TYPE Original Research PUBLISHED 29 September 2022 DOI 10.3389/fpls.2022.984100 Effects of impaired steryl ester OPEN ACCESS biosynthesis on tomato growth EDITED BY Rogelio Santiago, University of Vigo, Spain and developmental processes REVIEWED BY Eric Soupene, Alma Burciaga-Monge 1†, Joan Manel López-Tubau 1†, University of California, San Francisco, United States Natalie Laibach 1, Cuiyun Deng 1, Albert Ferrer 1,2* Esther Novo-Uzal, Instituto Nacional de Investigación y and Teresa Altabella 1,3* Tecnologı´a Agroalimentaria 1 Plant Synthetic Biology and Metabolic Engineering Program, Centre for Research in Agricultural (INIA), Spain Genomics (CRAG), CSIC-IRTA-UAB-UB, Barcelona, Spain, 2 Department of Biochemistry and *CORRESPONDENCE Physiology, Faculty of Pharmacy and Food Sciences, Universitat de Barcelona, Barcelona, Spain, Albert Ferrer 3 Department of Biology, Healthcare and the Environment, Faculty of Pharmacy and Food Sciences, albertferrer@ub.edu Universitat de Barcelona, Barcelona, Spain Teresa Altabella taltabella@ub.edu † These authors have contributed equally to this work Steryl esters (SE) are stored in cytoplasmic lipid droplets and serve as a reservoir SPECIALTY SECTION of sterols that helps to maintain free sterols (FS) homeostasis in cell membranes This article was submitted to throughout plant growth and development, and provides the FS needed to Plant Metabolism and Chemodiversity, a section of the journal meet the high demand of these key plasma membrane components during Frontiers in Plant Science rapid plant organ growth and expansion. SE are also involved in the recycling of RECEIVED 01 July 2022 sterols and fatty acids released from membranes during plant tissues ACCEPTED 08 September 2022 senescence. SE are synthesized by sterol acyltransferases, which catalyze the PUBLISHED 29 September 2022 transfer of long-chain fatty acid groups to the hydroxyl group at C3 position of CITATION FS. Depending on the donor substrate, these enzymes are called acyl-CoA: Burciaga-Monge A, López-Tubau JM, Laibach N, Deng C, Ferrer A and sterol acyltransferases (ASAT), when the substrate is a long-chain acyl-CoA, Altabella T (2022) Effects of impaired and phospholipid:sterol acyltransferases (PSAT), which use a phospholipid as a steryl ester biosynthesis on tomato donor substrate. We have recently identified and preliminary characterized the growth and developmental processes. Front. Plant Sci. 13:984100. tomato (Solanum lycopersicum cv. Micro-Tom) SlASAT1 and SlPSAT1 enzymes. doi: 10.3389/fpls.2022.984100 To gain further insight into the biological role of these enzymes and SE COPYRIGHT biosynthesis in tomato, we generated and characterized CRISPR/Cas9 single © 2022 Burciaga-Monge, López-Tubau, knock-out mutants lacking SlPSAT1 (slpsat1) and SlASAT1 (slasat1), as well as the Laibach, Deng, Ferrer and Altabella. This is an open-access article distributed double mutant slpsat1 x slasat1. Analysis of FS and SE profiles in seeds and under the terms of the Creative leaves of the single and double mutants revealed a strong depletion of SE in Commons Attribution License (CC BY). The use, distribution or reproduction slpsat1, that was even more pronounced in the slpsat1 x slasat1 mutant, while in other forums is permitted, provided an increase of SE levels was observed in slasat1. Moreover, SlPSAT1 and SlASAT1 the original author(s) and the inactivation affected in different ways several important cellular and copyright owner(s) are credited and that the original publication in this physiological processes, like leaf lipid bo1dies formation, seed germination journal is cited, in accordance with speed, leaf senescence, and the plant size. Altogether, our results indicate that accepted academic practice. No use, distribution or reproduction is SlPSAT1 has a predominant role in tomato SE biosynthesis while SlASAT1 would permitted which does not comply mainly regulate the flux of the sterol pathway. It is also worth to mention that with these terms. some of the metabolic and physiological responses in the tomato mutants lacking functional SlPSAT1 or SlASAT1 are different from those previously reported in Arabidopsis, being remarkable the synergistic effect of SlASAT1 inactivation in the absence of a functional SlPSAT1 on the early germination and premature senescence phenotypes. Frontiers in Plant Science 01 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 KEYWORDS CRISPR-Cas9, germination, lipid droplets, phytosterols, senescence, sterol acyltransferase, steryl ester, tomato Introduction Regulation of free sterols (FS) levels in the membrane can be achieved by modulating the carbon flux of the sterol biosynthesis Sterols are essential structural and functional components of pathway and/or by converting excess FS into steryl esters (SE), most eukaryotic cell membranes and, as such, they play an which are a non-toxic form of storage sterols in the cytoplasm. important role in several physiological processes (Wollam and Very recently, HISE1 (HIGH STEROL ESTER 1) has been Antebi, 2011; Zhang et al., 2020). Cholesterol and ergosterol are identified as a key regulatory factor in Arabidopsis sterol the sole end products of sterol biosynthesis in animals and fungi, homeostasis (Shimada et al., 2019). It has been proposed that respectively. In plants, however, over 250 different sterol species, HISE1 suppresses excess sterol formation by downregulating the also named phytosterols, have been identified. Although their protein levels of HMG-CoA reductase, the major rate-limiting amounts and relative proportions vary between plant species enzyme of the sterol pathway whose activity is tightly controlled and tissues, the most abundant sterols are usually b-sitosterol at different levels from transcriptional to post translational and stigmasterol, the end products of the 24-ethylsterols branch, (Wentzinger et al., 2002; Nieto et al., 2009; Leivar et al., 2011; and campesterol, the final product of the 24-methylsterols Doblas et al., 2013; Robertlee et al., 2017). In the event that branch. Some members of the Solanaceae, Liliaceae, and excess sterols are still present, they can be acylated to form SE, Scrophulariaceae families accumulate also significant amounts which are subsequently segregated and stored in cytoplasmic of cholesterol (Hartmann, 2004). lipid droplets (LD). The buffering effect of SE formation is In addition to their role as structural components of cell especially evident when sterol biosynthesis is enhanced, since membranes, and independently of their function as precursors in these conditions FS levels remain essentially unchanged while of brassinosteroid hormones (BR), phytosterols can also act as excess sterols accumulate as SE in LDs (Maillot-Vernier et al., signaling molecules that regulate plant growth and development. 1991; Gondet et al., 1994; Wilkinson et al., 1994; Schaller et al., Mutants defective in genes encoding enzymes of the post- 1995; Bouvier-Navé et al., 2010; Shimada et al., 2019). These squalene segment of the phytosterol pathway, such as dry2/ intracellular structures are present in the seed and leaf cells of sqe1, smt1, smt2, cpi1, cyp51a, hyd1, and hyd2/fackel, show many land plants, and their function and content may differ developmentally impaired phenotypes (Jang et al., 2000; Peng depending on the plant tissue and species. Thus, seed LDs et al., 2002; Clouse, 2002; Posé et al., 2009; Carland et al., 2010), function as storage compartments for lipids, while in leaves and specific inhibitors of sterol biosynthetic enzymes suppress they seem to act as subcellular factories with a role in senescence cell growth (Yates et al., 1991; Wentzinger et al., 2002; Gas- and fungal infection (Mckegney et al., 1995; Bouvier-Navé et al., Pascual et al., 2015; Shimada et al., 2019). But not only sterol 2010; Shimada et al., 2014; Shimada et al., 2015). Regarding the deficiency has deleterious effects on plant growth and neutral lipid composition of these vesicular-structures, although development, an excess of sterols is also harmful for plants, TAGs are the major component in most seed LDs (Tzen et al., although the mechanisms underlying the toxic effects of sterol 1993), some of them, like those from canola, lotus or maize, are overaccumulation in plant cells are still poorly understood. It has enriched in SE (Kemp and Mercer, 1968; Herná ndez-Pinzó n been recently suggested that excess sterols might inhibit cellular et al., 1999; Harker et al., 2003), while leaf LDs accumulate activity of vegetative organs by inducing the expression of genes mainly SE or wax esters (Parker and Murphy, 1981). LDs are related to biotic and abiotic stress (Shimada et al., 2020). Cell dynamic organelle which size and number can be modulated by membranes with enhanced levels of sterols may lose the ability genetic or environmental factors (Tatenaka et al., 2019). to retain water and leak iron, thus resulting in iron deficiency The proposed model for maintaining membrane sterol (Shimada et al., 2020). Interestingly, depletion of membrane homeostasis (Shimada et al., 2019) highlights a central role of sterols also leads to misregulation of genes related to biotic and sterol esterification in this process (Lewis et al., 1987; Dyas and abiotic stress responses and alters iron homeostasis (Manzano Goad, 1993; Chang et al., 1997; Sturley, 1997; Schaller, 2004). SE et al., 2016). Altogether, these and other observations support formation implies the esterification of a long-chain fatty acid the well-established view that membrane sterol levels must be group to the free hydroxyl at C3 position of the FS four-ring tightly controlled to avoid the negative effects of perturbed sterol backbone. This reaction is catalyzed by a family of enzymes homeostasis on cell viability and overall plant performance. known as sterol acyltransferases, which can be classified into Frontiers in Plant Science 02 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 acyl-CoA:sterol acyltransferases (ASAT; EC 2.3.1.26) and et al., 2010). More recently, it has been reported that the phospholipid:sterol acyltransferases (PSAT; EC 2.3.1.43) Arabidopsis hise1 mutant shows normal growth phenotype depending on whether the acyl donor substrate is a long-chain and similar sterol content than WT leaves, because excess fatty acyl-CoA or a phospholipid, respectively (Korber et al., sterols are esterified by PSAT1 and stored in LDs (Shimada 2017). The sterol moiety of SE includes the