*3.3. lpa Mutants in SULTR Transporters*

The first mutants affected in SULTR3;3 and SULTR3;4 genes were described in *Arabidopsis thaliana*. They have been characterized for phenotypic alterations related to sulfate translocation between seed compartments [87]. Moreover, using the quintuple mutant defective in all SULTR3 subfamily members, it was recently shown that they have functional redundancy in chloroplast sulfate uptake and consequent influence on Cys, glutathione, and ABA biosynthesis, with the resulting growth retardation and altered stress responses in the multiple mutants [88]. Otherwise, no evidence of the involvement of these Arabidopsis genes in PA metabolism has been reported so far, with the only exception of AtSULTR3;4 for which contrasting results have been reported. In fact, very recently, Ding and co-workers [89] demonstrated that AtSULTR3;4/SPDT functions as a high-affinity Pi transporter, being able to mediate Pi uptake when injected in the Xenopus oocyte. Furthermore, it has been shown to localize to the plasma membrane, while Chen et al. reported a chloroplast localization [89]. On the other hand, these data are in agreemen<sup>t</sup> with those reported below on mutations affecting the HvST, OsSULTR3;3 and OsSULTR3;4 genes which confer grain *lpa* phenotype and in which the relevant proteins are localized in the endoplasmic reticulum and plasma membranes, respectively [19,20,84].

### 3.3.1. Mutants A ffected in the *SULTR3;3* Genes

In the case of *HvST* a nonsense mutation (*M422*) in the last exon of the barley *lpa1-1* gene was isolated from a sodium azide mutagenized population [19], and in the case of *OsSULTR3;3* the two di fferent *Os-lpa-Z9B-1* and *Os-lpa-MH86-1* mutations were a 6 bp deletion in the first exon and a 1 bp deletion in the 12th exon, identified through screening of a gamma-ray irradiation mutagenized population [28,45,84]. These barley and rice mutants exhibit a decrease in phytic acid-P like other *lpa* mutants, but also a decrease in total P in the seed (about 15% in barley mutant and 27.5–18.9% in rice mutants) [19,84,90], di fferently from *lpa* mutants a ffected in biosynthetic or *MRP*-transporter genes. Particularly, an endosperm-specific total P reduction [90] was reported that is not due to a reduction in the uptake of P in the maternal plant, suggesting that HvST functions as a seed-specific or filial determinant of barley endosperm total P. Moreover, OsSULTR3;3 disruption dramatically alters the grain metabolite profile. In fact, an increase was observed in the concentration of sugars involved in the close biosynthetic pathway leading to PA, sugar alcohols, free fatty acids, organic acids, biogenic amine GABA, serine, and lysine. However, the concentration of cysteine was decreased [84]. These traits were also stably maintained in the homozygous *lpa* progeny of generations F4 to F7 of crosses between the original *Os-lpaMH86-1* mutant with a commercial rice cultivar [91]. In addition, the metabolic profiles of the *lpa* progenies were strongly influenced by the lipid profiles of the wild type cultivar used as the crossing parent [92].

The OsSULTR3;3 mutants also show a significant increase in seed total sulfur and in sulfate concentration in embryo and pericarp/aleuronic layers. The mutations also increase root and leaf P and Pi concentrations and decrease root and leaf sulfate concentration in comparison to their corresponding wild type parents. Moreover, the analysis performed on developing seeds of the *MH86* mutant showed that the expression of genes coding for the last steps of PA biosynthesis was altered: Generally, an up-regulation was shown, and the expression of genes for sulfur metabolism and sulfate transport was di fferent in the mutant compared to the corresponding wild type. However, the most dramatic e ffects on gene expression concern several genes involved in P signaling and homeostasis [84]. A redistribution of Pi in endosperm and a reduction of lysophospholipid content were also observed in the rice mutant [93].

As previously mentioned, the role of the SULTR3;3 transporter is not clear, as, when expressed in heterologous systems, such as yeas<sup>t</sup> or Xenopus oocytes, it is unable to transport either sulfate, or phosphate, or PA precursors [84]. However, it cannot be excluded that in plant systems OsSULTR3;3 may transport these molecules as well as PA. In plant cells OsSULTR3;3 is ER-localized. Previous studies have suggested that the final steps of PA synthesis (from InsP3 to InsP6) take place in the ER [94]. Zhao and collaborators suggested that OsSULTR3;3 may have a specific role in the existing cross-talk between sulfate and phosphate homeostasis and/or signaling, as it has e ffects on phosphate as well as on sulfate concentrations in both vegetative tissues and grain [84].

Unfortunately, from an agronomic point of view, these mutants show some negative pleiotropic effects. In the rice mutant, grain weight reduction and yield per plant reductions have been shown [45]. In barley, only in rain-fed locations and not in irrigated ones, the mutation is associated with reduced test weight and percentage of plump kernels [95].

Interestingly, the mutant barley straws, although not showing significant di fferences in terms of fiber composition, compared to the wild type, after an acidic pre-treatment, showed increased fiber hydrolysibility, thus representing a promising material for cellulosic ethanol production [96].
