Molecular and cellular basis of cell wall invertase-mediated fruit set

Research Doctorate - Doctor of Philosophy (PhD)Fruit and seed (grain) are major components of world crop production that need to be doubled over the next 50 years to meet population demand and hence secure food security. Moreover, the issue has become increasingly urgent and challenging by increased frequencies of global warming-associated events imposing abiotic stresses (e.g. drought; heat) and biotic stresses that have caused severe yield losses. During fruit development, fruit set is the most important stage for the realization of crop yield potential. Sinks (e.g. fruit and seed) rely on import of sucrose (Suc) produced from photosynthetic leaves (sources). Invertases (Inv), and particularly cell wall invertases (CWIN), are the enzymes responsible for hydrolysing phloem-imported Suc into, glucose (Glc) and fructose (Fru) during early seed and fruit development. The CWIN-derived hexoses function as both signalling molecules regulating gene expression and substrates to fuel energy production and provide carbon skeletons for biosynthetic and storage processes. The mechanistic basis as to how CWIN regulates fruit set is unknown. The question is technically difficult to address using CWIN-knockout or silenced plants as they often grow poorly with shrivelled or non-viable seeds. By silencing its inhibitor gene (INH), the endogenous CWIN activities were elevated, causing significant increases in tomato fruit hexose levels and seed weights (Jin et al 2009). This discovery provides a unique opportunity to explore the mechanism by which CWIN regulates fruit and seed development without potential complications from misexpression of foreign CWIN genes as the case for many over-expression studies. Fruit set is the most important stage for the realization of crop yield potentials. In our study, we focused on ovaries and fruitlets at 2 days before and 2 days after anthesis respectively (2dba ovaries and 2daa fruitlets). These two developmental stages span the fruit set process in tomato. To understand the role of CWIN and related genes in assimilate distribution in fruit set, potential regulators or functionally related proteins controlling CWIN activity and sugar transporters potentially involved in Suc unloading were investigated. Using a bioinformatics approach, we identified no new functional CWINs, but a defective CWIN (<i>SldeCWIN1</i>), as well as a new CWIN inhibitor (<i>SlINVINH2</i>) in the tomato genome. SldeCWIN1 and inhibitors (INHs) are both potential regulators of CWIN activity. Transcript expression profile revealed that SlCWIN1 is the dominant CWIN during fruit set. The transcript levels of <i>SlCWIN1</i> in WT plants decreased sharply in transiting from 2 dba ovaries to 2 daa fruitlets. The two potential regulators in controlling CWIN activity showed different expression patterns during fruit set. <i>SlINVINH1</i> was expressed in both 2 dba ovaries and 2 daa fruitlets and its expression levels were not changed significantly during fruit set. However, the transcripts of the newly-identified <i>SlINVINH2</i> and <i>SldeCWIN1</i> were only detected in 2- dba ovaries, but not in 2 daa fruitlets. The activity of CWIN was elevated significantly while VIN and CIN activities were not changed in 2 daa fruitlets as compared to 2 dba ovaries. The results indicate that CWIN, but not VIN and CIN, plays an important role during fruit set. Also, it is clear that there is a discrepancy between the transcripts of CWIN (decreased) and the CWIN activity (elevated) in 2 daa fruitlets when compared to 2 dba ovaries. Silencing the expression of <i>SlINVINH1</i> did not affect the expression of <i>SlCWIN1</i> and <i>SldeCWIN1</i>. This indicates that elevated CWIN activity is regulated by SlINVINH1 at the post-translational level. The defective CWIN has been shown to enhance inhibition of CWIN activity in the presence of INH in tobacco extracts. Thus, we propose that a high level of SldeCWIN1 expression may help SlINVINH1 and SlINVINH2 to reduce SlCWIN1 activity in 2 dba ovaries. In contrast, the near absence of SldeCWIN1 and SlINVINH2 expression in 2 daa fruitlets could alleviate the inhibition, leading to an increase in CWIN activity through post-translational regulation. In regard to sugar transporters, 2 SlSUTs (SlSUT2 and SlSUT4), 2 SlHTs (SlHT2, SlHT3), 2 putative Glc-transporting SWEETs (SlSWEET1a, 1b) and 2 putative Suc-transporting SWEETs (SlSWEET12b, SlSWEET12c) may involve in Suc unloading during fruit set. Among these genes, only <i>SlSWEET12c</i> was responsive to elevated CWIN activity. The expression level of <i>SlSWEET12c</i> was increased in 2 daa fruitlets of <i>SlINVINH1</i>-RNAi as compared to WT plants. The results indicate that expression of <i>SlSWEET12c</i> may be functionally coupled with that of <i>SlCWIN1</i>. CWIN, located in cell walls of sieve elements of 2 dba ovaries and 2 daa fruitlets, degrades Suc into hexoses to create a steep Suc concentration between sieve elements and the apoplasm of recipient cells. Thus, the role of SWEET12c might be involved in releasing Suc into the apoplasm of recipient cells from