Leaf Rubisco turnover variation in a perennial ryegrass (Lolium perenne L.) population : analysis of quantitative trait loci, implications for productivity, and potential for manipulation : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Plant Science, Massey University, Institute of Natural Resources, College of Sciences, Palmerston North, New Zealand
The Grasslands II perennial ryegrass (Lolium perenne L.) mapping population comprising F1 progeny with the two parents (one plant each from the cultivars Samson and Impact) (Crush et al., 2007) was used to detect putative quantitative trait loci (QTL) for leaf Rubisco turnover and herbage yield traits. Rubisco turnover was described by three mathematical parameters: d (maximum Rubisco content), g (time of d) and f (a measure of curve width). All three parameters exhibited continuous variation among the F1 progeny. Sixteen QTL were detected, seven for Rubisco turnover and nine for herbage yield traits. Support interval overlap on linkage group (LG) 1 and close location on LG2 for plant dry weight (DW) QTL in this study and in a previous analysis (Sartie, 2007) of the same mapping population suggests DW QTL stability across environments. Some QTL identified by Sartie (2007) were not re-confirmed in this study, but new QTL were identified. This suggests genotype x environment interaction generated by variable expression of genes in different environments. Clusters of QTL with overlapping support intervals were found on LG2 and LG5. The cluster on LG2 included QTL for herbage yield traits leaf lamina length (LL), tiller number (TN), productivity index (PI) and DW. The cluster on LG5 included QTL for DW, PI, TN, and d. These two regions offer potential for plant breeding applications. Apart from the QTL for d on LG5, there was no co-location of Rubisco turnover and herbage yield QTL. However, principal component analysis indicated plants with lower d tended to have higher DW; thus Rubisco turnover effects on plant productivity may relate to energy cost of Rubisco synthesis rather than photosynthetic capacity. DW was generally unrelated to f and g; therefore, hypothesised nitrogen use inefficiencies arising from premature Rubisco degradation, or retention of Rubisco at leaf senescence, were not confirmed. LG5 and LG7 on which QTL for d were located have conserved syntenic regions with rice chromosomes 8 and 9 where QTL for Rubisco content at different stages during heading were mapped by Ishimaru et al (2001a).