Identification and characterisation of mutations in homoeologues of GA20OX1 and GA20OX2
We screened an ethyl methanesulphonate (EMS)-mutagenized wheat population using CelI digestion and high-resolution melting analysis and identified lines carrying mutations introducing premature stop codons in GA20OX1-B1 and GA20OX2-A1, mutations in the splice acceptor sites of GA20OX1-A1 and GA20OX2-B1, and several mis-sense mutations in GA20OX1-D1 and GA20OX2-D1 predicted to have deleterious effects on the activity of the encoded enzyme (Fig. 2, Table 1 and Additional file 1, Table S3).
Table 1
Induced mutations in GA20OX1 and GA20OX2 evaluated in this study. Sorting Intolerant From Tolerant (SIFT) scores for amino acid substitutions in GA20OX1-D1 and GA20OX2-D1 are provided in brackets, where values less than 0.05 indicates the mutation is predicted to have a deleterious effect on protein function. * indicates the introduction of a premature stop codon in the coding sequence.
Gene
|
IWGSCv1.1 ID
|
Allele ID
|
Effect
|
TaGA20OX1-A1
|
TraesCS4A02G319100
|
ga20ox1-A1b
|
Splice acceptor
|
TaGA20OX1-B1
|
TraesCS5B02G560300
|
ga20ox1-B1b
|
Q213*
|
TaGA20OX1-D1
|
TraesCS5D02G566200
|
ga20ox1-D1b
|
G187D (0.02)
|
|
|
ga20ox1-D1c
|
R190H (0.01)
|
TaGA20OX2-A1
|
TraesCS3A02G406200
|
ga20ox2-A1b
|
Q136*
|
TaGA20OX2-B1
|
TraesCS3B02G439900
|
ga20ox2-B1b
|
Splice acceptor
|
TaGA20OX2-D1
|
TraesCS3D02G401400
|
ga20ox2-D1b
|
H152V (0)
|
|
|
ga20ox2-D1c
|
G318E (0)
|
|
|
ga20ox2-D1d
|
R359H (0)
|
The GA20OX1-B1b and GA20OX2-A1b alleles carry point mutations resulting in the complete loss of enzyme activity as they both cause premature termination of translation at a position upstream of the Fe- and 2-oxoglutarate-binding residues in the GA20OX protein sequences that are essential for protein function (Additional file 1, Figure S2).
Although mutations in conserved splice acceptor sites are assumed to result in aberrant splicing, latitude in the splicing mechanism or the presence of alternative, in-frame splice acceptor sites may result in the production of transcripts encoding a functional protein. To investigate the effects of the splice acceptor site mutations GA20OX1-A1b and GA20OX2-B1b (Fig. 2) on gene function we carried out deep sequencing of RT-PCR products spanning intron 1 of all homoeologues of the two genes, using cDNA generated from peduncular ligule tissue of single homozygous mutants and segregating wild-type lines (Fig. 3, Additional file 1, Table S4). In segregating wild-type plants, the majority of transcripts from all three homoeologues of GA20OX1 and GA20OX2 were correctly spliced, with at most 7% of transcripts being un-spliced. By contrast, in plants carrying the GA20OX1-A1b allele just 8% of transcripts from GA20OX1-A1 were correctly spliced, whereas 73% were mis-spliced at positions + 1 or + 4 relative to the correct site, resulting in a translational frameshift (Fig. 3). The remainder of the transcripts (18%) were un-spliced (Fig. 3). Transcripts of all other GA20OX1 or GA20OX2 homoeologues in this mutant were correctly spliced as in the wild-type line (Additional file 1, Table S4). Similarly, in lines carrying the GA20OX2-B1b allele, 8% of transcripts from GA20OX2-B1 were correctly spliced, while the majority (61%) of transcripts were un-spliced, with the remainder mis-spliced at positions + 1 and + 13 (Fig. 3, Additional file 1, Table S4).