bulk sterol species et al., 2019). However, the hise1-3 psat1-2 double mutant present in the FS fraction and also many other less abundant exhibited a severe growth phenotype due to high sterol sterols and biosynthetic intermediates, the latter being in some content in the leaves, in which the presence of LDs was not cases the predominant sterols in the SE fraction (Ferrer et al., detected, while these structures were present in WT and hise1 2017). Different fatty acids may also be found esterified at the C3 mutant leaves (Shimada et al., 2019). Thus, PSAT1 plays a position, covering a wide range of lengths (from C12 to C22), but critical role in the suppression of sterol overaccumulation the most common are palmitic, stearic, oleic, linoleic and through their esterification and subsequent storage in LDs. In linolenic acids (Dyas and Goad, 1993). The structural diversity a previous work, we identified, cloned and characterized the of plants SE reflects the relevance of the esterification process PSAT1 and ASAT1 enzymes of tomato (Lara et al., 2018). This and their specialized functions in different tissues and organs. allowed us to undertake the study reported here, in which we Thus, SE are especially abundant in anther tapetal cells, pollen have generated and characterized tomato (cv. Micro-Tom) grains, seeds and the phloem (Dyas and Goad, 1993; Herná ndez- CRISPR-Cas9-edited loss-of-function mutans lacking PSAT1 Pinzó n et al., 1999; Wu et al., 1999; Harker et al., 2003; Behmer and ASAT1. The results obtained contribute to deepen the et al., 2013; Villette et al., 2015). High levels of SE are also current understanding on the role of SE and the enzymes accumulated in senescent leaves (Bouvier-Navé et al., 2010), in involved in their synthesis in tomato plant growth mutant plants overproducing sterols (Gondet et al., 1994; and development. Schaller et al., 1995; Shimada et al., 2020), and when the flux through the sterol pathway is forced to increase by supplying cell cultures with pathway intermediates (Wilkinson et al.,1994). Materials and methods Thus, the existing data indicate that SE serve as a reversible storage pool of sterols that helps to maintain the proper levels of Generation of CRISPR/Cas9 vectors and FSs in cell membranes during normal plant growth and plant transformation development, thus preventing the potential destabilizing effect that changes of FS levels above or below a certain threshold may Recombinant plasmids for CRISPR/Cas9-mediated genome have on membrane structure and function. As this is likewise editing were generated as described by Schiml et al. (2016) using important under special circumstances, FS homeostasis is crucial a modified version of the pDE-Cas9 plasmid (Karlsruhe Institute in stress conditions (Kumar et al., 2018), during senescence and of Technology), in which the phosphinothricin resistance gene aging (Bouvier-Navé et al., 2010) and also when free sterols are (BAR) was replaced by the kanamycin resistance gene (NPTII) required immediately in important quantities, for instance by restriction digest with HindIII and subsequent ligation, during seed germination and early seedling development and yielding the plasmid pDE-Cas9-KanR. The sgRNAs targeting pollen tube growth (Dyas and Goad, 1993). To meet these SlPSAT1 (Solyc09g072710) and SlASAT1 (Solyc11g012260) were demands, it is necessary to achieve the correct balance designed using the on-line Breaking-Cas tool (http://bioinfogp. between SE formation and mobilization, two biochemical cnb.csic.es/tools/breakingcas/) (Oliveros et al., 2016). The processes that are still poorly understood in plants. oligonucleotide pairs encoding the selected sgRNAs Plant sterol acyltransferase activity has been reported to be (Supplementary Table S1) were annealed and cloned into the primarily associated with membrane fractions in different BbsI site of the pEn-Chimera entry vector (Addgene). The species (Garcia and Mudd, 1978; Zimowski and resulting pEn-Chimera/ASAT1 and pEn-Chimera/ASAT1 Wojciechowski, 1981a; Zimowski and Wojciechowski, 1981b; plasmids were sequenced to exclude cloning artifacts. Plasmids Kalinowska et al., 1989; Bouvier-Navé and Benveniste, 1995; pEn-Chimera/ASAT1 and pEn-Chimera/PSAT1 were Banas et al., 2005; Chen et al., 2007), but so far only ASAT and recombined with pDE-Cas9-KanR destination vector yielding PSAT enzymes from Arabidopsis and tomato have been cloned plasmids pDE-Cas9 ASAT1 and pDE-Cas9 PSAT1 (Figure 1), and functionally characterized (Banas et al., 2005; Chen et al., which were transferred to the A. tumefaciens strain GV3101:: 2007; Lara et al. 2018). In Arabidopsis, the characterization of T- pMP90 prior to transform tomato (cv. Micro-Tom) cotyledons DNA insertion mutants defective in the expression of PSAT1 (Fernandez et al., 2009). The presence of the transgenes in and ASAT1 genes revealed that PSAT1 plays a major role in the kanamycin-resistant tomato plants (T0) was checked by PCR synthesis of SE in seeds and in the maintenance of leaf viability amplification of a 442-bp fragment of the AtU6-26 promoter during aging, since a premature senescence phenotype using primer pairs shown in Supplementary Table 1 and leaf associated to decreased SE levels was observed in leaves of the genomic DNA isolated with the cetyltrimethyl ammonium psat1 mutant lines, but not in the asat1 ones (Bouvier-Navé bromide (CTAB) method (Richards et al., 1994). Frontiers in Plant Science 03 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B C FIGURE 1 Map of the T-DNA region in the binary vectors pDE-Cas9 PSAT1 and pDE-Cas9 ASAT1 (A). It includes the parsley ubiquitin 4-2 gene promoter (pUbi4-2), the Arabidopsis U6 snRNA-26 gene promoter (pAt6-26), the sequence coding for the Streptococcus pyogenes CAS9 nuclease (Cas9), the single guide RNA sequences targeting the SlPSAT1 and SlASAT1 genes (sgRNA), and the kanamycin resistance gene (nptII). Exons of the SlPSAT1 gene (B) and the coding region of the SlASAT1 gene (C) are represented by blue boxes. The sgRNA target sequences in the SlPSAT1 and SlASAT1 genes are also shown. The protospacer adjacent motif (PAM) is shown in blue and predicted CAS9 nuclease cleavage sites are indicated by red arrowheads. Mutation analysis in CRISPR/Cas9-edited expose the stigma and style, and all open flowers near the slpsat1 and slasat1 mutants emasculated flowers were also removed to minimize the risk of self-fertilization. Pollen was collected from open flowers of Genomic DNA fragments of 303 and 293 bp spanning the homozygous slpsat31 plants, and pollination of slasat34 SlASAT1 and SlPSAT1 target sites, respectively, were amplified flowers was performed 24 h after emasculation by dipping the by PCR using DNA from transgenic plants as template, exposed stigma into the pollen (Chetelat and Peacock, 2013). AccuPrime™ Taq DNA Polymerase High Fidelity (Thermo Seeds from the resulting fruits were collected and double Fisher Scientific), and specific primer pairs flanking the target heterozygous mutant plants were initially selected by PCR- sites (Supplementary Table S1). The PCR products from T0 based genotyping of the two mutations as described above. transgenic plants were cloned into pGEM®-T Easy (Promega) Additional generations were obtained through self-pollination and the inserts of at least 10 independent clones were Sanger until slpsat31 x slasat34 plants carrying both mutant alleles in sequenced and analyzed with Serial Cloner version 2.6.1 (Perez, homozygosis were identified as described above. 2004). In subsequent generations, the bulk PCR products were directly Sanger sequenced after purification with the Gel and PCR clean-up kit (Macherey-Nagel) using primers sgRNA RT-qPCR expression analysis ASAT1 Fw and sgRNA PSAT1 Fw (Supplementary Table 1). The resulting sequence trace files were analyzed with the For RT-qPCR gene expression analysis, cDNA samples were Tracking of Indels by Decomposition (TIDE) software (http:// prepared from leaf DNA-free RNA obtained using the Maxwell tide.nki.nl) (Brinkman et al., 2014). RSC Plant RNA Kit with the Maxwell RSC Instruments (Promega) according to manufacturer’s instructions. Real-time PCR assays were performed in triplicate with SYBR Green I Generation of slpsat1 x slasat1 Master (Roche Diagnostics) using a Light Cycler 480 detection double mutant system (Roche Diagnostics) as described in Ramirez-Estrada et al. (2017). Specific primer pairs for target genes SlPSAT1, To generate the slpsat1 x slasat1 double knock-out mutant, SlASAT1 and SlSAG12 (Solyc02g076910), and the actin gene sepals, petals and the anther cone of unopened flower buds of (Solyc03g078400) used as reference gene, are shown in homozygous slasat34 plants were removed with forceps to Supplementary Table 1. Frontiers in Plant Science 04 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 Imaging lipid droplets Cas9 cassettes (Figure 1) designed to target the nucleotide sequences coding for the amino acid sequences GLARGLLG Lipid droplets were stained with TopFluor® cholesterol (positions 267 to 274 in SlASAT1) and GDYSKLSG (positions essentially as described in Zhou et al. (2019). Leaf discs of about 27 to 34 in SlPSAT1). Primary transformants (T0) were first 0.5 cm in diameter were excised from fresh leaves, transferred into screened for the presence of the CRISPR/Cas9 constructs by PCR a 20 mL syringe containing 2 mL 10 mM TopFluor Cholesterol amplification of the AtU6-26 promoter that drives the expression (Avanti Polar Lipids) and 30 mM methyl-b-ciclodextrin (Fischer of the sgRNAs. Next, genomic fragments encompassing the Scientific), and infiltrated under negative pressure by pulling the corresponding sgRNA target sites were amplified by PCR, plunger for 30s. After the atmospheric pressure was recovered, cloned and sequenced. Comparison of the obtained sequences leaf discs were kept in the staining solution for an additional 4 h in with those of the equivalent genomic regions of WT plants darkness. LD images were acquired by confocal scanning revealed the presence of a variety of short indels ranging from 1 fluorescence microscopy using a Olympus FV1000 microscope to 16 nucleotides in 14 slasat1 and 7 slpsat1 T0 mutant plants. A and the FV10-ASW software (Olympus) in Z-stack mode. similar analysis conducted in the respective T1 and T2 progenies TopFluor® cholesterol was excited at 488 nm and fluorescence allowed the identification of plant lines with CRISPR/Cas9- was detected at 490-520 nm. LD