Metabolite Quantitative Trait Loci for flavonoids …...2020/03/20 · 98 A pseudo full-sibling F 1 population of 165 individuals obtained from a cross between the variety 99 ‘

Metabolite Quantitative Trait Loci for flavonoids provide new insights into the genetic architecture of 1

strawberry (Fragaria x ananassa) fruit quality 2

Marc Labadie1§, Guillaume Vallin1§, Aurélie Petit1,2, Ludwig Ring3, Thomas Hoffmann3, Amèlia Gaston1, Aline 3

Potier1, Wilfried Schwab3, Christophe Rothan1*, Béatrice Denoyes1* 4

1 INRAE, Univ. Bordeaux, UMR BFP, F-33140, Villenave d’Ornon, France 5

2 INVENIO, MIN de Brienne, 110 quai de Paludate, 33800 Bordeaux, France 6

3 Biotechnology of Natural Products, Technical University Munich, Liesel-Beckmann-Str. 1, 85354 7

Freising, Germany 8

§ These authors contributed equally to this work 9

10

* Corresponding authors: 11

Béatrice Denoyes 12

https://orcid.org/0000-0002-0369-9609 13

Email: [email protected] 14

UMR BFP – INRAE 15

71 avenue Edouard Bourlaux 16

33140 Villenave d’Ornon, France 17

Phone: +335 57 12 24 60 18

Christophe Rothan 19

https://orcid.org/0000-0002-6831-2823 20

Email: [email protected] 21

UMR BFP – INRAE 22

71 avenue Edouard Bourlaux 23

33140 Villenave d’Ornon, France 24

Phone: +335 57 12 25 32 25

26

(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 22, 2020. ; https://doi.org/10.1101/2020.03.20.000455doi: bioRxiv preprint

https://orcid.org/0000-0002-0369-9609

mailto:[email protected]

https://orcid.org/0000-0002-6831-2823

mailto:[email protected]

https://doi.org/10.1101/2020.03.20.000455

27

Abstract 28

Flavonoids are products from specialized metabolism that contribute to fruit sensorial (colour) and 29

nutritional (antioxidant properties) quality. Here, using a pseudo full-sibling F1 progeny previously 30

studied for fruit sensorial quality of cultivated strawberry (Fragaria x ananassa), we explored over two 31

successive years the genetic architecture of flavonoid-related traits using LC-ESI-MS (13 compounds 32

including anthocyanins, flavonols and flavan-3-ols) and colorimetric assays (anthocyanins, flavonoids, 33

phenolics, FRAP and TEAC antioxidant capacity). Network correlation analysis highlighted the high 34

connectivity of flavonoid compounds within each chemical class and low correlation with colorimetric 35

traits except anthocyanins. Mapping onto the female and male linkage maps of 152 flavonoid metabolic 36

QTLs (mQTLs) and of 26 colorimetric QTLs indicated co-localization on few linkage groups of major 37

flavonoid- and taste-related QTLs previously uncovered. These results pave the way for the discovery of 38

genetic variations underlying flavonoid mQTLs and for marker-assisted selection of strawberry varieties 39

with improved sensorial and nutritional quality. 40

41

Keywords. Fragaria x ananassa, fruit, specialized metabolism, flavonoids, anthocyanins, mQTL, colour, 42

antioxidant, nutritional quality 43

44


https://doi.org/10.1101/2020.03.20.000455

Introduction 45

Consumers are more and more aware that consumption of fruits and vegetables has long-term effects 46

on human health.(1,2,3). In this respect, small fruits including strawberry are an important source of 47

phytochemicals with proven health-benefits.(4-7). The most consumed small fruit worldwide is 48

cultivated strawberry (Fragaria x ananassa), which is issued from the hybridization few centuries ago 49

between F. chiloensis and F. virginiana.(8). Since then, cultivated strawberry has undergone intense 50

breeding activity for traits related to yield and cultural practices (9) but also for sensorial fruit quality 51

traits such as fruit weight and shape, shininess and colour and, more recently, flavor.(10). In the last 52

decades, with the increased interest in nutritionally active phytochemicals from fruits, major strawberry 53

fruit antioxidants such as ascorbic acid (vitamin C), ellagic acid and other polyphenolic compounds have 54

been studied more deeply.(3,5,6,11). 55

In contrast to ascorbate, for which QTLs have recently been mapped and candidate genes have 56

been identified (11), the genetic architecture of polyphenol-dependent nutritional traits remains poorly 57

known in cultivated strawberry. To date, detailed analyses of flavonoid composition have been 58

performed on both cultivated and wild diploid Fragaria vesca strawberry species.(12-14). In addition, 59

major strawberry flavonoid genes have been isolated and the activity of a few corresponding enzymes 60

have been characterized.(15-17). In the wild diploid strawberry species Fragaria nilgerrensis, a natural 61

mutation in the MYB10 transcription factor regulating anthocyanin biosynthesis is responsible for the 62

white fruit phenotype.(18). However, with a few exceptions such as the mapping of a peroxidase that 63

controls the trade-off between lignin and anthocyanins biosynthesis to a locus responsible for fruit color 64

variations (19), the extent to which natural genetic variations may control flavonoid content in 65

cultivated strawberry remains poorly known. Such studies are further complicated by the octoploid 66

status of cultivated strawberry (2n = 8x = 56) in which variations in a given trait at a single locus can be 67

controlled by up to eight homoeoalleles located on four linkage groups (20) corresponding to four 68

different subgenomes (F. iinumae, F. nipponica, F. viridis and F. vesca).(21). 69

To get more insights into the genetic control of flavonoid content, the first step is to isolate and 70

identify the most abundant flavonoids present in strawberry fruit and then to map the corresponding 71


https://doi.org/10.1101/2020.03.20.000455

metabolic Quantitative Trait Loci (mQTL) onto cultivated strawberry genetic map. State-of-the art 72

methods now allow the exhaustive analysis of flavonoids, which are derived from the phenylpropanoid 73

pathway.(22). Flavonoids (anthocyanins, flavonols, flavan-3-ols) found in strawberry fruit are antioxidant 74

molecules that have proven dietary health-benefits.(2,23,24). Flavonols in cultivated strawberry are 75

mainly glycosides of quercetin and kaempferol. Most common flavan-3-ols are catechin and epicatechin 76

as well as their monohydroxylated equivalents afzlechin and epiafzlechin and glycosylated derivatives. 77

Anthocyanins are mainly glycosides of pelargonidin and cyanidin.(25). In addition to their antioxidant 78

properties, anthocyanins are water-soluble pigments that give to the strawberry fruit its attractive 79

bright red (pelargonidin derivatives) to dark red (cyanidin derivatives) colour. Pelargonidin derivatives 80

are the most abundant anthocyanins in cultivated strawberry (19,23) while pelargonidin and cyanidin 81

derivatives are equally found in F. vesca.(14). Although several colour-related QTLs have been detected 82

(20), the genomic regions responsible for quantitative variations of the various flavonoid compounds 83

found in cultivated strawberry fruit have not been uncovered to date. 84

In this study, to further explore the genetic architecture of polyphenolic- and flavonoid-dependent 85

traits in cultivated strawberry, we analyzed over two successive years (with three repeats each year) a 86

pseudo full-sibling F1 progeny previously studied for the genetic control of fruit sensorial quality.(20). 87

The population also segregates for flavonoids (13 compounds from three chemical classes) and for 88

colour- and antioxidant-related traits (5 traits assayed by colorimetry). In total, we detected 152 mQTLs 89

for the flavonoid compounds, all of which were identified for the first time from cultivated strawberry, 90

and 26 QTLs for the colour- and antioxidant-related traits. The largest number of flavonoid mQTLs and 91

of colour- and antioxidant-related QTLs was detected on the LG VIa linkage group that also harbors 92

major fruit sweetness and acidity-related traits (20) and is therefore a likely target for the improvement 93

by Marker-Assisted Selection (MAS) of the sensorial and nutritional quality of strawberry fruit. 94

95

Material and methods 96

Plant materials and preparation. 97


https://doi.org/10.1101/2020.03.20.000455

A pseudo full-sibling F1 population of 165 individuals obtained from a cross between the variety 98

‘Capitola’ (CA75.121-101 x Parker, University of California, Davis, USA) and the advanced line ‘CF1116’ 99

([‘Pajaro’ x (‘Earlyglow’ x ‘Chandler’)], reference from Ciref, France) was developed. The two parents, 100

‘Capitola’ and ‘CF1116’, display many contrasting fruit quality traits.(20,26). For each of the two 101

consecutive study years (2010 and 2011), six cold-stored strawberry plants per genotype, which were 102

planted in 2009 and 2010, were grown in soil-free pine bark substrate under plastic tunnel with daily 103

ferti-irrigation and control of biotic stresses. Mapping population included a total of 165 individuals over 104

the two study years. 105

Within this progeny, 72 and 131 individuals including parents were respectively phenotyped in 2010 106

and in 2011. Fruits were harvested at red ripe stage, when red coloration of the fruit is homogeneous. 107

To ensure conformity of the fruit samples and to avoid undesirable effects on antioxidant and fruit 108

polyphenolic contents, several precautions were taken. Fruits were collected in the morning and those 109

showing abnormal shape and size, or injuries were not harvested. For each genotype, two harvests of 4-110

8 fruits each were performed at the peak of fruit production (8-16 fruits in total). Fruits were 111

immediately frozen in liquid nitrogen and stored at -80°C, in order to avoid degradation of antioxidants 112

and polyphenolics. They were then ground into a fine powder in liquid nitrogen and frozen powders 113

from two successive harvests were pooled (volume/volume) and blended. Three samples of pooled 114

frozen powder were used for the various extractions. 115

Antioxidant measurements 116

Extraction of hydrophilic antioxidants was adapted from Capocasa et al. (27). For each sample, 0.7 117

+/- 0.02 g of fruit powder was dissolved in the dark in 7 ml of methanol/water (80/20 v/v) extraction 118

solution. The mixture was then vortexed for 30 s, agitated at 160 rpm for 30 min and centrifuged for 10 119

min at 4,500 g. All steps were performed at ~5°C in the dark. Aliquotes of 250 µl of supernatant (fruit 120

extract) were then stored at -80°C in microtubes until colorimetric analyses. 121

Five antioxidant-related traits were measured on fruit extracts by colorimetric assays including 122

anthocyanin content (ANTHc), flavonoid content (FLAVc), total phenolic content (PHENc) and the two 123

antioxidant-related traits FRAP (Ferric Reducing Antioxidant Power) and TEAC (Trolox Equivalent 124


https://doi.org/10.1101/2020.03.20.000455

Antioxidant Capacity). ANTHc was estimated according to the protocol of Villarreal et al. (28). Results 125

are expressed as mg pelargonidin-3-glucoside equivalents/100 g fresh weight. FLAVc was estimated by 126

the method of Dewanto et al. (29). Results are expressed as µg catechin equivalents/g fresh weight. 127

PHENc was estimated by the method of Sinklard and Singleton.(30). Results are expressed as mg gallic 128

acid equivalents/g fresh weight. FRAP and TEAC were determined according to Benzie and Strain (31) 129

and Re et al. (32), respectively. Results are expressed as µm Trolox equivalents/g fresh weight for both 130

FRAP and TEAC. For each genotype and colorimetric assay, 4 technical repeats from the pooled two-131

harvest-fruit-powder were performed except for FRAP in 2010 (2 technical repeats). 132

LC–ESI–MSn analysis of polyphenolic metabolites 133

A total of 13 individual phenolic metabolites [anthocyanins (pelargonidin-3-glucoside; pelargonidin-134

3-glucoside-malonate; pelargonidin-3-rutinoside; cyanidin-3-glucoside (kuromanin); (epi)afzelechin-135

pelargonidin-glucoside), flavonols (kaempferol-glucoside; kaempferol-glucuronide; kaempferol-136

cumaroyl-glucoside; quercetin-glucuronide), flavan-3-ols (catechin; (Epi)catechin dimers; (epi)afzelechin-137

(epi)catechin dimers; (epi)afzelechin-glucoside)] were measured by LC–ESI–MS for the two years. 138

Extraction and LC–ESI–MSn analysis of polyphenolic metabolites were done as described in Ring et al. 139

(19). Briefly, 500 mg of frozen powder was extracted twice with methanol from each biological replicate, 140

adding biochanin A as internal standard. After centrifugation, the supernatants were combined, dried 141

under vacuum, dissolved in water and the samples were injected twice as a technical replicate. Samples 142

were analysed on an Agilent 1100 HPLC/UV system (Agilent Technologies, Waldbronn, Germany) 143

equipped with a reversed phase column (Luna® 3 μm C18(2) 100Å 150 × 2 mm, Phenomenex, 144

Aschaffenburg, Germany) and connected to a Bruker Esquire 3000 plus ion trap mass spectrometer 145

(Bruker Daltonics, Bremen, Germany). Data were analysed with Data analysis 5.1 software (Bruker 146

Daltonics, Bremen, Germany). Metabolites were identified by comparing their retention times and mass 147

spectra (MS and MS2) with those of measured authentic reference compounds. The major known 148

phenolic metabolites were quantified in the positive and negative MS mode by the internal standard 149

method, using QuantAnalysis 2.0 (Bruker Daltonics, Bremen, Germany). Results are expressed as mg 150

equivalents /100 g fresh weight assuming a response factor of 1. Analyses of pooled frozen powder 151


https://doi.org/10.1101/2020.03.20.000455

samples were carried out in 2010 and 2011 on 6 replicates for parents and on 3 replicates for individuals 152

from the progeny. 153

Data analysis 154

Exploratory analyses were performed on all individuals (72 in 2010 and 131 in 2011) and parents 155

('Capitola' and 'CF1116') using R software (R 3.5.0) in the interface RStudio (RStudio 1.2.1572). A Kruskal-156

Wallis test (ANOVA on the rank and appropriate post hoc test) was used to compare the mean values 157

between the parents (agricolae 1.3.1 and PMCMR 4.3 R packages). Trait segregation was declared 158

transgressive when at least one progeny had a value that was higher or lower than that of the highest or 159

lowest parent, by at least twice the standard deviation of the parents.(20). Phenotypic correlations were 160

estimated as Pearson correlations between each trait and represented by heatmap using stats 3.5.0 and 161

corrplot 0.84 R packages and by network correlation using ggnet 0.1.0 and network 1.15 R packages 162

(correlation values r > 0.3 for highlighting the strong correlations). Where a genotypic effect within 163

progeny was found, broad sense heritability was evaluated from variance analysis as follows: 164

165

With VarG: Genetic variance, Vare: residuals' variance, n= number of repetitions (in our case n=3). VarG 166

and Vare were determined using ANOVA on linear regression of each trait. 167

QTL detection. 168

To allow comparison with previously published results, QTL detection and analysis were performed 169

as described in Lerceteau-Köhler et al. (20). Briefly, QTL detection was performed by composite interval 170

mapping (CIM) (33,34) using model 6 of QTL Cartographer software, in which five co-factors were 171

included. The statistical significance thresholds (LOD value) for both female and male linkage maps, and 172

for declaring a putative QTL were 2.8 or 3.1 after 1.000 permutation times and significance level of α = 173

