Narrow-sense heritability and among-family and within-family variance components were estimated for antioxidant activity (AA), total phenolic content (TPH), and anthocyanin content (ACY) in blueberry (Vaccinium L. sp.) fruit. AA, TPH, and ACY were determined in the parents and in 10 offspring from each of 20 random crosses for each of 2 years at Becker, Minn. Offspring-midparent regression analysis provided combined-year heritability estimates of 0.43 ± 0.09 (P ≤ 0.0001) for AA, 0.46 ± 0.11 (P ≤ 0.0001) for TPH, and 0.56 ± 0.10 (P ≤ 0.0001) for ACY. Analyses of variance delineated variation among and within families for AA, TPH, and ACY (P ≤ 0.001). Year-to-year variation in the means for all offspring genotypes was not significant for AA or TPH, but there were changes in rank between years for families and for offspring within families for these traits. Year-to-year variation in the mean for all offspring genotypes was significant for ACY, but rank changes were observed only among offspring within families, not among families. In total, 18 of 200 offspring from 7 of the 20 crosses were transgressive segregants for AA, exceeding the higher parent of the cross by at least two sds. Estimates of variance components showed that variation among families accounted for 24% to 27% of total variance for the three traits. However, variation within families was greater than that among families, accounting for 38% to 56% of total variance for the three traits. These results suggest that increasing antioxidant activity in blueberry through breeding is feasible, and that the breeding strategies utilized should exploit the large within-family variation that exists.
Ann Marie Connor, James J. Luby, and Cindy B.S. Tong
Ann Marie Connor, James J. Luby, and Cindy B.S. Tong
Variation in antioxidant activity (AA), total phenolic content (TPH), and total anthocyanin content (ACY) was examined in 1998 and 1999 in fruit of 52 (49 blue-fruited and 3 pink-fruited) genotypes from a blueberry breeding population. The species ancestry included Vaccinium corymbosum L. (northern highbush blueberry), V. angustifolium Ait. (lowbush blueberry), V. constablaei Gray (mountain highbush blueberry), V. ashei Reade (rabbiteye blueberry), and V. myrtilloides Michx. (lowbush blueberry). Using a methyl linoleate oxidation assay (MeLO) on acidified methanolic extracts of the berries, a 5-fold variation was found in AA in 1998 and a 3-fold variation in 1999 among the blue-fruited genotypes. Analyses of variance (ANOVA) revealed variation among genotypes (P < 0.0001) in single and combined years, regardless of inclusion of pink-fruited selections and adjustment for berry size. While mean AA of all genotypes did not change between the 2 years, ranking of some genotypes for AA changed significantly between 1998 and 1999. Of the 10 genotypes that demonstrated the highest AA in 1998, four were among the 10 genotypes that demonstrated highest AA in 1999. Similarly, of the 15 genotypes with the highest AA, 10 were the same both years. As with AA, mean TPH of all genotypes did not change between years and ANOVA demonstrated genotypic variation regardless of adjustment for berry size/weight or exclusion of pink-fruited selections. Changes in genotype rank occurred between years. The difference in TPH between lowest- and highest-ranking blue-fruited genotypes was ≈2.6-fold in both 1998 and 1999. Seven of the 10 highest-ranking genotypes were the same both years and TPH correlated with AA (r = 0.92, P < 0.01) on a genotype mean basis for combined years. ACY correlated less well with AA (r = 0.73, P < 0.01 for combined years). When genotypes were categorized into six groups according to species ancestry, V. myrtilloides and V. constablaei × V. ashei crosses ranked highest and second highest, respectively, for AA in both years. The groups comprised of V. corymbosum genotypes, V. angustifolium genotypes, and those with both V. corymbosum and V. angustifolium in their lineage were indistinguishable from each other. Samples from some of the genotypes were analyzed for oxygen radical absorbance capacity and ferric-reducing antioxidant power, and these aqueous-based antioxidant assays correlated well with the lipid emulsion-based MeLO (all r ≥ 0.90, P < 0.01). The three antioxidant assays may be equally useful for screening in a blueberry breeding program and the choice of assay may depend on the goal of the program and the resources available.