were counted from each image induced homozygous mutations in the SlPSAT1 and SlASAT1 using ImageJ 1.48v software and averaged by leaf area. genes, among which, mutant lines slpsat28, slpsat31, slasat34 and slasat35 were selected for further characterization. Lines slpsat28 and slasat34 showed single nucleotide insertions at the predicted Aging of detached leaves Cas9 editing sites, while lines slpsat31 and slasat35 had deletions of 4 and 16 nucleotides, respectively (Figure 2). All these Leaflets of fully green second and third leaves detached from 7 frameshift mutations were expected to result in truncated forms week-old-plants were placed in distilled water (20 mL) in Petri of both enzymes predicted to be catalytically inactive. The native dishes and stored in the dark at 24°C for ten days. Leaf images were SlPSAT1 protein consists of 630 amino acid residues, in sharp taken with a Nikon Z50 camera. The non-necrosed area of senescing contrast to the predicted sequences of 43 (slpsat28) and 71 leaves was measured as pixels by image processing. Necrosed vs non (slpsat31) amino acid residues of proteins encoded by the necrosed leaf surface was measured using Fiji (https://imagej.net/ slpsat1 mutant alleles. In the case of SlASAT1, the WT enzyme software/fiji/) image processing package of ImageJ2. consists of 444 amino acids, while the truncated proteins encoded by the mutant alleles are predicted to have 271 (slasat34) and 282 (slasat35) amino acids, thus lacking the key catalytic amino acid Chlorophyll measurements residues, Asn299 and His343, found in all MBOATs (Hofmann, 2000; Huang et al., 2014). Chlorophyll was measured as described by Porra (2002). Briefly, total chlorophyll from fresh leaves (250 mg) was extracted using aqueous 80% acetone. After centrifugation at 13,000 x g for 2 min, Effect of SlPSAT1 and SlASAT1 the chlorophyll was quantified spectrophotometrically in the inactivation on the profile of free supernatant as described in Zhang et al. (2009). and esterified sterols in tomato seeds and leaves Sterol analysis The impact of SlPSAT1 and SlASAT1 inactivation on free Free and esterified sterols were quantified by GC-MS in extracts and esterified sterol levels was investigated in seeds and leaves of from tomato seeds (200-400 mg) and the third and fourth leaves the slpsat1 and slasat1 mutant plants. SlPSAT1 inactivation led (3 g) as described in Lara et al. (2018). Normalization of the SE to a drastic depletion of total SE levels in the seeds of both slpsat1 levels to the FS levels to determine the esterified fraction of sterols lines to less than 10% (8.9% in slpsat28 and 8.2% in slasat31) (EF) was calculated as 100*SE/(SE+FS). compared to those in WT seeds (Figure 3A and Supplementary Table 2). A significant but less severe SE reduction was also observed in the leaves, as SE levels were reduced below 30% Results (26.7% in slpsat28 and 29% in slasat31) compared to WT leaves (Figure 3B and Supplementary Table 3). Rather unexpectedly, Generation of tomato slpsat1 and slasat1 the levels of SE in seeds and leaves of both slasat1 mutants knock-out mutants increased instead of decreasing. Indeed, the total SE content in the slasat1 seeds increased by about 50% (48% in slasat34 and To investigate the biological role of SlASAT1 and SlPSAT1, 61% in slasat35) compared to WT seeds (Figure 3A and tomato (cv. Micro-Tom) plants were transformed with CRISPR/ Supplementary Table 2), while a much stronger effect was Frontiers in Plant Science 05 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 FIGURE 2 CRISPR/Cas9-induced knock-out mutations in the SlPSAT1 and SlASAT1 genes. Nucleotide sequences surrounding the sgRNA target site in SlPSAT1 (upper box) and SlASAT1 genes (lower box), and the resulting mutated sequences after imperfect NHEJ repair are shown. The sgRNA target sequences are depicted in blue, including the PAM motif highlighted in bold. The single nucleotide insertions in slpsat28 and slasat34 mutant alleles are shown in red capital letters, while deletions of 4 and 16 nucleotides in slpsat31 and slasat35 alleles, respectively, are shown in red above the dashed lines. Premature stop codons as a result of frameshift mutations are shown in red. Nucleotides are numbered on the left. The amino acids encoded by the wt and mutant alleles are shown below the respective nucleotide sequences and are numbered on the right. Amino acids in the mutant proteins that are different from those in the wild type protein sequences are underlined. The wild type SlPSAT1 and SlASAT1 proteins as well as their corresponding truncated versions are shown by blue boxes. Dark blue boxes represent the six conserved domains A-F found in LCAT proteins, and the eight predicted transmembrane sequences (I – VIII) of SlASAT1. The amino acid residues (Ser, Asn and His) of the catalytic triad in regions C, E, and F of SlPSAT1, and the SlASAT1 catalytic amino acid residues (Asn and His) are indicated by red vertical lines. The acyl-CoA binding site in SlASAT1 is shown in yellow. White boxes in the truncated proteins represent amino acid sequences not found in the wild type proteins arising from the frameshift mutations introduced by the NHEJ repair mechanism. Frontiers in Plant Science 06 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B C D FIGURE 3 Total levels of free and esterified sterols in seeds and leaves of slpsat1, slasat1, and slpsat1 x slasat1 mutans. Quantification of steryl esters (A, C) and free sterols (B, D) was carried out in seeds and the third and fourth leaf of one-month-old plants. Data are presented as mean ± SEM from three (seeds) and four (leaves) biological replicates per genotype. Asterisks indicate values that are significantly different compared to those in the control plants determined by t-test (*P<0.05; **P<0.01; ***P<0.005). observed in the leaves, which showed approximately a three-fold Supplementary Table 2), the leaves of slpsat1 and slasat1 increase in SE levels (284% in slasat34 and 308% in slasat35) mutants showed a moderate but significant enhancement of FS compared to WT leaves (Figure 3B and Supplementary Table 3). levels ranging from 1.3- to 1.7-fold the WT levels (Figure 3D and Taken together, these results demonstrated that loss of function Supplementary Table 3). All these changes were reflected on the of SlPSAT1 and SlASAT1 has the same qualitative effects on the total EF values (Table 1), which were clearly lower in both slpsat1 SE levels in seeds and leaves, although there are some seeds (5.3% and 4.1% in slpsat28 and slpsat31, respectively) and quantitative differences. On the contrary, the effects of leaves (11.6% and 12% in slpsat28 and slpsat31, respectively) impaired sterol esterification on total FS levels were clearly than in the corresponding WT organs (37.4% and 41% in seeds different in both organs. Thus, while SlPSAT1 inactivation had and leaves, respectively), but markedly increased in slasat1 seeds no significant impact on seeds FS content and SlASAT1 knock- (about 60% in both mutant lines) and leaves (61.5% and 68.9% out caused a reduction in FS levels of about 60% (58% in slasat34 in slasat34 and slasat35, respectively). These data reinforce the and 63% in slasat35) those in WT seeds (Figure 3C and contrasting effects of SlPSAT1 and SlASAT1 inactivation on the Frontiers in Plant Science 07 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 TABLE 1 Esterified sterol fractions in seeds and leaves of wt, slasat1, slpsat1 and slpsat1 x slasat1 mutants. Esterified sterol fraction (%) SEEDS WT slasat34 slasat35 slpsat28 slpsat31 slpsat31 x slasat34 Total sterols 37.4 60.5 60.2 5.3 4.1 3.0 Cholesterol 76.3 89.9 87.0 11.1 n.d. 16.7 Campesterol 40.3 62.8 46.9 n.d. n.d. n.d. Stigmasterol 13.9 38.4 26.7 0.3 n.d. 2.7 Sitosterol 28.0 45.5 53.2 4.3 5.1 2.2 Isofucosterol 50.3 77.3 75.2 n.d. n.d. n.d. LEAVES WT slasat34 slasat35 slpsat28 slpsat31 slpsat31 x slasat34 Total sterols 41.2 61.5 68.9 11.6 12.0 5.2 Cholesterol 39.7 65.8 64.8 16.5 12.0 n.d. Campesterol 51.4 52.1 51.8 n.d. n.d. n.d. Stigmasterol 7.4 26.7 23.1 n.d. n.d. n.d. Sitosterol 35.5 59.8 64.8 7.8 10.1 17.3 Values were calculated according to 100*SE/(SE+FS). n.d. stands for not detected. Values shown in bold correspond to the esterified fraction of the total sterols. total SE content in tomato seeds and leaves. A detailed analysis SlPSAT1 expression in seeds and the opposite trend of changes of SE and FS profiling in both organs showed that inactivation of in SlPSAT1 expression (decreased) and SE levels (enhanced) in both tomato sterol acyltransferases affected to a greater or lesser leaves, indicate that the increased levels of SE in the slasat1 extend the contents of all measured free and esterified sterol mutants cannot be attributed to a transcriptional up-regulatory species in seeds (Supplementary Table 2) and leaves response of the intact SlPSAT1 gene. Instead, the possibility that (Supplementary Table 3), which resulted in a strong reduction the reduced expression of SlASAT1 in the slpsat1 mutant seeds of the EF of all the analyzed species in seeds and leaves of slpsat1 might contribute to the strong reduction of SE levels cannot mutants compared to WT (Table 1). In fact, only esterified be excluded. b-sitosterol could be detected in seeds of both mutant lines, with 5.5- and 6.5-fold reductions of the EF in slpsat28 and slpsat31, respectively, while in leaves, esterified b-sitosterol and Generation and sterol profiling of cholesterol were detected with reductions of their EF ranging tomato slpsat1 x slasat1 double from 2.5- to 4.5-fold in both slpsat1 mutants. In the slasat1 knock-out mutant mutants the higher values of the total EF compared to the WT were primarily due to an increase of the stigmasterol EF in seeds The highly contrasting profiles of FS and SE in the slpsat1 (2- and 2.8-fold in slasat34 and slasat35, respectively) and leaves and slasat1 single knock-out mutants prompted us to generate a (more than 3-fold in both slasat1 mutants), while increases slpsat1 x slasat1 double knock-out mutant, in order to assess the ranging from 1.5- and 1.9-fold were observed in the EF of impact of simultaneous loss of function of both enzymes other sterol species, with the exception of cholesterol in seeds involved in SE biosynthesis. To this end, we crossed slpsat31 and campesterol in leaves, which were only slightly enhanced or and slasat34 single mutants to obtain a tomato line carrying both remained unchanged, respectively (Table 