VPP5 expression pattern
The spatio-temporal expression pattern of VPP5 indicates its role in the reproductive tissue as it is highly expressed in anthers, ovaries, and embryos (Fig. 1a). In the developing seeds, VPP5 is strongly upregulated in the endosperm during 7–14 days after fertilization (Fig. 1b). This expression pattern coincides with rapid grain filling, suggesting a role for VPP5 in regulating starch biosynthesis. However, VPP5 expression in germinating seeds that undergo starch hydrolysis has not been reported. Transcriptomic analysis of the germinating seeds found a large set of differentially expressed genes during seed imbibition (Li et al., 2022; Yang et al., 2020), indicating massive transcriptomic changes within a few hours of the initiation of seed germination. We analyzed these data and found that VPP5 is differentially expressed in the imbibed seeds during early stages of seed germination. However, its expression pattern in the later stages of seed germination has not been investigated. We determined VPP5 expression in the germinating seeds after the emergence of the seedling (prophyll from coleoptile) and found that it is expressed in the germinating seeds at 3, 6, and 9 days after germination (DAG). The peak expression was observed at 3 DAG with a gradual decline at 6 DAG and a diminished expression at 9 DAG (Fig. S1). This expression pattern coincides with rapid conversion of starch to sucrose in the endosperm. To determine the effect of vpp5 mutation created by CRISPR/Cas9 targeting of the VPP5 promoter (Srivastava et al., 2017), two mutant lines, vpp5-1 and vpp5-2, were analyzed at 3 and 6 DAG. The VPP5 expression was significantly lower in these mutants with a remarkable downregulation at 3 DAG compared to the wildtype (WT) (Fig. 1c). This expression pattern correlated with a significantly higher level of PPi at 3 DAG in vpp5 seeds (~ 2-fold), while no disturbance was observed in the homeostasis of inorganic phosphate (Pi) (Fig. 1d-e). In summary, VPP5 is expressed in germinating seeds and targeted mutagenesis of VPP5 promoter resulted in its downregulation during germination. As a result, PPi accumulated in the mutant seeds while Pi homeostasis was maintained.
Growth characteristics of vpp5 mutants
In potting mix, emergence of vpp5 seedlings was delayed by 2–3 days compared to the wildtype (WT), and the seedling length after 12 days of sowing was shorter by ~ 1 inch (Fig. 2a). Although, ~ 90% germination was observed within 10 days of sowing, speed of germination (days to germinate) was slower in vpp5 lines compared to the WT (Fig. 2b). Earlier we reported that the impacted post-germinative growth of vpp5 could be rescued by the addition of 1% sucrose in the media (Gann et al., 2023). Here, we analyzed the effect of 1% sucrose supplementation on the seedling length and fresh weight in 9-d-old seedlings grown in the media plates. A significant positive-effect of sucrose was observed on the seedling length and fresh weight of vpp5 but not on the WT seedlings (Fig S2). Further, measurement of seedling length at 3-day intervals between 9 and 21 DAG showed a significant positive-effect of sucrose on the seedling growth (p ≤ 0.05) in vpp5 lines (Fig. 2c). Overall, vpp5 lines grew faster on sucrose-containing media compared to the sucrose-free media. Next, the plant height of vpp5 lines in the greenhouse was significantly shorter (p ≤ 0.05) compared to WT as measured in 2-wk-old and 2-mo-old plants (Fig. 2d-e) or in the R0 plants (Fig. S3). As expected, the slower growth rate led to a lower canopy cover, where vpp5 lagged WT throughout the vegetative phase from 10 to 70 days after emergence (Fig. S3). Interestingly, grain numbers (GN) and weight of 1000 grains (W1000G) were not significantly different (Fig. 2f-g), while significant differences were observed in panicle number, panicle length, and percent sterility (Fig. S4).
Reduced rate of starch utilization in vpp5 germinating seeds
Conversion of starch to sucrose involves several enzymatic steps through a dedicated starch degradation/utilization pathway (Fig. 3a). The process of starch degradation in the seed is initiated soon after the imbibition of water and continues until post-germinative growth (emergence and elongation of the seedling). Previously we reported that sucrose content in vpp5 germinating seeds was lower compared to WT that correlated with reduced rate of starch utilization (Gann et al., 2023). We measured the sucrose and starch contents in the germinated seeds at 3 and 6 DAG. The sucrose content was highly variable within the genotype; however, signifcant differences were observed between the genotypes (p ≤ 0.05). At both stages, WT seeds contained the highest concentration, and vpp5 lines contained 20–35% lower sucrose in the seed (Fig. 3b). The starch utilization rate was indicated by the difference in starch concentrations in 3 DAG and 6 DAG seeds. This analysis showed a substantially reduced rate of starch utilization in vpp5 lines compared to the WT (p ≤ 0.05) (Fig. 3c).