0.10 or 0.05 respectively. The principal characteristics of mQTLs and colour- and antioxidant-related 174

traits' QTL included the chromosome, marker, position, LOD value, confidence interval (LODmax ± 1 175

LOD), and the proportion in % of phenotypic variance (R2) explained by a single QTL. When QTLs from 176

two or three replicates of a same year overlapped, they were summarized in a single QTL. 177


https://doi.org/10.1101/2020.03.20.000455

178

Results 179

180

Flavonoid and antioxidant- and colour-related traits in the ‘Capitola’ and ‘CF1116’ parents. 181

Flavonoid compounds were identified by LC–ESI-MS in the cultivated variety ‘Capitola’ (female), the 182

breeding genotype ‘CF1116’ (male) and in the pseudo full-sibling F1 progeny issued from a cross 183

between them. The two parents are contrasted for several fruit quality traits including fruit shape and 184

weight, firmness and fruit sweetness, acidity and colour-related traits. (20). Among the putative 185

metabolites observed, 13 flavonoid compounds were unambiguously identified in 2010 and 2011 by 186

comparison with commercial standard run under the same conditions. The identified compounds 187

included five anthocyanins (pelargonidin-3-glucoside, pelargonidin-3-glucoside-malonate, pelargonidin-188

3-rutinoside, cyanidin-3-glucoside and (epi)afzelechin-pelargonidin-glucoside), four flavonols 189

(kaempferol-glucoside, kaempferol-glucuronide, kaempferol-coumaryl-glucoside and quercetin-190

glucuronide) and four flavan-3-ols (catechin, (epi)catechin dimer, (epi)afzelechin-(epi)catechin dimer 191

and (epi)afzelechin-glucoside). Corresponding abbreviations are indicated in Table 1. The metabolic 192

compounds identified were further summed up by chemical family to give total anthocyanins (Ant), total 193

flavonols (Fvo) and total flavan-3-ols (F3ol). We also analyzed five traits by colorimetric assays: 194

anthocyanin content (ANTHc), flavonoid content (FLAVc), phenolic content (PHENc) and antioxidants 195

(FRAP and TEAC). In 2011, the FRAP trait was not measured. The general distribution parameters of the 196

metabolites and of the additional antioxidant- and colour-related traits were evaluated for the two 197

parents (‘Capitola’ and ‘CF1116’) and for the progeny in 2010, and for only ‘Capitola’ and for the 198

progeny in 2011. In order to facilitate comparison of the progeny with the two parents, the results of 199

metabolite profiling and colorimetric assays for 2010 are shown in Table 1. Results for 2011 are 200

provided as Supplemental Table S1. 201

202

Mean values were significantly different between the two parents for less than half of the traits 203

analyzed (Table 1). Total flavonoid content, obtained by summation of the anthocyanins, flavonols and 204


https://doi.org/10.1101/2020.03.20.000455

flavan-3-ols, was remarkably similar in ‘Capitola’ and ‘CF1116’ (28.2 vs. 27.8 mg equ/100 g fresh weight 205

in 2010). Anthocyanins were by far the major contributors (~93 % of the flavonoids) followed by flavan-206

3-ols (~2.9 to 3.7 %). Close examination of the results showed small to large variations for individual 207

compounds within a given chemical family. The content in pelargonidin-3-glucoside, which contributes 208

to the total anthocyanin content for as much as ~90 %, was similar in both parents. In contrast, 209

pelargonidin-3-rutinoside content (~2.4 to 8.7 % of the total anthocyanins) was more than 3-fold higher 210

in ‘Capitola’ at the opposite of cyanidin-3-glucoside (~0.7 to 1.8 % of the total anthocyanins) that was 211

almost 3-fold lower in ‘Capitola’. The main flavonol kaempferol-glucoside (~37.5 to 43.8 % of the total 212

flavonols) displayed similar values in both parents. Contents in the main flavan-3-ol (epi)catechin dimers 213

(~44.9 to 57.3 % of the total flavan-3-ols) and in catechin (~17.8 to 24.4 %) were not much different 214

between the two parents while (epi)afzelechin-glucoside (~13.4 to 33.6 % of the total flavan-3-ols) was 215

2.5-fold higher in ‘Capitola’. Surprisingly, given the results from LC-MS analyses that showed similar 216

anthocyanins values for both parents, the colorimetric estimation of anthocyanin content produced a 217

mean ANTHc value 1.8-fold higher in ‘Capitola’ than in ‘CF1116’, in agreement with previously published 218

results.(20). FLAVc, PHENc, FRAP and TEAC mean values were all higher in 'Capitola', with FLAVc and 219

PHENc being respectively 1.6-fold and 1.5-fold higher than in ‘CF1116’. 220

221

Distribution of metabolites and antioxidant- and colour-related traits in the ‘Capitola’ x ‘CF1116’ 222

population 223

Metabolite profiling and measurements of antioxidant- and colour-related traits were next 224

performed on the progeny in order to delineate genomic regions responsible for trait variations. Trait 225

analysis was carried out on 72 individuals in 2010 and 131 genotypes in 2011. Distribution parameters 226

(mean, standard deviation, range and heritability) are shown in Table 1 for 2010. In both years, 227

considerable variations were observed for most of the traits, which segregated in the progeny. 228

Compared to the parents, the range was considerably extended for metabolites belonging to the various 229

chemical classes and for colorimetric traits, thus indicating a genotype-specific control. Most of the 230

variations among the most extreme genotypes were in the 4- to 10-fold range, an example of which is 231


https://doi.org/10.1101/2020.03.20.000455

the 5-fold variation for the most abundant compound pelargonidin-3-glucoside. However, the 232

anthocyanins cyanidin-3-glucoside and pelargonidin-3-rutinoside displayed 23-fold and 114-fold 233

variations in the progeny, respectively, while the flavan-3-ol (epi)afzelechin-glucoside exhibited a 17-fold 234

variation. Similar results were reported in 2011 (Supplemental Table S1). 235

Calculated broad sense heritability displayed high values (h2 > 0.5) for all the 21 traits analyzed in 236

2010 except FRAP (h2=0.49). Very high heritability values (from h2 > 0.8) were even observed for 14 traits 237

associated with anthocyanins (3), flavonols (5) and flavan-3-ols contents (4) and with ANTH and TEAC 238

(Table 1). Very similar results were obtained in 2011 (Supplemental Table S1). Transgressions were 239

detected in 2010 for all traits. They were both positive and negative for 16 of the 21 traits, but only 240

negative for PgRs and positive for AfPgGs, AfGs and PHENc (Table 1). 241

242

Correlation of metabolites and antioxidant- and colour-related traits in the ‘Capitola’ x ‘CF1116’ 243

population. 244

Pearson phenotypic correlations between the various traits are shown in Figure 1A for 2010 and in 245

Supplemental Figure S1A for 2011. In 2010, Pearson correlation analysis highlighted the strong 246

correlation (r=0.99) between total anthocyanins and PgGs, which is not surprising given that PgGs is the 247

preponderant anthocyanin in cultivated strawberry. It also highlighted significant correlations with 248

minor anthocyanin compounds including PgGsM (r=0.36), PgRs (r=0.48), CyGs (r=0.57) and AfPgGs 249

(r=0.52). In addition, anthocyanin content was positively correlated with that of flavonoids from various 250

chemical classes including flavonols (r=0.30 for QGn to r=0.54 for KGs) and flavan-3-ols (r=0.40 for 251

CatCAt and r=0.38 for AfCat). As could be expected, correlations were also significant between flavonoid 252

compounds belonging to the same chemical classes e.g. between the flavonols KGn and QGn (r=0.53) 253

and the flavan-3-ols CatCat and AfCat (r=0.75). No negative correlation was found except for the weak 254

correlation (r=-0.18) between the anthocyanin PgRs and the flavan-3-ol Cat. 255

Correlations between phenotypic values obtained by colorimetric assays commonly used for 256

measuring anthocyanins (ANTHc), flavonoids (FLAVc), phenolics (PHENc) and antioxidants (FRAP and 257

TEAC) and values obtained for individual flavonoid compounds measured by LC-ESI-MS were also 258


https://doi.org/10.1101/2020.03.20.000455

calculated for 2010 (Figure 1A) and for 2011 (Supplemental Figure S1A). For 2010, ANTHc showed 259

indeed high correlations with PgGs (r=0.52) and CyGs (r=0.45), but weak correlations (r=0.15 to 0.35) 260

with other anthocyanins and weak or no significant correlations with flavonols or flavan-3-ols. Strikingly, 261

FLAVc was poorly correlated with all the flavonoid compounds (r=0.14 to 0.23) while PHENc showed a 262

high correlation (r=0.54) only with PgRs and with FRAP and TEAC (r=0.43 and 0.51, respectively). FRAP 263

and TEAC values were highly correlated (r=0.67) but more weakly correlated with all the other traits 264

except PHENc and FLAVc. Similar results were obtained in 2011 for all the traits, except for the eventual 265

odd results such as the high correlation between PHENc and PgRs which was not reproduced. 266

Interestingly, while correlations between compounds and traits were almost exclusively positive in 267

2010, weak but significant negative correlations could be observed in 2011, for example between the 268

flavan-3-ol AfGs and some anthocyanins (PgGsM, PgRs) and between ANTHc and some flavonols (KCoGs, 269

QGn) and flavan-3-ols (Cat, CatCat). 270

We further represented the most significant Pearson correlations between the traits (r > 0.3) as a 271

phenotypic correlation network for 2010 (Figure 1B) and for 2011 (Supplemental Figure S1B). The 272

correlation network highlighted the strong relationships between the compounds in each chemical class 273

and between anthocyanins and flavonols. It also pinpointed the weak correlations between colorimetric 274

and metabolic traits. 275

Scatter plot analysis of trait phenotypic values in 2010 and 2011 (Figure 2) further revealed that 276

ANTHc values had the highest correlation between the two years (r=0.72). High correlations were also 277

observed for the anthocyanin PgGs (r=0.45), the flavonols KGs (r=0.49) and KGn (r=0.42), the flavan-3-ol 278

AfGs (r=0.74) and the ANTHc (r=0.52). Noteworthy, poor correlations (r<0.1) were observed for 279

compounds such as CyGs, QGn and AfCat, underlining the high incidence of environmental conditions on 280

flavonoid accumulation in strawberry fruit. 281

282

QTL analysis 283

We used the same linkage maps as those previously described in Lerceteau-Köhler et al. (20) with 284

the additional markers described in Ring et al. (19). The markers covered the 28 expected linkage groups 285


https://doi.org/10.1101/2020.03.20.000455

(LG) for the female (f) and male (m) linkage maps with two linkage groups (IV-d-f and III-c-m) 286

represented each by two groups (IV-d1-f and IV-d2-f, III-c1-m and III-c2-m) and with five linkage groups 287

anchored (IV-X1 and IV-X2) or not (F30, M41, M44) to one of the seven homoeology groups (HG). To 288

decipher the genetic architecture of the flavonoid compounds, we performed QTL analyses on each of 289

the three replicates of the two successive years (2010 and 2011). Colorimetric traits were analyzed for 290

each year. 291

Linkage analysis on female and male maps revealed a total of 152 mQTLs for the flavonoid 292

compounds, all of which were identified for the first time from cultivated strawberry, and of 26 QTLs for 293

the antioxidant- and colour-related traits measured by colorimetry (Table 2 and Figure 3). Similar 294

number of mQTLs was identified for each chemical class of the 13 flavonoid compounds analyzed. The 295

anthocyanins (5 compounds) displayed the highest number of mQTLs (57 mQTLs) followed by the 296

flavan-3-ols (4 compounds; 53 mQTLs) and by the flavonols (4 compounds; 42 mQTLs). Remarkably, the 297

number of mQTLs and QTLs mapped on the female and male maps was almost identical for the 298

anthocyanins (28 for the female vs. 29 for the male) and for the flavan-3-ols (26 for the female vs. 27 for 299

the male) but was uneven for the flavonols (25 for the female vs. 17 for the male) and for the 300

colorimetric traits (11 for the female vs. 15 for the male). 301

Because cultivated strawberry is an octoploid species, each of the seven homoeology groups (HGs) 302

has four linkage groups (LGs). LGs were assigned to one of the seven HGs and arbitrarily labelled (e.g. LG 303

Ia, b, c or d) as previously described (35). On both female and male maps, mQTLs were identified for all 304

the HGs (Table 3, Figure 3 and Supplemental Tables S2, S3, S4). Interestingly, the colorimetric QTLs were 305

mostly localized on a unique HG (HG VI) in which 4 QTLs (female) and 10 QTLs (male) were detected 306

whereas none, one or at most two QTLs were mapped on the other HGs (Figure 3). Moreover, within a 307

given HG, mQTLs and colorimetric QTLs were unequally distributed on the various LGs. For example, on 308

the female map, 9 mQTLs among which 6 mQTLs for flavonols (kaempferol derivatives) were detected 309

on LG IIIb while only two were detected on LG IIIa and one on LG IIId. At the opposite, on the male map, 310

only one mQTL was detected on LG IIIb while 9 mQTLs, among which 5 mQTLs for flavan-3-ols, were 311

detected on LG IIIa. A striking result was that in both female and male maps, the largest number of 312


https://doi.org/10.1101/2020.03.20.000455

flavonoid mQTLs and colour- and antioxidant-related QTLs (11 on female and 14 on male maps) was 313

detected on the LG VIa (Figure 3) making this linkage group a likely target for the improvement of 314

nutritional quality of strawberry fruit. 315

Interestingly, the analysis of the distribution and effects of the mQTLs (Table 2 and Supplemental 316

Tables S2, S3, S4) highlighted large differences between the compounds belonging to a same chemical 317

class. For the most abundant anthocyanin compound, which is pelargonidin-3-glucoside, 6 mQTLs were 318

detected on female map and 4 on male map. Among them, one major PgGs mQTL localized on LG IIIa 319

from male map (LOD values of 3.7-7.4) could explain 14 to 27% of the PgGs variance (R²) in 2010 320

(Supplemental Tables S2, S3). For the second most abundant anthocyanin compound, pelargonidin-3-321

rutinoside, which displayed considerable variations in the progeny (Table 1 and Supplemental Table S1), 322

large number of mQTLs was detected on both female (8 mQTLs) and male (11 mQTLs) maps. Though the 323

effect of most PgRs mQTLs on the PgRs phenotypic variation was low, one major mQTL detected on LG 324

VIIb from the male map (LOD value of 4.4) explained 36% of the variance (R²) in 2010. In addition, this 325