1). Overall, these mutant alleles in homozygosis. Sterol profiling in the double results demonstrated that SlPSAT1 has a predominant role in mutant revealed a metabolic phenotype very similar to that of SE biosynthesis while SlASAT1 would mainly have a regulatory the slpsat1 mutants, though there was an additional reduction of role of the sterol biosynthesis pathway. total SE levels compared to the single slpsat1 mutants, which To investigate the possible existence of transcriptional were further depleted to 5% and 7% the content in WT seeds crosstalk between the two sterol acyltransferases, we measured (Figure 3A and Supplementary Table 2) and leaves (Figure 3B the expression of the SlPSAT1 and SlASAT1 genes in the slasat1 and Supplementary Table 3), respectively, while FS levels were and slpsat1 mutants, respectively. A significant decrease (1.6- to restored to WT levels in both cases (Figures 3C, D, and 1.7-fold) of SlASAT1 expression was observed in slpsat1 seeds Supplementary Tables 2 and 3). These observations confirmed (Figure 4C) but not in leaves (Figure 4D), while the SlPSAT1 the predominance of SlPSAT1 over SlASAT1 in tomato SE expression in slasat1 seeds was similar to WT (Figure 4A) and a biosynthesis. Interestingly, the effects of the simultaneous significant decrease (1.7-fold) was observed in slasat34 leaves inactivation of SlPSAT1 and SlASAT1 on individual sterol (Figure 4B). The lack of effect of SlASAT1 inactivation on species EF were disparate, since the b-sitosterol EF in leaves Frontiers in Plant Science 08 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B C D FIGURE 4 SlPSAT1 and SlASAT1 mRNA levels in seeds and leaves of slasat1 and slpsat1 mutants. RT-qPCR analyses were performed using RNA samples obtained from the same tissue samples used for sterol analysis (seeds and the third and fourth leaf of one-month-old WT, slpsat1 and slasat1 plants). The mRNA levels of the actin gene were used to normalize transcript levels of SlPSAT1 (A, B) and SlASAT1 (C, D). Data are presented as mean ± SEM from two biological replicates per genotype with three technical replicates each (n=6). Asterisks indicate values that are significantly different compared to those in the WT plants determined by t-test (*P ≤ 0.05). and cholesterol and stigmasterol in seeds were increased in leaves of slpsat28, slpsat31 and slpsat1 x slasat1 mutants, compared to those in corresponding organs of the slpsat1 respectively, compared to the WT leaves (Figure 5B). single mutants, in contrast to the lower values of other sterol Unexpectedly, the abundance of LDs in both slasat1 mutants species (Table 1). Taken together, the results of the sterol was also lower that in the WT plants, with reductions of ~18% profiling in the single and double mutants support the view (slasat34) and ~35% (slasat35) (Figure 5B) despite the leaves of that different regulatory mechanisms operate in tomato seeds these plants accumulate much higher levels of SE than the WT and leaves and also in the different branches of the sterol (Figure 3B). Thus, the direct correlation between ES levels biosynthesis pathway leading to the major sterols, to maintain (Figure 3B) and LD abundance (Figure 5B) observed in leaves FS homeostasis. of the slpsat1 and slpsat1 x slasat1 mutants does not seem to exist in the slasat1 leaves. Effect of SlPSAT1 and SlASAT1 inactivation on the lipid droplet Slpsat1 and slpsat1 x slasat1 knock-out abundance in tomato leaves mutants exhibit reduced size an early leaf senescence phenotype To investigate the possible correlation between SE levels and the abundance of cytosolic LD, discs from the fourth leaf of WT Tomato CRISPR-Cas9 mutant lines described in this work and mutant plants were stained with TopFluor® cholesterol were examined for morphological phenotypes associated with (Gocze and Freeman, 1994) and observed with confocal the loss of function of SlPSAT1, SlASAT1 or both. Seven- and microscopy. As displayed in Figure 5A, leaves of slpsat1 and ten-week-old slasat1 mutant plants grew almost identically than slpsat1 x slasat1 mutants contain less LDs than the WT and the WT plants (Figures 6A-C), whereas slpsat1 plants were slasat1 single mutants. This is in agreement with the results of approximately 50% shorter than WT plants (Figures 6A-C), the corresponding quantitative analysis of stained LDs, which and a 25% reduction in height was observed in the double revealed reductions of ~63, ~51 and ~74% in the amount of LDs mutant slpsat1 x slasat1 compared to WT (Figures 6A-C). These Frontiers in Plant Science 09 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B FIGURE 5 Intracellular lipid droplet distribution in leaf epidermal cells of slpsat1, slasat1, and slpsat1 x slasat1 knock-out mutants. (A) Representative confocal scanning fluorescence microscopy images showing the distribution of lipid droplets stained with TopFluor® cholesterol. (B) Quantification of lipid droplet abundance using ImageJ 1.48v software averaged by leaf area. Data are presented as mean ± SD from three biological replicates per genotype, with n=15-23 images per genotype analyzed. Asterisks indicate values that are significantly different compared to those in the control leaves determined by t-test (* P ≤ 0.05; ***P ≤ 0.005: ****P ≤ 0.001). Scale bars = 10 mM. Frontiers in Plant Science 10 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B C D FIGURE 6 Phenotypic characterization of slpsat1, slasat1, and slpsat1 x slasat1 mutant plants. Representative images of 7-week-old (A) and 10-week-old (B) slpsat1, slasat1, and slasat1 x slpsat1 knock-out mutants grown under greenhouse conditions. (C) Quantitative analysis of plant height to first inflorescence of 10-week-old plants. (D) SlSAG12 mRNA levels measured by RT-qPCR in RNA samples from the 2nd and 3rd leaves of 7-week- old plants. The mRNA levels of the actin gene were used to normalize SlSAG12 transcript levels. Data are presented as mean ± SEM from three biological replicates per genotype with three technical replicates each (n=9). Asterisks indicate values that are significantly different compared to those in the WT plants determined by t-test (****P ≤ 0.001). results demonstrated that a functional SlPSAT1 is essential for slpsat1 mutants (Figure 6C). On the other hand, when mutant normal tomato plant growth while, at least under normal growth plants were about 2.5-month-old, the oldest (lower) leaves of conditions, the lack of functional SlASAT1 has no evident effects slpsat1 mutants started to yellow, while those of the slpsat1 x on plant growth. Interestingly, the loss of function of SlASAT1 slasat1 double mutant turned brown and dried out as typical partially reverses the dwarf phenotype caused by SlPSAT1 senescent leaves. In contrast, all leaves in the slasat1 mutants, inactivation, since a compensatory size effect was observed in like those of WT plants, were still green (Figure 6B). These the slpsat1 x slasat1 double mutant compared with the single observations indicated that a decrease of SE content in tomato Frontiers in Plant Science 11 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 leaves was associated to an early senescence phenotype, which synergistic effect in the premature induction of this was more pronounced the lower the SE content was (compare physiological process. Figures 3B and 6B). The positive correlation between reduced levels of SE and premature leaf senescence was confirmed by measuring the expression of the senescence marker gene SAG12 Tomato slpsat1 and slasat1 mutations (Noh and Amasino, 1999). To this end, total RNA was extracted increase the germination speed from leaves of two different ages (second and third leaves) harvested from seven-week-old WT and mutant plants. At this It is generally accepted that mature seeds accumulate SE to developmental stage the third leaves of the different mutants and meet the rapidly increasing demand for sterols when they the WT plants were completely green, whereas the second leaves germinate and during the early stages of seedling growth of slpsat1 mutants showed senescence symptoms, which were (Ferrer et al., 2017). Since loss of function of SlPSAT1 and more pronounced in the case of the slpsat1 x slasat1 double SlASAT1 leads to altered levels of SE in seeds, being particularly mutant (Figure 6B). As expected, Similar basal levels of the remarkable the strong depletion in slpsat1 and slpsat1 x slasat1 SlSAG12 mRNA were detected in the third leaves of WT and the mutants (Figure 3A and Table 1), we monitored the effects of mutant plants (Figure 6D). However, the expression of this SlPSAT1 and SlASAT1 loss of function on the rate and speed of senescence-marker gene was drastically induced in the second germination. As displayed in Figure 8, the double mutant slpsat1 leaves of both slpsat1 mutants compared to the equivalent leaves x slasat1 seeds germinated significantly faster than those from of WT plants (about 28- and 16-fold higher in slpsat28 and WT and the single mutants slpsat1 and slasat1, whose seeds slpsat31, respectively). Interestingly, the induction of SlSAG12 germinated also faster than WT (Figure 8). Thus, two days after expression was more pronounced in the slpsat1 x slasat1 double seed imbibition, the germination percentage of the double mutant than in the slpsat1 single mutants (34-fold compared to mutant was about 10-fold higher than the WT. However, these WT). On the contrary, no significant differences in the SlSAG12 differences decreased to 4-fold at day 3, when a significant transcript levels were observed in the second leaves of both increase of about 3-fold was also observed in the germination slasat1 mutants compared to the WT leaves (Figure 6D). To percentage of the single slpsat1 mutant seeds and, to a lower investigate if the early senescence phenotype was simply due to extent (about 2-fold), in that of the slasat1 seeds (Figure 8A). some developmental alteration affecting leaf ontogeny (i.e. These differences were also reflected in the calculated t50 values, natural senescence) or it could involve some other regulatory which were only 1.92 days for slpsat1 x slasat1 mutant seeds process, an induced leaf senescence assay was performed. compared to 3.56 days for WT seeds. A slight decrease of this Leaflets of equivalent fully green leaves detached from WT value was also observed in the single mutants compared to WT and the mutant plants (fourth leaf of 7-week-old plants) were (Figure 8B). It is worth to mention that