Skip to main content
. 2020 Nov 19;9:e61701. doi: 10.7554/eLife.61701

Figure 2. Mycorrhizal symbiosis promotes maize growth under cultivation.

(A) Susceptible (AMF-S) and resistant (AMF-R) families segregate genomic content from two founder parents, shown as red and blue bars, but are homozygous for the wild-type (green) or mutant (yellow) allele at Castor (black arrow), respectively, blocking AM symbiosis in AMF-R. (B), Border between representative AMF-S and AMF-R plots, Ameca, Mexico, 2019. (C), Representative AMF-S ears. (D), Representative AMF-R ears. (E), Plant height (PH) of 73 AMF-S (green) and 64 AMF-R (yellow) families. (F), Anthesis-silking interval (ASI). (G), Kernel shape (KS) based on the analysis of scanned kernel images . (H), Total kernel number (TKN). The violin plots in (E, F and H) are a hybrid of boxplot and density plot. The box represents the interquartile range with the horizontal line representing the median and whiskers extending 1.5 times the interquartile range. The shape of the violin plot represents the probability density at different values along the y-axis. ***, difference between AMF-S and AMF-R significant at p<0.001 (Wilcoxon test; Bonferonni adjustment based on the total trait number).

Figure 2.

Figure 2—figure supplement 1. Location of experimental fields.

Figure 2—figure supplement 1.

(A) Satellite view of the Ameca site used for the principal field evaluation (20.57 lat, −104.04 lon, 1268 masl). Yellow rectangle shows the field area. (B) Satellite view of the Puerto Vallarta (PV) site used for population generation and preliminary high-input evaluation (20.78 lat, −105.243 lon, 41 masl). Red rectangle shows the field area. (C) Mexico showing the two sites.
Figure 2—figure supplement 2. Representative AMF-S and AMF-R families in high- and medium- input sites.

Figure 2—figure supplement 2.

(A) Winter PV field in 2020, located in Nayarit, Mexico. (B) Summer Ameca field in 2019, located in Jalisco, Mexico.
Figure 2—figure supplement 3. Plan of the Ameca field design.

Figure 2—figure supplement 3.

Summer experimental field 2019, located in Ameca, Jalisco. Three complete blocks were evaluated. Forty-five seeds were sown per three-row plot (15 seed per row). AMF-S and AMF-R families were alternated with the order of the families within each subpopulation randomized within each block. The rows marked with red, blue, and green correspond to blocks 1, 2, and 3, respectively. A commercial UNISEM hybrid was planted on the border of the experiment and used as a check throughout the field. The field was fertilized at planting with 250 kg / Ha of diammonium phosphate (DAP) as a source of nitrogen and phosphorus (18-46-00, NPK). A further application of 250 kg / Ha of urea was given (46-00-00, NPK) at 40 days after planting.
Figure 2—figure supplement 4. Comparison of plant phenotypic traits between AMF-S (wt) and AMF-R (castor) families.

Figure 2—figure supplement 4.

The box in violin plots represents the interquartile range with the horizontal line representing the median and whiskers representing 1.5 times the interquartile ranges. The shape of the violin plot represents the probability density of data at different values along the y-axis. Results were based on two-group Wilcoxon tests with Bonferonni adjusted p-values. Note: *: p<0.05; **: p<0.01; ***: p<0.001; NS: not significant. Trait codes as in Table 1.
Figure 2—figure supplement 5. Ear and kernel image analysis.

Figure 2—figure supplement 5.

The shape of maize ear (A), cob (B), and kernel (C) of AMF-S (wt) and AMF-R (castor) families. Median shapes of ear, cob, and kernel from two families were shown in thick green and yellow lines, whereas individual genotypes of two groups were shown in thin, semi-transparent green, and yellow lines. Shapes were extracted from scanned images from 3 ears per plot, 50 kernels per ear.
Figure 2—figure supplement 6. Principal component analysis.

Figure 2—figure supplement 6.

A, Phenotypic variation captured by the first five principal components (PCs). The first two principal components (PCs) explained more than 50% of the phenotypic variance. PC1 (40.1%) was primarily associated with ASI and maize yield-related traits, such as ear weight (EW), total kernel number (TKN), and total kernel weight (TKW). B, Loading of AMF-S and AMF-R families on PC1 and PC2. The position of trait names (as Table 1) with respect to the origin indicates contribution to each PC.