PgRs mQTL was co-localized with a major pelargonidin-3-glucoside malonate (PgGsM) (LOD value of 4.5) 326

that also explained 36% of the phenotypic variation (R²) in 2010. Noteworthy, several PgRs mQTLs were 327

localized on each of the four different linkage groups from HG VII on the male map (Figure 3), thus 328

highlighting the possible contribution of each F. x ananassa subgenome from HG VII to the control of 329

PgRs content. The analysis of the co-localization of anthocyanin mQTLs additionally pinpointed several 330

genomic regions to be considered when breeding for strawberry fruit colour improvement. On the male 331

map, mQTLs for four anthocyanin compounds derived from pelargonidin (PgGs, PgRs, PgGsM, AfPgGs) 332

were co-localized on LG Ia together with the total anthocyanin content measured by colorimetric assay 333

(ANTHc) (Figure 3). Likewise, on both the female and the male maps, different clusters of anthocyanin-334

related mQTLs were identified on LG VIa (Figure 3). Among them was a cluster of mQTLs for PgGs and 335

PgRs (female map) or for PgGs, PgRs and AfPgGs (male map) that co-localized with ANTHc QTLs and a 336

cluster of overlapping mQTLs for PgGsM and CyGs (female map). 337

Regarding the flavonols, the most complex genetic architecture was that of the kaempferol-338

coumaryl-glucoside (KCoGs), with 10 mQTLs on the female map and 8 mQTLs on the male map, followed 339


https://doi.org/10.1101/2020.03.20.000455

by the kaempferol-glucoside (KGs), with 7 mQTLs on the female map and 4 mQTLs on the male map 340

(Table 2). Among them, a remarkable mQTL cluster was identified on female LG IIIb where mQTLs for 341

kaempferol derivatives (KGs, KGn) overlapped with those of the anthocyanin PgGs and AfPgGs (Figure 342

3). Moreover, the female KGs mQTL on LG IIIb was a major mQTL (LOD value of 4.6-5.7) that accounted 343

for 18 to 23% of the explained variance (R²). Major mQTLs were also identified in 2010 for the quercetin-344

glucuronide (QGn) on the female map on LG IIa (LOD value of 4.5; R² of 20%) and on LG IVb (LOD value 345

of 3.9; R² of 20%) and on male map on LG VIb (LOD value of 4.8; as much as 44% of the variance 346

explained). 347

As for the flavan-3-ols, the most numerous mQTLs were detected for epiafzelechin derivatives, 348

including 8 female and 8 male mQTLs for (epi)Afzelechin-(epi)catechin (AfCat) dimers and 5 female and 349

8 male mQTLs for (epi)Afzelechin-glucoside (AfGs). None of the flavan-3-ols mQTLs exhibited large 350

explained phenotypic variance (R² > 20%) in 2010 except the female AfCat mQTL located on LG IIIc (LOD 351

value of 3.6; R² of 22%) and Cat mQTL located on LG Vd (LOD value of 4.8; R² of 22%). The most 352

interesting flavan-3-ols mQTL cluster was that of catechin (Cat, CatCat) and epiafzlechin (AfGs, AfCat) 353

derivatives that overlapped on male map on a narrow genomic region from LG IIIa. 354

Surprisingly, while 8 ANTHc QTLs and 8 PHENc QTLs were identified, few QTLs were discovered for 355

the fruit antioxidant capacity measured by FRAP (2 in total) and TEAC (5 in total) assays. Noteworthy, 356

the colorimetric QTLs, PHENc, FLAVc and TEAC were clustered on LG VIa on female map (Figure 3). 357

358

Discussion 359

360

Small berry fruits, including strawberry, are an importance source of antioxidants in our diet.(4,36). 361

Antioxidant capacity of strawberry fruit depends not only on antioxidants such as vitamin C (11) but also 362

on its composition in polyphenolics, which contribute to both its sensorial and nutritional 363

quality.(5,6,23). Thanks to the recent advances in the analysis of specialized metabolism (22), it is now 364

possible to breakdown complex traits such as fruit colour and antioxidant capacity into more discrete 365

traits controlling variations in individual chemical compounds. We investigated here the fruit flavonoid 366


https://doi.org/10.1101/2020.03.20.000455

and antioxidant content of a pseudo full-sibling F1 population obtained from a cross between the variety 367

‘Capitola’ and the advanced line ‘CF1116’. The two parents, ‘Capitola’ and ‘CF1116’, display many 368

contrasting fruit quality traits.(26). ‘Capitola’ produces large fruits with low sugar and high acidity while 369

‘CF1116’ produces small fruits with high sugar and lower acidity. In addition, the segregating population 370

issued from ‘Capitola’ and ‘CF1116’ displays considerable variation in fruit colour, as shown by the 371

differences in the Lab colour space values and total anthocyanin content.(20). In a previous work, we 372

mapped quantitative trait loci (QTLs) for various fruit quality traits related to fruit development, texture, 373

sugar and organic acid contents as well as fruit colour.(20). Mapping flavonoid metabolic QTLs (mQTLs) 374

to specific linkage groups in the octoploid cultivated strawberry (21), as done in this study, and 375

comparative analysis of their localization with that of other fruit quality QTLs previously described (20) 376

will facilitate the identification of genomic regions that can be targeted through marker-assisted 377

selection (MAS) for breeding superior strawberry varieties with enhanced sensorial and nutritional 378

quality. 379

380

The ‘Capitola’ x ‘CF1116’ population displays a large variability in flavonoid composition. 381

Anthocyanins are water-soluble flavonoid pigments that constitute the major flavonoid compounds 382

found in strawberry fruit. They are responsible for the red colour of strawberry fruit and strongly 383

contribute to its antioxidant capacity.(23). In the parents and progeny studied, we found that the major 384

anthocyanin compound in the fruit is pelargonidin-3-glucoside (~84-90 % of the anthocyanins) followed 385

by pelargonidin-3-rutinoside while minor anthocyanin compounds (cyanidin-3-glucoside, pelargonidin-3-386

glucoside-malonate, (epi)Afzelechin-pelargonidin-glucoside) were also detected (Table 1 and 387

Supplemental Talbe S1). While anthocyanins were by far the major contributors (~95-97 %) to the 388

flavonoids detected in the ‘Capitola’ and ‘CF1116’ parents and progeny, other flavonoids belonging to 389

the flavan-3-ols (~2 to 4 %) and flavonols chemical classes were also detected and quantified (Table 1 390

and Supplemental Table S1). Although flavonols represent a very minor fraction (1 %) of the flavonoids 391

in cultivated strawberry (23), mainly in the form of kaempferol and quercetin derivatives (Table 1 and 392


https://doi.org/10.1101/2020.03.20.000455

Supplemental Table S1), their intake has a health-promoting effect and reduces the risk of several 393

diseases.(5,7). 394

The relative contributions of individual compounds to their respective chemical classes are in 395

agreement with previous findings for cultivated strawberry.(12,19,23). However, they may contrast with 396

other results published for the wild diploid woodland strawberry F. vesca.(14). In this species, for 397

example, the major anthocyanin compounds are both pelargonidin-3-glucoside and cyanidin-3-398

glucoside, which may respectively contribute to 50% and 40% of the fruit anthocyanins.(12,14). In our 399

segregating population, most of the variations among the most extreme genotypes in the progeny were 400

in the 4 to 10-fold range, an example of which is the ~5-fold variation in pelargonidin-3-glucoside 401

content and the ~4-fold variation in flavonols; it is worth mentioning that values for these compounds 402

were similar in the parents regardless of the large variations in the progeny. Much higher variations 403

were observed in the progeny for the anthocyanins cyanidin-3-glucoside and pelargonidin-3-rutinoside 404

(23-fold and 114-fold, respectively) and for the flavan-3-ol (epi)afzelechin-glucoside (17-fold). Such 405

variations in individual flavonoids may affect the nutritional value of the fruit because of the synergetic 406

effect of anthocyanins and flavonols on health.(37). They may also affect fruit sensorial quality because 407

anthocyanin composition will likely affect fruit colour intensity and hue.(38,39). 408

409

High connectivity between individual flavonoid compounds within each chemical class but low 410

correlation with antioxidant traits measured by colorimetric assays. 411

Not surprisingly given that they share common biosynthetic pathways, the various anthocyanins were 412

positively correlated with each other, as were the flavonol and the flavan-3-ol compounds (Figure 1A 413

and B and Supplemental Figure S1). In addition, high positive correlation values were observed between 414

anthocyanins and flavonols, highlighting the tight connection between these chemical flavonoid classes, 415

which share naringenin and dihydrokaempferol as common precursors. Co-regulations of different 416

enzymes involved at various steps of the flavonoid pathway may additionally explain the parallel 417

variations of the various flavonoids in the progeny studied. Such co-regulations have been well 418

documented in various plant species, including strawberry.(40,41). 419


https://doi.org/10.1101/2020.03.20.000455

The correlations between the total flavonoids, phenolics, anthocyanins and antioxidant trait values 420

estimated by colorimetric methods (FLAVc, PHENc, ANTHc, FRAP and TEAC) and the various flavonoids 421

determined by LC-MS-ESI produced some unexpected results. While anthocyanins contents were well 422

correlated with ANTHc, the correlations with FLAVc, FRAP, and TEAC were poor (r < 0.3). The weak 423

correlations of anthocyanins and other flavonoids with FRAP and TEAC can be explained by the 424

differences in the metabolites detected by colorimetric assays and by LC–ESI–MS analysis. In addition, 425

both FRAP and TEAC measure total antioxidants, among which the flavonoids, but also additional 426

phenolics such as the ellagic acid that has high antioxidant and nutritional values.(42). Indeed, 427

pelargonidin-3-glucoside, which is the main flavonoid compound, only accounts for ~25% of the total 428

strawberry antioxidant capacity.(23). Likewise, PHENc, which is highly correlated with FRAP (Figure 1A 429

and B), is not restricted to flavonoids but also estimates total phenolics.(30). More surprising are the 430

weak correlations between the total flavonoid content measured by LC–ESI–MS and that estimated by 431

FLAVc assay, which questions the use of colorimetric methods commonly used for assessing fruit 432

nutritional quality and the interpretations of their results. 433

In addition to measuring the variations in individual flavonoid compounds in the ‘Capitola’ x 434

‘CF1116’ progeny for a given year, we measured for each trait the correlation values between the two 435

successive years (Figure 2). High correlations were observed for several traits, including colour-related 436

traits (ANTHc, PgGs content) but also flavonols (kaempferol derivatives KGs and KGn) and flavan-3-ol 437

(AfGs), highlighting the strong genetic control of flavonoid composition already observed in F. 438

vesca.(14). In our experimental conditions where plants were cultivated under plastic tunnel and 439

therefore subjected to natural climatic variations, a likely explanation for the poor correlations observed 440

for some traits, an example of which is the CyGs anthocyanin (Figure 2), is the effect of environmental 441

conditions. Sensitivity of flavonoid metabolism to environmental conditions is well documented.(43). 442

Additionally, because poor correlations were mostly due to few individuals of the progeny showing 443

contrasted and extreme phenotypic values in 2010 (e.g. the individuals no. 6 and 195), a strong 444

genotypic effect in the phenotypic plasticity i.e. the response to environmental variations (here year of 445

study), is likely.(44). 446


https://doi.org/10.1101/2020.03.20.000455

447

Flavonoid mQTL mapping and comparison with fruit quality QTLs previously identified. 448

Evaluation and genetic mapping of both specialized (polyphenolics, flavor components) and primary 449

(sugars, organic acids) metabolites are indispensable to identify the genetic architecture of fruit 450

sensorial and nutritional quality in cultivated strawberry.(20,45-47). In addition, genetic studies allow 451

pinpointing genotypes that can be used in a breeding scheme for the genetic improvement of 452

strawberry. In the post-genomic era, to efficiently harness the available strawberry diversity and 453

translate these findings into crop improvement, genetic studies have to be extended to whole 454

populations displaying large phenotypic and genetic diversity.(10), which are typically bi-parental 455

populations in cultivated strawberry.(20,45-47). Using such a bi-parental population issued from a cross 456

between two parents with contrasted sensory trait values (20) and by additionally breaking down the 457

flavonoid composition to its individual chemical components (22), we were able to map mQTLs for each 458

chemical compound from the three flavonoid chemical classes detected in strawberry fruit. The analysis 459

of three replicates over two successive years for each trait allowed us to identify some robust QTLs for 460

specialized metabolism. Additional QTLs that were not significantly detected in the two successive years 461

nevertheless displayed a significant effect from one year to another. As mentioned above for trait 462

correlations between years, such QTLs are likely more sensitive to the environmental conditions, which 463

may considerably vary when plants are grown in natural conditions. 464

An example of a major mQTL (14 to 27% of the variance explained) that is possibly sensitive to 465

environmental conditions is the mQTL for pelargonidin-3-glucoside, the most abundant anthocyanin 466

compound, which was detected on male linkage group LG IIIa for the three replicates in 2010 but not in 467

2011. Because this mQTL is also co-localized with robust QTLs for the colour physical parameters L and b 468

(colour space values), which were previously detected over three successive years (20), it can 469

nevertheless be targeted for fruit colour improvement. Additionally, because we used the same linkage 470

mapping approach and set of markers as in our previous study (20), we could identify a LG IIIa-m 471

gata165c marker linked to both PgGs content and ANTHc and L colour-related traits. This marker is also 472

adjacent to the SSR EFMv029 (allele v029205c) marker, which can therefore be used for the marker-473


https://doi.org/10.1101/2020.03.20.000455

assisted selection (MAS) screening of genotypes-of-interest displaying fruits with enhanced colour and 474

sweetness. Thus, by combining fruit flavonoid mQTL with QTLs previously detected for diverse fruit 475

quality traits (20), we could detect a genomic region that can be targeted for both fruit sensorial and 476

nutritional quality. 477

Another genomic region that appears on both female and male maps as a major linkage group for 478

the control of nutritional traits related to specialized metabolism and of sensorial traits is LG VIa. In our 479

previous study on strawberry fruit sensorial traits, LG VIa (male and/or female) already appeared as an 480

important linkage group for the control of traits related to primary metabolism such as Soluble Solid 481

Content (SSC), sucrose and glucose contents, pH and malic and citric acid contents. In this case as well, 482

markers common to taste and colour-related traits could be identified. These include the SSR EMF006 483

(allele v006205c) marker, which is close to the AFLP markers tgta383 and gata170 in male, and the SSR 484

Fvi020 (allele g020175) marker, which is close to the AFLP ccta303 marker in female. Both markers are 485

linked to several taste- and colour-related QTLs (SSC, sucrose, a and b color space values, ANTH) 486

previously identified (20) and to the colour-related PgGs mQTL and ANTHc QTL detected in the present 487

study. These markers can be further used for the early selection of genotypes for improved sensorial 488