at the end of the incubated in water in permanent darkness. The leaflets from WT experimental process (7 days), no significant differences were and slasat1 mutant leaves remained completely green after detected between the germination percentage values of WT incubation for ten days, while clear senescence symptoms seeds and those from the different mutant genotypes (yellow and necrotic areas) were observed on leaflets from (Figure 8A). All these data demonstrated that changes in SE slpsat1 and slpsat1 x slasat1 mutant leaves (Figure 7A). In fact, content have no effect on tomato seeds final germination rate but slasat1 mutant leaflets showed the same percentage of non- cause an early germination phenotype observed not only when necrotic area (~97% in slasat34 and ~94% in slasat35) than the the SE levels are decreased, but also, although more slightly, WT (~96%), while this percentage decreased significantly in when they are increased. This unexpected result was confirmed slpsat1 mutants (~66% in slpsat28 and ~70% in slpsat31) and was by the stronger early germination phenotype shown by slpsat1 x even lower in the double mutant slpsat1 x slasat1 (~52%) slasat1 seeds, which moreover demonstrated a synergistic effect (Figure 7B). These results were fully supported by the of the two mutations (Figure 8). chlorophyll levels measured in the corresponding leaflet samples. Indeed, no differences in chlorophyll content were observed between slasat1 and WT plants (0.59 and 0.57 mg/g Discussion FW in slasat34 and slasat35, respectively, compared to 0.58 mg/g FW in WT), while this pigment content was significantly lower The information currently available on plant SE biosynthesis in the slpsat1 mutants (0.40 and 0.37 mg/g FW in slpsat28 and and the sterol acyltransferases responsible for their formation is slpsat31, respectively) than in WT. In the double mutant the still limited. Hence, to expand the knowledge about this content was only about 0.30 mg/g FW (Figure 7C). Taken metabolic process and its biological significance, we generated together, all the results above demonstrated that WT levels of and characterized tomato (cv. Micro-Tom) CRISPR/Cas9- SlPSAT1, in contrast to SlASAT1, are required for normal leaf induced single slpsat1 and slasat1 loss-of-function mutants senescence progression, even though in the absence of a (Figures 1, 2) as well as the corresponding slpsat1 x slasat1 functional SlPSAT1, the loss of function of SlASAT1 has a double mutant, which was obtained by crossing selected single Frontiers in Plant Science 12 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B C FIGURE 7 Induced senescence assay of slpsat1, slasat1, and slpsat1 x slasat1 detached leaves. (A) Representative image showing leaves detached from the indicated plants after 10 days of incubation at 24°C on water, in total darkness. The leaves of slpsat28 and slasat35 mutants, which are not shown in this figure, presented the same phenotype as those detached from slpsat31 and slasat35 mutants, respectively. (B) The percentage of non-necrotic leaf area was calculated by dividing the non-necrotic area by the total leaf area, which were quantified using the ImageJ2 software. (C) Chlorophyll content was determined in the leaves after incubation for ten days under the conditions described above. Asterisks indicate values that are significantly different compared to those in the WT plants determined by the t-test (**P < 0.01; ****P < 0.001). mutants. Analysis of the total SE contents in slpsat28, slpsat31, suggests that SlASAT1-dependent sterol esterification plays a slasat34 and slasat35 mutants demonstrated that SE formation regulatory role in tomato FS biosynthesis, as previously in seeds and leaves depends almost solely on SlPSAT1 activity, suggested in Arabidopsis (Chen et al., 2007). It has been since depletion of SE levels was observed only in the slpsat1 proposed that ASAT1 regulates the flux through the post- mutants while SE levels in the slasat1 mutants were increased squalene segment of the sterol pathway, and therefore the instead of being depleted (Figure 3, Table 1). These results, along amount of major sterol end products, by modulating the with the observation that the total amount of SE, as well as the quantity of cycloartenol (Chen et al., 2007; Zhou et al., 2019), total EF, in the slpsat1 x slasat1 double mutant were even lower which is the first cyclic intermediate in the plant sterol pathway than in the slpsat1 single mutants (Figure 3, Table 1), strongly (Ohyama et al., 2009). A detailed analysis of the SE profiling data Frontiers in Plant Science 13 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 A B FIGURE 8 Germination rate of seeds from slpsat1, slasat1, and slpsat1 x slasat1 knock-out mutants. Sterilized seeds (n=60 per genotype) were sown on MS plates and allowed to germinate under long day conditions (16h light/8h dark) at 24°C. The percentage of germinated seeds at the indicated time-points (A) was calculated by dividing the number of seeds showing radicle emergence by the total number of seeds. Data are presented as mean ± SEM from n=4-5 replicates per genotype. The t50 values (B) were calculated using the formula of Coolbear et al. (1984). Asterisks indicate values that are significantly different compared to those in the control plants determined by t-test (*P < 0.05; **P < 0.01; ***P < 0.005). in seeds and leaves of the mutant plants revealed that the (~8-fold vs ~3.5-fold reduction, respectively, compared to WT) contents of all measured sterol species were enhanced in the and, conversely that SlASAT1 inactivation actually had a very slasat1 mutants and depleted in both the slpsat1 and the slpsat1 x similar effect in both organs (~1.5-fold increase compared to slasat1 mutants (Table 1, Supplementary Tables 2 and 3), thus WT) (Table 1). A differential response was also observed in the indicating that tomato PSAT1, like its Arabidopsis counterpart seeds and leaves of the double mutant (~12-fold vs ~8-fold (Banas et al., 2005), has a broad sterol substrate specificity. reduction, respectively, compared to WT). Furthermore, this The qualitative impact of SlPSAT1 and SlASAT1 analysis unveiled a contrasting impact on the EF of specific sterol inactivation on the profile of total SE in seeds and leaves was species, such as cholesterol in seeds and b-sitosterol in leaves of almost identical, although important quantitative differences the double mutant, compared to the total EF and that of other were detected. The relative effect of SlPSAT1 inactivation on individual sterols. It is also worth noting that the stigmasterol EF total SE levels was more pronounced in seeds than in leaves was particularly sensitive to SlASAT1 inactivation compared to (~90% vs ~70% reduction, respectively, compared to WT), while the EF of the remaining sterols (Table 1). These observations that of SlASAT1 inactivation in seeds was far smaller than in indicate that the regulatory mechanisms controlling SE- leaves (~50% vs ~300% increase, respectively, compared to WT) dependent FS homeostasis are different not only in seeds and (Figure 3). However, normalization of SE levels to the FS levels leaves but also in the different branches of the sterol biosynthesis to calculate the total EF, revealed that the effect of SlPSAT1 pathway leading to the main sterol end products. This cannot be inactivation was indeed more pronounced in seeds than in leaves considered completely unexpected taking into account the Frontiers in Plant Science 14 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 distinct roles of sterols in these two organs and the differences in previously reported in Arabidopsis, there are also some relevant the biological properties of 24-ethylsterols (stigmasterol and b- differences. The dramatic increase of SE in leaves of the slasat1 sitosterol), 24-methylsterols (campesterol) and sterols with no tomato mutants (Figure 3B) was not observed in the rosette alkyl substituent at this position (cholesterol). leaves of the Arabidopsis asat1-1 mutant, whose SE levels are It is also worth noting that the increased amounts of ES even lower than those in the control leaves (Bouvier-Navé et al., found in seeds and leaves of the slasat1 mutants compared to 2010; Lara et al. 2018). Similarly, the increase of SE levels their WT counterparts cannot be attributed to SlPSAT1 observed in tomato slasat1 seeds (Figure 3A) is also in transcriptional upregulation, since its mRNA levels remained contrast to the unchanged content of SE in Arabidopsis seeds either unaltered in seeds (Figure 4A) or even diminished in (Bouvier-Navé et al., 2010). Remarkable differences were also leaves (Figure 4B). Thus, it remains to be established whether the observed in the effects on plant growth and important observed accumulation of SE levels in the slasat1 mutant seeds developmental processes such as leaf senescence and seed and leaves is due to post-transcriptional upregulation of the germination. A functional SlPSAT1, but not SlASAT1, is SlPSAT1 enzyme activity, simply to a greater availability of the necessary for proper tomato plant growth (Figures 6A, B), FS substrates of PSAT1, or both. In fact, it has been proposed which is consistent with the suggested role of SE as a source of that PSAT1 is not rate-limiting for SE formation in Arabidopsis sterols to sustain rapid expansion of vegetative organs (Zhou (Noiriel et al., 2004; Bouvier-Navé et al., 2010) and the tea tree et al., 2019). The reason why a similar phenotype was not (Zhou et al., 2019). On the other hand, the possibility that the observed in the Arabidopsis mutants lacking a functional enhanced levels of SE in the slasat1 mutants could result from PSAT1 (Bouvier-Navé et al., 2010) is unclear, although it the upregulation of some of the seven additional genes in the might be speculated that it could be due, at least in part, to the tomato genome coding for SlASAT-like proteins (Lara et al. relative lower levels of SE in the vegetative organs of tomato 2018) cannot be completely ruled out, although the observation slpsat1 and slpsat1 x slasat1 mutants (Figure 3B) compared to that the levels of SE in leaves and seeds of the double slpsat1 x those in the equivalent Arabidopsis mutants (Bouvier-Navé slasat1 mutant are lower than those in the single slpsat1 mutants et al., 2010), so that the necessity of a minimum SE threshold (Figures 3A, B) argues against this hypothesis. level for proper plant growth cannot be