ORIGINAL RESEARCH published: 07 June 2019 doi: 10.3389/fpls.2019.00731 Rice Seeds as Biofactories of Rationally Designed and Cell-Penetrating Antifungal PAF Peptides Mireia Bundó1†, Xiaoqing Shi1†, Mar Vernet1, Jose F. Marcos 2, Belén López-García1 and María Coca1* 1 Centre for Research in Agricultural Genomics (CRAG, CSIC-IRTA-UAB-UB), Barcelona, Spain, 2 Institute of Agrochemistry and Food Technology (IATA, CSIC), Paterna, Spain Edited by: Suvi Tuulikki Häkkinen, PAFs are short cationic and tryptophan-rich synthetic peptides with cell-penetrating VTT Technical Research Centre of antifungal activity. They show potent and selective killing activity against major fungal Finland Ltd, Finland pathogens and low toxicity to other eukaryotic and bacterial cells. These properties make Reviewed by: Eva Stoger, them a promising alternative to fulfill the need of novel antifungals with potential applications University of Natural Resources and in crop protection, food preservation, and medical therapies. However, the difficulties of Life Sciences Vienna, Austria István Pócsi, cost-effective manufacturing of PAFs by chemical synthesis or biotechnological production University of Debrecen, Hungary in microorganisms have hampered their development for practical use. This work explores *Correspondence: the feasibility of using rice seeds as an economical and safe production system of PAFs. María Coca The rationally designed PAF102 peptide with improved antifungal properties was selected maria.coca@cragenomica.es for assessing PAF biotechnological production. Two different strategies are evaluated: (1) † These authors have contributed equally to this work the production as a single peptide targeted to protein bodies and (2) the production as an oleosin fusion protein targeted to oil bodies. Both strategies are designed to offer Specialty section: stability to the PAF peptide in the host plant and to facilitate its downstream purification. This article was submitted to Plant Metabolism and Our results demonstrate that PAF does not accumulate to detectable levels in rice seeds Chemodiversity, when produced as a single peptide, whereas it is successfully produced as fusion protein a section of the journal Frontiers in Plant Science to the Oleosin18, up to 20 μg of peptide per gram of grain. We show that the expression Received: 03 December 2018 of the chimeric Ole18-PAF102 gene driven by the Ole18 promoter results in the specific Accepted: 16 May 2019 accumulation of the fusion protein in the embryo and aleurone layer of the rice seed. Published: 07 June 2019 Ole18-PAF102 accumulation has no deleterious effects on seed yield, germination Citation: capacity, or seedling growth. We also show that the Oleosin18 protein serves as carrier Bundó M, Shi X, Vernet M, Marcos JF, López-García B and to target the fusion protein to oil bodies facilitating PAF102 recovery. Importantly, the Coca M (2019) Rice Seeds as recovered PAF102 is active against the fungal phytopathogen Fusarium proliferatum. Biofactories of Rationally Designed and Cell-Penetrating Antifungal Altogether, our results prove that the oleosin fusion technology allows the production of PAF Peptides. PAF bioactive peptides to assist the exploitation of these antifungal compounds. Front. Plant Sci. 10:731. doi: 10.3389/fpls.2019.00731 Keywords: PAF, antifungal, pathogens, fungi, rice, seed, oil bodies, protein bodies Frontiers in Plant Science | www.frontiersin.org 1 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production INTRODUCTION as they are economical and easy to grow, as well as of safety because of the low risk of contamination with human and Infections caused by fungi pose a serious threat to human animal pathogens (Twyman et al., 2003). Particularly, rice seeds and animal health and to food security and safety (Fisher have been reported as efficient bioreactors of AMPs, including et al., 2012). Invasive fungal diseases have significantly increased natural or rational designed peptides (Bundó et al., 2014; in recent decades and are important causes of mortality, Montesinos et al., 2016, 2017). AMP production is favored by particularly in immunocompromised patients, killing about limiting their accumulation to seeds that avoids the negative one and a half million people every year. This value exceeds impacts on plant performance reported in some cases when the death rate for malaria or breast cancer (Brown et al., accumulated in vegetative tissues (Coca et al., 2006; Nadal 2012). Plant disease epidemics caused by fungi and fungal- et al., 2012; Company et al., 2014). Several seed-specific promoters like oomycetes are an old problem that have been further are now available to drive strong expression of AMP genes exacerbated by intensive agricultural practices, globalization, either in the rice endosperm or embryo (Qu and Takaiwa, and climate change (Bebber and Gurr, 2015). Today, crop- 2004). Among endosperm-specific promoters are the ones from destroying fungi account globally for yield losses of ~20%, the seed storage proteins glutelins and globulins, including the with further 10% postharvest losses (Fisher et al., 2018). In GluB1, GluB4, and Glb1 promoters; and among embryo-specific addition, food safety is challenged by mycotoxigenic fungi promoters are the ones from the oleosin proteins, such as the that contaminate food and feed with detrimental toxins for Ole18 promoter. Another factor that favors the production of human health. The limited number of licensed antifungals AMPs in seeds is their confinement into storage organelles, currently available, together with the unprecedented rise of such as protein bodies (PBs) or oil bodies (OBs) (Bundó et al., multidrug-resistant pathogenic fungi, makes crucial the 2014; Montesinos et al., 2016, 2017). Storage organelles offer development of novel antifungal compounds to combat fungal a stable environment for packing a large amount of AMPs, infections (Perfect, 2016; Fisher et al., 2018). Antimicrobial together with host cell protection from AMP exposure. Proteins peptides (AMPs) are being actively explored to fulfill the can be targeted to PBs through signal peptides linked at their need of novel antifungals with potential applications in crop N-terminus and/or KDEL sequence at their C-terminus, together protection, food preservation, and medical therapies. through intrinsic physicochemical properties in certain storage AMPs are peptides and small proteins produced by most proteins (Khan et al., 2012; Takaiwa et al., 2017). OB targeting living organisms that exhibit lytic or inhibitory activity is achieved using oleosin proteins as carriers (van Rooijen against microorganisms (Zasloff, 2002; Zhang and Gallo, and Moloney, 1995; Montesinos et al., 2016). Oleosins are the 2016). However, most AMPs are obtained at low yields