(sweetness, colour) and nutritional (anthocyanin content) quality. 489

In addition to its interest for accelerating breeding through MAS, the QTL mapping strategy can also 490

help deciphering which pathway(s) or enzyme(s) are likely involved in the natural variation of a given 491

trait. In the last decades, reverse genetic studies have successfully been used in strawberry and in other 492

fruit species to investigate the role of regulatory or structural genes in the control of specific steps of the 493

phenylpropanoid pathway. Such studies often rely upon the silencing or overexpression of a candidate 494

gene whose possible role is inferred from its already known function. Among the many studies 495

published for strawberry are the demonstration of the crucial role of MYB transcription factors and 496

associated MYB-bHLH-WD40 complex in the regulation of anthocyanin biosynthesis (48), of the 497

anthocyanidin glucosyltransferase in flavonoid biosynthesis (16), and of the anthocyanidin reductase 498

(ANR) enzyme in the trade-off between anthocyanidin and flavan-3-ols.(17). Precise QTL mapping now 499

gives access to candidate genes underlying natural quantitative variation in a given metabolite, an 500


https://doi.org/10.1101/2020.03.20.000455

example of which is the discovery of the role of a peroxidase in the trade-off between lignin and 501

anthocyanin biosynthesis.(19). The analysis of the clustering of mQTLs for compounds derived from the 502

same pathway, e.g. the clusters of pelargonidin derived compounds on female LG VIa, or from different 503

pathways, e.g. the clusters of pelargonidin (anthocyanins) and kaempferol (flavonols) derived products 504

on female LG IIIb, additionally provide cues on which step of the flavonoid pathway is worth 505

investigating. The recent sequencing of the octoploid cultivated strawberry (21) will now give access to 506

the tools necessary to identify the underlying candidate genes and the genetic variations responsible for 507

the disparities in flavonoid-related traits in strawberry. 508

509

Acknowledgements 510

Thanks for their help to Karine Tallès and Gabriel Jousseaume for fruit harvests and colorimetric 511

measurements. 512

513

Funding sources 514

The authors gratefully acknowledge support from Région Nouvelle-Aquitaine (AgirClim project N°2018-515

1R20202), the European Union’s ERANET (FraGenomics N°PCS-08-TRIL-00) and European Union’s 516

Horizon 2020 research and innovation program (GoodBerry project N° 679303). 517

518

Author contribution statement: BD conceived and designed the experiments. AuP conducted hands-on 519

experiments and data collection. AuP, AlP and AG participated in the data collection. LR, TH and WS 520

generated LC-LS data. ML, GV, CR and BD conducted data analysis and performed statistical analysis. CR 521

wrote the original draft. All authors read and approved the final manuscript. 522

523

Conflict of interest: On behalf of all authors, the corresponding author states that there is no conflict of 524

interest. 525

526


https://doi.org/10.1101/2020.03.20.000455

References 527

528

1. Gasparrini, M.; Giampieri, F.; Forbes-Hernandez, T. Y.; Afrin, S.; Cianciosi, D.; Reboredo-Rodriguez, P.; 529

Varela-Lopez, A.; Zhang, J.; Quiles, J. L.; Mezzetti, B.; Bompadre, S.; Battino, M. Strawberry extracts 530

efficiently counteract inflammatory stress induced by the endotoxin lipopolysaccharide in Human Dermal 531

Fibroblast. Food Chem. Toxicol. 2018, 114, 128−140. 532

2. Giampieri, F.; Alvarez-Suarez, J. M.; Battino, M. Strawberry and human health: effects beyond 533

antioxidant activity. J. Agric. Food Chem. 2014, 62, 3867–3876. 534

3. Giampieri, F.; Alvarez-Suarez, J. M.; Cordero, M. D.; Gasparrini, M.; Forbes-Hernandez, T. Y.; Afrin, S.; 535

Santos−-Buelga, C.; González-Paramás, A. M.; Astolfi, P.; Rubini, C.; Zizzi, A.; Tulipani, S.; Quiles, J. L.; 536

Mezzetti, B.; Battino, M. Strawberry consumption improves aging-associated impairments; mitochondrial 537

biogenesis and functionality through the AMP-activated protein kinase signaling cascade. Food Chem. 2017, 538

234, 464−471. 539

4. Battino, M.; Beekwilder, J.; Denoyes-Rothan, B.; Laimer, M.; McDougall, G. J.; Mezzetti, B. Bioactive 540

compounds in berries relevant to human health. Nutr. Rev. 2009, 67, S145−S150. 541

5. Afrin, S.; Gasparrini, M.; Forbes-Hernandez, T. Y.; Reboredo-Rodriguez, P.; Mezzetti, B.; Varela-López, A.; 542

Giampieri, F.; Battino, M. Promising health benefits of the strawberry: A focus on clinical studies. J. Agric. 543

Food Chem. 2016, 64, 4435−49. 544

6. Forbes-Hernandez, T. Y.; Gasparrini, M.; Afrin, S.; Bompadre, S.; Mezzetti, B.; Quiles, J. L.; Giampieri, F.; 545

Battino, M. The healthy effects of strawberry polyphenols: which strategy behind antioxidant capacity? Crit. 546

Rev. Food Sci. Nutr. 2016, 56, S46−S59. 547

7. Miller, K.; Feucht, W.; Schmid, M. Bioactive compounds of strawberry and blueberry and their potential 548

health effects based on human intervention studies: A brief overview. Nutrients 2019, 11, 1510. 549

8. Darrow, G. The strawberry: History, breeding and physiology. The New England Institute for Medical 550

Research, Eds. Publisher: Holt, Rinehart & Winston, New-York, 1966; pp 447. 551


https://doi.org/10.1101/2020.03.20.000455

9. Perrotte, J.; Gaston, A.; Potier, A.; Petit, A.; Rothan, C.; Denoyes, B. Narrowing down the single 552

homoeologous FaPFRU locus controlling flowering in cultivated octoploid strawberry using a selective 553

mapping strategy. Plant Biotechnol. J. 2016, 14, 2176−2189. 554

10. Gaston, A.; Osorio, S.; Denoyes, B.; Rothan, C. Applying the Solanaceae strategies to strawberry crop 555

improvement. Trends Plant Sci. 2020, 25, 130−140. 556

11. Amaya, I.; Osorio, S.; Martinez-Ferri, E.; Lima-Silva, V.; Doblas, V. G.; Fernández-Muñoz, R.; Fernie, A. R.; 557

Botella, M. A.; Valpuesta, V. Increased antioxidant capacity in tomato by ectopic expression of the 558

strawberry D-galacturonate reductase gene. Biotechnol. J. 2015, 10, 490−500. 559

12. Muñoz, C.; Sánchez-Sevilla, J. F.; Botella, M. A.; Hoffmann, T.; Schwab, W.; Valpuesta, V. Polyphenol 560

composition in the ripe fruits of Fragaria species and transcriptional analyses of key genes in the pathway. 561

J. Agric. Food Chem. 2011, 59, 12598−604. 562

13. Crecelius, A. C.; Hölscher, D.; Hoffmann, T.; Schneider, B.; Fischer, T. C.; Hanke, M. V.; Flachowsky, H.; 563

Schwab, W.; Schubert, U. S. Spatial and temporal localization of flavonoid metabolites in strawberry fruit 564

(Fragaria × ananassa). J. Agric. Food Chem. 2017, 3, 65:3559−3568. 565

14. Urrutia, M.; Schwab, W.; Hoffmann, T.; Monfort, A. Genetic dissection of the (poly)phenol profile of 566

diploid strawberry (Fragaria vesca) fruits using a NIL collection. Plant Sci. 2016, 242, 151−168. 567

15. Almeida, J. R.; D'Amico, E.; Preuss, A.; Carbone, F.; de Vos, C. H.; Deiml, B.; Mourgues, F.; Perrotta, G.; 568

Fischer, T. C.; Bovy, A. G.; Martens, S.; Rosati, C. Characterization of major enzymes and genes involved in 569

flavonoid and proanthocyanidin biosynthesis during fruit development in strawberry (Fragaria x ananassa). 570

Arch. Biochem. Biophys. 2007, 465, 61−71. 571

16. Griesser, M.; Hoffmann, T.; Bellido, M. L.; Rosati, C.; Fink, B.; Kurtzer, R.; Aharoni, A.; Muñoz-Blanco, J.; 572

Schwab, W. Redirection of flavonoid biosynthesis through the down-regulation of an anthocyanidin 573

glucosyltransferase in ripening strawberry fruit. Plant Physiol. 2008, 146, 1528−39. 574

17. Fischer, T. C.; Mirbeth, B.; Rentsch, J.; Sutter, C.; Ring, L.; Flachowsky, H.; Habegger, R.; Hoffmann, T.; 575

Hanke, M.V.; Schwab, W. Premature and ectopic anthocyanin formation by silencing of anthocyanidin 576

reductase in strawberry (Fragaria × ananassa). New Phytol. 2014, 201, 440−451. 577


https://doi.org/10.1101/2020.03.20.000455

18. Zhang, J.; Lei, Y.; Wang, B.; Li, S.; Yu, S.; Wang, Y.; Li, H.; Liu, Y.; Ma, Y.; Dai, H.; Wang, J.; Zhang, Z. The 578

high-quality genome of diploid strawberry (Fragaria nilgerrensis) provides new insights into anthocyanin 579

accumulation. Plant Biotechnol. J. 2020. doi: 10.1111/pbi.13351. Epub ahead of print. 580

19. Ring, L.; Yeh, S. Y.; Hücherig, S.; Hoffmann, T.; Blanco-Portales, R.; Fouche, M.; Villatoro, C.; Denoyes, B.; 581

Monfort, A.; Caballero, J. L.; Muñoz-Blanco, J.; Gershenson, J.; Schwab, W. Metabolic interaction between 582

anthocyanin and lignin biosynthesis is associated with peroxidase FaPRX27 in strawberry fruit. Plant Physiol 583

2013, 163, 43−60. 584

20. Lerceteau-Köhler, E.; Moing, A.; Guérin, G.; Renaud, C.; Petit, A.; Rothan, C.; Denoyes, B. Genetic 585

dissection of fruit quality traits in the octoploid cultivated strawberry highlights the role of homoeo-QTL in 586

their control. Theor. Appl. Genet. 2012, 124, 1059−1077. 587

21. Edger, P. P.; Poorten, T. J.; VanBuren, R.; Hardigan, M. A.; Colle, M.; McKain, M. R.; Smith, R.D.; Teresi, S. 588

J.; Nelson, A. D. L.; Wai, C. M.; Alger, E. I.; Bird, K. A.; Yocca, A. E.; Pumplin, N.; Ou, S.; Ben-Zvi, G.; Brodt, A.; 589

Baruch, K.; Swale, T.; Shiue, L.; Acharya, C. B.; Cole, G. S.; Mower, J. P.; Childs, K. L.; Jiang, N.; Lyons, E.; 590

Freeling, M.; Puzey, J. R.; Knapp, S. J. Origin and evolution of the octoploid strawberry genome. Nat. Genet. 591

2019, 51, 541−547. 592

22. Haugeneder, A.; Trinkl, J.; Härtl, K.; Hoffmann, T.; Allwood, J. W.; Schwab, W. Answering biological 593

questions by analysis of the strawberry metabolome. Metabolomics 2018, 14, 145. 594

23. Tulipani, S.; Mezzetti, B.; Capocasa, F.; Bompadre, S.; Beekwilder, J.; de Vos, C. H.; Capanoglu, E.; Bovy, 595

A.; Battino, M. Antioxidants, phenolic compounds, and nutritional quality of different strawberry 596

genotypes. J. Agric. Food Chem. 2008, 56, 696−704. 597

24. Winter, A. N.; Ross, E. K.; Wilkins, H. M.; Stankiewicz, T. R.; Wallace, T.; Miller, K.; Linseman, D. A. An 598

anthocyanin-enriched extract from strawberries delays disease onset and extends survival in the hSOD1G93A 599

mouse model of amyotrophic lateral sclerosis. Nutr. Neurosci. 2018, 21, 414−426. 600

25. Hanhineva, K.; Rogachev, I.; Aura, A. M.; Aharoni, A.; Poutanen, K.; Mykkänen, H. Qualitative 601

characterization of benzoxazinoid derivatives in whole grain rye and wheat by LC-MS metabolite profiling. J. 602

Agric. Food Chem. 2011, 59, 921−927. 603


https://doi.org/10.1101/2020.03.20.000455

26. Moing, A.; Renaud, C.; Gaudillere, M.; Raymond, P.; Roudeillac, P.; Denoyes−Rothan, B. Biochemical 604

changes during fruit development of four strawberry cultivars. J. Am. Soc. Hortic. Sci. 2001, 126, 394−403. 605

27. Capocasa, F.; Scalzo, J.; Mezzetti, B.; Battino, M. Combining quality and antioxidant attributes in the 606

strawberry: The role of genotype. Food Chem. 2008, 111, 872−878. 607

28. Villarreal, N. M.; Martínez, G. A.; Civello, P. M. Influence of plant growth regulators on 608

polygalacturonase expression in strawberry fruit. Plant Sci. 2009, 76, 749−757. 609

29. Dewanto, V.; Wu, X.; Adom, K. K.; Liu, R. H. Thermal processing enhances the nutritional value of 610

tomatoes by increasing total antioxidant activity. J. Agric. Food Chem. 2002, 50, 3010−3014. 611

30. Sinklard, K.; Singleton, V. L. Total phenol analysis: automation and comparison with manual methods. 612

Am. J. Enol. Viticult. 1977, 28, 49−55. 613

31. Benzie, I. F.; Strain, J. J. The ferric reducing ability of plasma (FRAP) as a measure of "antioxidant 614

power": the FRAP assay. Anal. Biochem. 1996, 239, 70−76. 615

32. Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.; Rice-Evans, C. Antioxidant activity applying 616

an improved ABTS radical cation decolorization assay. Free Radic. Biol. Med. 1999, 26, 1231−1237. 617

33. Zeng, Z. B. Theoretical basis for separation of multiple linked gene effects in mapping quantitative trait 618

loci. Proc. Natl. Acad. Sci. USA 1993, 90, 10972−10976. 619

34. Zeng, Z. B. Precision mapping of quantitative trait loci. Genetics 1994, 136, 1457−1468. 620

35. Rousseau-Gueutin, M.; Lerceteau-Köhler, E.; Barrot, L.; Sargent, D. J.; Monfort, A.; Simpson, D.; Arús, P.; 621

Guérin, G.; Denoyes-Rothan, B. Comparative genetic mapping between octoploid and diploid Fragaria 622

species reveals a high level of colinearity between their genomes and the essentially disomic behavior of 623

the cultivated octoploid strawberry. Genetics 2008, 179, 2045−2060. 624

36. Tulipani, S.; Mezzetti, B.; Battino, M. Impact of strawberries on human health: insight into marginally 625

discussed bioactive compounds for the Mediterranean diet. Public Health Nutr. 2009, 12, 1656−1662. 626