For GA20OX1-D1 and GA20OX2-D1 we failed to find either non-sense or splice-site mutations (Additional file 1, Table S3). However, in each case we identified mis-sense mutations in conserved residues that Sorting Intolerant From Tolerant (SIFT) analysis suggested were highly deleterious (Fig. 2, Additional file 1, Figure S2, Table 1). To directly assess the effect of these mutations on enzyme activity we selected two mis-sense alleles of GA20OX1-D1 (GA20OX1-D1b and GA20OX1-D1c) and three mis-sense alleles of GA20OX2-D1 (GA20OX2-D1b, GA20OX2-D2c and GA20OX2-D1d) (Table 1). We replicated each mutation in a cDNA copy of the respective gene by synthesis and expressed the wild-type and mutant versions in E. coli. Cell lysates producing wild-type GA20OX1-D1 catalysed conversion of 14C-labelled GA12 to GA9, involving multiple oxidation steps and loss of C-20 (Additional file 1, Figure S3A). In contrast, expression products containing the mis-sense mutations GA20OX1-D1b and GA20OX1-D1c were almost completely inactive, producing only a small amount of GA15, with a single oxidation at C-20 (Additional file 1, Figure S3A). It was therefore concluded that both GA20OX1-D1b and GA20OX1-D1c encode null alleles of this gene. Due to similar results from heterologous expression and limited resources, only ga20ox1-D1b was taken forward into crosses with GA20OX1-A1b and GA20OX1-B1b to generate the ga20ox1 mutant.
Despite numerous attempts to express GA20OX2-D1 in E. coli, we could not demonstrate the full range of GA 20-oxidase activities for this protein: cell lysates from the wild-type GA20OX2-D1 converted GA12 only to GA15 (Additional file 1, Figure S3B). However, cell lysates from E. coli expressing cDNAs containing the mis-sense mutations GA20OX2-D1b, GA20OX2-D1c and GA20OX2-D1d had no detectable GA 20-oxidase activity (Additional file 1, Figure S3B), suggesting that these alleles should also be regarded as significantly impaired. All three alleles were taken forward for crossing with GA20OX2-A1b and GA20OX2-B1b alleles to develop three ga20ox2 mutant lines with different GA20OX2-D1 mutant alleles. Henceforth, we refer to these three mutant lines as ga20ox2(b), ga20ox2(c) and ga20ox2(d).
Stacking mutations in GA20OX1 and GA20OX2 homoeologues
To generate ga20ox1, ga20ox2(b), ga20ox2(c) and ga20ox2(d) triple mutants, mutations in all three homoeologues were stacked and then backcrossed a minimum of three times to reduce the number of background EMS mutations (Additional file 1, Figure S1). Analysis of allele frequencies in the F2 generations of selfed triple heterozygotes (AaBbDd) in both GA20OX paralogues showed evidence of distorted segregation. All mutant alleles of GA20OX1 and GA20OX2 showed significantly reduced transmission, with the frequency of each mutant allele in the F2 generation being 8.1–15.7% lower than expected (Additional file 1, Table S5).
Glasshouse phenotyping of GA20OX1 and GA20OX2 mutants
Initial phenotypic characterisation of the ga20ox1 and ga20ox2 triple mutants was performed in the glasshouse using homozygous lines at the BC4 − 5F3 or subsequent generation combined across backcrossing schemes (Additional file 1, Figure S1). Compared to their respective wild-type segregant, plant height was significantly reduced by 11.4% in the ga20ox1 triple mutant (P < 0.001), by 12.0% in the ga20ox2(b) mutant (P < 0.001) and by 8.8% in the ga20ox2(c) mutant (P < 0.01) (Fig. 4A, Additional file 1, Table S6, Additional file 1, Figure S4A). Although the ga20ox2(d) mutant was 5.5% shorter than the wild-type segregant, this difference was not significant (P > 0.05) (Fig. 4A, Additional file 1, Table S6). Representative photos showing the height phenotypes of all lines are shown in Additional file 1, Figure S4A.
To determine the contribution of individual GA20OX1 and GA20OX2 homoeologues, we measured plant height in lines carrying different allelic combinations in segregating BC0F2 and BC0F3 populations in the glasshouse (Additional file 1, Table S7). For GA20OX1, all three double mutant combinations exhibited intermediate height between the wild-type and ga20ox1 triple mutant, although only the aabbDD and AAbbdd double mutant lines were significantly shorter than the wild-type (P < 0.05) (Fig. 4B). The aaBBdd double mutant was significantly taller than the triple mutant (P < 0.05), suggesting that GA20OX1-B1 is the most active GA20OX1 homoeologue in developing wheat stems (Fig. 4B). In a population segregating for the ga20ox2(b) mutant alleles, the aaBBdd and AAbbdd double mutants were both significantly shorter than the wild-type (P < 0.05), but were not significantly different in height from the triple mutant (Fig. 4C). The aabbDD double mutant was not significantly different from the wild-type (P > 0.05) suggesting that GA20OX2-D1 is the most active GA20OX2 homoeologue in these tissues (Fig. 4C).