from most abundant structural proteins of plant seed OBs, whose natural sources, and some of them show properties, such lipophilic character and secondary structure determine their as low stability or low specificity, that might compromise association to OBs (Abell et al., 1997). Both PBs and OBs their applications. Their peptidic nature enables the rational have served to stabilize AMPs in rice seeds and to reach high design of novel molecules with improved properties to yields (Bundó et al., 2014; Montesinos et al., 2016). be produced at high yields through biotechnological systems. This work explores the feasibility of using rice seeds as a PAF102 was designed as a novel antifungal peptide with platform for the production of cell-penetrating antifungal PAF improved properties and optimized to be produced in peptides, exemplified as PAF102. Two different strategies are biofactories (López-García et al., 2015). PAF102 is a modified evaluated: (1) the production as a single peptide targeted to concatemer of the hexapeptide PAF26 (RKKWFW), which PBs or (2) the production as an oleosin fusion protein targeted was identified through a combinatorial approach as a Peptide to OBs. Here, we report that PAF103, a His-tagged and KDEL- with specific AntiFungal activity (PAF) (López-García et al., extended PAF102, was not accumulated to detectable levels 2002, 2015; Muñoz et al., 2013). PAF102 shows potent in rice seeds, whereas PAF102 was successfully produced as antifungal activity against economically relevant fusion to the Ole18 protein in rice seeds. We demonstrate phytopathogens and very low toxicity to other eukaryotic that the Ole18-PAF102 fusion protein was accumulated in cells, including human erythrocytes. The antifungal mechanism OBs without affecting seed yield or germination capacity. of PAF102 is similar to that of the parental PAF26, and it We also show that biologically active PAF102 can be recovered involves the interaction with the fungal cell envelope, followed from rice OBs. Our results demonstrate that the oleosin fusion by cell penetration and intracellular effects that cause cell technology is a good strategy for the production of PAF death (Muñoz et al., 2013; López-García et al., 2015). This antifungal peptides. mode-of-action is different to the one of licensed antifungal drugs; thus, PAF peptides might be an alternative to combat fungal pathogens. However, the difficulties of cost-effective MATERIALS AND METHODS manufacturing of PAFs by chemical synthesis, or by conventional microbial-based production systems associated Preparation of Plant Expression Vectors to host toxicity, have hampered their development for Four different constructs were prepared for the expression of practical use. the synthetic PAF genes in rice seeds (Figure 1A). Three of Plants provide a platform for the production of AMPs that them were designed for the production of a PAF as an individual offer advantages in terms of cost-effectiveness and scalability peptide, and the last one as a fusion to the rice Oleosin 18 kDa Frontiers in Plant Science | www.frontiersin.org 2 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production A B FIGURE 1 | Gene constructs for PAF production in rice seeds. (A) Schematic representation of the constructs in which the expression of the synthetic PAF103 gene was controlled by the 2.3 kb GluB1, or the 1.4 kb GluB4, or 0.9 kb Glb1 promoters, and the expression of the chimeric Ole18-PAF102 fusion gene by the Ole18 promoter. The corresponding signal peptides (SP) were included in the PAF103 constructs. The Nos terminator (Nos-t) was present in all the constructs. PRS corresponds to the TEV protease recognition sequence that links the Ole18 protein to PAF102. Relevant restriction enzyme sites for cloning purposes are indicated. (B) Amino acid sequences of PAF103 peptide and Ole18-PAF102 fusion protein. Gray sequences correspond to SPs of the corresponding storage proteins, which are post-translationally cleaved; black sequences to the mature PAF103 or PAF102 peptides; red sequence to the 18 kDa oleosin (Ole18) protein; and blue sequence to PRS. Underlined sequences correspond to the differential residues between PAF peptides. protein (Ole18). The individual peptide was His-tagged in The fusion protein corresponds to the Ole18 protein linked N-terminal and KDEL-extended in C-terminal resulting in a to the PAF102 peptide through a Tobacco Etch Virus NIa new PAF that was named PAF103 (Figure 1B). The PAF103 protease recognition site (TEV protease) (Figure 1B), without gene was synthesized by GenScript based on the codon usage His-tag or KDEL extension but with an extra glycine residue bias in Oryza sativa and flanked in both ends by BamHI at N-terminus that remains after TEV protease digestion. The restriction sites (Supplementary Figure S1). To drive the construct was prepared by PCR amplification of the Ole18 expression of PAF103, three different endosperm-specific promoter and the Ole18 protein coding sequence (pOle18:Ole18) promoters were used, namely Glutelin B1 (GluB1), Glutelin B4 from the pC::pOle18:Ole18-CecA:tNos vector (Montesinos et al., (GluB4), and the 26 KDa Globulin (Glb1) (Qu and Takaiwa, 2016) using the primers in Supplementary Table S1, which 2004). Additionally, the sequence encoding the signal peptide introduce an EcoRI and a SalI site at 5′ and 3′ ends of the of the corresponding seed storage proteins was fused to the fragment, respectively. The PAF102 coding sequence extended N-terminus of the PAF103 gene for internalization into in frame at N-terminus with the TEV protease recognition the endoplasmic reticulum (ER) system (Figure 1B). The site (PRS), and flanked by SalI and PstI restriction sites, was vectors containing the pGluB1:PAF103:tNos and pGluB4: synthesized by GenScript (Supplementary Figure S1). The PAF103:tNos constructs were prepared replacing the BamHI- Nopaline Synthase (Nos)-terminator sequence was introduced BamHI CecA-KDEL DNA fragment by the BamHI-BamHI into the cloning vector containing the PRS-PAF102 PAF103 DNA fragment into the previously described (pUC57::PRS-PAF102) as a PstI-HindIII fragment, which was pC::pGluB1:CecAKDEL:tNos and pC::pGluB4:CecAKDEL:tNos amplified by PCR using the primers that add these restriction vectors (Bundó et al., 2014). The pC::pGlb1:PAF103:tNos vector sites (Supplementary Table S1). Into this plasmid, the was prepared by replacing the NarI-NarI BP178 fragment by pOle18:Ole18 fragment was also introduced as an EcoRI-SalI a NarI-NarI PAF103 fragment into the previously described restriction fragment. Finally, the whole construct was mobilized pC::pGlb1:BP178:tNos vector (Montesinos et al., 2017). The to the binary vector pCAMBIA1300 as an EcoRI-HindIII fragment NarI-NarI PAF103 fragment was obtained by PCR amplification to generate