37. Scarano, A.; Butelli, E.; De Santis, S.; Cavalcanti, E.; Hill, L.; De Angelis, M.; Giovinazzo, G.; Chieppa, M.; 627

Martin, C.; Santino, A. Combined dietary anthocyanins; flavonols; and stilbenoids alleviate inflammatory 628

bowel disease symptoms in mice. Front. Nutr. 2018, 4, 75. 629


https://doi.org/10.1101/2020.03.20.000455

38. Yoshida, Y.; Koyama, N.; Tamura, H. Color and anthocyanin composition of strawberry fruit: changes 630

during fruit development and differences among cultivars, with special reference to the occurrence of 631

pelargonidin 3-malonylglucoside. J. Japan Soc. Hort. Sci. 2002, 71, 355−361. 632

39. Roy, S.; Wu, B.; Liu, W.; Archbold, D. D. Comparative analyses of polyphenolic composition of Fragaria 633

spp. color mutants. Plant Physiol. Biochem. 2018, 125, 255−261. 634

40. Jaakola, L. New insights into the regulation of anthocyanin biosynthesis in fruits. Trends Plant Sci., 2013, 635

18, 477−483. 636

41. Härtl, K.; Denton, A.; Franz-Oberdorf, K.; Hoffmann, T.; Spornraft, M.; Usadel, B.; Schwab, W. Early 637

metabolic and transcriptional variations in fruit of natural white-fruited Fragaria vesca genotypes. Sci. Rep. 638

2017, 7, 45113. 639

42. Schulenburg, K.; Feller, A.; Hoffmann, T.; Schecker, J. H.; Martens, S.; Schwab, W. Formation of 640

β−glucogallin, the precursor of ellagic acid in strawberry and raspberry. J. Exp. Bot. 2016, 67, 2299–2308. 641

43. Blancquaert, E. H.; Oberholster, A.; Ricardo-da-Silva, J. M.; Deloire, A. J. Grape flavonoid evolution and 642

composition under altered light and temperature conditions in Cabernet Sauvignon (Vitis vinifera L.). Front. 643

Plant. Sci. 2019, 10, 1062. 644

44. Schlichting, C. D. Phenotypic plasticity in plants. Plant Species Biol. 2002, 17, 85–88. 645

45. Zorrilla-Fontanesi, Y.; Cabeza, A.; Domínguez, P.; Medina-Minguez, J.; Valpuesta, V.; Denoyes-Rothan, 646

B.; Sánchez-Sevilla, J.; Amaya, I. Quantitative trait loci and underlying candidate genes controlling 647

agronomical and fruit quality traits in octoploid strawberry (Fragaria × ananassa). Theor. Appl. Genet. 2011, 648

123, 755–778. 649

46. Zorrilla-Fontanesi, Y.; Rambla, J. L.; Cabeza, A.; Medina, J. J.; Sánchez-Sevilla, J. F.; Valpuesta, V.; Botella, 650

M. A.; Granell, A.; Amaya, I. Genetic analysis of strawberry fruit aroma and identification of O-651

methyltransferase FaOMT as the locus controlling natural variation in mesifurane content. Plant Physiol. 652

2012, 159, 851–870. 653


https://doi.org/10.1101/2020.03.20.000455

47. Vallarino, J. G.; Pott, D. M.; Cruz-Rus, E.; Miranda, L.; Medina-Minguez, J. J.; Valpuesta, V.; Fernie, A. R.; 654

Sánchez-Sevilla, J. F.; Osorio, S.; Amaya, I. Identification of quantitative trait loci and candidate genes for 655

primary metabolite content in strawberry fruit. Hortic. Res. 2019, 6(4), 1−17. 656

48. Schaart, J. G.; Dubos, C.; Romero De La Fuente, I.; van Houwelingen, A. M.; de Vos, R. C.; Jonker, H. H.; 657

Xu, W.; Routaboul, J. M.; Lepiniec, L.; Bovy, A. G. Identification and characterization of MYB−bHLH−WD40 658

regulatory complexes controlling proanthocyanidin biosynthesis in strawberry (Fragaria × ananassa) fruits. 659

New Phytol. 2013, 197, 454–467. 660

661

662


https://doi.org/10.1101/2020.03.20.000455

Figure legends 663

664

Figure 1. Phenotypic correlations for the traits measured in 2010. 665

A) Pearson phenotypic correlations (P < 0.05). Correlation values 'r' are represented with colour circle in 666

upper right triangle. Scale combines circle size (small circle: correlation near 0; large circle: correlation 667

near 1) and colour gradient from red (negative correlation) to blue (positive correlation). Pearson 668

correlation values are indicated in lower left triangle. Only significant correlations are represented (P < 669

0.05). A white box represents a non-significant correlation. Diagonal values are not represented. 670

Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-malonate; 671

PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside; AfPgGs, (epi)Afzelechin-pelargonidin-3-672

glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-glucuronide; KCoGs, 673

Kaempferol-coumaryl-glucoside; QGn, Quercetin-glucuronide; F3ol, total flavan-3-ols; Cat, Catechin; 674

CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin dimers; AfGs, (epi)Afzelechin-675

glucoside; ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids (colorimetry); PHENc, phenolics 676

(colorimetry); FRAP, antioxidant (FRAP assay); TEAC, antioxidant (TEAC assay). 677

B) Pearson phenotypic correlation network (P < 0.05). Traits are represented as nodes and coloured 678

according to chemical families [Anthocyanins (grey), Flavonols (orange), Flavan-3-ols (yellow)] or to 679

colorimetric assays (red). Positive (green) and negative (red) correlations with absolute values r > |0.3| 680

are represented as links between nodes. The thickness of the links depends on the correlation values; 681

the more the correlation value is high, the more the link is thick. Only significant correlations are 682

represented (P < 0.05). 683

684

Figure 2. Correlations between the two years. 685

Scatter plots for all traits measured on the two years. The blue line represents the linear regression with 686

associated 95% confidence interval. Correlation ratio (R), percentage of variance (R²) and p (p) value of 687

the Pearson test. When p < 0.05, metabolite contents between the two years are correlated. Blue circles 688

indicate examples of genotypes showing extreme values in 2010. 689


https://doi.org/10.1101/2020.03.20.000455

Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-malonate; 690

PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside; AfPgGs, (epi)Afzelechin-pelargonidin-3-691

glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-glucuronide; KCoGs, 692

Kaempferol-coumaryl-glucoside; QGn, Quercetin-glucuronide; F3ol, total flavan-3-ols; Cat, Catechin; 693

CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin dimers; AfGs, (epi)Afzelechin-694

glucoside; ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids (colorimetry); PHENc, phenolics 695

(colorimetry); TEAC, antioxidant (TEAC assay). 696

697

Figure 3. Distribution of QTLs detected in ‘Capitola’ (female) and ‘CF1116’ (male). 698

(A), female linkage map; (B), male linkage map. QTLs were detected by QTL Cartographer and are 699

represented by bars of different colours. The different classes of flavonoids and the colour- and 700

antioxidant-related traits are each represented by a single colour. Names of mQTLs and of colorimetric 701

QTLs are detailed in Supplemental Table S2 for female and S3 for male. 702

703


https://doi.org/10.1101/2020.03.20.000455

Table 1. Trait values for ‘Capitola’ and ‘CF1116’ parents and related progeny in 2010. 704

705 For each trait, the mean value, standard deviation (s.d.), range and heritability (h²) are described. 706 The significance levels of the differences of the means between ‘Capitola’ and ‘CF1116’ are indicated in the Cap vs CF column and were analyzed by a test of 707 ANOVA-on the rank (Kruskal-Wallis test and post-hoc test). n.s., *, ** and *** indicate values that are not significant (n.s) or are significant at 0.01 <P< 0.05, 708

Traits Abbr. ‘Capitola’ ‘CF1116’

Progeny h² Tr

mean s.d. range mean s.d. Range Cap vs CF mean s.d. range

Anthocyanins

Total anthocyanins Ant 26.88 6.81 19.37-32.63 26.62 4.29 21.76-29.84 n.s. 24.18 6.94 10.55-57.64 0.77 + & -

Pelargonidin-3-glucoside PgGs 24.15 5.40 18.00-28.15 25.17 3.92 20.71-28.04 n.s. 22.37 6.32 10.19-53.47 0.81 + & -

Pelargonidin-3-glucoside-malonate PgGsM 0.19 0.09 0.10-0.26 0.31 0.14 0.21-0.47 n.s. 0.17 0.30 0.00-2.12 0.93 + & -

Pelargonidin-3-rutinoside PgRs 2.35 1.47 1.13-3.99 0.64 0.15 0.47-0.74 * 1.25 0.89 0.04-4.56 - -

Cyanidin-3-glucoside CyGs 0.18 0.04 0.14-0.23 0.48 0.12 0.35-0.57 * 0.36 0.24 0.07-1.63 0.92 + & -

(epi)Afzelechin-pelargonidin-glucoside AfPgGs 0.001 0.002 0.000-0.003 0.017 0.002 0.015-0.019 * 0.01 0.01 0.00-0.03 0.57 +

Flavonols

Total Flavonols Fvo 0.24 0.05 0.18-0.28 0.32 0.04 0.28-0.34 n.s. 0.30 0.09 0.13-0.73 0.94 + & -

Kaempferol-glucoside KGs 0.09 0.03 0.06-0.12 0.14 0.01 0.12-0.15 * 0.10 0.04 0.03-0.22 0.96 + & -

Kaempferol-glucuronide KGn 0.05 0.01 0.04-0.06 0.045 0.003 0.04-0.05 n.s. 0.06 0.02 0.02-0.15 0.88 + & -

Kaempferol-coumaryl-glucoside KCoGs 0.07 0.01 0.05-0.08 0.09 0.01 0.08-0.10 n.s. 0.09 0.03 0.03-0.19 0.92 + & -

Quercetin-glucuronide QGn 0.03 0.01 0.02-0.04 0.05 0.01 0.04-0.07 * 0.05 0.04 0.02-0.27 0.98 + & -

Flavan-3-ols

Total Flavan-3-ols F3ol 1.07 0.23 0.81-1.22 0.82 0.07 0.75-0.90 n.s. 1.21 0.36 0.58-2.70 0.94 + & -

Catechin Cat 0.19 0.04 0.15-0.23 0.20 0.02 0.19-0.23 n.s. 0.22 0.05 0.10-0.42 0.78 + & -

(epi)Catechin dimers CatCat 0.48 0.13 0.34-0.56 0.47 0.06 0.40-0.52 n.s. 0.47 0.11 0.21-0.97 0.85 + & -

(epi)Afzelechin-(epi)catechin dimers AfCat 0.04 0.01 0.03-0.04 0.05 0.01 0.04-0.06 * 0.05 0.02 0.01-0.13 0.81 + & -

(epi)Afzelechin-glucoside AfGs 0.36 0.09 0.29-0.46 0.11 0.03 0.08-0.14 * 0.47 0.31 0.11-1.95 0.98 +

Colorimetry

Anthocyanins (colorimetry) ANTHc 832.82 7.63 824.5-839.5 461.05 5.39 454.9-464.7 * 652.96 191.92 249.8-1131.4 1 + & -

Flavonoids (colorimetry) FLAVc 625.62 13.39 611.2-637.7 392.81 36.68 366.9-418.8 n.s. 465.77 164.15 0.0-788.2 0.66 + & -

Phenolics (colorimetry) PHENc 1.71 0.55 1.28-2.33 1.12 0.56 0.66-1.75 n.s. 1.37 0.48 0.54-3.24 - +

Antioxidant FRAP FRAP 7.77 1.39 6.79-8.75 6.76 1.48 5.71-7.81 n.s. 7.59 1.44 4.60-12.06 0.49 +

Antioxidant TEAC TEAC 14.37 1.21 12.98-15.19 11.79 0.49 11.29-12.28 * 13.62 2.15 7.98-19.55 0.8 + & -


https://doi.org/10.1101/2020.03.20.000455

0.001 <P< 0.01, and P <0.001, respectively. Trait segregation was declared transgressive (Tr) when at least one progeny had a value higher (+), lower (-) or both 709 (+ & -) than the highest or lowest parental value by at least twice the s.d. of the parents. 710 Flavonoid metabolites values are expressed as mg-equ/100 g FW assuming a response factor of 1. 711 Colour- and antioxidant-related traits: ANTHc results are expressed as mg pelargonidin-3-glucoside equivalents/100 g FW. FLAVc results are expressed as μg 712 catechin equivalents/g FW. PHENc results are expressed as mg gallic acid equivalents/g FW. FRAP and TEAC results are expressed as μm Trolox equivalents/g 713 FW. Values are the means of n = 3 replicates per genotype. 714 715 716

717


https://doi.org/10.1101/2020.03.20.000455

Table 2. Location of significant QTLs detected for all traits based on CIM analysis with LOD > LOD threshold (significance level α = 0.10). 718 719

Traits Abbr. Female Linkage Groupsa Nbb Male Linkage Groupsa Nbb

Anthocyanins

Pelargonidin-3-glucoside PgGs IIa-f, IIIb-f, Vc-f, VIa-f (2), VIIa-f 6 Ia-m, Id-m, IIIa-m, VIa-m 4 Pelargonidin-3-glucoside-malonate

PgGsM Id-f, IIId-f, IVX2-f, VIa-f, VIa-f 5 Ia-m, IIa-m, VIa-m, VIIc-m 4

Pelargonidin-3-rutinoside PgRs IIa-f, IIb-f, IVb-f, Vb-f, Vc-f, Vd-f, VIa-f, VId-f 8 Ia-m, IIa-m, VIa-m, VIb-m, VId-m, VIIa-m (2), VIIb-m, VIIc-m, VIId-m, M41-m 11

Cyanidin-3-glucoside CyGs IIa-f, VIa-f 2 IIb-m, IIIa-m (2), Va-m, Vd-m 5

(epi)Afzelechin-pelargonidin-glucoside

AfPgGs IIIa-f, IIIb-f, Va-f, Vd-f, VIb-f, VId-f, VIIa-f 7 Ia-m, IIa-m (2), IIIb-m, VIa-m 5

Flavonols

Kaempferol-glucoside KGs IIIb-f (3), IVa-f, IVc-f, VIa-f, VIIa-f 7 IIIa-m, IIIc1-m, IVb-m, IVd-m 4

Kaempferol-glucuronide KGn IIIb-f, IVc-f (2) 3 IVb-m, Va-m, Vb-m, VIId-m 4

Kaempferol-coumaryl-glucoside

KCoGs Ib-f (2), IIIb-f (2), IIIc-f, Vb-f, VIa-f, VIa-f, VIIb-f, VIIc-f

10 IIIc2-m, IVb-m, Vc-m (3), Vd-m, VIb-m, VIId-m 8

Quercetin-glucuronide QGn IIa-f, IIc-f, IIIa-f, IIIc-f, IVb-f 5 VIb-m 1

Flavan-3-ols

Catechin Cat IVb-f, IVc-f, IVd2-f, Vd-f (2), VIIa-f 6 IIIa-m, Vb-m, Vd-m, VIb-m (2), VIIc-m, VIId-m 7