There was evidence of delayed flowering in some lines: the ga20ox1 and ga20ox2(b) triple mutants both headed approximately 3 days later than their corresponding wild-type segregants (P < 0.01) (Additional file 1, Table S6). However, heading date was not significantly different between ga20ox2(c) and ga20ox2(d) triple mutants and their respective wild-type segregants (P > 0.05) (Additional file 1, Table S6).
Taken together, these results suggest that both GA20OX1 and GA20OX2 contribute to wheat stem elongation in glasshouse conditions and that for both genes, all three homoeologues contribute additively to GA biosynthesis in stem tissues.
Field phenotyping of GA20OX1 and GA20OX2 mutants
Initial field phenotyping was carried out in 1 m2 replicated plots sown in spring 2020, precluding accurate yield measurements. Within each genotype, plants from different backcrossing streams (three streams for ga20ox1 and ga20ox2(b), two streams for ga20ox2(c) and ga20ox2(d)) all showed similar height phenotypes (Additional file 1, Table S8), so we analysed the combined data across streams. The ga20ox1 homozygous mutant conferred a 4.1% reduction in height and was significantly shorter than the wild-type segregating line (P < 0.05, Fig. 5A) but the difference was much smaller than that observed in glasshouse conditions (Fig. 4A). All three ga20ox2 homozygous mutants conferred significant reductions in plant height of between 20.9% and 32.1% compared to their wild-type segregant lines (P < 0.001) (Fig. 5A, Additional file 1, Table S8). The reduction in height conferred by ga20ox2 mutations was slightly greater than that of Rht-D1b, which was 17.9% shorter than the wild-type Rht-D1a allele in the ‘Cadenza’ background (P < 0.001) (Fig. 5A). Representative photos of all lines grown in field conditions are shown in Additional file 1, Figure S4B.
The seeds harvested from these plots were pooled by genotype and used to sow large-scale field experiments in spring 2021. The ga20ox2(d) mutant was not included in this experiment. The ga20ox1 triple homozygous mutant was just 3.0% shorter than its wild-type segregant and the difference was not significant (P > 0.05) (Table 2, Fig. 5B). The ga20ox2(b) and ga20ox2(c) stacked mutations conferred significant reductions in height of 12.1% (P < 0.001) and 18.4% (P < 0.001) respectively, comparable to the 19.4% reduction conferred by the Rht-D1b allele (P < 0.001) (Table 2, Fig. 5B). Measurements of individual internodes showed that in both ga20ox2(b) and ga20ox2(c) mutants the three uppermost internodes, including the peduncle, were all significantly shorter than in their wild-type segregants and were affected approximately equally (Additional file 1, Table S9). By contrast, no stem internode was significantly different in length between the ga20ox1 mutant and its segregating wild-type line (Additional file 1, Table S9).
Heading date was delayed by between 2 and 4 days in all mutant lines compared to their respective wild-type segregants, although the difference was only significant for the ga20ox2(c) mutant (P < 0.001) (Table 2).
Table 2
Phenotypic data of Rht-D1b, ga20ox1 and ga20ox2 mutations in the field in 2021. Grain yield measurements adjusted to 85% dry matter. Significant differences are highlighted in orange. *** = P < 0.001 contrasting the wild-type and mutant segregant lines from each genotype based on a two-tailed Student’s t-test.