the pC::pOle18:Ole18-PAF102:tNos vector for rice from the GenScript clone using the oligonucleotides NarIPAF103_ transformation (Figure 1A). All the constructs were then verified fwd and PAF103NarI_rev in Supplementary Table S1. by nucleotide sequencing. An additional vector for the expression of the chimeric gene encoding an Ole18-PAF102 fusion protein was prepared Production of Transgenic Plants (Figure 1A). In this case, the gene expression was driven by Transgenic rice plants (Oryza sativa cv. Ariete) were produced the embryo-specific Ole18 own promoter (Montesinos et al., 2016). by Agrobacterium-mediated transformation of embryogenic calli Frontiers in Plant Science | www.frontiersin.org 3 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production as previously described (Sallaud et al., 2003). Transgene insertion Lab™ Software (Version 5.2.1) included in the ChemiDoc™ was confirmed in the regenerated plants by PCR analysis using Touch Imaging System (Bio-Rad, USA). leaf genomic DNA as template. The positive plants were selected to obtain homozygous lines in the T2 generation. The homozygous lines were identified by segregation of hygromycin resistance RT-PCR Analysis of Transgene Expression Transgene expression was determined by RT-PCR analysis of afforded by the htpII marker gene in the T-DNA region of total RNA isolated from a pool of 10 immature seeds (before pCAMBIA1300-derived vectors. The transgene copy number seed desiccation, around 20–25 days after flowering). Total was estimated by quantitative PCR (qPCR) using the Sucrose RNA was extracted using the method previously described Phosphate Synthase (SPS) reference gene as previously described (Chang et al., 1993). DNase-treated RNA (1 μg) was (Yang et al., 2005; Bundó et al., 2014). Rice plants transformed retrotranscribed using the High Capacity cDNA Reverse with the empty vector (pCAMBIA 1300) were also produced Transcription kit (Applied Biosystems) using the oligo(dT) as a control for this study. All rice plants were grown at primer. PCRs were carried out using specific primers 28 ± 2°C with a 14/10 h light/dark photoperiod. (Supplementary Table S1) that annealed to GluB1, GluB4, and Glb1 SP encoding sequences (forward primers) and to Protein Extraction and the PAF103 gene sequence (reverse primers). Amplified Immunoblot Analysis transcripts encoded by the three different transgenes were Protein extracts were prepared from dehulled mature seeds compared to the OsEF1a housekeeping transcripts (Os03g08060). (10 seeds, ~200 mg) imbibed in water for 1 h. Seeds were ground and homogenized in a sucrose-containing buffer In situ Immunodetection of PAFs in (10 mM phosphate buffer pH 7.6, 0.6 M sucrose). After Whole Seeds filtration with miracloth, homogenates were centrifuged at PAF103 and PAF102 accumulation in the transgenic rice seeds low speed (200g) to remove cellular debris and starch. Clarified was analyzed by in situ immunodetection using the anti-His homogenates were then centrifuged at high speed (2,000g) tag (dil 1:1,000) and anti-PAF102 antibodies (dil 1:500), to obtain PB-enriched fractions, as the precipitated dense respectively, and the fluorescent labeled AlexaFluor448 anti- fractions (Bundó et al., 2014), or the OB-enriched fractions, mouse secondary antibody (Molecular Probes, 1:5,000 dilution) as the floating fractions (Montesinos et al., 2016). PB-enriched as previously described (Bundó et al., 2014). fractions were resuspended directly in SDS-loading buffer, separated on tricine-SDS-PAGE (16.5%), transferred to a nitrocellulose membrane (Amersham Protran 0.2 μm), and Fungal Infection Assays on Rice Seeds Transgenic rice seeds were evaluated for resistance to the seed immunodetected using commercial monoclonal antibodies fungal pathogen Fusarium proliferatum as previously described anti-His tag (A00186 GeneScript). (Bundó et al., 2014). Briefly, 12 surface-sterilized seeds per Immunoblot analysis of OB-associated proteins was done line and treatment were placed on MS medium without sucrose after solubilization in SDS-loading buffer, separation in and then inoculated with 50 μl of sterile water (control) or SDS-PAGE, transfer to nitrocellulose membranes (Amersham of F. proliferatum spore suspension (103 spores/ml). Seeds were Protran 0.4 μm), and immunodetection with antibodies allowed to germinate for 7 days to determine the percentage against the PAF102 (1:1,000 dilution, this work) and the of germination under control or infection conditions. Three rice Ole18 [1:2,000 dilution, (Montesinos et al., 2016)]. Mouse independent assays were performed. polyclonal antibodies against synthetic PAF102 (GeneScript) were produced at the Laboratory Animal Facilities (registration number B9900083) of the Center for Research and OBs Isolation and PAF102 Purification Development (CID) from the Spanish National Research OBs were isolated from the OB-enriched fractions by two Council (CSIC), in strict accordance with the bioethical consecutive cycles of flotation-centrifugation on a sucrose principles established by the Spanish legislation following containing buffer. The integrity of the isolated OBs was tested international guidelines. The protocol was approved by the by selective staining with Nile red (1 ng/ml, Sigma) and confocal Committee on Bioethics of Animal Experimentation from fluorescent visualization. PAF102 was recovered from the OBs CID and by the Department of Agriculture, Livestock, containing the Ole18-PAF102 protein by digestion with TEV Fisheries, Food and Environment of the Government of protease (Invitrogen, 1:100). Proteolytic digestion was conducted Catalonia (permit number DAAM:7461). All efforts were overnight at 30°C in the TEV protease buffer supplemented made to minimize suffering of the animals. Four injections with 0.25 M sucrose. TEV protease efficiency was estimated of synthetic PAF102 (0.5 mg each) in a three-weekly basis based on the disappearance of the Ole18-PAF102 signal in were applied to mice, which were bleed 1 week after the immunoblot analysis by quantification of signal intensities. last injection to obtain the PAF102 antiserum. The amount of PAF102 accumulation per seed was estimated Antifungal Assays on immunoblot by quantification of signal intensities of Growth inhibition assays of F. proliferatum were performed in Ole18-PAF102 to known amounts of synthetic PAF102. Signal 96-well flat-bottom microtiter plates, as previously described intensities were quantified using the Quantity Tools Image (López-García et al., 2015). Basically, 70 μl of fungal conidia Frontiers in Plant Science | www.frontiersin.org 4 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production (1.4 × 103 conidia/ml) in half strength of potato dextrose reticulum (ER) retention sequence (KDEL) remains in the broth (PDB) containing 