marker-assisted selection for these traits, a process that is more efficient than conventional selection.

New QTLs for milling-quality traits identified and mapped

The grain of durum wheat (Triticum turgidum L. var durum) is used to produce economically important foods (such as pasta, couscous, and burghul). Therefore, grain qualities which aid in milling and processing are being selected for. To increase the efficiency of conventional empirical breeding efforts, the CIMMYT/ICARDA durum project has produced a genetic linkage map of durum wheat. In 2002, scientists focused on identifying and mapping quantitative trait loci (QTLs) for the milling-quality traits of test weight (TW, hecto-liter weight) and 1000-kernel weight (TKW). A high TKW is desirable for easy processing and milling; and, high TKWs and TWs are both associated with the production of high quality semolina.

CIMMYT/ICARDA Durum Wheat Breeder, Dr Miloudi Nachit (pointing to the screen) and his colleague score molecular markers linked to durum grain quality for use in the marker-assisted breeding program.

Fig. 4. Kernel weight scan for QTL main effect (top axis) and QTL x environment interaction (bottom axis). The 14 MDM chromosomes and g15 are shown from left to right (starting with short arm). Horizontal lines show thresholds with 5000 permutations. Arrows show positions of QTLs.

     The mean test weight for the recombinant inbred lines was 72.4 g (ranging from 66.6 g to 77.6 g), whereas the means for the parents ('Omrabi5' and T. dicoccoides600545) were 79.2g and 70.6 g, respectively. Test weights from all environments closely followed the normal distribution. The main chromosomal regions controlling the test weight in durum were found on chromosomes 7A and 6B. Across this 6B genomic region, Xgwm88a on 6BS was selected as the QTL for the test weight trait; this explained 9% of total variation in test weight. The other QTL identified for test weightXgwm60c on 7ASexplained 17% of the total variation observed. These findings are in agreement with earlier studies showing a high genotypic effect on test weight.  

dicoccoides600545//Omrabi5, and the genetic map constructed by researchers, will be used as the basis for the future detection of QTLs linked to other grain-quality traits, as well as to other agronomic, physiological, and biotic traits. The mapping population may also be of help when assessing environmental impacts on trait expression, as it is already genetically fixed.
     For TW and TKW, the QTLs determined here will be validated using populations with diverse genetic backgrounds and will be used in marker-assisted breeding. Consequently, researchers will be able to conduct selection at the genetic level, rather than having to rely on the phenotypic expression of these traits.

Protein-content QTLs and their inter-actions with different environments

Protein content is a major grain-quality trait controlled by a complex genetic system and strongly influenced by environmental factors. The major influences on grain crude protein concentration are environmental in nature (for example, rainfall and soil nitrogen levels). Triticum dicoccoides, a wild relative of wheat, is used as a source of genes for high protein content, so this species was used to breed a specific mapping population (Omrabi5/ Triticum dicoccoides600545//Omrabi5) for use in recent grain-quality studies. These studiespart of the CIMMYT/ICARDA Mediterranean durum programaimed to determine the QTLs linked to protein content in durum, and to assess the QTL x environment (QTL x E) effect. A genetic linkage map for the mapping population was constructed using three types of markers: microsatellites, AFLPs, and seed storage protein components. The recombinant inbred lines (RILs) derived from the mapping population were planted in 18 different environments; their grain protein contents were measured at harvest.