Genotype
|
Days to heading
|
Plot height (cm)
|
Thousand Grain Weight (g)
|
Grain yield (t/ha)
|
Longitudinal grain area (mm2)
|
Cadenza
|
103.8 ± 0.5
|
773.8 ± 5.0
|
47.1 ± 0.1
|
7.7 ± 0.3
|
20.2 ± 0.1
|
Rht-D1b
|
106.5 ± 0.6
|
624.1 ± 7.2***
|
40.6 ± 0.3***
|
6.6 ± 0.5
|
19.0 ± 0.1***
|
GA20OX1 WT
|
103.8 ± 0.8
|
784.7 ± 10.3
|
46.3 ± 0.6
|
7.6 ± 0.5
|
20.1 ± 0.1
|
ga20ox1
|
105.8 ± 0.8
|
761.3 ± 18.5
|
46.8 ± 0.9
|
7.6 ± 0.8
|
19.8 ± 0.2
|
GA20OX2(b) WT
|
104.3 ± 1.0
|
784.7 ± 6.3
|
46.7 ± 0.5
|
8.1 ± 0.4
|
19.7 ± 0.2
|
ga20ox2(b)
|
106.8 ± 0.5
|
690.0 ± 10.3***
|
42.7 ± 0.6***
|
7.4 ± 0.5
|
19.3 ± 0.1
|
GA20OX2(c) WT
|
104.3 ± 0.6
|
788.8 ± 6.9
|
43.6 ± 1.1
|
7.9 ± 0.3
|
19.2 ± 0.1
|
ga20ox2(c)
|
108.0 ± 0.4***
|
634.4 ± 6.3***
|
42.1 ± 0.2
|
7.4 ± 0.1
|
19.1 ± 0.1
|
Grain yield from the ga20ox2(b) and ga20ox2(c) field plots were 9.0% and 6.4% lower than their respective wild-type plots, although these differences were not significant (Table 2). Yield was also reduced by 13% in Rht-D1b plots compared to ‘Cadenza’ wild-type plots, and marginally higher in the ga20ox1 mutant, but neither of these differences were significant (Table 2). The ga20ox2(b), ga20ox2(c) and Rht-D1b plots that had slight reductions in yield also exhibited reduced thousand grain weight (TGW), although the difference was significant only in ga20ox2(b) (P < 0.001) (Table 2). Longitudinal grain area was lower in the mutant alleles of all four genotypes but was only significantly different in Rht-D1b (P < 0.001) (Table 2).
To confirm these results, the field assessment was repeated as autumn-sown large-scale plots, but with the segregating wild-type lines omitted. Total height of the ga20ox1 mutant was just 1.2% shorter than ‘Cadenza’ and was not significantly different (P > 0.05) (Table 3). Both ga20ox2(b) (13.7%, P < 0.01) and GA20ox2(c) (29.9%, P < 0.001) mutants were significantly shorter than ‘Cadenza’, comparable to the Rht-D1b mutant (15.5% shorter, P < 0.01) (Table 3). Consistent with the previous field evaluations, the dwarfing effect of ga20ox2 and Rht-D1b mutations affected the lengths of the peduncle and upper three internodes approximately equally (Additional file 1, Table S10). Yields of ga20ox1, ga20ox2(b) and Rht-D1b plots were slightly higher than ‘Cadenza’, while the yield of ga20ox2(c) plots was slightly lower, but none of these differences were significant (P > 0.05; Table 3). Thousand grain weights of ga20ox1 and Rht-D1b were also slightly greater than ‘Cadenza’, but not significantly so, whereas TGWs of ga20ox2(b) and ga20ox2(c) were significantly reduced by 8.4% and 12.4%, respectively (Table 3). Reflecting this, grain area was also reduced by 4.4% in ga20ox2(b) mutants and by 5.5% in ga20ox2(c) mutants, but these differences were not significant (Table 3). Neither spikelet number, nor grain number per spike was significantly affected in ga20ox1 or any ga20ox2 mutant, suggesting these mutations have a minimal effect on inflorescence development. As in the spring-sown experiment, heading date was delayed by 2–3 days in ga20ox1, ga20ox2(b) and Rht-D1b mutants and by 7.5 days in the ga20ox2(c) mutant (Table 3).
Table 3
Field data from field experiment sown in autumn 2022. Results are means from 4 plots of each genotype +/- standard error of the mean. Values highlighted in yellow indicate a significant difference from ‘Cadenza’ (P < 0.001) based on a two-tailed Student’s t-test.