0.02% chloramphenicol were mixed mature proteins, and although it is not strictly required for in each well with 30 μl of samples in OB resuspension buffer PB deposition, it is reported to favor accumulation levels (Takagi (0.2 M sucrose; 10 mM Tris-HCl pH 7.5; 0.02% Tween-20). et al., 2005b). In addition to the KDEL sequence, the PAF102 Samples were prepared in triplicate. Plates were incubated with was His-tagged at the N-terminus to facilitate its purification agitation for 72 h at 28°C. Fungal growth was monitored every from rice endosperm, resulting in a new PAF peptide that 24 h by measuring the optical density (OD) at 600 nm using was named PAF103 (Figure 1B). Growth inhibitory activity a Spectramax M3 reader (Molecular Devices), and mean values against the F. proliferatum fungal pathogen revealed equivalent and standard deviation (SD) were calculated. Experiments were antifungal activity for both peptides, namely PAF102 and repeated twice. PAF103, with a minimal inhibitory concentration (MIC) value of 3.5 μM. One additional construct was prepared to produce PAF102 RESULTS as a fusion protein to the Ole18 (Figure 1A). In this case, the expression of the chimeric fusion gene Ole18-PAF102 is Generation and Characterization of directed by the embryo specific promoter of the Ole18 gene Transgenic Rice Plants (Qu and Takaiwa, 2004). This strategy intends to target PAF102 Three different constructs were prepared for the expression of to OBs. a PAF synthetic gene in rice seed endosperm (Figure 1A). Transformation of embryogenic rice calli was performed via Two of them contain the glutelin promoters (pGluB1 and Agrobacterium tumefaciens. Using the hygromycin resistance for pGluB4) to drive expression in the peripheral region of the selecting transformed calli, a similar number of plants was endosperm, and the other one contains the 26 kDa globulin regenerated for each transformation event (around 10 independent promoter (pGlb1) to drive expression in the inner starchy lines obtained from independent calli). The presence of the endosperm tissue (Qu and Takaiwa, 2004). All three constructs transgenes was confirmed in most of the regenerated plants by were designed to produce an individual PAF peptide targeted PCR analysis using leaf genomic DNA as template. The positive to PBs by including the signal peptides (SP) of the corresponding plants were grown under containment greenhouse conditions storage protein and the KDEL signal. The N-terminal signal to obtain homozygous transgenic lines in T2 generation. Four peptides target proteins to the secretory pathway, are to five independent homozygous lines per construct were identified co-translationally cleaved, and are indispensable for PB sorting based on the segregation of the hygromycin resistance marker. (Takagi et al., 2005b). However, the C-terminal endoplasmic No apparent adverse effects on growth, flowering, or grain yield FIGURE 2 | Detection of the transgenes in the rice plant genome. PCR analysis on genomic DNA purified from leaves of wild-type (WT) or transgenic lines carrying the empty vector (EV) or the indicated transgenes. Plasmid DNA was used as a positive control (+). Arrows indicate the position of the specific oligonucleotides used for PCR amplification. The size of amplified fragments showed full length transgene insertion. Molecular markers are shown on the left in kpb. Frontiers in Plant Science | www.frontiersin.org 5 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production were observed on the transgenic PAF plants across generations; all of them performed similar to the plants transformed with the empty vector grown simultaneously under the same conditions. Stability and inheritance of transgenes in T3 plants were then confirmed by PCR analysis of genomic DNA, as done in the T0 plants (Figure 2). The copy number of transgene insertions in the different lines was estimated by qPCR analysis in comparison to the SPS single copy gene in the rice genome. The results FIGURE 3 | PAF103 transgene expression in rice seeds. RT-PCR analysis of show that the transgenes were present in a single copy in all indicated transgenes in immature rice seeds. PAF103 transcripts (upper the selected homozygous lines, in agreement with the segregation bands) were compared to the OsEF1a housekeeping gene transcripts in previous generations of the antibiotic resistance marker encoded (lower bands). in the T-DNA (Supplementary Table S2). PAF103 Does Not Accumulate in PBs of the rice seeds, the corresponding product PAF103 does not Rice Seeds accumulate to detectable levels. The accumulation of the single PAF103 peptide was evaluated on T2 homozygous seeds carrying the constructs pGluB1:PAF103, PAF102 Accumulates as Fusion to the pGluB4:PAF103, or pGlb1:PAF103. Seed protein extracts enriched Ole18 Protein in Rice Seeds in dense organelles were successfully used previously for detection The accumulation of the PAF102 when fused to the Ole18 of cationic antimicrobial peptides deposited in PBs, such as protein was evaluated in the T2 homozygous seeds of transgenic cecropin A or BP178 (Bundó et al., 2014; Montesinos et al., lines carrying the pOle18:Ole18-PAF102. We first analyzed 2017). Accordingly, we prepared PB-enriched extracts from PAF102 accumulation by in situ immunodetection in whole mature seeds of all the homozygous lines carrying the constructs mount seeds. As shown in Figure 4A, PAF102 was for PAF103 expression. Western blot analysis using anti-His immunodetected in the seed embryo and aleurone layer of tag antibodies for PAF103 detection revealed no differential pOle18:Ole18-PAF102 seeds but not in wild-type seeds. This band on protein extracts from transgenic lines in comparison distribution pattern corresponds to the expression pattern of to wild type, whereas PAF103 was clearly immunodetected on the Ole18 promoter (Qu and Takaiwa, 2004) and correlates a wild-type extract supplemented with the synthetic PAF103 with OB accumulation in rice seeds (Montesinos et al., 2016). peptide (Supplementary Figure S2). This result suggests that In order to confirm that PAF102 was produced as a fusion PAF103 does not accumulate in the transgenic seeds. To discard protein and retains the natural targeting of Ole18, we isolated an extraction problem, we conducted an in situ immunodetection OBs from seeds of five independent pOle18:Ole18-PAF102 assay in whole seeds that equally failed to detect PAF103 on homozygous lines. After solubilization, OB-associated proteins the transgenic seeds. were separated by SDS-PAGE and immunodetected using anti- A different approach to assess PAF103 accumulation is PAF102 antibodies. A polypeptide of apparent molecular mass through the detection of its antifungal activity. Thus, transgenic of 23 kDa (the expected mass of the fusion protein is 23.14 kDa, PAF103 