Ann Marie Connor, Chad E. Finn, and Peter A. Alspach
Antioxidant compounds absorbed from our diet are thought to have a role in preventing chronic diseases that result from oxidative damage. Berry fruit have high levels of antioxidants, and further increases in antioxidant activity (AA) might be possible through breeding. We determined the AA, total phenolic content (TPH), and fruit weight in 16 blackberry and hybridberry (Rubus L.) cultivars harvested in New Zealand and Oregon in 2002 and 2003, to assess genetic and environmental variation. Both AA and TPH varied significantly between years within location, but not among cultivars or between locations per se. However, cultivar interactions with both location and year within location contributed to variation in both variates. In contrast, both cultivar and location contributed to variation in fruit weight, but years within location did not. However, the cultivar × year within location interaction was significant for this trait. Variance component distributions confirmed that cultivar and location effects together contributed little (<20%) to the total variation in either AA or TPH, while cultivar × environment interactions accounted for >50% of total variation in these traits. Cultivar and location effects together contributed ≈70% of the total variation observed in fruit weight. Phenotypic correlations were significant between AA and fruit weight (r = -0.44), and between TPH and fruit weight (r = -0.51). When adjusted for fruit weight, analyses for AA and TPH demonstrated that cultivar effects approached significance (P = 0.06) and accounted for ≈25% of total variance, while location effects accounted for none. Although the cultivars in this study had diverse interspecific backgrounds, utilization of various Rubus species in blackberry and hybridberry breeding is not uncommon, and our results demonstrating significant cultivar × environment interaction for AA and TPH should be applicable to breeding for high AA genotypes.
Ann Marie Connor, James J. Luby, Cindy B.S. Tong, Chad E. Finn, and James F. Hancock
Dietary antioxidants may have a role in preventing some of the chronic diseases in humans resulting from free radical oxidation of lipids and other cellular components. Blueberries (Vaccinium L. sp.) are considered one of the best fresh fruit sources of antioxidants, and there is the potential to increase the antioxidant activity further through breeding. Thus, the variability of fruit antioxidant activity (AA) was examined among a set of 16 highbush and interspecific hybrid cultivars grown at locations in Minnesota (MN), Michigan (MI), and Oregon (OR) over 2 years (1998 and 1999) to determine effects of genotype, year, and location. Nine cultivars were common to all three locations in both years. Antioxidant activity, total phenolic content (TPH), and total anthocyanin content (ACY), were determined in triplicate samples from each genotype. Cultivars differed significantly (P ≤ 0.05) in AA, TPH, and ACY both within and over locations. The single location mean AA for all cultivars changed significantly between the 2 years in OR and in MI, while the single location mean for TPH differed between the 2 years in MN and MI. Changes in cultivar rank were significant for AA, TPH, and ACY between years within each location. Significant changes in rank for TPH and ACY were also noted between pairs of locations as well. Pearson's correlation for AA (based on cultivar means) appeared highest between MN and OR (r = 0.90) and MN and MI (r = 0.69) in 1998; correlations between locations for the combined years were 0.74 for MN and OR, 0.55 for MN and MI and 0.45 for MI and OR. For the group of nine cultivars, AA correlated well with TPH within each location, with r ranging from 0.67 to 0.95 for data from individual and combined years. Correlation of AA with ACY at each location was lower than that for AA with TPH, in both individual and combined years. This study demonstrates significant genotype× environment interaction for AA in blueberry.
