*4.2. Dehydration Stress in Allium spp. and Pectin Proxy*

In addition to causing structural changes within plant tissue, the formation of calcium– HG cross-linkages has also been tied to reduced cell wall permeability [12]. Thus, it was hypothesized the application of calcium would increase resistance to dehydration stress tolerance by reducing percentage water loss.

Even though calcium was found to significantly (*p* < 0.05) reduce percentage water loss in high methylated citrus pectin (proxy for *Allium*), the role calcium played on improving dehydration stress resistance in *A. fistulosum* and *A. cepa* was non-significant (*p* > 0.05) and is indicative of the complex nature of plants compared to pure systems. Nevertheless, previous studies have observed the efficacy of calcium application in enhancing drought stress tolerance/resistance in species such as *Beta vulgaris*, *Nicotiana tabacum* and wheat [62–64].

The discrepancy between these previous findings and those within this paper may be the result of a multitude of reasons, one of them being pore size. While the exact diameter of pores within the *A. fistulosum* epidermal cell layer is not known, previous research has found that, under cold acclimation, the diameter is reduced to less than 1.3 nm; however, the diameter of a water molecule at 25 ◦C is ~0.27 nm (2.7 Å) [65,66]. Therefore, even if permeability was reduced as a result of calcium application, the pore size was still larger than that of a water molecule. The large volume of water within the sheaths of *A. fistulosum* and *A. cepa* may also mask the effect of the calcium treatment.

Calcium is a complex element within plant systems, playing a role in a wide range of functions, including acting as a signaling molecule [67]. In its role as a signaling molecule, calcium has also been found to influence tolerance and resistance to drought stress in a variety of species [63,64,68]. For example, Knight et al. (1997) observed changes in the concentration of free calcium within cytosol during drought stress treatment in Arabidopsis [68]. More generally, the three main calcium sensors (1) calcineurin B-like protein; (2) calmodulin, and calmodulin-like proteins; and (3) calcium-dependent protein kinases) transduce Ca2+ signals, in turn causing the phosphorylation of a downstream target [69–76]. This ultimately results in a response to drought stress.

While calcium was unable (*p* > 0.05) to improve the overall resistance to dehydration in either *A. fistulosum* or *A. cepa*, *A. fistulosum* was significantly (*p* < 0.05) more tolerant to dehydration stress. *A. fistulosum* has also been found to be more tolerant to freezinginduced dehydration as compared with *A. cepa* [34,35]. *A. fistulosum* tolerates temperatures around −40 ◦C, while *A. cepa* lacks the ability to cold acclimate beyond −11 ◦C [34,35]. The connection between dehydration stress resistance and freezing stress tolerance is critical in the face of climate change. While initially drought stress and freezing stress may seem unrelated, both are tied to dehydration stress. During both drought stress and freezing stress, loss of turgor pressure can occur as the cell loses water [77–79]. As a result, cytorrhysis may occur beyond a tolerable limit. Oertli (1986) and Oertli et al. (1990) reported

that the ability for the cell wall to resist cytorrhysis depends on the mechanical strength of the cell wall [80,81]. Therefore, the ability to resist dehydration stress and to tolerate freezing stress may be related to the mechanical strength of the wall within *A. fistulosum*. Pectin is known to play a large role in the strength and stability of the cell wall [4].

## *4.3. Dehydration Stress in Arabidopsis and Pectin Proxy*

Similar to calcium and HG, boron and RG-II have the ability to form cross-linkages, which results in the formation of RG-II dimers [4–6]. Previous research by Panter et al. (2019) examining the *MUR1* gene in *Arabidopsis* found that the formation of these RG-II dimers improves tolerance to freezing stress [24]. Since dehydration stress is associated with freezing stress, it is plausible that RG-II dimers positively influence the water holding capacity of pectin. Thus, the addition of boron to GENU BETA likely resulted in the formation of RG-II dimers since the 8% GENU BETA pectin with boron lost significantly (*p* < 0.05) less water after 6 h compared to 8% GENU BETA pectin without boron.