seeds were then evaluated for resistance to F. corresponding to 18 kDa of oleosin + 1.2 kDa of TEV protease proliferatum. However, PAF103 seeds showed similar, or even recognition size + 3.94 kDa of PAF102) was clearly detected increased, susceptibility to the fungal pathogen than wild-type in the OBs of four out of five pOle18:Ole18-PAF102 lines and or empty vector seeds (Supplementary Figure S3). All the was absent in the empty vector and wild-type OBs (Figure 4B). seeds from the different lines showed reduced germination This protein was also immunoreacting with the anti-Ole18 capacity and reduced seedling growth after inoculation with antibodies as an additional and less intense band than the fungal spores, whereas they germinated and grew normally one corresponding to the Ole18 protein. The accumulation of under control conditions. Therefore, the antifungal activity the fusion protein seems not to alter the protein profile of of PAF103 was not detected in the transgenic rice seeds, OBs as visualized by protein Coomassie blue staining. These providing additional support that the PAF103 peptide is not results demonstrate that PAF102 accumulates in rice OBs when accumulated in these seeds. fused to the Ole18 protein. Given that transgenes were integrated in the genome of all The amount of produced PAF102 was estimated in T3 seeds the independent transgenic PAF103 lines but the transgene in comparison to known amounts of PAF102. The highest products were not detected, we evaluated the transgene expression value was found in the line #6 producing 20 ± 3 μg/g of in immature seeds at the developmental stage where GluB1, seed, and the mean value for the four independent lines was GluB4, and Glb1 promoters are active (Qu and Takaiwa, 2004). 15 ± 6 μg/g of seed. We amplified the corresponding mRNAs in the tested independent In addition to be stably produced across plant generations, lines (Figure 3) by RT-PCR analysis on total RNA isolated the fusion protein remains stable in seeds during long-time from seeds and using specific primers (Supplementary Table S1). storage at room temperature. The fusion protein was still These results indicate that although pGluB1:PAF103, detected in seeds after 3 years of storage in the laboratory pGluB4:PAF103, or pGlb1:PAF103 transgene is expressed in with a mean value of 16 ± 6 μg/g of seed. Frontiers in Plant Science | www.frontiersin.org 6 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production of the Ole18-PAF102 has no impact in seed filling. A Additionally, seeds accumulating the recombinant fusion protein germinated at the same rate and timing as the control seeds, and their seedlings showed similar appearance (Figures 5D,E). Thus, the presence of Ole18-PAF102 in the OBs seems not to affect the viability of rice seeds and seedling growth. Altogether, these results suggest that the expression of pOle18:Ole18-PAF102 does not alter the fitness of the rice plants. B Ole18-PAF102 Accumulation Does Not Protect Rice Seeds Against Fungal Infection To investigate whether the fusion protein Ole18-PAF102 retains the antifungal activity of the single PAF102 peptide, we evaluated the pOle18:Ole18-PAF102 seeds for resistance to F. proliferatum. These transgenic seeds showed similar susceptibility to the fungal pathogen than the wild-type or empty vector seeds (Figure 6). All the seeds from the different lines showed reduced germination capacity and reduced seedling growth after inoculation with fungal spores. These data suggest that the Ole18-PAF102 does not protect plants in situ against the fungal infection. Biologically Active PAF102 Is Recovered From pOle18:Ole18-PAF102 Seeds We next assessed the recovery of the single PAF102 peptide FIGURE 4 | PAF102 accumulates stably in rice seeds as an oleosin fusion from rice seed OBs carrying Ole18-PAF102. For that, we digested protein. (A) In situ immunolocalization of PAF102 in pOle:Ole18-PAF102 the recombinant OBs with the TEV protease, since Ole18 and transgenic seeds (line #6) in comparison to wild-type (WT). Immunoreaction PAF102 polypeptides were linked through the protease recognition was detected using a fluorescent-labeled secondary antibody. (B) Immunoblot analysis of OB fractions purified from seeds of WT, empty site. The immunoblot analysis of OB fractions before and after vector (EV), and five independent transgenic lines carrying the pOle18:Ole18- proteolytic digestion is shown in Figure 7A. We observed that PAF102 transgene, using anti-PAF102 or anti-oleosin18 antibodies as the fusion protein nearly disappeared after protease digestion indicated. Protein profile of OB fractions is shown by Coomassie blue staining of the OBs from two pOle18:Ole18-PAF102 transgenic lines (#3, of SDS gel. Proteins were purified from recent harvested seeds. #6). Subtle differences were detected among lines and experiments, and TEV protease efficiency was calculated at 87.5 ± 6.5% on average. These data indicate a high efficiency of proteolytic Ole18-PAF102 Accumulation Has processing of the fusion protein Ole18-PAF102 on intact OBs. No Negative Impact in Rice Next, we investigated the presence of the PAF102 single peptide Plant Performance in the protease-digested fractions. We immunodetected a PAF102 is a cell-penetrating peptide that kills fungal cells polypeptide in the fractions of lines #3 and #6 with a higher intracellularly (López-García et al., 2015). In order to evaluate electrophoretic mobility to the synthetic PAF102 peptide, but potential toxicity of PAF102 fused to Ole18 and accumulated that was absent in the WT fractions (Figure 7B). Equally, the inside the embryonic rice cells, we characterized phenotypically polypeptide immunodetected in the EV fractions supplemented the transgenic rice plants expressing the Ole18-PAF102 under with synthetic PAF102 peptide showed a different mobility than the control of the Ole18 promoter. These plants showed the synthetic PAF102 peptide alone. These results indicate that, normal phenotypical appearance during the vegetative phase, in the presence of plant extracts enriched with OBs, PAF102 similar to the wild-type and empty vector plants (Figure 5A). exhibits different electrophoretic mobility, consistent with the Interestingly, they did not show a penalty in grain yield altered electrophoretic mobility that has been previously reported (Figure 5B), indicating that the accumulation of the for other small cationic peptides (Coca et al., 2006; Bundó et al., Ole18-PAF102 does not affect seed production. Although 2014; Montesinos et al., 2016, 2017). We also observed a couple differences in seed weight were observed among lines, those of immunoreactive bands for the pure synthetic PAF102 indicating seeds accumulating the highest levels of Ole18-PAF102 showed a tendency to form multimers (Figure 7B). Thus, our