excluded. Interestingly, the inactivation of SlPSAT1 and SlASAT1 also The involvement of SE in tomato senescence regulation, as led to changes in the levels of FS, albeit less pronounced than previously reported in Arabidopsis (Bouvier-Navé et al., 2010), those of SE, which also differ depending on the organ and the would be mainly mediated by SlPSAT1, since an early mutant genotype. While in the leaves there was a moderate but senescence phenotype was observed in both slpsat1 mutant significant increase of FS contents in both mutants, the content of lines, but not in the slasat1 ones (Figures 6A and 7A). FS in seeds remained unchanged in the slpsat1 mutants and was Interestingly, in contrast to that observed in Arabidopsis, significantly reduced in the slasat1 mutants (Figure 3). These where the asatpsat mutants showed the same early leaf organ-dependent and highly contrasting metabolic responses to senescence phenotype than the single psat1 mutants (Bouvier- SlPSAT1 and SlASAT1 inactivation likely reflect the differences Navé et al., 2010), a synergistic effect of slpsat1 and slasat1 in the physiological role and the regulatory mechanisms mutations was observed in tomato, since the senescence controlling SE levels in these organs. Seeds contain about 5 phenotype of the tomato double mutant slpsat1 x slasat1 was times more SE than leaves (Figure 3 and Supplementary clearly stronger than that of slpsat1 (Figures 6A and 7A). This Tables 2 and 3), and their suggested primary role is to serve as observation is fully consistent with the expression profile of the a storage pool of sterols that can be easily mobilized to meet the senescence-marker gene SlSAG12 detected in leaves of different increasing demand of plasma membrane components during the ages from WT and mutant plants (Figure 6D). Moreover, the early stages of seedling growth. In the leaves, SE are also senescence-promoting effect of the slpsat1 mutation and the suggested to play a role in providing FS to support rapid synergistic effect of both mutations were also observed in an expansion (Zhou et al., 2019), in addition to their involvement induced senescence assay using detached leaves (Figure 7). In in maintaining sterol homeostasis in the cell membranes during aging tissues, sterols and fatty acids released from disorganized growth and development, including the recycling of membrane membranes would be converted to SE (Holmer et al., 1973; Chen sterols as senescence progresses or during stress (Bouvier-Navé et al., 2007). This explains the increase of SE content observed et al., 2010; Kumar et al., 2018). Lastly, it is also worth to mention during senescence in several plant species and organs, including that residual levels of SE were still detected in the slpsat1 x slasat1 leaves (Dyas and Goad, 1993; Mckegney et al., 1995; Bouvier- double knock mutant, suggesting that other acyltransferases may Navé et al., 2010; Li et al., 2016). Thus, the impaired capacity of contribute to SE formation in both seeds and leaves. In fact, as slpsat1 and slpsat1 x slasat1 mutants to esterify FS in leaves mentioned above, the tomato genome contains up to seven genes (Figure 3B) is likely to be one of the reasons underlying the early coding for ASAT-like proteins (Lara et al., 2019). senescence phenotype observed in the leaves of these mutants Despite a number of metabolic responses in tomato plants which, on the other hand, is not detected in the slasat1 mutants lacking functional SlPSAT1 or SlASAT1 are similar to those (Figures 6 and 7) that are able to produce and accumulate high Frontiers in Plant Science 15 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 levels of SE (Figure 3B). As mentioned previously, SE that SE are important to supply the necessary FS to sustain seed accumulate in cytosolic LDs, whose primary role in leaves germination and seedling growth during the early development seems to be the removal of potentially toxic FS and other lipid stages (Dyas and Goad, 1993; Harker et al., 2003). In this work catabolites scavenged from damaged membranes during stress we demonstrate that changes in tomato seeds SE contents, either or senescence (Huang, 2018). The reduced amounts of LDs a moderate increase or a strong reduction (Figure 3A), do not detected in leaves of tomato slpsat1, and slpsat1 x slasat1 mutant affect significantly the germination rate, as evidenced by the plants compared to WT (Figure 5) are in close correlation with similar percentage of WT and the different mutant seeds the temporal senescence differences observed between these germinated after 7 days (Figure 8). On the contrary, the genotypes and the WT. Thus, it is reasonable to assume that slpsat1 x slasat1 mutant seeds, which are almost completely the reduced abundance of LDs in these mutants reflects the devoid of SE, germinated significantly faster than the single impaired capacity to esterify sterols which, in turn, prevents slpsat1 and slasat1 mutant seeds, whose germination speed was proper maintenance of FS membrane homeostasis and slightly higher than that of WT seeds (Figure 8). The difference accelerates the onset of senescence. in germination speed between slpsat1 x slasat1 and the slpsat1 A number of studies support the notion that LD formation is seeds is intriguing given the small differences in their SE content strongly associated with increasing levels of SE (Maillot-Vernier (Figure 3A and Supplementary Table 2), although the possibility et al., 1991; Gondet et al., 1994; Bouvier-Navé et al., 2010; that SE levels in the slpsat1 x slasat1 seeds are below a given Shimada et al., 2019). However, such positive correlation threshold necessary for normal germination cannot be excluded. between SE overaccumulation and enhanced LD abundance In any case, this different germination behavior cannot be was not observed in the case of the slasat1 leaves, in which the attributed to differences in FS levels (Figure 3C and number of LD was lower than in WT leaves (Figure 5) despite Supplementary Table 2) and, moreover, reinforces the their much higher levels of ES (Figure 3B). In this respect, it is synergistic effect of both mutations previously observed in the interesting to note that a certain lack of correlation between LD early senescence phenotype (Figures 6–8). abundance and SE contents has also been described during the Sterol mobilization during seed germination has been diurnal cycle in tea leaves (Zhou et al., 2019). Either way, the reported in several plant species, including tomato (Duperon enhanced capacity of slasat1 leaves to form SE might explain et al., 1984; Garg and Nes, 1985; Huang and Grunwald, 1988; why these mutants do not show any symptom of premature Guo et al., 1995; Zhang et al., 2007). The early germination senescence, but at the same time raises the question about the phenotype of slpsat1 x slasat1 seeds could result from a better reason underlying the unexpected reduced accumulation of LDs. reserve mobilization rather than from changes in its content, It has been recently suggested that altered membrane sterol which, as mentioned above, is primarily related with the composition may compromise endoplasmic reticulum (ER)- germination rate. Moreover, germination is inversely based LD biogenesis in Arabidopsis seeds and leaves due to proportional to a weaker endosperm and seed coat the capacity of sterols to modulate cell membranes (Steinbrecher and Leubner-Metzger, 2017). It is thus tempting physicochemical properties and the organization of putative to hypothesize that early seed germination in tomato SE PSAT1-localized ER microdomains where ES would be formed biosynthetic mutants could be due to the potential effect of the and segregated into LDs (Yu et al., 2021). Thus, it might be altered FS and ES profiles in the seed coat formation. Sterols speculated that the altered profile of FS in leaves of the slasat1 might also affect directly the germination signaling pathway, mutants (Figure 3D and Supplementary Table 3) may hamper since during plant growth and development phytosterols can proper formation of LDs in their leaves through a similar interact with different hormone pathways (Vriese et al., 2021), mechanism. Also, the possibility that SE content/concentration including gibberellins (He et al., 2003), which have been in the LDs of slasat1 mutants is higher compared to that in LDs reported as essential factors for tomato seed germination of WT leaves to accommodate excess SE without increasing the (Groot and Karssen, 1987; Bassel et al., 2004). Work is number of LDs cannot be either excluded. currently in progress to shed light on the biochemical, The rate of seed germination has been correlated with a molecular and physiological mechanisms underlying the proper seed reserve content (Sun et al., 2015; Shrestha et al., tomato phenotypes associated to altered profiles of SE. 2016; Zhao et al., 2018). This physiological process involves the breakdown of storage reserves accumulated in the endosperm, like carbohydrates, lipids and proteins, into soluble molecules Data availability statement that are mobilized through the embryo to allow seedling development (Bewley and Black, 1994; Pritchard et al., 2002). The original contributions presented in the study are Previous studies indicate that changes in seed sterol levels affect included in the article/Supplementary Material. Further seed germination (Du et al., 2022) and it is generally believed inquiries can be directed to the corresponding authors. Frontiers in Plant Science 16 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 Author contributions Scholarship Council (China). We also thank Pilar Fontanet for her invaluable help in the generation of the tomato transgenic AB-M and JL-T performed most of the experimental work. lines characterized in this study, and the CRAG greenhouse staff NL performed the germination assay. NL and CD contributed to members for the maintenance of plants. the analysis of senescence parameters. TA and AF conceived the research, managed and supervised the project, and wrote the manuscript. All the authors analyzed data. All authors Conflict of interest contributed to the article and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could Funding be construed as a potential conflict of interest. This work was supported by grants AGL2017-88842-R and PID2021-126591OB-I100 funded by MCIN/AEI/10.13039/ Publisher’s note 501100011033, and by FEDER, a way of making Europe (Spain). This work was also supported by grants CEX2019-000902-S funded All claims expressed in this article are solely those of the by MCIN/AEI/10.13039/501100011033 (Spain) and by the CERCA authors and do not necessarily represent those of their affiliated Programme/Generalitat de Catalunya. organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed Acknowledgments or endorsed by the publisher. JL-T received a doctoral fellowship (Ref. PRE2018-084652) from the Ministerio de Ciencia e Innovació n (Spain). NL has Supplementary material financial support from the European Union’s Horizon 2020 research and innovation program under the Marie The Supplementary Material for this article can be found Skłodowska-Curie grant agreement No 945043. CD received a online at: https://www.frontiersin.org/articles/10.3389/ Ph.D. fellowship (CSC202109110067) from the China fpls.2022.984100/full#supplementary-material References Banas, A., Carlsson, A. S., Huang, B., Lenman, M., Banas, W., Lee, M., et al. nonbrassinosteroid products. Plant Physiol. 