M. Joseph Stephens, Jessica Scalzo, Peter A. Alspach, Ron A. Beatson, and Ann Marie Connor
Breeding for high yield is a major objective of most small fruit breeding programs worldwide. In recent years, research associated with health benefits of berry fruit has resulted in some breeding programs looking for material with improved health properties with the goal of incorporating these into genotypes with high yield and other favorable agronomic characters. In this study, we estimated variance components, heritabilities, and phenotypic and genotypic correlations for yield components and phytochemicals [total phenolics (TPH), antioxidant activity (FRAP), and total anthocyanins (TACY)] from 828 genotypes in a red raspberry (Rubus idaeus) factorial mating design based on 42 full-sib families derived from seven female and six male parents harvested in one season in New Zealand. Narrow sense heritability estimates varied from moderately low [0.23 for percent budburst (PCBB)] to high [0.82 for mean berry weight (BWT)]. Highest genotypic correlations with total yield (TYLD) were found for mean cane length (0.60) followed by moderate correlations for PCBB (0.38), BWT (0.34) and fruit number per unit lateral (0.31). For these four components, the correlation between the product of the empirical breeding values (eBV) and TYLD was as good (0.67) as the correlation between the product of the seven yield components measured (0.68). Moderately high negative genotypic correlations were found between TYLD and TPH (–0.67), FRAP (–0.68), and TACY (–0.64), suggesting that breeding for high-yielding genotypes may result in reduced phytochemical levels. A pigment-deficient R. parvifolius × R. idaeus hybrid derivative parent (R. parv deriv) had the most influence on berry weight as a yield component. Removal of the R. parv deriv progeny from the analysis increased the contribution of berry weight to total yield. Heritability estimates were generally lower when the R. parv deriv crosses were omitted, particularly for number of canes BWT, TYLD, and the phytochemicals. Implications of R. parv deriv crosses on these results are discussed. Of the 828 genotypes, only five had an eBV for total yield greater than that of the best yielding parent, whereas 49 had larger berries than the best parent. None of the 828 genotypes had an eBV for yield components or phytochemicals less than that of the lowest parent, whereas 13 had a higher eBV associated with phytochemicals than that of the highest parent.
Ann Marie Connor, Tony K. McGhie, M. Joseph Stephens, Harvey K. Hall, and Peter A. Alspach
We determined variance components and narrow-sense heritability estimates for total and individual anthocyanin (ACY) content and antioxidant activity (AA) in fruit from 411 genotypes in a red raspberry (Rubus idaeus L.) factorial mating design based on 42 full-sib families derived from seven female and six male parents, harvested in 2002 and 2003. Within half-sib family total ACY content ranged from ≈1-60+ mg/100 g fruit in both seasons. The four major ACYs quantified by high-performance liquid chromatography also showed wide ranges each year. Female and male parent contributions to variation in total and individual ACYs were significant (P ≤ 0.001) in combined year analysis, and together accounted for 29% to 48% of the total variation. A substantial proportion of the female contribution was attributed to the use of a pigment-deficient R. parvifolius L. × R. idaeus hybrid derivative as a female parent. Female × male interaction was nonsignificant and contributed negligibly to total variance. Year effects accounted for <2.5% of variation in ACYs and were only marginally significant. Year interactions were negligible. Within family variation (among plots and within plot) accounted for ≈50% of the variation in total ACY and 62% to 69% of the variation in individual ACYs. Combined year narrow-sense heritability estimates were high (h 2 = 0.54-0.90 for individual ACYs, 1.00 for total ACY) among all factorial genotypes, but moderate when the progeny of the R. parvifolius derivative were excluded (h 2 = 0.45-0.78 for individual ACYs, 0.74 for total ACY). The latter estimates are applicable to breeding programs in which pigment-deficient genotypes are rarely or never used in breeding. Parental main effects were significant for AA, together accounting for 19% of total variance; female × male interaction was nonsignificant. Year effects were marginally significant and year interactions nonsignificant; together these sources of variation contributed <2% of total variation in AA. The majority of AA variation was found within- and among-plots within family. The phenotypic correlation between AA and total ACY was r = 0.53, and ranged from r = 0.21-0.46 between AA and individual ACYs; genetic correlations between AA and the ACYs were similar to the phenotypic correlations, suggesting predominantly additive genetic effects accounted for the phenotypic correlations. Linear modelling for AA based on individual ACYs and their interactions explained ≈0.53 of AA variation, substantially less than that explained by total phenolic content (R 2 = 0.88). Our results show substantial variation and moderate to high narrow-sense heritability estimates for red raspberry ACYs, but ACY content and profile information are ineffective proxies and predictors for AA in red raspberry fruit.