Next, we analyzed gene expression of α-Amylase 1a (AMY1a), UDP-Glc Pyrophosphorylase (UGP), and Sucrose Phosphate Synthase (SPS) in the germinating seeds of vpp5 and WT. These genes play critical roles in the starch utilization pathway (Fig. 3a). Rice genome contains eight members of α-amylase genes in three subfamilies, AMY1, AMY2, AMY3 (Huang et al., 1990; Mitsui et al., 1996; Ranjhan et al., 1991). Of these, AMY1a, b and AMY3a, d, e are abundantly upregulated during seed germination along with β-amylases (Li et al., 2022; Yang et al., 2020). We chose AMY1a as the representative α-amylase gene for this study. Next, rice contains two UGP genes that are ubiquitously expressed (Abe et al., 2002; Chen et al., 2007). We analyzed both UGP1 and UGP2. Finally, rice contains five SPS genes which are differentially expressed. We chose SPS11 as it is strongly expressed in germinating seeds (Okamura et al., 2011; Li et al., 2022; Yang et al., 2020).
Compared to WT, AMY1a was downregulated in vpp5 at 3 DAG but expressed at similar levels at 6 DAG. Thus, in vpp5, upregulation of AMY1a was delayed to 6 DAG (Fig. 3d). UGP1 was slightly upregulated at 3 DAG but > 2 fold upregulated at 6 DAG in vpp5 compated to WT (Fig. 3e). UGP2 expression, on the other hand, was downregulated in vpp5 at both 3 DAG and 6 DAG; however, expression changes in UGP2 is probably not metabolically relevant as its expression is substantially lower (100-fold) than that of UGP1 (Fig. S4). This finding corroborates with a previous study that reported a higher expression of UGP1 and a predominance of UGP1 encoded UGPase isoform in rice (Chen et al., 2007). Finally, SPS11 was 7–14 fold downregulated at 3 DAG, but 3–5 fold upregulated at 6 DAG in vpp5 (Fig. 3f). To summarize, the key starch pathway genes were differentially expressed in vpp5 and upregulated at 6 DAG, suggesting a delayed onset of robust starch to sucrose conversion.
Metabolome profile
The effect of VPP5 suppression on the metabolome of the germinating seeds was investigated by LC-MS analysis. Comparing vpp5-1 line against the WT, this analysis showed signifcant differences (p ≤ 0.05) in the metabolome of the germinating seeds (Fig. 4a-b). A total of 44 metabolites in 3 DAG and 91 metabolites in 6 DAG were up- or down-regulated (log2 FC ≥ 1.5 or -1.5) in vpp5 (Table S1-S2) and several KEGG pathways were differentially expressed, including starch and sucrose metabolism in 3 DAG (Fig. 4c-d). Notably, D-Glucose (Glc), D-Fructose 6-phosphate (F6P), and UDP-Glucose (UDP-Glc) were significantly downregulated in 3 DAG correlating with a reduced starch to sucrose conversion in vpp5 lines (Table S1). In the 6 DAG seeds, on the other hand, UDP-Glc was significantly upregulated (Table S2); however, signifcant differences were not observed in Glc and F6P levels. Next, in a 3 DAG vs 6 DAG comparison, a clear difference in the metabolome of WT was observed but not in that of vpp5 (Fig. S6). Further, although UDP-Glc was significantly different between 3 and 6 DAG in each genotype, opposite trends were observed. In WT, UDP-Glc decreased as the germination progressed from 3 to 6 DAG and in vpp5, it increased in 6 DAG compared to 3 DAG. This pattern correlates with the UGP1 and SPS11 expression patterns in vpp5 that are expressed at higher levels in 6 DAG seeds compared to 3 DAG (Fig. 3e-f). In summary, the metabolomic analysis points to a slower metabolic changes in vpp5 germinating seeds including a slower rate of starch utilization.