data similar weight on average to the control wild-type and empty suggest that PAF102 is released from the fusion protein and vector seeds (Figure 5C). These data indicate that accumulation associates with other compounds in OB fractions or multimerizes. Frontiers in Plant Science | www.frontiersin.org 7 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production A B C D E FIGURE 5 | Production of the Ole18-PAF102 fusion protein in seeds has no negative impact in rice plant performance. (A) Phenotypical appearance of the wild- type (WT) and the transgenic rice plants carrying the empty vector (EV) or the pOle18:Ole18-PAF102 gene at 30 days after sowing. (B) Average grain yield per plant calculated from four plants per line in three independent assays (n = 12). (C) Average weight of 100 seeds per line (n = 12). (D) Phenotypical appearance of seedlings at 7 days post imbibition. (E) Percentage of germinated seeds at 2 and 4 days after imbibition. Values correspond to the mean value of three independent assays. Error bars represent standard deviation. To characterize the released PAF102 peptide, we tested whether the synthetic PAF102 was active against F. proliferatum the antifungal activity of the different OB fractions in in vitro in the OB isolation buffer (Figure 7C). We observed total fungal growth inhibitory assays (Figures 7C–E). First, we checked fungal growth inhibition at the concentration of 4 μM PAF102. Frontiers in Plant Science | www.frontiersin.org 8 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production A B FIGURE 6 | Ole18-PAF102 accumulation does not protect rice seeds against the fungal pathogen F. proliferatum. (A) Phenotypical appearance of wild-type (WT), empty vector (EV), and pOle18:Ole18-PAF102 transgenic seedlings (lines #1, #3, #6, and #5) at 7 days after inoculation with F. proliferatum spore suspensions (103 spores/ml). Pictures are representative of three independent experiments. (B) Percentage of seed germination upon infection in comparison to control conditions (see Figure 5D). The graph shows mean and standard deviation values of the indicated lines from three independent assays. A B C D E FIGURE 7 | Recovery of active PAF102 from rice seed OBs. (A,B) Immunoblot analysis of equivalent amount of OB fractions from wild-type (WT), empty vector (EV), and pOle18:Ole18-PAF102 lines #3 and #6 seeds, before (−) or after (+) TEV protease digestion, and using anti-PAF102 or anti-Ole18 antibodies. Digested fractions were run in a tricine-SDS gel in parallel to synthetic PAF102 (50 and 200 ng) or to synthetic PAF102 (200 ng) added to EV fractions (B). (C–E) In vitro growth inhibitory activity of F. proliferatum by synthetic PAF102 peptide at indicated concentrations (C), by OB fractions (D), or by OB fractions digested with TEV protease (E) from indicated lines. WT fractions were supplemented with 5 μM of synthetic PAF102. Graphs show the mean of OD600nm of triplicate samples ± SD. This inhibitory concentration agrees with reported values has no antifungal activity, as suggested by the fungal infection (López-García et al., 2015). Then, we tested intact OBs carrying assays of the seeds accumulating the fusion protein (Figure 6). the Ole18-PAF102 fusion protein from two independent lines Finally, OB fractions digested with TEV protease containing in comparison to OBs from empty vector lines, and we did the released PAF102 showed clear growth inhibitory activity not detect fungal growth inhibitory activity (Figure 7D). This against F. proliferatum, whereas the empty vector fractions is an additional evidence that the fusion protein Ole18-PAF102 did not possess antifungal activity (Figure 7E). OB fractions Frontiers in Plant Science | www.frontiersin.org 9 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production from line pOle18:Ole18-PAF102 #6 showed higher inhibitory have been reported to direct expression to the outer endosperm capacity than those from line #3, in agreement with the and have been successfully used to express the Cecropin A protein accumulation levels. The inhibitory activity depicted gene, encoding a small, linear, and cationic antimicrobial by fractions from line #6 was similar to wild-type fractions peptide (Bundó et al., 2014). The globulin promoter directs supplemented with the synthetic PAF102. According to the the expression to the inner endosperm and worked better activity, the amount of PAF102 in OB fractions was around than pGluB1 and pGluB4 for the expression of the BP178 3–4 μM (30 μl from a total of 100 μl obtained from 10 gene, encoding also a synthetic small, linear, and cationic seeds), which represents 13.3–17.6 μg/g of seed. These values antimicrobial peptide (Montesinos et al., 2017). Although agree with the estimation for the fusion protein Ole18-PAF102, we showed that all the three promoters directed the expression which was 15 ± 6 μg/g of seed. Therefore, the results indicate of the gene to the rice seeds, we could not detect the product that most of the PAF102 was released after the proteolytic using different methods. Failed detection supports that the digestion of the Ole18-PAF102 fusion protein as an active peptide did not accumulate in the rice seed. Given that all antifungal peptide. the rice plants expressing the PAF transgenes showed normal growth and development, and no altered seed filling and yield, cytotoxic effects seem not to be responsible for the DISCUSSION lack of PAF accumulation. The most plausible reason is peptide instability in plant tissues. Further experiments are Our study demonstrates that rice seeds can be used as needed to understand why these peptides are not accumulated biofactories of rationally designed antifungal peptides, in rice seeds. Whatever was the reason, our results clearly exemplified in the PAF102. The production of this small show difficulties to produce PAFs as single peptides in bioactive peptide was only feasible when fused to the Ole18 rice seeds. protein, and its accumulation was not detected when expressed A better strategy for the production of this type of peptides as a single peptide. Since the identification of PAF26 (López- is the fusion to the Ole18 protein. The fusion protein guided García et al., 2002), this and other PAF-derived peptides by the oleosin is embedded in OBs where peptides are have been recalcitrant to be produced through biotechnology immobilized and inactivated. Our assays clearly show that (either in bacteria, yeast, or higher eukaryotes). Therefore, Ole18-PAF102 does not exhibit in vivo or in vitro activity one major point of novelty of the current study is to manage against fungi, but as soon as it is released from the Ole18 the biotechnological production of PAF peptides through it becomes active. The immobilization in OBs confers protection fusion to oleosin. We