153, 741–756. doi: 10.1104/ (2005). Cellular sterol ester synthesis in plants is performed by an enzyme pp.109.152587 (phospholipid:sterol acyltransferase) different from the yeast and mammalian Chang, T. Y., Chang, C. C., and Cheng, D. (1997). Acyl-coenzyme a:cholesterol acyl-CoA:sterol acyltransferases. J. Biol. Chem. 280, 34626–34634. doi: 10.1074/ acyltransferase. Annu. Rev. B iochem. 66, 613–638. doi: 10.1146/ jbc.M504459200 annurev.biochem.66.1.613 Bassel, G. W., Zielinska, E., Mullen, R. T., and Bewley, J. D. (2004). Down- Chen, Q., Steinhauer, L., Hammerlindl, J., Keller, W., and Zou, J. (2007). regulation of DELLA genes is not essential for germination of tomato, soybean, and Biosynthesis of phytosterol esters: identification of a sterol O-acyltransferase in arabidopsis seeds. Plant Physiol. 136, 2782–2789. doi: 10.1104/pp.103.034876 arabidopsis. Plant Physiol. 145, 974–984. doi: 10.1104/pp.107.106278 Behmer, S. T., Olszewski, N., Sebastiani, J., Palka, S., Sparacino, G., Sciarrno, E., Chetelat, R., and Peacock, S. (2013). Guidelines for emasculating and pollinating et al. (2013). Plant phloem sterol content: forms, putative functions, and tomatoes (Univ. California, Davis, Tomato Genet. Resource Ctr). Available at: implications forphloem-feeding insects. Front. Plant Sci. 4. doi: 10.3389/ http://tgrc.ucdavis.edu/Guidelines_Emasculating_and_Pollinating_Tomatoes.pdf. fpls.2013.00370 Clouse, S. D. (2002). Arabidopsis mutants reveal multiple roles for sterols in Bewley, J. D., and Black, M. (1994). Seeds: Physiology of development and plant development. Plant Cell 14, 1995–2000. doi: 10.1105/tpc.140930 germination.(New York: Plenum Press) doi: 10.1007/978-1-4899-1002-8 Coolbear, P., Francis, A., and Grierson, D (1984). The effect of low temperature Bewley, J. D., and Black, M. (2013). Seeds: physiology of development and pre-sowing treatment on the germination performance and membrane integrity of germination (New York: Springer Science & Business Media). artificially aged tomato seeds. J. Exp. Bot 35, 1609–1617. doi: 10.1093/jxb/ Bouvier-Navé , P., and Benveniste, P. (1995). Sterol acyl transferase and steryl 35.11.1609 ester hydrolase activities in a tobacco mutant which overproduces sterols. Plant Sci. Doblas, V. G., Amorim-Silva, V., Posé , D., Rosado, A., Esteban, A., Arró , M., 110, 11–19. doi: 10.1016/0168-9452(95)04181-S et al. (2013). The SUD1 gene encodes a putative E3 ubiquitin ligase and is a positive Bouvier-Navé, P., Berna, A., Noiriel, A., Compagnon, V., Carlsson, A. S., Banas, A., et al. regulator of 3-hydroxy-3-methylglutaryl coenzyme a reductase activity in (2010). Involvement of the phospholipid sterol acyltransferase1 in plant sterol arabidopsis. Plant Cell 25, 728–743. doi: 10.1105/tpc.112.108696 homeostasis and leaf senescence. Plant Physiol. 152, 107–119. doi: 10.1104/pp.109.145672 Du, Y., Fu, X., Chu, Y., Wu, P., Liu, Y., Ma, L., et al. (2022). Biosynthesis and the Brinkman, E. K., Chen, T., Amendola, M., and Van Steensel, B. (2014). Easy roles of plant sterols in development and stress responses. Int. J. Mol. Sci. 23, 2332. quantitative assessment of genome editing by sequence trace decomposition. doi: 10.3390/ijms23042332 Nucleic Acids Res. 42, e168. doi: 10.1093/nar/gku936 Duperon, R., Thiersault, M., and Duperon, P. (1984). High level of glycosylated Carland, F., Fujioka, S., and Nelson, T. (2010). The sterol methyltransferases sterols in species of solanum and sterol changes during the development of the SMT1, SMT2, and SMT3 influence arabidopsis development through tomato. Phytochemistry 23, 743–746. doi: 10.1016/S0031-9422(00)85016-5 Frontiers in Plant Science 17 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 Dyas, L., and Goad, L. J. (1993). Steryl fatty acyl esters in plants. Phytochemistry Lara, J. A., Burciaga-Monge, A., Chá vez, A., Revé s, M., Lavilla, R., Arró , M., et al. 34, 17–29. doi: 10.1016/S0031-9422(00)90777-5 (2018). Identification and characterization of sterol acyltransferases responsible for Fernandez, A. I., Viron, N., Alhagdow, M., Karimi, M., Jones, M., Amsellem, Z., steryl ester biosynthesis in tomato. Front. Plant Sci. 9. doi: 10.3389/fpls.2018.00588 et al. (2009). Flexible tools for gene expression and silencing in tomato. Plant Leivar, P., Antolı́n-Llovera, M., Ferrero, S., Closa, M., Arro, M., Ferrer, A., et al. (2011). Physiol. 151, 1729–1740. doi: 10.1104/pp.109.147546 Multilevel control of arabidopsis 3-hydroxy-3-methylglutaryl coenzyme a reductase by Ferrer, A., Altabella, T., Arró , M., and Boronat, A. (2017). Emerging roles for protein phosphatase 2A. Plant Cell 23, 1494–1511. doi: 10.1105/tpc.107.052142 conjugated sterols in plants. Prog. Lipid Res. 67, 27–37. doi: 10.1016/ Lewis, T. A., Rodriguez, R. J., and Parks, L. W. (1987). Relationship between j.plipres.2017.06.002 intracellular sterol content and sterol esterification and hydrolysis in Saccharomyces Garcia, R. E., and Mudd, J. B. (1978). Fatty acid and sterol specificity in the cerevisiae. biochim. Biophys. Acta 921, 205–212. doi: 10.1016/0005-2760(87)90020-8 biosynthesis of steryl esters by enzyme preparations from spinach leaves. Arch. Li, L., Zhao, J., Zhao, Y., Lu, X., Zhou, Z., Zhao, C., and Xu, G. (2016). Biochem. Biophys. 190, 315–321. doi: 10.1016/0003-9861(78)90281-3 Comprehensive investigation of tobacco leaves during natural early senescence Garg, V. K., and Nes, W. R. (1985). Changes in D5- and D7-sterols during via multiplatform metabolomics analyses. Sci. Rep 6, 37976. doi: 10.1038/srep37976 germination and seedling development of Cucurbita maxima. Lipids 20, 876–883. Maillot-Vernier, P., Gondet, L., Schaller, H., and Belliard, G. (1991). Genetic doi: 10.1007/BF02534771 study and further biochemical characterization of a tobacco mutant that Gas-Pascual, E., Simonovik, B., Schaller, H., and Bach, T. J. (2015). Inhibition of overproduces sterols. Molec. Gen. Genet. 231, 33–40. doi: 10.1007/BF00293818 cycloartenol synthase (CAS) function in tobacco BY-2 cells. Lipids 50, 761–772. Manzano, D., Andrade, P., Caudepó n, D., Altabella, T., Arró , M., and Ferrer, A. doi: 10.1007/s11745-015-4036-6 (2016). Suppressing farnesyl diphosphate synthase alters chloroplast development Gocze, P. M., and Freeman, D. A. (1994). Factors underlying the variability of and triggers sterol-dependent induction of jasmonate-and fe-related responses. lipid droplet fluorescence in MA-10 leydig tumor cells. Cytometry J. Int. Soc Anal. Plant Physiol. 172, 93–117. doi: 10.1104/pp.16.00431 Cytol. 17, 151–158. doi: 10.1002/cyto.990170207 Mckegney, G., Yao, K., Ghosh, S., Huff, A., Mayak, S., and Thompson, J. E. Gondet, L., Bronner, R., and Benveniste, P. (1994). Regulation of sterol content (1995). The lipid composition of cytosolic particles isolated from senescing bean in membranes by subcellular compartmentation of steryl-esters accumulating in a cotyledons. Phytochemistry 39, 1335–1345. doi: 10.1016/0031-9422(95)00111-J sterol-overproducing tobacco mutant. Plant Physiol. 105, 509–518. doi: 10.1104/ Nieto, B., Foré s, O., Arró , M., and Ferrer, A. (2009). Arabidopsis 3-hydroxy-3- pp.105.2.509 methylglutaryl-CoA reductase is regulated at the post-translational level in Groot, S. P. C., and Karssen, C. M. (1987). Gibberellins regulate seed response to alterations of the sphingolipid and the sterol biosynthetic pathways. germination in tomato by endosperm weakening: a study with gibberellin- Phytochemistry 70, 53–59. doi: 10.1016/j.phytochem.2008.10.010 deficient mutants. Planta 171, 525–531. doi: 10.1007/BF00392302 Noh, Y. S., and Amasino, R. M. (1999). Identification of a promoter region Guo, D. A., Venkatramesh, M., and Nes, W. D. (1995). Developmental responsible for the senescence-specific expression of SAG12.. Plant Mol Biol 41, regulation of sterol biosynthesis in Zea mays. Lipids 30, 203–219. doi: 10.1007/ 181–194. doi: 10.1023/A:1006342412688 BF02537823 Noiriel, A., Benveniste, P., Banas, A., Stymne, S., and Bouvier-Navé , P. (2004). Harker, M., Holmberg, N., Clayton, J. C., Gibbard, C. L., Wallace, A. D., Rawlins, Expression in yeast of a novel phospholipase A1 cDNA from Arabidopsis thaliana. S., et al. (2003). Enhancement of seed phytosterol levels by expression of an n- FEBS J. 271, 3752–3764. doi: 10.1111/j.1432-1033.2004.04317.x terminal truncated Hevea brasiliensis (rubber tree) 3-hydroxy-3-methylglutaryl- Ohyama, K., Suzuki, M., Kikuchi, J., Saito, K., and Muranaka, T. (2009). Dual CoA reductase. Plant Biotechnol. J. 1, 113–121. doi: 10.1046/j.1467- biosynthetic pathways to phytosterol via cycloartenol and lanosterol in arabidopsis. 7652.2003.00011.x Proc. Natl. Acad. Sci. U. S. A. 106, 725–730. doi: 10.1073/pnas.0807675106 Hartmann, M. A. (2004). “Sterol metabolism and function in higher plants,” in Oliveros, J. C., Franch, M., Tabas-Madrid, D., San-Leó n, D., Montoliu, L., Cubas, Lipid metabolism and membrane biogenesis. Ed. G. Daum (Heidelberg: Springer), P., et al. (2016). Breaking-cas-interactive design of guide RNAs for CRISPR-cas 183–211. experiments for ENSEMBL genomes. Nucleic Acids Res. 44 (W1), W267–W271. He, J. X., Fujioka, S., Li, T. C., Kang, S. G., Seto, H., Takatsuto, S., et al. (2003). doi: 10.1093/nar/gkw407 Sterols regulate development and gene expression in arabidopsis. Plant Physiol. Parker, M. L., and Murphy, G. J. P. (1981). Oleosomes in flag leaves of wheat: 131, 1258–1269. doi: 10.1104/pp.014605 their distribution, composition and fate during senescence and rust-infection. Herná ndez-Pinzó n, I., Ross, J. H. E., Barnes, K. A., Damant, A. P., and Murphy, Planta 152, 36–43. doi: 10.1007/BF00384982 D. J. (1999). Composition and role of tapetal lipid bodies in the biogenesis of the Peng, L., Kawagoe, Y., Hogan, P., and Delmer, D. (2002). Sitosterol beta- pollen coat of Brassica napus. Planta 208, 588–598. doi: 10.1007/s004250050597 glucoside as primer for cellulose synthesis in plants. Science 295, 147–150. Hofmann, K. (2000). A superfamily of membrane-bound O-acyltransferases doi: 10.1126/science.1064281 with implications for wnt signaling. Trends Biochem. Sci. 25, 111–112. doi: 10.1016/ Perez, F. (2004). Serial cloner v. 2.6. 