(epi)Catechin dimers CatCat Ic-f, IIb-f, IVc-f, IVd2-f, Vd-f, VIIa-f, VIIb-f 7 IIc-m, IIIa-m, Vd-m, VIId-m 4 (epi)Afzelechin-(epi)catechin dimers

AfCat Ib-f, IIa-f, IIb-f, IIIb-f, IIIc-f, IVb-f, Vd-f (2) 8 IIa-m, IIIa-m (2), Vb-m, Vd-m, VIa-m, VIa-m, VIId-m 8

(epi)Afzelechin-glucoside AfGs IVX2-f, IVb-f, IVb-f, VId-f (2) 5 IIIa-m, IIIc1-m, IVc-m (2), Va-m (2), VIa-m (2) 8

Colorimetry

Anthocyanins (colorimetry)

ANTHc IVd2-f, VIa-f (2), VIb-f 4 Ia-m, IId-m, VIa-m, VIa-m 4

Flavonoids (colorimetry) FLAVc IIa-f 1 VIId-m, VIa-m 2

Phenolics (colorimetry) PHENc IIb-f, IIIc-f 2 IIIc1-m, VIa-m (2), VIb-m, VIb-m, VId-m 6

Antioxidant FRAP FRAP VIc-f 1 Va-m 1

Antioxidant TEAC TEAC Ic-f, VIc-f, VIIb-f 3 VIa-m, VIb-m 2


https://doi.org/10.1101/2020.03.20.000455

TOTAL 90 88

720

721 a, Linkage groups (LGs) were assigned to one of the seven HGs and arbitrarily labelled (e.g. LG Ia, b, c or d) as previously described (Rousseau-Gueutin et al. 2008). 722 In bracket, the number of QTLs identified in the corresponding LG. 723 b, Number of QTLs (mQTL and colour- and antioxidant related trait QTLs) 724

725 726


https://doi.org/10.1101/2020.03.20.000455

Table 3. Distribution of QTLs according to the homoeology groups (HG) and linkage groups. 727

728

HGs Linkage groups Other HG

HGI Ia Ib Ic Id

5 3 2 2 HGII IIa IIb IIc IId

11 5 2 1 HGIII IIIa IIIb IIIc IIId

11 10 8 1 HGIV IVa IVb IVc IVd

1 9 7 4

2

HGV Va Vb Vc Vd

6 5 5 12 HGVI VIa VIb VIc VId

25 10 2 6 HGVII VIIa VIIb VIIc VIId

7 4 4 7

HG unknown 1 729

The QTLs found on the 28 expected linkage groups are distributed on the seven homoeology groups 730 (HGs). Several QTLs were found on one linkage group anchored on the HGIV (IVX2) or not anchored on 731 any HG (M41). 732 733

734

735


https://doi.org/10.1101/2020.03.20.000455

Figure 1A. Phenotypic correlations for the traits measured in 2010.(A) Pearson phenotypic correlations (P < 0.05). Correlation values 'r' are represented with colour circle in upper right triangle. Scale combines circle size (small circle:correlation near 0; large circle: correlation near 1) and colour gradient from red (negative correlation) to blue (positive correlation). Pearson correlation values areindicated in lower left triangle. Only significant correlations are represented (P < 0.05). A white box represents a non-significant correlation. Diagonal values are notrepresented.Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-malonate; PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside; AfPgGs,(epi)Afzelechin-pelargonidin-3-glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-glucuronide; KCoGs, Kaempferol-coumaryl-glucoside; QGn,Quercetin-glucuronide; F3ol, total flavan-3-ols; Cat, Catechin; CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin dimers; AfGs, (epi)Afzelechin-glucoside;ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids (colorimetry); PHENc, phenolics (colorimetry); FRAP, antioxidant (FRAP assay); TEAC, antioxidant (TEAC assay).

(A)


https://doi.org/10.1101/2020.03.20.000455

AntPgGs

PgGsM

PgRs

CyGs

AfPgGs

Fvo

KGs

KGn

KCoGsQGnF3ol

Cat

CatCat

AfCat

AfGs

ANTHc

FLAVc

PHENc

FRAP

TEAC

Category

Anthocyanins

Flavonols

Flavan-3-ols

Colorimetry

Figure 1B. Phenotypic correlations for the traits measured in 2010.(B) Pearson phenotypic correlation network (P < 0.05). Traits are represented as nodes and coloured according to chemical families[Anthocyanins (grey), Flavonols (orange), Flavan-3-ols (yellow)] or to colorimetric assays (red). Positive (green) and negative (red) correlationswith absolute values r > |0.3| are represented as links between nodes. The thickness of the links depends on the correlation values; the more thecorrelation value is high, the more the link is thick. Only significant correlationsare represented (P < 0.05).


https://doi.org/10.1101/2020.03.20.000455

6, 195

Figure 2. Correlations between the two years.Scatter plots for all traits measured on the two years. The blue line represents the linear regression with associated 95% confident interval. Correlation ratio (R),percentage of variance (R²) and p (p) value of the Pearson test. When p < 0.05, metabolites contents between the two years are correlated. Blue circles indicateexamples of genotypes showing extreme values in 2010.Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-malonate; PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside;AfPgGs, (epi)Afzelechin-pelargonidin-3-glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-glucuronide; KCoGs, Kaempferol-coumaryl-glucoside; QGn, Quercetin-glucuronide; F3ol, total flavan-3-ols; Cat, Catechin; CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin dimers; AfGs,(epi)Afzelechin-glucoside; ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids (colorimetry); PHENc, phenolics (colorimetry); TEAC, antioxidant (TEAC assay).

6, 195


https://doi.org/10.1101/2020.03.20.000455

KG

n_0

PgG

s_0

KG

s_0

IVa-f

PgR

s_0

AfC

at_

0A

fGs_0

AfG

s_1

Cat_

0

QG

n_1

IVb-f

Cat_

0

CatC

at_

0

KG

n_0

KG

n_1

KG

s_1

IVc-f IVd1-f

AN

TH

c_0

Cat_

1

CatC

at_

1IVd2-f

IVx1-f

PgG

sM

_0

AfG

s_1

IVx2-f

AfP

gG

s_1

Va-f

PgR

s_0

KC

oG

s_0

Vb-f

PgG

s_0

PgR

s_1

Vc-f

AfP

gG

s_0

PgR

s_0

AfC

at_

0A

fCat_

1

Cat_

0C

at_

1

CatC

at_

1

Vd-f

AN

TH

c_0

AN

TH

c_1

CyG

s_1

PgG

s_0

PgG

s_1

PgG

sM

_0

PgG

sM

_1

PgR

s_1

KC

oG

s_0

KC

oG

s_1K

Gs_0

VIa-f

AN

TH

c_1

AfP

gG

s_1

VIb-f

FR

AP

_0

TE

AC

_0

VIc-f

AfP

gG

s_0

PgR

s_1

AfG

s_0

AfG

s_1

VId-f

AfP

gG

s_1

PgG

s_1

Cat_

0C

atC

at_

0

KG

s_1

VIIa-f

TE

AC

_0

CatC

at_

0

KC

oG

s_0

VIIb-f

KC

oG

s_0

VIIc-f VIId-f

Ia-f

AfC

at_

1K

CoG

s_0

KC

oG

s_1

Ib-f

TE

AC

_1

CatC

at_

0

Ic-f

PgG

sM

_0

Id-f

FLA

Vc_0

CyG

s_1

PgG

s_0

PgR

s_1

AfC

at_

1Q

Gn_0

IIa-f

PH

EN

c_0

PgR

s_0

AfC

at_

1

CatC

at_

1

IIb-f

QG

n_1

IIc-f IId-f

KC

oG

s_1

AfP

gG

s_1

QG

n_0

IIIa-f

AfP

gG

s_0

AfC

at_

0

KC

oG

s_0

KG

s_0

KG

s_1

KG

s_1

IIIb-f

PH

EN

c_0

AfC

at_

0

KC

oG

s_0

QG

n_1

IIIc-fP

gG

sM

_1

IIId-f

(A) Female mQTLs and colorimetric QTLs

Figure 3A. Distribution of QTLs detected in ‘Capitola’ (female).QTLs were detected by QTL Cartographer and are represented by bars of different colours. The different classes of flavonoids and the colour- and antioxidant-related traits are each represented by a single colourNames of mQTLs and colorimetric QTLs are detailed in Supplemental Table S2.


https://doi.org/10.1101/2020.03.20.000455

IVa-m

KC

oG

s_0

KG

n_0

KG

s_1

IVb-m

AfG

s_0

AfG

s_1

IVc-m

KG

s_1

IVd-m

FR

AP

_0

CyG

s_1

AfG

s_0

AfG

s_1

KG

n_0

Va-m

AfC

at_

1

Cat_

1K

Gn_0

Vb-m

KC

oG

s_0

KC

oG

s_1

KC

oG

s_1

Vc-m

CyG

s_1

AfC

at_

1

Cat_

1

CatC

at_

1

KC

oG

s_0

Vd-m

AN

TH

c_0

AN

TH

c_1

TE

AC

_1

FLA

Vc_1

PH

EN

c_0

PH

EN

c_1

PgG

sM

_1

AfP

gG

s_1

PgG

s_1

PgR

s_1

AfC

at_

0A

fCat_

1

AfG

s_0

AfG

s_1

VIa-m

TE

AC

_1

PH

EN

c_0

PH

EN

c_1

PgR

s_1

Cat_

0C

at_

0

KC

oG

s_0

QG

n_0

VIb-m VIc-m

PH

EN

c_1

PgR

s_1

VId-m

KG

n_1

CatC

at_

1

PgR

s_0

PgR

s_1

VIIa-m

PgR

s_0

VIIb-m

PgG

sM

_0

PgR

s_0

Cat_

0

VIIc-m

FLA

Vc_0

PgR

s_1

AfC

at_

1C

at_

1

KC

oG

s_0

VIId-m PgR

s_1

M41-m

M44-m

AN

TH

c_0

AfP

gG

s_1

PgG

s_1

PgG

sM

_1

PgR

s_1

Ia-m Ib-mIc-m

PgG

s_1

Id-m

AfP

gG

s_0

AfP

gG

s_1

PgG

sM

_1

PgR

s_0

AfC

at_

0

IIa-m CyG

s_1

IIb-m

CatC

at_

1

IIc-m

AN

TH

c_0

IId-m

CyG

s_1

CyG

s_0

PgG

s_0

AfC

at_

0A

fCat_

1

AfG

s_1

Cat_

0

CatC

at_

0

KG

s_0

IIIa-m

AfP

gG

s_1

IIIb-m

PH

EN

c_0

AfG

s_1

KG

s_1

IIIc1-m

KC

oG

s_1

IIIc2-m

IIId-m

(B) Male mQTLs and colorimetric QTLs

Figure 3B. Distribution of QTLs detected in ‘CF1116’ (male).QTLs were detected by QTL Cartographer and are represented by bars of different colours. The different classes of flavonoids and the colour- and antioxidant-related traits are each represented by a single colour. Names of mQTLs and colorimetric QTLs are detailed in Supplemental Table S3.


https://doi.org/10.1101/2020.03.20.000455

1

SUPPORTING INFORMATION Metabolite Quantitative Trait Loci for flavonoids provide new insights into the genetic architecture

of strawberry (Fragaria ananassa) fruit quality

Marc Labadie1§, Guillaume Vallin1§, Aurélie Petit1,2, Ludwig Ring3, Thomas Hoffmann3, Amèlia Gaston1,

Aline Potier1, Wilfried Schwab3, Christophe Rothan1*, Béatrice Denoyes1*

1 INRAE, Univ. Bordeaux, UMR BFP, F-33140, Villenave d’Ornon, France

2 INVENIO, MIN de Brienne, 110 quai de Paludate, 33800 Bordeaux, France

3 Biotechnology of Natural Products, Technical University Munich, Liesel-Beckmann-Str. 1, 85354

Freising, Germany

§ These authors contributed equally to this work

* Corresponding authors

E-mail: [email protected] (B. Denoyes), [email protected] (C. Rothan)

Contents

Supplemental Figure S1. Phenotypic correlations for the traits measured in 2011.

A) Pearson phenotypic correlations (P < 0.05). 2

B) Pearson phenotypic correlation network (P < 0.05). 3

Supplemental Table S1. Trait values for ‘Capitola’ and ‘Capitola’ x ‘CF1116’ progeny in 2011. 4

Supplemental Table S2. Significant QTLs detected for ‘Capitola’ (female). 6

Supplemental Table S3. Significant QTLs detected for ‘CF1116’ (male). 9

Supplemental Table S4. Distribution of the QTLs according to HGs and LGs. 12


https://doi.org/10.1101/2020.03.20.000455

2

Supplemental Figure S1. Phenotypic correlations for the traits measured in 2011.

A) Pearson phenotypic correlations (P < 0.05). Correlation values 'r' are represented with colour

circle in upper right triangle. Scale combines circle size (small circle: correlation near 0; large circle:

correlation near 1) and colour gradient from red (negative correlation) to blue (positive correlation).

Pearson correlation values are indicated in lower left triangle. Only significant correlations are

represented (P < 0.05). A white box represents a non-significant correlation. Diagonal values are

not represented.

Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-

malonate; PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside; AfPgGs, (epi)Afzelechin-

pelargonidin-3-glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-

glucuronide ; KCoGs, Kaempferol-coumaryl-glucoside; QGn, Quercetin-glucuronide; F3ol, total

flavan-3-ols; Cat, Catechin; CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin

dimers; AfGs, (epi)Afzelechin-glucoside; ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids

(colorimetry); PHENc, phenolics (colorimetry); FRAP, antioxidant (FRAP assay) ; TEAC, antioxidant

(TEAC assay).


https://doi.org/10.1101/2020.03.20.000455

3

Figure S1. Phenotypic correlations for the traits measured in 2011.

B) Pearson phenotypic correlation network (P < 0.05). Traits are represented as nodes and

coloured according to chemical families [Anthocyanins (grey), Flavonols (orange), Flavan-3-ols

(yellow)] or to colorimetric assays (red). Positive (green) and negative (red) correlations with

absolute values r > |0.3| are represented as links between nodes. The thickness of the links

depends on the correlation values; the more the correlation value is high, the more the link is thick.

Only significant correlations are represented (P < 0.05).

Ant, total anthocyanins; PgGs, Pelargonidin-3-glucoside; PgGsM, Pelargonidin-3-glucoside-

malonate; PgRs, Pelargonidin-3-rutinoside; CyGs, Cyanidin-3-glucoside; AfPgGs, (epi)Afzelechin-

pelargonidin-3-glucoside; Fvo, total flavonols; KGs, Kaempferol-glucoside; KGn, Kaempferol-

glucuronide ; KCoGs, Kaempferol-coumaryl-glucoside; QGn, Quercetin-glucuronide; F3ol, total

flavan-3-ols; Cat, Catechin; CatCat, (epi)Catechin dimers; AfCat, (epi)Afzelechin-(epi)catechin

dimers; AfGs, (epi)Afzelechin-glucoside; ANTHc, anthocyanins (colorimetry); FLAVc, flavonoids

(colorimetry); PHENc, phenolics (colorimetry); FRAP, antioxidant (FRAP assay) ; TEAC, antioxidant

(TEAC assay).


https://doi.org/10.1101/2020.03.20.000455

4

Supplemental Table 1. Trait values for ‘Capitola’ and ‘Capitola’ x ‘CF1116’ progeny in 2011.