Genotype
|
Plant height (cm)
|
Days to heading
|
Thousand Grain Weight (g)
|
Longitudinal grain area (mm2)
|
Grain yield (t/ha)
|
Spikelet number
|
Seeds per spike
|
Cadenza
|
935.4 ± 15.5
|
223.5 ± 0.3
|
50.4 ± 0.6
|
21.1 ± 0.1
|
10.0 ± 0.5
|
23.2 ± 0.2
|
69.6 ± 1.9
|
Rht-D1b
|
790.4 ± 6.0
|
226 ± 0.3
|
51.5 ± 0.4
|
21.5 ± 0.2
|
10.2 ± 0.4
|
23.2 ± 0.3
|
76.0 ± 1.8
|
ga20ox1
|
923.5 ± 13.5
|
225.5 ± 0.3
|
51.8 ± 0.6
|
21.0 ± 0.2
|
10.5 ± 0.3
|
22.9 ± 0.1
|
72.1 ± 2.0
|
ga20ox2(b)
|
807.0 ± 16.3
|
226.3 ± 0.3
|
46.1 ± 0.8
|
20.2 ± 0.2
|
10.1 ± 0.5
|
22.9 ± 0.3
|
71.2 ± 1.9
|
ga20ox2(c)
|
656.1 ± 13.9
|
231.0 ± 0.4
|
44.1 ± 0.5
|
20.0 ± 0.3
|
9.9 ± 0.5
|
22.9 ± 0.2
|
66.6 ± 2.2
|
Gibberellin and transcript quantification in GA20OX1 and GA20OX2 mutants
Endogenous 13-OH and 13-H GAs were quantified in peduncles and peduncular nodes of the primary and secondary tillers of ga20ox1, ga20ox2(b) and ga20ox2(c) mutants and their respective wild-type lines grown in glasshouse and field conditions. In the glasshouse experiment, where both ga20ox1 and ga20ox2(b) mutants exhibited reduced height (Fig. 4A), both mutants showed reductions in bioactive GA1 and GA4 levels compared to their wild-type segregants, although only the reduction in GA4 in the ga20ox2(b) mutant was significant (P < 0.05, Table 4). Both lines had significantly lower levels of the 2β-hydroxylated C19-GA GA29 (Table 4) and in the ga20ox1 mutant, there was also a significant increase in the C20-GA substrates of GA 20-oxidase GA19 (P < 0.05, Table 4). In the 2021 field experiment, most GAs showed no significant changes in the ga20ox1 mutant, consistent with the lack of a height phenotype in this environment (Fig. 5B). Levels of bioactive GA1 and GA4 were lower in both the ga20ox2(b) and ga20ox2(c) mutants compared to their wild-type segregants, although the difference was significant only for GA1 in the ga20ox2(c) mutant (P < 0.05, Table 4). Both mutants showed a significant increase in GA53, the C20 substrate of GA 20-oxidases (Table 4).
Table 4
GA quantification in GA20OX1 and GA20OX2 wild-type and mutant genotypes in the glasshouse and field. Significant differences are highlighted in orange. * = P < 0.05, ** = P < 0.01, *** = P < 0.001 contrasting the wild-type and mutant segregant lines from each genotype. Genotype
|
GA53
|
GA44
|
GA19
|
GA20
|
GA1
|
GA29
|
GA8
|
GA4
|
GA34
|
Glasshouse
|
|
|
|
|
|
|
|
|
|
GA20OX1 WT
|
1.60 ± 0.29
|
16.49 ± 1.40
|
18.72 ± 0.75
|
1.60 ± 0.12
|
3.16 ± 0.14
|
1.11 ± 0.06
|
7.22 ± 1.26
|
1.20 ± 0.29
|
0.09 ± 0.01
|
ga20ox1
|
1.97 ± 0.31
|
15.50 ± 0.18
|
21.03 ± 0.39*
|
1.62 ± 0.12
|
2.55 ± 0.23
|
0.75 ± 0.07**
|
4.65 ± 0.57
|
0.44 ± 0.14
|
0.05 ± 0.01**
|
GA20OX2(b) WT
|
2.75 ± 1.13
|
14.12 ± 1.56
|
18.18 ± 1.12
|
2.23 ± 0.15
|
2.81 ± 0.24
|
0.96 ± 0.06
|
5.05 ± 0.44
|
0.53 ± 0.08
|
0.08 ± 0.01
|
ga20ox2(b)
|
1.92 ± 0.18
|
11.55 ± 0.36
|
19.72 ± 0.40
|
1.36 ± 0.25*
|
2.07 ± 0.22
|
0.62 ± 0.06**
|
4.29 ± 0.45
|
0.27 ± 0.04*
|
0.03 ± 0.00
|
Field
|
|
|
|
|
|
|
|
|
|
GA20OX1 WT
|
0.11 ± 0.02
|
3.41 ± 0.19
|
7.38 ± 0.45
|
2.33 ± 0.18
|
1.27 ± 0.11
|
1.86 ± 0.13
