These studies focused on the influence of leaf morphology on fruit sugar level, and revealed important correlations between leaf shape and fruit sugar accumulation. However, how leaf shape contributes to sugar accumulation in fruit is not yet known . The impact of leaf traits such as leaf complexity and leaf veins on fruit sugar levels is currently uninvestigated. Tomato bipinnate is a classic leaf mutant, with highly increased leaf complexity resulting from the loss of function of a BEL-LIKE HOMEODOMAIN gene called BIPINNATE , and provides ideal material to investigate the influence of leaf traits on fruit sugars. To investigate the links between leaf complexity and fruit sugar accumulation, we performed an analysis of leaflet shape, leaf complexity, vein density, yield, and fruit BRIX on bip mutants and their isogenic backgrounds . Results suggested that leaf vein density, and not leaf complexity, was highly correlated to fruit BRIX in bip mutants. RNA-Seq analysis of vegetative apices and leaves of bip mutants and the wild type shows an association between the expression of genes regulating leaf vein development and carbohydrate metabolism and transport. These data suggest that alterations in leaf vein development would also modulate processing of carbohydrate metabolism and transport which, in turn, draining pots affects the accumulation of fruit sugar. Our analysis offers insight into how leaf morphology may influence fruit sugar and provides a new direction to improve tomato fruit sugar content.
To investigate whether changes of leaf complexity caused by mutations at the BIP gene could influence fruit BRIX, we measured leaf complexity and fruit BRIX of bip0663 mutants and the corresponding isogenic cultivar, ‘Lukullus’. Leaf complexity and fruit BRIX of bip0663 mutants were both significantly increased in comparison to that of ‘Lukullus’ . However, another bip mutant, bip2, had significantly increased leaf complexity , but fruit BRIX similar to its isogenic cultivar, M82 . These results indicate that changes in leaf complexity are not directly correlated with fruit BRIX. Additionally, fruit BRIX of both bip0663 and ‘Lukullus’ were respectively higher than that of bip2 and M82. Fruit yield was also measured, since yield has an impact on fruit BRIX . Fruit yield in bip0663 was similar to ‘Lukullus’, while bip2 fruit yield was lower than M82, suggesting that the difference between bip0663 and ‘Lukullus’ BRIX was not directly related to variation in yield. Similar to fruit BRIX, fruit yield of bip0663 and ‘Lukullus’ was also significantly higher than that of bip2 and M82. This result was confirmed by sampling across two seasons in the field and in greenhouse experiments .To validate whether changes in leaf traits contributed to the changes in fruit BRIX between bip0663 and ‘Lukullus’, we performed a grafting assay where the ‘Lukullus’ and bip0663 separately functioned as reciprocal scion and rootstock. During the flowering and fruiting stage, leaves on the scion, and flowers and fruit on the rootstock were completely removed. Thus, the flowering and fruiting of the scion were driven by photosynthate transfer only from the leaves onthe rootstock portion. In the control group, the ‘Lukullus’ and bip0663 were self-grafted and generated a fruit BRIX that resemble their nongrafted lines . In the treatment group, where bip0663 served as the rootstock for ‘Lukullus’ , fruit BRIX of ‘Lukullus’ was significantly increased compared with its selfgrafted line , and when ‘Lukullus’ served as the rootstock for bip0663 , fruit BRIX of bip0663 was decreased compared with to its self-grafted line .