Traits Abbr. Capitola CF1116 Progeny h² Tr mean s.d. range mean s.d. range Cap vs CF mean s.d. range

Anthocyanins

Total anthocyanins Ant 38.52 8.37 31.65-47.84 NA NA NA NA 29.34 7.23 13.23-53.99 0.55 NA

Pelargonidin-3-glucoside PgGs 32.26 6.91 26.65-39.98 NA NA NA NA 24.87 6.06 11.19-45.31 0.54 NA

Pelargonidin-3-glucoside-malonate PgGsM 0.23 0.06 0.19-0.30 NA NA NA NA 0.15 0.34 0.00-4.12 0.97 NA

Pelargonidin-3-rutinoside PgRs 4.97 1.15 3.96-6.22 NA NA NA NA 3.58 1.24 1.16-7.62 0.8 NA

Cyanidin-3-glucoside CyGs 0.65 0.18 0.52-0.85 NA NA NA NA 0.41 0.19 0.11-1.01 0.91 NA

(epi)Afzelechin-pelargonidin-glucoside AfPgGs 0.41 0.08 0.32-0.49 NA NA NA NA 0.33 0.16 0.05-0.85 0.9 NA

Flavonols

Total Flavonols Fvo 0.27 0.05 0.21-0.31 NA NA NA NA 0.28 0.07 0.11-0.53 0.65 NA

Kaempferol-glucoside KGs 0.11 0.02 0.09-0.13 NA NA NA NA 0.13 0.04 0.05-0.26 0.74 NA

Kaempferol-glucuronide KGn 0.08 0.01 0.06-0.09 NA NA NA NA 0.06 0.02 0.02-0.16 0.82 NA

Kaempferol-coumaryl-glucoside KCoGs 0.06 0.02 0.04-0.08 NA NA NA NA 0.05 0.03 0.01-0.15 0.74 NA

Quercetin-glucuronide QGn 0.02 0.01 0.02-0.03 NA NA NA NA 0.03 0.01 0.01-0.10 0.88 NA

Flavan-3-ols

Total Flavan-3-ols F3ol 0.95 0.14 0.80-1.09 NA NA NA NA 1.32 0.42 0.52-3.02 0.87 NA

Catechin Cat 0.20 0.03 0.17-0.22 NA NA NA NA 0.24 0.06 0.12-0.49 0.64 NA

(epi)Catechin dimers CatCat 0.47 0.08 0.39-0.54 NA NA NA NA 0.55 0.16 0.24-1.11 0.77 NA

(epi)Afzelechin-(epi)catechin dimers AfCat 0.06 0.01 0.05-0.08 NA NA NA NA 0.07 0.02 0.02-0.15 0.81 NA

(epi)Afzelechin-glucoside AfGs 0.22 0.02 0.19-0.24 NA NA NA NA 0.47 0.32 0.06-1.95 0.96 NA

Colorimetry

Anthocyanins (colorimetry) ANTHc 870.64 14.42 854.2-881.4 NA NA NA NA 831.1 209.0 393.6-1456.0 1.0 NA

Flavonoids (colorimetry) FLAVc 301.51 19.60 280.9-319.9 NA NA NA NA 422.7 95.5 230.6-858.4 1.0 NA

Phenolics (colorimetry) PHENc 2.13 0.55 1.72-2.76 NA NA NA NA 2.38 0.39 1.29-4.70 0.89 NA

Antioxidant FRAP FRAP NA NA NA NA NA NA NA NA NA NA NA NA

Antioxidant TEAC TEAC 26.83 2.53 24.94-29.70 NA NA NA NA 34.41 4.77 21.45-49.15 0.95 NA

For each trait, the mean value, standard deviation (s.d.), range and heritability (h²) are described. ‘CF1116’ and transgressive values were not studied in 2011.


https://doi.org/10.1101/2020.03.20.000455

5

Flavonoid metabolites values are expressed as mg-equ/100 g FW assuming a response factor of 1. Colour- and antioxidant-related traits: ANTHc results are expressed as mg pelargonidin-3-glucoside equivalents/100 g FW. FLAVc results are expressed as μg catechin equivalents/g FW. PHENc results are expressed as mg gallic acid equivalents/g FW. FRAP and TEAC results are expressed as μm Trolox equivalents/g FW. Values are the means of n = 3 replicates per genotype.


https://doi.org/10.1101/2020.03.20.000455

6

Supplemental Table 2. Significant QTLs detected for ‘Capitola’ (female). QTLs were identified for all the traits based on CIM analysis with LOD>LOD threshold (2.8 and 3.1 for respectively α = 0.10 or 0.05). QTLs identified in the two years of the study are in boxes. LOD in bold when significance level of the QTL is for α = 0.05.

Trait Abb. Yeara QTL

nameb Repc LGd

Marker namee

Pos. (cM)

LODe R²

(%)e

Pos. LOD-

1 (cM)

e

Pos. LOD+1 (cM) e

Anthocyanins

Pelargonidin-3-glucoside

PgGs

2010 PgGs_0 1 IIa-f ggat194c 155.9 3.4 0.11 146.0 165.0

2010 PgGs_0 2 IIIb-f gaac466r-i092260r

51.4-60.1

2.9-3.6

0.10-0.13 42.5 64.1

2010 PgGs_0 1 Vc-f Pe28150r 149.2 3.1 0.09 136.0 149.2

2010 PgGs_0 2 VIa-f g020175c 36.1 4.0-4.2

0.12-0.15 16.0 44.0

2011 PgGs_1 1 VIa-f gtgx540c 26.7 3.6 0.09 22.0 38.1

2011 PgGs_1 1 VIIa-f caat159r 153.3 3.2 0.09 114.0 161.3

Pelargonidin-3-glucoside-malonate

PgGsM

2010 PgGsM_0 1 IVX2-f bx18270c 0.0 2.9 0.12 0.0 14.0

2010 PgGsM_0 1 Id-f gtat121c 81.4 3.3 0.14 65.0 87.8

2011 PgGsM_1 1 IIId-f gtgx127c 94.0 3.0 0.07 75.5 104.0

2010 PgGsM_0 2 VIa-f tgag650r 73.3 3.0-3.7

0.13-0.16 67.0 94.5

2011 PgGsM_1 3 VIa-f gtaa362r 82.5 5.4-6.1

0.15-0.17 80.0 88.5

Pelargonidin-3-rutinoside

PgRs

2011 PgRs_1 1 IIa-f tgag360r 6.0 3.1 0.10 0.0 22.0

2010 PgRs_0 1 IIb-f caaa261c 108.0 3.0 0.13 99.5 112.1

2010 PgRs_0 1 IVb-f bx48345c 75.3 2.8 0.10 59.0 86.0

2010 PgRs_0 1 Vb-f gaat430r 9.4 3.0 0.12 1.0 19.5

2011 PgRs_1 1 Vc-f MY12280c 86.9 2.9 0.08 76.5 94.9

2010 PgRs_0 1 Vd-f tgtc700c 109.0 2.9 0.14 100.5 111.5

2011 PgRs_1 1 VIa-f g020175c 40.1 2.9 0.09 23.6 44.2

2011 PgRs_1 1 VId-f tgaa142r 57.4 2.8 0.10 39.4 81.0

Cyanidin-3-glucoside

CyGs

2011 CyGs_1 1 IIa-f gaat370r 100.0 3.8 0.10 88.0 107.0

2011 CyGs_1 2 VIa-f tgag650r-caaa195c

68.3-79.6

3.5-4.1

0.10-0.11 60.0 100.5

(epi)Afzelechin-pelargonidin-

glucoside AfPgGs

2011 AfPgGs_1 2 IIIa-f u001158c 4.3 3.1-3.6 0.08 1.0 26.0

2010 AfPgGs_0 2 IIIb-f i092260r 58.2 3.0-3.1

0.10-0.12 40.8 62.0

2011 AfPgGs_1 1 Va-f M142194c 67.9 3.0 0.12 54.0 99.5

2010 AfPgGs_0 1 Vd-f MY12270c 28.9 5.1 0.22 20.0 33.0

2011 AfPgGs_1 3 VIb-f gttt401r 147.9 3.1-5.8

0.07-0.16 139.5 158.9

2010 AfPgGs_0 1 VId-f catc255r 24.7 4.1 0.17 20.0 30.0


https://doi.org/10.1101/2020.03.20.000455

7

2011 AfPgGs_1 2 VIIa-f caat159r 155.3 3.0-4.5

0.08-0.12 142.0 161.3

Flavonols

Kaempferol-glucoside

KGs

2010 KGs_0 3 IIIb-f i092260r 54.5 4.6-5.7

0.18-0.23 44.4 63.0

2011 KGs_1 1 IIIb-f gaat170r 57.0 5.0 0.13 52.0 60.0

2011 KGs_1 1 IIIb-f k033113r 104.1 4.4 0.13 100.0 107.3

2010 KGs_0 2 IVa-f i175199r 38.3-42.3

2.8-3.5 0.11 35.0 55.6

2011 KGs_1 1 IVc-f k103138r 129.3 4.0 0.10 117.0 129.3

2010 KGs_0 1 VIa-f gttt492c 94.5 3.1 0.15 87.0 100.5

2011 KGs_1 1 VIIa-f b044128r 150.8 2.9 0.07 142.0 161.3

Kaempferol-glucuronide

KGn

2010 KGn_0 3 IIIb-f gaac466r 51.4 2.9-3.0

0.11-0.12 42.0 64.2

2010 KGn_0 1 IVc-f b008168r 119.9 3.5 0.16 100.0 126.0

2011 KGn_1 2 IVc-f b008168r 117.9 3.6-4.0 0.11 107.0 126.0


KCoGs

2010 KCoGs_0 1 Ib-f u002128c 15.5 3.2 0.11 12.0 30.7

2011 KCoGs_1 1 Ib-f v019228r 18.8 3.9 0.11 16.5 21.0

2010 KCoGs_0 1 IIIb-f tgtt286r 83.7 2.8 0.10 76.0 93.0

2011 KCoGs_1 2 IIIb-f b045163c 151.1 3.1-3.4

0.09-0.12 146.0 151.1

2010 KCoGs_0 1 IIIc-f gaac230c 18.8 3.5 0.17 9.4 34.4

2010 KCoGs_0 1 Vb-f gaaa167c 50.8 4.7 0.18 46.2 57.0

2010 KCoGs_0 1 VIa-f u316234c 21.4 3.5 0.13 0.0 32.2

2011 KCoGs_1 1 VIa-f tgag650r 66.3 3.8 0.10 57.0 74.5

2010 KCoGs_0 1 VIIb-f tcaa172c 20.4 5.0 0.20 18.0 24.0

2010 KCoGs_0 1 VIIc-f tgtg178r 37.0 3.5 0.14 28.0 44.0

Quercetin-glucuronide

QGn

2010 QGn_0 1 IIa-f tgtg173c 28.1 4.5 0.20 20.0 35.0

2011 QGn_1 2 IIc-f caaa095c-gctg285c

8.0-25.6

2.8-2.9

0.07-0.08 1.0 40.0

2010 QGn_0 1 IIIa-f tcaa312c 75.0 3.3 0.14 66.0 83.0

2011 QGn_1 1 IIIc-f gaac230c 12.8 3.6 0.09 0.0 19.5

2011 QGn_1 1 IVb-f catc139c 152.5 3.9 0.20 141.0 166.0

Flavan-3-ols

Catechin Cat

2010 Cat_0 1 IVb-f FaRemc 190.0 3.2 0.13 184.0 201.0

2010 Cat_0 1 IVc-f b008168r 123.9 3.2 0.16 117.9 129.3

2011 Cat_1 1 IVd2-f gaag490c 79.7 3.3 0.13 69.7 81.7

2010 Cat_0 1 Vd-f k017142c 58.1 4.8 0.22 55.0 64.0

2011 Cat_1 1 Vd-f gaaa227c 99.9 5.6 0.15 90.0 105.5

2010 Cat_0 1 VIIa-f caaa200r 106.4 3.0 0.13 91.0 112.0

(epi)Catechin dimers

CatCat

2010 CatCat_0 1 Ic-f b041100c 47.2 2.9 0.11 39.0 47.2

2011 CatCat_1 1 IIb-f tgac137c 91.6 4.0 0.10 78.0 96.0

2010 CatCat_0 1 IVc-f b008168r 123.9 3.7 0.17 116.0 129.3

2011 CatCat_1 1 IVd2-f gaag490c 81.7 2.9 0.09 70.0 81.7

2011 CatCat_1 2 Vd-f tggx164c-gaaa227c

63.5-99.8

4.3-5.5

0.12-0.14 60.0 108.0


https://doi.org/10.1101/2020.03.20.000455

8

2010 CatCat_0 1 VIIa-f i008298r 136.9 4.8 0.19 135.0 140.9

2010 CatCat_0 1 VIIb-f gaag320c 110.9 4.7 0.20 106.8 113.0

(epi)Afzelechin-(epi)catechin

dimers AfCat

2011 AfCat_1 1 Ib-f tgtt375c 61.9 4.1 0.10 43.0 69.6

2011 AfCat_1 1 IIa-f k046144r 99.2 4.8 0.14 94.0 100.0

2011 AfCat_1 2 IIb-f ggat550c-tgac137c

90.4-91.6 3.1 0.08 72.0 96.5

2010 AfCat_0 2 IIIb-f M021198r 25.9 3.0-4.0

0.12-0.18 14.0 39.0

2010 AfCat_0 1 IIIc-f gaac230c 16.8 3.6 0.22 5.0 21.0

2010 AfCat_0 1 IVb-f bx48345c 75.3 3.2 0.15 73.0 78.0

2010 AfCat_0 1 Vd-f ccta200c 29.5 3.0 0.12 18.5 39.0

2011 AfCat_1 1 Vd-f gaaa227c 99.9 3.0 0.08 88.0 110.0

(epi)Afzelechin-glucoside

AfGs

2011 AfGs_1 3 IVX2-f bx18270c 8.0 3.0-3.2

0.09-0.10 0.0 18.5

2010 AfGs_0 2 IVb-f ccaa410c 105.2 3.2-4.3

0.14-0.16 88.8 126.0

2011 AfGs_1 1 IVb-f gata295c 205.6 2.9 0.07 199.0 205.6

2010 AfGs_0 1 VId-f tggx160r 103.8 2.8 0.11 88.0 112.0

2011 AfGs_1 1 VId-f tgaa366r 81.0 3.6 0.09 63.5 96.0

Colorimetry


ANTHc

2010 ANTHc_0 - IVd2-f gaag490c 61.7 3.0 9.4 56.5 75.7

2010 ANTHc_0 - VIa-f g020175c 34.1 4.3 14.4 23.4 40.1

2011 ANTHc_1 - VIa-f MY12160c 32.6 4.2 9.8 25.2 42.1

2011 ANTHc_1 - VIb-f caat192r 152.0 3.8 9.0 142.8 154.9

Flavonoids (colorimetry)

