Page 180 - Breeding and regulatory opportunities, Renaud

Chapter 5

compound concentrations responded diferently. All compounds showed

genetic variation, but also a substantial proportion of variance components

were accounted for by high level interactions (Table 5.3; Supplemental Table

)
5.2. While M main efect was generally small, it had a substantial contribution

in three- and four-way interactions. In particular, many G×L×S×M interactions

were large relative to other variance components. This indicates that for

the phytochemicals, M did have an inluence on G, but that there were no

consistent patterns across locations and seasons that would have shown up

as signiicant G×M. Rather in each season and location, the paired organic and

conventional environments difered signiicantly from one another but each

situation was unique. In contrast to many comparisons between organic and

conventional production systems (De Ponti et al., 2012), it should be noted that

in our trials, yields averaged over the years did not difer signiicantly between

the organic and conventional management systems (Renaud et al., 2014).

Among the nine compounds, glucoraphanin was the most strongly inluenced

by genotype followed by lutein: supporting the indings of several other

broccoli studies where variation in concentrations for glucoraphanin (Brown

et al., 2002; Farnham et al., 2004; Charron et al., 2005ab) and lutein (Farnham

and Kopsell, 2009; Ibrahim and Juvik, 2009) was primarily due to genotype. For

γ-tocopherol, genotype was a large source of variation, but this compound was

equally inluenced by location and season (also found by Ibrahim and Juvik,

2009). For glucobrassicin and neoglucobrassicin the location was the largest

source of variation, but also L×S interaction was very inluential, particularly for

neoglucobrassicin, which is supported by Kushad et al. (1999) and Schonhof

et al. (2004). Jasmonic acid, a signal transduction compound in plants, is

up-regulated under conditions of plant stress, wounding, and herbivory.

Increased endogenous levels or exogenous application of this compound (or

methyl jasmonate) increases biosynthesis and transport of neoglucobrassicin

to broccoli lorets. This up-regulation was not observed for glucobrassicin

biosynthesis (Kim and Juvik, 2011). This could explain why neoglucobrassicin

was primarily under the control of Location and L×S interaction in our study.

Season was the largest variance component for δ-tocopherol and zeaxanthin,

which contrasts with the work of Ibrahim and Juvik (2009) who found genotype

had the largest inluence on these compounds, followed by genotype by

environment interaction although this study was constrained by the fact that

162