|
3.73 ± 0.16
|
0.62 ± 0.14
|
0.06 ± 0.00
|
ga20ox1
|
0.13 ± 0.03
|
3.07 ± 0.25
|
9.14 ± 0.56
|
1.67 ± 0.07*
|
1.21 ± 0.06
|
1.56 ± 0.12
|
3.22 ± 0.23
|
0.42 ± 0.05
|
0.05 ± 0.01
|
GA20OX2(b) WT
|
0.11 ± 0.02
|
3.43 ± 0.19
|
7.80 ± 0.27
|
1.75 ± 0.20
|
1.38 ± 0.06
|
1.96 ± 0.06
|
3.59 ± 0.14
|
0.43 ± 0.12
|
0.06 ± 0.00
|
ga20ox2(b)
|
0.26 ± 0.04*
|
3.75 ± 0.63
|
11.15 ± 1.45
|
2.24 ± 0.11
|
1.15 ± 0.08
|
1.59 ± 0.26
|
2.58 ± 0.29
|
0.40 ± 0.13
|
0.02 ± 0.00***
|
GA20OX2(c) WT
|
0.08 ± 0.01
|
3.31 ± 0.20
|
7.11 ± 0.67
|
1.91 ± 0.19
|
1.40 ± 0.09
|
1.94 ± 0.19
|
3.97 ± 0.20
|
0.49 ± 0.08
|
0.04 ± 0.00
|
ga20ox2(c)
|
0.15 ± 0.02*
|
2.55 ± 0.18*
|
6.29 ± 0.16
|
1.32 ± 0.05
|
1.03 ± 0.09*
|
1.55 ± 0.11
|
2.69 ± 0.22**
|
0.39 ± 0.04
|
0.04 ± 0.01***
|
To investigate whether the phenotypes of ga20ox1 and ga20ox2 mutants might be affected by homeostatic regulation of GA biosynthesis, we quantified transcript levels of selected GA biosynthesis and signalling genes in elongating peduncles. In the glasshouse experiment, expression of GA20OX1 was significantly reduced in the ga20ox1 mutant compared to the ‘Cadenza’ control and, conversely, GA20OX2 expression was lower in the ga20ox2(b) mutant, possibly reflecting the effects of the splice-site mutations in ga20ox1-A1b and ga20ox2-B1b and nonsense-mediated mRNA decay in ga20ox1-B1b and ga20ox2-A1b (Table 5). The ga20ox2(b) mutant also exhibited significantly higher transcript levels of GA20OX4, GA3OX2 and GID1 (P < 0.05, Table 5). These genes were also more highly expressed in the ga20ox1 mutant, but the increases were smaller and not statistically significant (Table 5). In the 2021 field experiment, there were no significant changes in the target genes in the ga20ox1 mutant, again correlating with its lack of a height phenotype in this environment (Table 4). In the ga20ox2(b) mutant, there was a small but significant increase in GA20OX1 expression and increased expression of GA20OX4 and a decrease in expression of GA2OX3 (Table 5).
Table 5
Transcript levels of selected GA biosynthetic and signalling genes in peduncle tissues of ‘Cadenza’, ga20ox1 and ga20ox2(b) mutants grown in glasshouse and field environments. Cells shaded in orange are significantly different from ‘Cadenza’ (P < 0.05). Relative transcript levels were calculated against the geometric mean of the expression levels of three reference genes and normalised to expression in ‘Cadenza’.
|
GA20OX1
|
GA20OX2
|
GA20OX4
|
GA3OX2
|
GA2OX3
|
GID1-1
|
Glasshouse
|
|
|
|
|
|
|
Cadenza
|
1.00 ± 0.16
|
1.00 ± 0.05
|
1.00 ± 0.03
|
1.00 ± 0.04
|
1.00 ± 0.05
|
1.00 ± 0.06
|
ga20ox1
|
0.71 ± 0.05
|
1.11 ± 0.07
|
1.09 ± 0.06
|
1.11 ± 0.07
|
0.88 ± 0.08
|
1.15 ± 0.06
|
ga20ox2(b)
|
1.06 ± 0.20
|
0.74 ± 0.06
|
1.51 ± 0.11
|
1.42 ± 0.10
|
0.75 ± 0.03
|
1.53 ± 0.07
|
Field
|
|
|
|
|
|
|
Cadenza
|
1.00 ± 0.18
|
1.00 ± 0.05
|
1.00 ± 0.06
|
1.00 ± 0.10
|
1.00 ± 0.12
|
1.00 ± 0.03
|
ga20ox1
|
0.95 ± 0.12
|
1.23 ± 0.05
|
1.13 ± 0.05
|
1.28 ± 0.08
|
0.83 ± 0.14
|
1.15 ± 0.10
|
ga20ox2(b)
|
1.15 ± 0.14
|
0.92 ± 0.06
|
1.43 ± 0.14
|
1.21 ± 0.11
|
0.51 ± 0.04
|
1.05 ± 0.09
|