Boron was also found to influence dehydration stress resistance in *Arabidopsis* in the various boron-transporter mutants. While none of the observed changes in dehydration stress resistance were found to be significant (*p* > 0.05) amongst the three boron-transporter mutants explored, *bor1* lost the most water. Decreased resistance to dehydration stress in *bor1* compared to *nip5;1* and *nip6;1* may be the result of differences in the roles of the boron transporters and further speaks to the importance of understanding the specific nature of the transporters.

Briefly, NIP5;1 and NIP6;1 are localized in the plasma membrane and are involved in borate transport [36,37]. NIP5;1 is also involved in arsenite transport [38]. In comparison, BOR1 is localized in the cytoplasm, endosome and vacuole in addition to the plasma membrane [38]. BOR1 is also involved in a wider range of functions, including detection of nutrients, ion homeostasis and transmembrane transport, as well as borate transport and response to boron-containing substances [38]. Therefore, a mutation in BOR1 may be more deleterious. However, there is a lack of additional research exploring NIP, BOR or other boron transporters in relation to dehydration stress resistance. Nonetheless, a range of species, including maize, wheat and tomatoes, had enhanced drought stress tolerance as a result of boron application [22,23,82,83]. This further implicates the importance of boron, and likely its structural role in the ability to retain water.

In addition to calcium and boron, PMEIs are another important aspect of pectin modifications and cell wall structure. PMEIs provide negative feedback on PMEs, in turn controlling the ability for "egg-box" structure formation in addition to altering the cell wall in a range of other structural ways [7]. The family of PMEIs within *Arabidopsis* is broad, and currently includes 71 putative genes [7,84,85]. Despite a lack of significance (*p* > 0.05), our results showed an over-expression of PMEI5 was beneficial in resistance to dehydration stress, by way of reducing percentage water loss and electrolyte leakage.

While seemingly counterintuitive, given that the over-expression of PMEI would reduce the formation of "egg-box" structures, research from An et al. (2008) supports our findings [86]. They found the over-expression of CaPMEI1, a PMEI from peppers, enhanced tolerance to drought stress [86]. This was attributable to an increased tolerance to oxidative stress, which was observed in the *CaPMEI1-OX Arabidopsis* line, which may in turn reduce the damage caused by stresses such as drought [86]. However, findings by An et al. (2008) and those observed in this study stand in direct contrast to findings by from Amsbury et al. (2016) and Yang et al. (2019), who found increased de-methylation of pectin enhanced drought stress tolerance [86–88]. This may be the result of differences amongst the various PMEIs.

#### *4.4. Fungal Pathogens*

Aside from abiotic stresses, biotic stresses also pose an immense threat to plants. Since both categories of stress may occur at the same time, a common mechanism of defense is critical to survival. The cell wall itself is the first line of defense and often serves as a signaling mechanism alerting plants to disease through pattern recognition receptors (PRRs) and damage-associated molecular patterns (DAMPs) [89]. Pectin methylesterase (PME) has long been associated with facilitated viral movement through the association between viral-encoded mobility proteins (MPs) and PME in which PME specifically recognizes binding domains of the viral MPs [90]. Lionetti et al. (2014) hypothesized once bound, the PME-MP increase the diameter of the plasmodesmata through the loosening of the callose and thereby facilitating greater viral movement [91]. They proposed that the inhibitor, PMEI, inhibits viral movement due to the reduction in PME-MP, thereby limiting the plasmodesmata dilation [91]. While the PMEI5 in our experiments did not show a significant reduction in fungal spread, it could be due to differences in homologous traits between the various PMEs and PMEIs. In Arabidopsis, there are 71 known PMEI genes and 66 known PME genes [7,85,92]. Given the large number of PMEs, it is highly plausible that only specific PMEs interact with MPs. The ability for the cell wall to remodel in response to viral pathogens is another critical component in defense, as it increases the strength of the cell wall [93]. This process has been found to be an important mechanism of defense against a variety of viral pathogens including halo blight disease in common bean, and *Potato virus* [93–95]. PMEI and PME have both been found to mediate cell wall remodeling [96,97]. While this set of experiments did not specifically focus on cell wall modeling in relation to viruses, the role of PMEI5 in connection with cell wall remodeling and defense against *B. cinerea* and *C. higginsianum* should be explored further. The analysis of lesion size over the course of 5 dpi for both *B. cinerea* and *C. higginsianum* demonstrated boron plays an integral role in resistance against both pathogens. *bor1* was significantly (*p* < 0.05) less resistant to *B. cinerea* compared to the other genotypes of interest (*nip5;1*, *nip6;1*, *p35S::PMEI5* and Col-0). This is likely the result of structural changes within pectin, and more specifically RG-II, given the findings of Petrasch et al. (2019) [28]. *bor1* was also significantly (*p* < 0.05) less resistant to *C. higginsianum* compared to Col-0. At 5 dpi, 100% of the total leaf area from *bor1* leaves was covered in a *B. cinerea* lesion. In comparison, leaves obtained from the other genotypes had less than 72% of their total leaf area covered lesions.

