Gender differences in CF phenotype associated with low bone mineral density

Adolescents and adults with CF have lower bone mineral density (BMD) than normal, but its relationship with phenotype is not well understood. Point FEV1% predicted (FEV) and rate of change of FEV are biased estimates of disease severity, because progressively older subjects represent a selected survivor population, with females at greater risk of death than males. To investigate the relationship between BMD and phenotype we used an index (predicted age at death) derived from Bayesian estimates of slope and intercept of FEV, age at last measurement and survival status. Predictive equations for the index were derived from 97 subjects (78 survivors) from the RCH CF clinic, and applied to a group of 102 comparable subjects who had BMD measured, classified as having‘mild’ ()75th), ‘moderate’ (25– 75th), or ‘severe’ (-25th centile) phenotype. Total body (TB) and lumbar spine (LS) BMD z-scores (Z) were compared, adjustingfor gender effects, using 2-way ANOVA. Annual mean change in FEV segregated, as expected, according to phenotype, ‘severe’ (ns25), ‘moderate’ (ns51) and ‘mild’ (ns25) y3.01(y3.73 to y2.30)%, y0.85(y1.36 to y0.35)%, 2.70(1.92 to 3.46)%, respectively, with no gender difference. LS and TB BMDZ were different in each phenotype (P-s 0.002), LS BMDZ for ‘severe’, ‘moderate’ and ‘mild’ y1.63(CI: y2.07 to y 1.19), y0.86(CI: y1.17 to y0.55), y0.06(CI: y0.54 to 0.41). Males had lower LS BMDZ than females overall (y1.22 (CI: y1.54 to y0.91) vs. y0.48(CI: y 0.84 to y0.12) Ps0.002). In the ‘severe’ group, males had lower TB BMDZ and LS BMDZ (PF0.002). Low BMD is associated with ‘moderate’ and ‘severe’ phenotypes, with relative preservation in females in the ‘severe’ group. Female biology (reproductive fitness) might promote resistance to bone resorption at a critical level of BMD loss.

Evaluating plant breeding strategies by simulating gene action and dryland environment effects

Functional genomics is the systematic study of genome-wide effects of gene expression on organism growth and development with the ultimate aim of understanding how networks of genes influence traits. Here, we use a dynamic biophysical cropping systems model (APSIM-Sorg) to generate a state space of genotype performance based on 15 genes controlling four adaptive traits and then search this spice using a quantitative genetics model of a plant breeding program (QU-GENE) to simulate recurrent selection. Complex epistatic and gene X environment effects were generated for yield even though gene action at the trait level had been defined as simple additive effects. Given alternative breeding strategies that restricted either the cultivar maturity type or the drought environment type, the positive (+) alleles for 15 genes associated with the four adaptive traits were accumulated at different rates over cycles of selection. While early maturing genotypes were favored in the Severe-Terminal drought environment type, late genotypes were favored in the Mild-Terminal and Midseason drought environment types. In the Severe-Terminal environment, there was an interaction of the stay-green (SG) trait with other traits: Selection for + alleles of the SG genes was delayed until + alleles for genes associated with the transpiration efficiency and osmotic adjustment traits had been fixed. Given limitations in our current understanding of trait interaction and genetic control, the results are not conclusive. However, they demonstrate how the per se complexity of gene X gene X environment interactions will challenge the application of genomics and marker-assisted selection in crop improvement for dryland adaptation.