have defined a model for the mode and offers stability to PAF102, allowing its accumulation. In of action of fungal-specific and PAF-derived peptides in three our experiments, the amount of PAF102 reached up to 20 μg/g steps: interaction with fungal cells, internalization, and of seed, which taking into account the low molecular weight intracellular killing (Muñoz et al., 2013; López-García et al., of the peptide (3.9 kDa), corresponds to 5.1 nmoles/g of 2015). This implies that the specificity of these peptides relies seed. This yield is a little lower than the one obtained with in the interaction with the fungal cell envelope and the cecropin A using the same production strategy (8 nmoles/g) subsequent internalization. Once the peptide is inside the (Montesinos et al., 2016), but still on the average of reported cell, it might be active and killing other cells different than yields for small peptides in rice seeds (0.03–10 nmoles/g of fungal cells, such as bacteria cells. We speculate that the seed) (Yasuda et al., 2005; Takagi et al., 2005a,b, 2008, 2010; peptide produced in any cell factory could be toxic when it Suzuki et al., 2011; Wakasa et al., 2011; Wang et al., 2013), accumulates intracellularly, unless fused to a carrier such as or even proteins, such as lysozyme yielding up to 80 μg/g the oleosin, as demonstrated in this study. The fusion of (5.6 nmoles/g of seed) when produced from a single expression PAF102 to the Ole18 protein targeted the peptide to the cassette or 150 μg/g (10.5 nmoles/g of seed) from two OBs, where it remains stable during long-time storage and independent expression cassettes (Hennegan et al., 2005). accumulates to high amounts of up to 20 μg/g of seeds for Most of these recombinant proteins were produced in the such small peptides (3.9 kDa). rice endosperm, which accounts for most of the grain volume The production of PAFs as single peptides was approached (90%), whereas our strategy of protein accumulation in OBs using three different promoters, namely pGluB1, pGluB4, and is restricted to the rice embryo and the aleurone layer that pGlb1, which drive strong endosperm-specific expression. represents only 10% of the rice grain volume. It would Our approach was based on targeting the peptide to PBs be interesting to explore the production of PAF102 in OB-rich with the three corresponding seed storage protein signal seeds such as safflower, sesame, rapeseed, soybean, or sunflower. peptides and the KDEL extension to produce the derivative These oily seeds have been used to produce recombinant PAF103. These SPs are known to guide protein sorting into proteins using the oleosin fusion technology (Parmenter et al., PBs, including antimicrobial peptides (Müntz, 1998; Yang 1995; Boothe et al., 2010; Nykiforuk et al., 2011). Reported et al., 2003; Ibl and Stoger, 2012; Bundó et al., 2014; Montesinos yield ranges from 0.13% of total protein for insulin in et al., 2017; Takaiwa et al., 2017). The KDEL extension was Arabidopsis thaliana seeds (Nykiforuk et al., 2006), 0.27% introduced because it normally increases protein accumulation for hirudin in Brassica napus seeds (Parmenter et al., 1995), levels in PBs (Takagi et al., 2005b; Takaiwa et al., 2017). to 0.55% for human growth factor protein in safflower The two glutelin promoters, namely pGluB1 and pGluB4, (Boothe et al., 2010). Taking into account that rice grain is Frontiers in Plant Science | www.frontiersin.org 10 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production not particularly rich in OBs, and our PAF yield in rice is ETHICS STATEMENT 0.025% of seed proteins, we predict that commercial relevant values might be reached in oily seeds. Therefore, our proof- Mouse polyclonal antibodies were produced at the Laboratory of-concept study indicates that the technology of oleosin Animal Facilities (registration number B9900083) of the Center fusion might be the best strategy to produce PAF peptides for Research and Development (CID) from Spanish National with the projection to improve yields in oily crops. Research Council (CSIC), in strict accordance with the bioethical In addition to offer stability, OB accumulation facilitates principles established by the Spanish legislation following the purification of PAFs from plant material by simple international guidelines. The protocol was approved by the flotation in dense sucrose solutions. However, the PAF102 Committee on Bioethics of Animal Experimentation from CID immobilized in the OBs was not active and required to and by the Department of Agriculture, Livestock, Fisheries, be released from the Ole18 for activity. Equally, the Food and Environment of the Government of Catalonia (permit antimicrobial peptide CecA did not exhibit activity while number DAAM:7461). All efforts were made to minimize immobilized on the OBs (Montesinos et al., 2016), whereas suffering of the animals. other proteins have been reported to be active while associated to OBs, such as the ß-glucuronidase (van Rooijen and Moloney, 1995), a xylanase (Liu et al., 1997; Hung et al., AUTHOR CONTRIBUTIONS 2008), the D-hydantoinase (Chiang et al., 2006), or the D-psicose-3-epimerase (Tseng et al., 2014), among others. JM, BL-G, and MC conceived and designed the study. BL-G The lack of activity shown by the antimicrobial peptide and MB prepared the gene constructs to be introduced in PAF102 while immobilized in OBs might be related to its rice. MB carried out all rice transformation experiments and mode of action. Being attached to OBs might prevent the the molecular characterization of transgenic plants. MB, MV, PAF102 internalization into fungal cells, a process that is and MC characterized phenotypically the generated transgenic required for its antifungal action (Muñoz et al., 2013; López- rice plants. XS, MV, and MC characterized PAF production García et al., 2015). Appropriately, antifungal activity was in rice seeds. MC coordinated the study and prepared the recovered upon release from OBs by TEV protease digestion, manuscript. All the authors read and approved the exhibiting equivalent activity against F. proliferatum to the final manuscript. synthetic peptide. Therefore, our results show that biologically active PAFs can be produced in planta. Although rice seeds are starchy more than oily, and they FUNDING might not be the best host for production of oleosin fusion proteins, they offer unique opportunities as bioreactors since This work was supported by SEPSAPE grants (Plant KBBE the rice gene transfer technology is well developed, cropping programme) EUI2008-03769 and EUI2008-03619 and by the conditions are easy and well-established worldwide, and high grant BIO2015-68790-C2-2-R; through the “Severo Ochoa grain yield can be obtained (Stoger et al., 2005; Takaiwa Programme for Centres of Excellence in R&D” (SEV-2015- et