Ann Marie Connor, Chad E. Finn, Tony K. McGhie, and Peter A. Alspach
Dietary anthocyanins (ACYs) may give health benefits through their antioxidant activity (AA) or other physiological effects. We examined AA and ACY profiles and contents in 16 blackberry and hybridberry (Rubus L. species) cultivars harvested in 2002 and 2003 in New Zealand and Oregon. Total ACY content varied widely among cultivars harvested from a single site in a single year (e.g., from 58 to 343 mg/100 g fruit Oregon in 2003). For the 12 cultivars common to both sites and years, cultivar and year within location significantly affected total ACY content, accounting for 40% and 10% of total variation, respectively. Cultivar interactions with both location and year within location were also significant, together accounting for 39% of variation. Cyanidin 3-O-glucoside and cyanidin 3-O-rutinoside were identified in all cultivars in both locations in at least 1 year. Compared with total ACY, cultivar accounted for more variation in these two ACYs (63% and 92%, respectively), while cultivar interactions together accounted for a smaller, but statistically significant, proportion of variation (23% and 7%, respectively). Cyanidin 3-O-sophoroside and cyanidin 3-O-(2G-glucosylrutinoside) were identified in only four cultivars. Cultivar effects accounted for 64% and 76% of variation in these ACYs, respectively, while cultivar interactions together contributed 18% and 24%, respectively. For AA, cultivar effects were not significant, contributing 11% of variation; in contrast, year effect and cultivar × environment interactions were significant, contributing 22% and 55% of total variation, respectively. Based on cultivar means for all 16 genotypes, the phenotypic correlation between AA and total ACY was positive but lower than that between AA and total phenolic content (TPH) (r = 0.63 and 0.97, respectively). Combinations of individual ACYs did not provide higher correlations with AA. Thus, ACY profiles and content are not as useful as TPH as a proxy measurement for AA.
Ann Marie Connor, M. Joseph Stephens, Harvey K. Hall, and Peter A. Alspach
Variance components and narrow-sense heritabilities were estimated for antioxidant activity (AA), total phenolic content (TPH), and fruit weight in red raspberry (Rubus idaeus L.) fruit from offspring of a factorial mating design. Forty-two full-sib families utilizing seven female and six male parents were evaluated in each of two years in Motueka, New Zealand. In a single year, values within individual half-sib families ranged as widely as 25.3-79.4 μg·g-1 fruit for AA, 205-597 mg/100 g fruit for TPH, and 1.06-7.69 g for fruit weight. Analyses of variance for these three variates demonstrated significant parental source variation in both individual and combined year analyses. For AA and TPH, female parental effects accounted for ≈7% to 19% of total variation, while male effects accounted for ≈6% to 8%. A partially pigment deficient R. parvifolius L. derivative female parent accounted for some of these differences. Female × male parent interaction was not significant for AA and TPH and was marginally significant for fruit weight in combined year analysis. Year had a significant effect on the overall mean AA and TPH, but contributed less than genetic effects to the overall variation in all three traits. Interactions of year with genetic effects were not statistically significant for AA or TPH, indicating that between-year rank or scale changes among families were negligible. The largest proportion of variation was found within rather than among full-sib families. However, variation among plots within full-sib families accounted for 12% to 19% of total variation, indicating environmental differences accounted for some of the observed within-family variation in AA and TPH. Antioxidant activity and TPH were highly phenotypically correlated (r = 0.93); their genetic correlation (r = 0.59) implies that substantial additive genetic factors underlie the phenotypic correlation, but that nonadditive genetic or environmental influences are also important. Both AA and TPH were weakly negatively phenotypically correlated with fruit weight (r = -0.34 and -0.33, respectively), but the corresponding genetic correlations were close to zero. Thus, selection for both high AA or TPH and high fruit weight is possible. Narrow-sense heritability estimates based on variance components from combined year data were h 2 = 0.54, 0.48, and 0.77 for AA, TPH, and fruit weight, respectively. These estimates imply a rapid response to selection is possible.