In addition, measurements of fruit yield show that there was no significant difference in fruit yield between the control and the treatment group . These results indicate that changes in fruit BRIX between bip0663 and ‘Lukullus’ were either caused by photosynthate production differences, or efflux from leaves. Another potential factor could be root sink strength. It was possible that the grafts containing ‘Lukullus’ rootstock could have reduced fruit sugar due to increased sink strength of ‘Lukullus’ roots. To determine if differences in root sink strength could have contributed to the differences in fruit BRIX, a second grafting assay was performed in which the roots of each genotype were reciprocally grafted. Root grafts with the bip0663 scion had the same high fruit BRIX and yield, regardless of the root genotype. Similarly, root grafts with the ‘Lukullus’ scion showed lower fruit BRIX and yield, regardless of the root genotype . There were no significant differences in fruit BRIX or yield when comparing bip0663/bip0663 to bip0663/’Lukullus’ or ‘Lukullus’/’Lukullus’ to ‘Lukullus’/bip0663 grafts . This supports the hypothesis that the presence of bip0663 source leaves led to increased fruit BRIX of the grafts. To further examine the differences between bip0663 and ‘Lukullus’ source leaves, the sugar and starch contents of leaves were assayed at dusk. The leaves of bip0663 had less sugarp=0.0939 and less starch than the leaves of ‘Lukullus’ . This suggests that bip0663 source leaves export photosynthate more efficiently, or produce more photosynthates, than those of ‘Lukullus’.Since the grafting experiment indicated that changes in leaf features could drive changes in fruit BRIX, and recent studies in an introgression population and a selected group of heirloom tomato varieties show leaf shape is strongly correlated with fruit BRIX and sugar accumulation , we did leaf shape analysis on bip mutants and their isogenic wild type cultivar to see whether changes in fruit BRIX between these genotypes were also correlated with variation in leaf shape.
Results from PCA analysis of a total of 1500 primary leaflets showed that PC1 contributes 65% of all variation of leaflet shape and is strongly correlated with leaflet area , suggesting leaf size was the largest source of variation among leaflets of the 4 genotypes. PC4 contributes 3.8% of all variation and is correlation with aspect ratio with an R 2 of 0.6 . PC2, PC3 and PC5 were not correlated with any traditional shape measures and may describe the overall variation in leaf symmetry . Fig.6C show the results of traditional leaf shape measurements. Both bip2 and bip0663 show a similar change in leaf shape compared with their corresponding isogenic wildtype backgrounds: leaflet area and solidity decreased, and leaflet aspect ratio and circularity increased. These results indicate that mutations in the BIP gene not only regulate leaf complexity, but also alter leaflet shape. ‘Lukullus’ and bip0663 displayed leaflet aspect ratio and circularity significantly higher than that of bip2 and M82. Compared to both isogenic wild genotypes, bothbip mutants displayed rounder leaflets, with significantly higher aspect ratios and circularity measures.To detect changes in the patterns of expression of the DEGs between phenotypes with different fruit BRIX, we performed DiffCorr analysis on all detected DEGs. DiffCorr networks were separately constructed on all DEGs between bip0633 and ‘Lukullus’ comparison and also between HB and LB comparisons with cutoff adjusted p-value<0.05. In these DiffCorr networks nodes with more connected edges represent genes which have the most differential co-expressions between the two compared co-expression networks. Of the top 20ranked Diffcorr genes, 11 genes were shared and highly differentially co-expressed between bipvsLu and HvsL Diffcorr networks . The expression patterns of these genes were highly different between bip0663 and ‘Lukullus’, and also between HB and LB comparisons. Six of these 11 genes were reported to be involved in regulating cellular processes, cell walls, large plastic garden pots leaf vein development or sugar metabolism. During leaf development, LTP4 , AHL3 and FUT11 were all down regulated in bip0663 and ‘Lukullus’, in young leaves and mature leaves compared with bip2 and M82 . PGL1 was up regulated in bip0663 and ‘Lukullus’ in ML . Thus, these genes might play an important role in regulating fruit BRIX and leaf vascular density between bip0663, ‘Lukullus’, bip2, and M82. In addition, many DEGs detected between the HB and LB groups were involved in regulating both development and carbohydrate metabolism . We therefore hypothesize that these genes might be co-expressed during leaf vascular development and involved in regulating leaf sugar metabolism either directly or indirectly through vascular transport processes. To validate this hypothesis, we performed correlation analysis and constructed a gene coexpression network across all genotypes using DEGs enriched in “transcription and development” and “carbohydrate and biosynthesis” related GO terms. Correlation analysis showed most of “transcription and development” related DEGs and “carbohydrate and biosynthesis”- related genes are highly correlated, whereas GPT2 was shown to be positively correlated with many genes related to carbohydrate and bio-synthetic process , such as Solyc01g074030 .