Since the BOR1 transporter is more widely expressed within the plant compared to NIP5;1 and NIP6;1, the *bor1* genotype likely has a lower concentration of boron [36–38]. Thus, the reduced resistance to *B. cinerea* in *bor1* compared to *nip5;1* and *nip6;1* may be the result of a lower concentration of boron, as boron has been previously found to mitigate *B. cinerea* infections [31].

Similarly, at 5 dpi following inoculation with *C. higginsianum*, 63.1% of the total *bor1* leaf area were covered in *C. higginsianum* lesion, compared to an average of only 31.7% for the other genotypes of interest. In addition to having twice the total leaf area covered in a lesion at 5 dpi, as early as 1 dpi, the *bor1* total leaf area covered in a *C. higginsianum* lesion was over 30 times greater compared to the other genotypes of interest, suggesting boron transported through the BOR1 channel is vital in the resistance to this disease. It is well understood that boron plays a critical role in the structure of the cell wall and a mutation in BOR1 would likely reduce the concentration of boron within the cell wall [98].

The reduced function of NIP5;1 and NIP6;1, in addition to the likelihood of redundancy within the NIP family, may be responsible for differences in leaf tissue infections compared to BOR1.

Furthermore, the over-expression of PMEI1, PMEI2, PMEI10, PMEI11 and PMEI12 has been previously found to help to maintain the integrity of the cell wall in plants infected with *B. cinerea* [26,27]. However, the over-expression of PMEI5 in the *p35S::PMEI5* did not significantly (*p* > 0.05) alter resistance to *B. cinerea*. Instead, leaves obtained from the *p35S::PMEI5* had the second greatest leaf area covered in *B. cinerea* lesions, after *bor1*. The contrasting nature of these findings compared to those obtained during our experiments may be the result of variation amongst PMEIs. PMEI5 belongs to Group 3 of PMEIs while previously explored PMEIs belong to other groups [26]. Future studies should focus on gaining a greater understanding into the differences between Group 1, 2 and 3 PMEIs.

Moreover, the more substantial lesion spread at 1 dpi on *bor1* leaves suggests weaker cell walls within the *bor1* genotype. *C. higginsianum* penetrates the cell wall of the host through physical pressure without the use of enzymes, unlike *B. cinerea* [99]. The rapid growth of lesions on *bor1* leaves is indicative that *C. higginsianum* may have penetrated the cell walls with less pressure, signaling potential weakness of the *bor1* cell walls, likely the result of lower concentrations of boron and in turn, a reduction in the number of RG-II dimers. Differences in function for the BOR1 transporter in comparison to NIP5;1 and NIP6;1 may reflect the variation observed in the percentage leaf area infected. More generally, while boron has been previously reported to act as an antifungal agent for a range of species infected with *Colletotrichum graminicola*, to our knowledge, there are no studies exploring NIP, BOR or any other B-transporters in relation to *C. higginsianum* [33,34].