0. Molecular biology free sofware. (Software S0968-0004(99)01539-X SerialBasics). Holmer, G., Ory, R. L., and Høy, C. E. (1973). Changes in lipid composition of Peters, J. (2000) Tetrazolium testing handbook: Contribution no. 29 to the germinating barley embryo. Lipids 8, 277–283. doi: 10.1007/BF02531905 handbook on seed testing. Available at: www.aosaseed.com. Huang, A. H. C. (2018). Plant lipid droplets and their associated proteins: potential Porra, R. J. (2002). The chequered history of the development and use of for rapid advances. Plant Physiol. 176, 1894–1918. doi: 10.1104/pp.17.01677 simultaneous equations for the accurate determination of chlorophylls a and b. Huang, L. S., and Grunwald, C. (1988). Sterol and phospholipid changes during Photosynth. Res. 73, 149–156. doi: 10.1023/A:1020470224740 alfalfa seed germination. Phytochemistry 27, 2049–2053. doi: 10.1016/0031-9422 Posé , D., Castanedo, I., Borsani, O., Nieto, B., Rosado, A., Taconnat, L., et al. (88)80095-5 (2009). Identification of the arabidopsis dry2/sqe1-5 mutant reveals a central role Huang, L. H., Nishi, K., Li, S., Ho, T., Dong, R., Chang, C. C. Y., et al. (2014). for sterols in drought tolerance and regulation of reactive oxygen species. Plant J. Acyl-coenzyme a: cholesterol acyltransferase 1 - significance of single nucleotide 59, 63–76. doi: 10.1111/j.1365-313X.2009.03849.x polymorphism at residue 526 and the role of Pro347 near the fifth transmembrane Pritchard, S. L., Charlton, W. L., Baker, A., and Graham, I. A. (2002). domain. FEBS J. 281, 1773–1783. doi: 10.1111/febs.12739 Germination and storage reserve mobilization are regulated independently in Jang, J. C., Fujioka, S., Tasaka, M., Seto, H., Takatsuto, S., Ishii, A., et al. (2000). A arabidopsis. Plant J. 31, 639–647. doi: 10.1046/j.1365-313x.2002.01376.x critical role of sterols in embryonic patterning and meristem programming Ramirez-Estrada, K., Castillo, N., Lara, J. A., Arró , M., Boronat, A., Ferrer, A., revealed by the fackel mutants of Arabidopsis thaliana. Genes Dev. 14, 1485– et al. (2017). Tomato UDP-glucose sterol glycosyltransferases: A family of 1497. doi: 10.1101/gad.14.12.1485 developmental and stress regulated genes that encode cytosolic and membrane- Kalinowska, M., Zimowski, J., Bucholc, M., and Wojciechowski, Z. A. (1989). associated forms of the enzyme. Front. Plant Sci. 8. doi: 10.3389/fpls.2017.00984 Reversible enzymatic transesterification between wax esters ans sterols in sinapis Richards, E., Reichardt, M., and Rogers, S. (1994). Preparation of genomic dna alba roots. Phytochemistry 28, 2931–2935. doi: 10.1016/0031-9422 from plant tissue. Curr. Protoc. Mol. Biol. 27, 2.3.1–2.3.7. doi: 10.1002/ Kemp, R. J., and Mercer, E. I. (1968). The sterol esters of maize seedlings. 0471142727.mb0203s27 Biochem. J. 110, 111–118. doi: 10.1042/bj1100111 Robertlee, J., Kobayashi, K., Suzuki, M., and Muranaka, T. (2017). AKIN 10, a Korber, M., Klein, I., and Daum, G. (2017). Steryl ester synthesis, storage and representative arabidopsis SNF 1-related protein kinase 1 (Sn RK 1), hydrolysis: A contribution to sterol homeostasis. Biochim. Biophys. Acta - Mol. Cell phosphorylates and downregulates plant HMG-CoA reductase. FEBS Lett. 59, Biol. Lipids 1862, 1534–1545. doi: 10.1016/j.bbalip.2017.09.002 1159–1166. doi: 10.1002/1873-3468.12618 Kumar, M. S. S., Mawlong, I., Ali, K., and Tyagi, A. (2018). Regulation of Schaller, H. (2004). New aspects of sterol biosynthesis in growth and phytosterol biosynthetic pathway during drought stress in rice. Plant Physiol. development of higher plants. Plant Physiol. Biochem. 42, 465–476. doi: 10.1016/ Biochem. 129, 11–20. doi: 10.1016/j.plaphy.2018.05.019 j.plaphy.2004.05.012 Frontiers in Plant Science 18 frontiersin.org Burciaga-Monge et al. 10.3389/fpls.2022.984100 Schaller, H., Crausem, B., Benveniste, P., Chye, M., Tan, C., Song, Y., et al. Wentzinger, L. F., Bach, T. J., and Hartmann, M. A. (2002). Inhibition of (1995). Expression of the Hevea brasiliensis (H. b. k.) müll. arg. tobacco results in squalene synthase and squalene epoxidase in tobacco cells triggers an up-regulation sterol overproduction. Plant Physiol. 109, 761–770. doi: 10.1104/pp.109.3.761 of 3-hydroxy-3-methylglutaryl coenzyme a reductase. Plant Physiol. 130, 334–346. Schiml, S., Fauser, F., and Puchta, H. (2016). “CRISPR/Cas-mediated site- doi: 10.1104/pp.004655 specific mutagenesis in arabidopsis thaliana using Cas9 nucleases and paired Wilkinson, S. C., Powls, R., and Goad, L. J. (1994). The effects of excess nickases,” in Chromosome and genomic engineering in plants. Ed. M. Murata exogenous mevalonic acid on sterol and steryl ester biosynthesis in celery (New York, NY: Humana Press), 111–122. doi: 10.1007/978-1-4939-4931-1_8 (Apium graveolens) cell suspension cultures. Phytochemistry 37, 1031–1035. Shimada, T. L., Shimada, T., Okazaki, Y., Higashi, Y., Saito, K., Kuwata, K., et al. doi: 10.1016/S0031-9422(00)89523-0 (2019). HIGH STEROL ESTER 1 is a key factor in plant sterol homeostasis. Nat. Wollam, J., and Antebi, A. (2011). Sterol regulation of metabolism, homeostasis, Plants 5, 1154–1166. doi: 10.1038/s41477-019-0537-2 and development. Annu. Rev. Biochem. 80, 885–916. doi: 10.1146/annurev- Shimada, T. L., Takano, Y., and Hara-Nishimura, I. (2015). Oil body-mediated biochem-081308-165917 defense against fungi: From tissues to ecology. Plant Signal. Behav. 10, e989036. Wu, S. S. H., Moreau, R. A., Whitaker, B. D., and Huang, A. H. C. (1999). doi: 10.4161/15592324.2014.989036 Steryl esters in the elaioplasts of the tapetum in developing brassica anthers and Shimada, T. L., Takano, Y., Shimada, T., Fujiwara, M., Fukao, Y., Mori, M., et al. their recovery on the pollen surface. Lipids 34, 517–523. doi: 10.1007/s11745- (2014). Leaf oil body functions as a subcellular factory for the production of a 999-0393-5 phytoalexin in arabidopsis. Plant Physiol. 164, 105–118. doi: 10.1104/pp.113.230185 Yates, P. J., Haughan, P. A., Lenton, J. R., and Goad, L. J. (1991). Effects of Shimada, T. L., Yamaguchi, K., Shigenobu, S., Takahashi, H., Murase, M., terbinafine on growth, squalene, and steryl ester content of a celery cell Fukuyoshi, S., et al. (2020). Excess sterols disrupt plant cellular activity by suspension culture. Pestic Biochem. Physiol. 40, 221–226. doi: 10.1016/0048- inducing stress-responsive gene expression. J. Plant Res. 13, 383–392. 3575(91)90093-2 doi: 10.1007/s10265-020-01181-4 Yu, H., Sun, C., Gong, Q., and Feng, D. (2021). Mitochondria-associated Shrestha, P., Callahan, D. L., Singh, S. P., Petrie, J. R., and Zhou, X. R. (2016). endoplasmic reticulum membranes in breast cancer. Front. Cell Dev. Biol. 9. Reduced triacylglycerol mobilization during seed germination and early seedling doi: 10.3389/fcell.2021.629669 growth in arabidopsis containing nutritionally important polyunsaturated fatty Zhang, X., Lin, K., and Li, Y. (2020). Highlights to phytosterols accumulation acids. Front. Plant Sci. 7. doi: 10.3389/fpls.2016.01402 and equilibrium in plants: Biosynthetic pathway and feedback regulation. Plant Steinbrecher, T., and Leubner-Metzger, G. (2017). The biomechanics of seed Physiol. Biochem. 155, 637–649. doi: 10.1016/j.plaphy.2020.08.021 germination, J. Exp. Bot. 68, 765–783. doi: 10.1093/jxb/erw428 Zhang, H., Sun, Y., Xie, X., Kim, M. S., Dowd, S. E., and Paré , P. W. (2009). A soil Sturley, S. L. (1997). Molecular aspects of intracellular sterol esterification: the bacterium regulates plant acquisition of iron via deficiency-inducible mechanisms. acyl coenzyme a:cholesterol acyltransferase reaction. Curr. Opin. Lipidol. 8, 167– Plant J. 58, 568–577. doi: 10.1111/j.1365-313X.2009.03803.x 173. doi: 10.1097/00041433-199706000-00007 Zhang, H., Vasanthan, T., and Wettasinghe, M. (2007). Enrichment of Sun, J., Wu, D., Xu, J., Rasmussen, S. K., and Shu, X. (2015). Characterization of tocopherols and phytosterols in canola oil during seed germination. J. Agric. starch during germination and seedling development of a rice mutant with a high Food Chem. 55, 355–359. doi: 10.1021/jf060940o content of resistant starch. J. Cereal Sci. 62, 94–101. doi: 10.1016/j.jcs.2015.01.002 Zhao, M., Zhang, H., Yan, H., Qiu, L., and Baskin, C. C. (2018). Mobilization and Tatenaka, Y., Kato, H., Ishiyama, M., Sasamoto, K., Shiga, M., Nishitoh, H., et al. role of starch, protein, and fat reserves during seed germination of six wild (2019). Monitoring lipid droplet dynamics in living cells by using fluorescent grassland species. Front. Plant Sci. 9. doi: 10.3389/fpls.2018.00234 probes. Biochem. 56, 499–503 doi: 10.1021/acs.biochem.8b01071 Zhou, X., Chen, X., Du, Z., Zhang, Y., Zhang, W., Kong, X., et al. (2019). Tzen, J. T. C., Cao, Y. Z., Laurent, P., Ratnayake, C., and Huang, A. H. C. (1993). Terpenoid esters are the major constituents from leaf lipid droplets of camellia Lipids, proteins, and structure of seed oil bodies from diverse species. Plant Physiol. sinensis. Front. Plant Sci. 10. doi: 10.3389/fpls.2019.00179 101, 267–276. doi: 10.1104/pp.101.1.267 Zimowski, J., and Wojciechowski, Z. A. (1981a). Acyl donors for Villette, C., Berna, A., Compagnon, V., and Schaller, H. (2015). Plant sterol diversity sterolesterification by cell-free preparations from sinapis alba roots. in pollen from angiosperms. Lipids 50, 749–760. doi: 10.1007/s11745-015-4008-x Phytochemistry 20, 1795–1798. doi: 10.1016/0031-9422(81)84006-X Vriese, K. D., Pollier, J., Goossens, A., Beeckman, T., and Vanneste, S. (2021). Zimowski, J., and Wojciechowski, Z. A. (1981b). Partial purification and The use of mutants and inhibitors to study sterol biosynthesis in plants.. J. Exp. Bot. specificity oftriacylglycerol:sterol acyltransferase from sinapis alba. 72, 241–253. doi: 10.1093/jxb/eraa429 Phytochemistry 20, 1799–1803. doi: 10.1016/0031-9422(81)84007-1 Frontiers in Plant Science 19 frontiersin.org