Interestingly, LTP4 showed significantly positive correlation with GPT2. In the Co-expression network, community C3 shows significant enrichment of the shared top ranked DiffCorr genes listed in Table S5 and is connected to other two communities , which compose a core network . C1 was enriched for genes related to transcription initiation and cell fate specification GO terms, and C2 contained genes involved in glucose metabolic process, carbohydrate bio-synthetic process, plant organ development, hormone-mediated signaling pathway and shoot system development . GPT2 can be found on the border of the core network, within C1 and co-expressed with several genes with hubs >50, revealing it might be an important gene that connects leaf vein development and carbohydrate bio-synthetic processes together. C3 contained genes involved in glucose-6- phosphate metabolic process, and C5 was enriched for genes involved in cellular carbohydrate bio-synthetic processes.Since sugars synthesized in leaves are the primary source of energy for fruit development and ripening , we and others have investigated the link between leaf and fruit. Previous studies have shown leaf traits, such as leaf shape and size are strongly correlated with fruit quality and yield . In this study, we used mutant analysis and grafting experiments to analyze the influence of leaf traits on fruit sugar level. Leaf shape analysis showed that both bip mutants had rounder leaflets, with significantly higher aspect ratios and circularity measures, compared to their respective isogenic cultivars. This suggests bip mutations caused not only high leaf complexity but also rounder leaves. Results also show that leaflet roundness of bip0663 and ‘Lukullus’ was significantly higher than bip2 and M82 . These results agree with previous research that showed tomato with rounder and more circular leaves tend to have the highest sugar content in their fruit . Although bip2 has rounder leaves than M82 it does not have measurably higher fruit BRIX , suggesting that the roundness caused by the bip mutation does not, by itself, lead to increased BRIX. A previous study suggested the correlation between leaf shapes and fruit sugar content may be due to the impact of leaf shape on photosynthetic capacity. However, we found that across the whole season, average photosynthetic rates of bip2 and M82 were higher than those of bip0663 and ‘Lukullus’, perhaps resulting from high vein density in the LB genotypes . Thus, the variation in fruit BRIX between these genotypes was likely due to aspects other than leaf photosynthesis. In addition, while mutations in the BIP gene cause increased leaf complexity, only bip0663 has increased fruit BRIX compared with ‘Lukullus’. This suggests that leaf complexity is not the cause of changes in fruit BRIX between bip0663 and ‘Lukullus’. This is consistent with the results from path modeling connections between leaf traits and fruit BRIX in multiple heirloom cultivars . The same path modeling also indicated that photosynthesis was not a major contributor to fruit BRIX .In leaves, sugars synthesized through photosynthesis are first loaded into leaf veins and then transported out of the leaves to the rest of the plant sinks, such as fruit. Thus, leaf veins also play an important role in fruit sugar accumulation by contributing to transport of carbohydrates . Theoretically, higher leaf vein density will increase the contact between vascular and photosynthetic tissues and lower the distance of photosynthate transportation . Several studies have shown high leaf vein density can not only enable higher Kleaf and higher rates of gas exchange per leaf area , but also can improve phloem loading . Based on these studies, with all else being equal, plants with high leaf vein density should have higher fruit sugar levels. However, in this study leaf vein density was found to be negatively correlated with fruit BRIX in bip mutants and their corresponding isogenic cultivars. This may be caused by the following factors: Leaves with low leaf vein density can enhance mesophyll light capture in shade and result in increased photosynthate .