sieve elements to provide sugar supply to drive fruit set. The physiological role of CWIN is determined, in part, by Suc unloading pathways. Before examining the phloem-unloading pathway within ovaries or fruitlets, we firstly tested the hypothesis of the pedicel phloem of 2 dba ovaries imposes a limitation on Suc delivery to the ovaries. Estimated specific mass transfer (SMT) of biomass through the pedicel sieve elements of 2 dba ovaries indicate that the sieve elements did not impose a limitation to Suc delivery to 2 dba ovaries. The result indicates that the sieve elements in the pedicel have transport spare capacity in 2 dba ovaries. Thus, we focused on the phloem unloading pathways within the ovaries and fruitlets themselves. In 2 dba ovaries, Suc unloading follows apoplasmic pathway in 2 dba ovary wall and placenta connecting to the ovules while symplasmic pathway within placenta areas. In 2 daa fruitlets, phloem unloading switches to a symplasmic pathway in the pericarp, but remains apoplasmic in placenta connecting to seeds and symplasmic pathways within placenta areas. The phloem unloading pathway remained unchanged in 2 dba ovaries and 2 daa fruitlets of <i>SlINVINH1</i>-RNAi and was comparable to that of WT plants. This implies that plasmodesmata conductivity or formation is not influenced by CWIN activity. Thus, Glc, as a carbon source for callose deposition around the neck region of plasmodesmata, is likely derived from the sucrose synthase and not the Inv pathway. The analyses revealed different physiological roles CWIN may play in phloem unloading in these different locations within the ovaries and fruitlets. During tomato fruit set, the Suc import rate increases from 6 nmol Suc/h in 2 dba ovaries to 39 nmol Suc/h in 2 daa fruitlets. As a consequence, significant changes must take place at the cellular level to underpin such dramatic increases in assimilate delivery for fruit set. In this context, we found that sieve elements are structurally more advanced in 2 daa fruitlets when compared to 2 dba ovaries. However, the less developed phloem in the ovaries does not appear to represent a limitation to phloem unloading since calculation of SMT value shows phloem development did not impose any limitation on phloem transport. Lastly, we examined what genes and biochemical pathways respond to changes in CWIN activity during ovary-to-fruit transition using transcriptome and metabolome analyses. Comparing ovaries with fruitlets showed that these organs exhibited remarkably different response to elevated CWIN activity in the transgenic plants where SlINVINH1 was silenced. We found that there were only 7 genes whose transcript levels exhibited changes to an elevation of CWIN activity in 2 daa fruitlets. By contrast, 319 transcripts were differentially expressed in 2 dba ovaries of SlINVINH1-RNAi (CWIN activity elevated) compared to WT plants. The results revealed that 2 dba ovaries are far more responsive to elevated CWIN activity as compared to fruitlets at the transcript level. Gene functional classification analyses revealed that the transcript levels of biotic, ethylene synthesis and cell cyclin-related genes were generally increased, but the mRNA levels of genes for photosynthesis, protein degradation and receptor like kinase-related genes were decreased, along with a large number of cell wall-related genes exhibiting differential expression in 2 dba ovaries of SlINVINH1-RNAi when compared to WT plants. Of particular significance is the enhanced expression of pathogen resistance (R) genes and pathogenesis-related (PR) genes in 2 dba ovaries in response to elevation of CWIN activity. Some genes involved in R-gene mediated pathway were also differentially expressed in 2 dba ovaries in CWIN elevated plants. They include cytoplasmic serine/threonine kinase, pathogenesis-related ERFs and a small GTPase (ROP). These results indicate that elevated CWIN activity enhanced R gene-mediated pathways for biotic resistance in the ovary. As plant defense reactions consume considerable resources and energy, it may affect plant development. However, we did not find any negative phenotypic impact during fruit set in the transgenic plants. Interestingly, the levels of a number of transcripts encoding proteins involved in protein degradation were significantly decreased in the transgenic ovaries, suggesting a prolonged protein turnover rate, which could save metabolic energy for synthesis of new proteins to boost the defence capacity of the ovaries. Overall, the data presented in this thesis have unraveled several novel aspects of the mechanisms by which CWIN regulate fruit set. We found that CWIN activity is under tight post-translational regulation in ovaries as compared to that in fruitlets, likely regulated by INHs and defective CWIN. Also, the findings allow us to us to formulate a model on how CWIN and sugar transporters may work together to regulate phloem unloading of Suc during the ovary-to-fruitlet transition. Finally, we discovered that elevation of CWIN activity enhanced R and PR gene expression, thereby potentially increasing the defence capacity of the ovaries against pathogens for fruit set