FLAVc 2010 FLAVc_0 - IIa-f g011311d 137.2 4.3 13.6 126.6 144.1

Phenolics (colorimetry)

PHENc 2010 PHENc_0 - IIb-f tgac137c 91.6 2.8 8.1 71.2 97.6

2010 PHENc_0 - IIIc-f gaac230c 18.8 5.4 18.9 9.4 28.4

Antioxidant FRAP

FRAP 2010 FRAP_0 - VIc-f gttg111c 66.9 3.3 11.3 63.0 88.1

Antioxidant TEAC

TEAC

2011 TEAC_1 - Ic-f CAD4c 28.5 3.2 8.7 16.0 44.4

2010 TEAC_0 - VIc-f gatg228c 66.7 3.7 12.1 59.0 90.1

2010 TEAC_0 - VIIb-f b031162c 126.2 3.5 13.2 122.2 126.2

a, year of observation of the QTL. b, name of the QTL, which includes the names of the trait and the year (0 for 2010 and 1 for 2011) c, Rep indicates the number of replicates for which the QTL was identified. d, Linkage groups (LGs) were assigned to one of the seven HGs (from I to VII) and arbitrarily labelled (e.g. LGs Ia,I b, Ic or Id) as previously described.(20,35). e, The most-closely associated marker locus (Marker name) is indicated with its position (Pos.), the LOD, the R2 (phenotypic variance explained by the QTL in %) and the confidence interval (LODmax ± 1 LOD).


https://doi.org/10.1101/2020.03.20.000455

9

Supplemental Table 3. Significant QTLs detected for ‘CF1116’ (female). QTLs were identified for all the traits based on CIM analysis with LOD>LOD threshold (2.8 and 3.1 for respectively α = 0.10 or 0.05). QTLs identified in the two years of the study are in boxes. LOD in bold when significance level of the QTL is for α = 0.05.

Trait Abb. Year QTL

name Rep LG Marker name

Pos. (cM)

LOD R²

(%)

Pos. up LOD-1 (cM)

Pos. down LOD-1 (cM)

Anthocyanins

Pelargonidin-3-glucoside

PgGs 2011 PgGs_1 2 Ia-m cctc227c 27.0 3.4-5.6 9-14 19.5 29.5

PgGs 2011 PgGs_1 1 Id-m tcaa218c 24.5 2.8 7 15.5 35.3

PgGs 2010 PgGs_0 3 IIIa-m gata165c 27.5 3.7-7.4

14-27 25.0 29.0

PgGs 2011 PgGs_1 1 VIa-m tgta383c 54.0 2.8 7 47.7 63.6

Pelargonidin-3-glucoside-malonate

PgGsM 2011 PgGsM_1 1 Ia-m cctc227c 27.1 3.7 8 18.0 29.5

PgGsM 2011 PgGsM_1 1 IIa-m b037222r 27.8 3.3 8 12.0 32.0

PgGsM 2011 PgGsM_1 1 VIa-m MY66228c 92.0 3.0 8 92.0 98.1

PgGsM 2010 PgGsM_0 1 VIIc-m ccta282c 65.7 4.5 36 65.5 67.8

Pelargonidin-3-rutinoside

PgRs 2011 PgRs_1 2 Ia-m tgaa197c-cctc227c

23.8-27.1 7.3 16 18.4 29.5

PgRs 2010 PgRs_0 1 IIa-m gatt290r 83.5 2.8 15 77.5 99.4

PgRs 2011 PgRs_1 1 VIa-m gtgx530c 52.3 5.6 11 49.2 55.0

PgRs 2011 PgRs_1 2 VIb-m tcac235c 136.2 2.9-4.6 6-10 129.0 141.0

PgRs 2011 PgRs_1 1 VId-m b048297c 4.7 3.9 9 0.0 7.0

PgRs 2010 PgRs_0 2 VIIa-m b004220d-caag330c

32.3-33.1

4.1-4.2

16-16 30.0 33.0

PgRs 2011 PgRs_1 2 VIIa-m gatt560r-gaac170s 8.9

3.1-3.3 7 0.0 18.0

PgRs 2010 PgRs_0 1 VIIb-m AAT700c 10.3 4.0 17 0.1 14.7

PgRs 2010 PgRs_0 1 VIIc-m tgta285r 57.9 4.4 36 53.0 60.0

PgRs 2011 PgRs_1 1 VIId-m gtag158r 44.7 3.2 7 34.7 51.1

PgRs 2011 PgRs_1 1 M41-m gatc390c 4.0 3.1 9 0.0 4.0

Cyanidin-3-glucoside

CyGs 2011 CyGs_1 1 IIb-m cttg188c 7.8 4.1 10 3.8 9.7

CyGs 2010 CyGs_0 1 IIIa-m gatc215d 63.5 4.1 18 49.5 65.8

CyGs 2011 CyGs_1 1 IIIa-m gatc215d 69.5 2.9 8 58.0 76.9

CyGs 2011 CyGs_1 2 Va-m u006199r-gtgx171c

38.9-57.9 2.9 7-11 45.0 55.9

CyGs 2011 CyGs_1 1 Vd-m caag138c 27.2 3.2 9 19.2 29.2

(epi)Afzelechin-pelargonidin-

glucoside

AfPgGs 2011 AfPgGs_1 2 Ia-m tgaa197c 23.8 4.6-6.2

10-14 16.5 30.1

AfPgGs 2010 AfPgGs_0 1 IIa-m tgag330r 66.4 3.4 24 56.7 83.5

AfPgGs 2011 AfPgGs_1 2 IIa-m b037215c-c020120r

31.8-34.4

5.9-7.5

16-18 19.4 38.3

AfPgGs 2011 AfPgGs_1 1 IIIb-m gtag280r 53.9 3.1 7 51.0 63.2

AfPgGs 2011 AfPgGs_1 1 VIa-m b047143c 58.3 4.4 10 47.0 62.0

Flavonols Kaempferol-

glucoside KGs 2010 KGs_0 1 IIIa-m gata165c 27.5 3.0 12 22.5 34.5

KGs 2011 KGs_1 1 IIIc1-m tgtt390c 36.8 3.2 9 25.6 36.8


https://doi.org/10.1101/2020.03.20.000455

10

KGs 2011 KGs_1 1 IVb-m gatt252c 0.0 3.6 10 0.0 10.5

KGs 2011 KGs_1 1 IVd-m tgac408c 68.9 2.8 9 56.0 82.8

Kaempferol-glucuronide

KGn 2010 KGn_0 1 IVb-m tcaa295c 35.1 2.9 12 27.8 30.4

KGn 2010 KGn_0 1 Va-m cctg124r 108.4 4.3 19 106.2 123.7

KGn 2010 KGn_0 1 Vb-m catc192c 24.4 3.6 15 17.0 31.3

KGn 2011 KGn_1 1 VIId-m gtag158r 34.7 3.5 9 28.0 41.0


KCoGs 2011 KCoGs_1 1 IIIc2-m caaa338c 3.2 2.8 7 0.0 14.0

KCoGs 2010 KCoGs_0 2 IVb-m tcaa295c 35.1 3.6-4.1 14 23.8 39.0

KCoGs 2010 KCoGs_0 1 Vc-m k017156c 44.4 3.6 15 40.6 47.0

KCoGs 2011 KCoGs_1 1 Vc-m tgta163r 34.8 3.6 11 28.0 37.0

KCoGs 2011 KCoGs_1 1 Vc-m tgtt140c 75.4 3.7 10 61.5 75.4

KCoGs 2010 KCoGs_0 1 Vd-m caag138c 19.2 2.8 10 5.0 29.0

KCoGs 2010 KCoGs_0 3 VIb-m tgaa195r-tcac235c

104.5-136.2

3.0-5.9

12-22 93.8 128.0

KCoGs 2010 KCoGs_0 1 VIId-m gtta161r 73.6 3.5 16 62.0 73.6 Quercetin-glucuronide QGn 2010 QGn_0 1 VIb-m b010162d 152.2 4.8 44 148.0 158.2

Flavan-3-ols

Catechin

Cat 2010 Cat_0 1 IIIa-m b048280c 37.6 3.4 17 27.3 42.6

Cat 2011 Cat_1 2 Vb-m gtat373c 1.0 3.2 9 0.0 11.0

Cat 2011 Cat_1 1 Vd-m gaat525c 16.1 2.8 7 0.0 29.2

Cat 2010 Cat_0 1 VIb-m tgtc358r 8.5 3.0 14 7.0 11.0

Cat 2010 Cat_0 1 VIb-m b010162d 146.2 2.9 13 139.0 160.0

Cat 2010 Cat_0 1 VIIc-m ccta282c 93.7 3.0 13 78.2 93.7

Cat 2011 Cat_1 1 VIId-m i008311r 33.3 3.4 8 23.3 42.9

(epi)Catechin dimers

CatCat 2011 CatCat_1 1 IIc-m i146188c 90.3 4.5 11 85.5 93.0

CatCat 2010 CatCat_0 1 IIIa-m tgaa325c 34.1 4.5 20 30.0 42.0

CatCat 2011 CatCat_1 1 Vd-m caag138c 21.2 4.3 11 8.2 27.7

CatCat 2011 CatCat_1 1 VIId-m gtat380r 1.8 4.8 12 0.0 13.0

(epi)Afzelechin-(epi)catechin

dimers

AfCat 2010 AfCat_0 1 IIa-m gatg161r 48.7 3.4 14 43.0 51.0

AfCat 2010 AfCat_0 1 IIIa-m b048280c 41.6 3.3 14 36.6 43.7

AfCat 2011 AfCat_1 1 IIIa-m gatc620c 132.5 3.5 8 112.0 139.0

AfCat 2011 AfCat_1 2 Vb-m r5g1172r-gaat440c

29.2-38.0

3.1-3.8 6-9 1.0 34.0

AfCat 2011 AfCat_1 1 Vd-m caag138c 19.2 4.0 9 8.1 25.5

AfCat 2010 AfCat_0 1 VIa-m gttg124r 38.4 2.8 12 29.0 51.1

AfCat 2011 AfCat_1 2 VIa-m v006205c 59.6 3.6-5.5 9-12 56.2 62.0

AfCat 2011 AfCat_1 1 VIId-m gtat380r 3.8 3.3 7 0.0 14.7

(epi)Afzelechin-glucoside

AfGs 2011 AfGs_1 3 IIIa-m v004260c-b048280c

43.6-43.9

3.2-3.3 7 32.1 52.0

AfGs 2011 AfGs_1 2 IIIc1-m v029211r-i075224r

6.0-10.2

2.9-3.2 6-8 0.0 18.0

AfGs 2010 AfGs_0 1 IVc-m tggx318c 44.6 4.3 18 38.6 63.6

AfGs 2011 AfGs_1 2 IVc-m gtgx380r-tgat165c

52.0-55.6

3.0-3.5 8 45.0 65.6

AfGs 2010 AfGs_0 1 Va-m gaat270c 21.6 3.2 13 0.1 31.6


https://doi.org/10.1101/2020.03.20.000455

11

AfGs 2011 AfGs_1 3 Va-m gtgx171c 32.9-40.9

2.9-4.0 6-9 20.3 46.9

AfGs 2010 AfGs_0 1 VIa-m gata170c 71.0 3.0 13 65.0 79.8

AfGs 2011 AfGs_1 3 VIa-m ccta102r-

MY66190c 112.4-113.9

2.9-5.5 9-15 102.3 122.0

Colorimetry


ANTHc 2010 ANTHc_0 - Ia-m tgaa197c 21.8 2.8 8 17.4 35.9

ANTHc 2010 ANTHc_0 - IId-m g011313c 31.7 3.4 10 25.6 38.6

ANTHc 2010 ANTHc_0 - VIa-m v006205c 59.6 3.7 13 58.3 63.6

ANTHc 2011 ANTHc_1 - VIa-m gata170c 71.0 4.8 14 64.5 75.0

Flavonoids (colorimetry)

FLAVc 2010 FLAVc_0 - VIId-m i008311r 25.3 3.2 12 20.6 33.3

FLAVc 2011 FLAVc_1 - VIa-m ccta102r 100.0 5.9 16 90.8 108.4

Phenolics (colorimetry)

PHENc 2010 PHENc_0 - IIIc1-m tcag430r 25.4 3.6 13 16.7 31.0

PHENc 2010 PHENc_0 - VIa-m MY66228c 92.0 3.6 11 79.8 102.4

PHENc 2010 PHENc_0 - VIb-m ggat388d 125.8 4.8 17 119.8 129.8

PHENc 2011 PHENc_1 - VIa-m MY66228c 96.0 3.8 9 82.6 102.4

PHENc 2011 PHENc_1 - VIb-m n228278c 13.7 2.8 6 2.0 29.5

PHENc 2011 PHENc_1 - VId-m i175224c 38.4 3.8 9 30.6 46.4

Antioxidant FRAP FRAP 2010 FRAP_0 - Va-m b037227c 53.7 3.6 12 42.9 74.9

Antioxidant TEAC TEAC 2011 TEAC_1 - VIa-m ccta102r 102.4 5.1 12 92.0 110.4

TEAC 2011 TEAC_1 - VIb-m tcta380r 27.5 2.8 8 22.3 36.3

a, year of observation of the QTL. b, name of the QTL, which includes the names of the trait and the year (0 for 2010 and 1 for 2011) c, Rep indicates the number of replicates for which the QTL was identified. d, Linkage groups (LGs) were assigned to one of the seven HGs (from I to VII) and arbitrarily labelled (e.g. LGs Ia,I b, Ic or Id) as previously described.(20,35). e, The most-closely associated marker locus (Marker name) is indicated with its position (Pos.), the LOD, the R2 (phenotypic variance explained by the QTL in %) and the confidence interval (LODmax ± 1 LOD).


https://doi.org/10.1101/2020.03.20.000455

12

Supplemental Table 4. Distribution of the number of QTLs according to the homoeology groups and the linkage groups. Linkage groups (LGs) were assigned to one of the seven HGs and arbitrarily labelled (e.g. LG Ia, b, c or d) as previously described (Rousseau-Gueutin et al. 2008).


https://doi.org/10.1101/2020.03.20.000455

13


https://doi.org/10.1101/2020.03.20.000455

Documents

Metabolite Quantitative Trait Loci for flavonoids …...2020/03/20 · 98 A pseudo full-sibling F 1 population of 165 individuals obtained from a cross between the variety 99 ‘