al., 2017). Moreover, the accumulation in seeds provides 0533) from Spanish Ministerio de Ciencia, Innovación y long-term stability during storage at room temperature, up Universidades (co-financed FEDER funds); and by the CERCA to 3 years in the case of PAF102. Seeds can be stockpiled Programme/Generalitat de Catalunya. without the need to synchronize production with product demand. Additionally, the OBs are restricted to the embryo cells and aleurone layer in the rice grain. Thus, the production ACKNOWLEDGMENTS of the Ole18-PAF fusion protein driven by the Ole18 promoter was only found in these specific tissues. Along with the We thank Izar Achaerondio for help with parts of this work embryo, the seed coats, including aleurone and pericarp, are and Blanca San Segundo for scientific advice. separated during rice milling to obtain the white refined grain and remain as the rice bran by-product. Consequently, downstream purification is facilitated using the PAF102- SUPPLEMENTARY MATERIAL enriched rice bran as the starting plant material. The use of rice bran for the production of PAFs could add an extra The Supplementary Material for this article can be found online value to this by-product assisting their exploitation to bring at: https://www.frontiersin.org/article/10.3389/fpls.2019.00731/ them to market. full#supplementary-material REFERENCES Bebber, D. P., and Gurr, S. J. (2015). Crop-destroying fungal and oomycete pathogens challenge food security. Fungal Genet. Biol. 74, 62–64. doi: 10.1016/j. Abell, B. M., Holbrook, L. A., Abenes, M., Murphy, D. J., Hills, M. J., and fgb.2014.10.012 Moloney, M. M. (1997). Role of the proline knot motif in oleosin endoplasmic Boothe, J., Nykiforuk, C., Shen, Y., Zaplachinski, S., Szarka, S., Kuhlman, P., reticulum topology and oil body targeting. Plant Cell 9, 1481–1493. doi: et al. (2010). Seed-based expression systems for plant molecular farming. 10.1105/tpc.9.8.1481 Plant Biotechnol. J. 8, 588–606. doi: 10.1111/j.1467-7652.2010.00511.x Frontiers in Plant Science | www.frontiersin.org 11 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production Brown, G. D., Denning, D. W., Gow, N. A. R., Levitz, S. M., Netea, M. G., Nadal, A., Montero, M., Company, N., Badosa, E., Messeguer, J., and Montesinos, L. and White, T. C. (2012). Hidden killers: human fungal infections. Sci. Transl. (2012). Constitutive expression of transgenes encoding derivatives of the Med. 4, 1–9. doi: 10.1126/scitranslmed.3004404 synthetic antimicrobial peptide BP100: impact on rice host plant fitness. Bundó, M., Montesinos, L., Izquierdo, E., Campo, S., Mieulet, D., Guiderdoni, E., BMC Plant Biol. 12:159. doi: 10.1186/1471-2229-12-159 et al. (2014). Production of cecropin A antimicrobial peptide in rice seed Nykiforuk, C. L., Boothe, J. G., Murray, E. W., Keon, R. G., Goren, H. J., endosperm. BMC Plant Biol. 14:102. doi: 10.1186/1471-2229-14-102 Markley, N. A., et al. (2006). Transgenic expression and recovery of biologically Chang, S., Puryear, J., and Cairney, J. (1993). A simple and efficient method active recombinant human insulin from Arabidopsis thaliana seeds. Plant for isolating RNA from pine trees. Plant Mol. Biol. Report. 11, 113–116. Biotechnol. J. 4, 77–85. doi: 10.1111/j.1467-7652.2005.00159.x doi: 10.1007/BF02670468 Nykiforuk, C. L., Shen, Y., Murray, E. W., Boothe, J. G., Busseuil, D., Rhéaume, E., Chiang, C.-J., Chen, H.-C., Kuo, H.-F., Chao, Y.-P., and Tzen, J. T. C. (2006). et al. (2011). Expression and recovery of biologically active recombinant A simple and effective method to prepare immobilized enzymes using Apolipoprotein AIMilano from transgenic safflower (Carthamus tinctorius) artificial oil bodies. Enzym. Microb. Technol. 39, 1152–1158. doi: 10.1016/j. seeds. Plant Biotechnol. J. 9, 250–263. doi: 10.1111/j.1467-7652.2010.00546.x enzmictec.2006.02.024 Parmenter, D. L., Boothe, J. G., van Rooijen, G. J., Yeung, E. C., and Moloney, Coca, M., Peñas, G., Gómez, J., Campo, S., Bortolotti, C., Messeguer, J., et al. M. M. (1995). Production of biologically active hirudin in plant seeds using (2006). Enhanced resistance to the rice blast fungus Magnaporthe grisea oleosin partitioning. Plant Mol. Biol. 29, 1167–1180. doi: 10.1007/BF00020460 conferred by expression of a cecropin A gene in transgenic rice. Planta Perfect, J. R. (2016). Is there an emerging need for new antifungals? Expert 223, 392–406. doi: 10.1007/s00425-005-0069-z Opin. Emerg. Drugs 21, 129–131. doi: 10.1517/14728214.2016.1155554 Company, N., Nadal, A., La Paz, J. L., Martínez, S., Rasche, S., Schillberg, S., Qu, L., and Takaiwa, F. (2004). Evaluation of tissue specificity and expression et al. (2014). The production of recombinant cationic α-helical antimicrobial strength of rice seed component gene promoters in transgenic rice. Plant peptides in plant cells induces the formation of protein bodies derived Biotechnol. J. 2, 113–125. doi: 10.1046/j.1467-7652.2003.00055.x from the endoplasmic reticulum. Plant Biotechnol. J. 12, 81–92. doi: 10.1111/ Sallaud, C., Meynard, D., van Boxtel, J., Gay, C., Bès, M., Brizard, J. P., et al. pbi.12119 (2003). Highly efficient production and characterization of T-DNA plants Fisher, M. C., Hawkins, N. J., Sanglard, D., and Gurr, S. J. (2018). Worldwide for rice (Oryza sativa L.) functional genomics. Theor. Appl. Genet. 106, emergence of resistance to antifungal drugs challenges human health and 1396–1408. doi: 10.1007/s00122-002-1184-x food security. Science 360, 739–742. doi: 10.1126/science.aap7999 Stoger, E., Ma, J. K. C., Fischer, R., and Christou, P. (2005). Sowing the seeds Fisher, M. C., Henk, D. A., Briggs, C. J., Brownstein, J. S., Madoff, L. C., of success: pharmaceutical proteins from plants. Curr. Opin. Biotechnol. 16, McCraw, S. L., et al. (2012). Emerging fungal threats to animal, plant and 167–173. doi: 10.1016/j.copbio.2005.01.005 ecosystem health. Nature 484, 186–194. doi: 10.1038/nature10947 Suzuki, K., Kaminuma, O., Yang, L., Takai, T., Mori, A., Umezu-Goto, M., Hennegan, K., Yang, D., Nguyen, D., Wu, L., Goding, J., Huang, J., et al. et al. (2011). Prevention of allergic asthma by vaccination with transgenic (2005). Improvement of human lysozyme expression in transgenic rice grain rice seed expressing mite allergen: induction of allergen-specific oral tolerance by combining wheat (Triticum aestivum) puroindoline b and rice (Oryza without bystander suppression. Plant Biotechnol. J. 9, 982–990. doi: 10.1111/j. sativa) Gt1 promoters and signal peptides. Transgenic Res. 14, 583–592. doi: 1467-7652.2011.00613.x 10.1007/s11248-004-6702-y Takagi, H., Hiroi, T., Hirose, S., Yang, L., and Takaiwa, F. (2010). Rice seed Hung, Y. J., Peng, C. C., Tzen, J. T. C., Chen, M. J., and Liu, J. R. (2008). ER-derived protein body as an efficient delivery vehicle for oral tolerogenic Immobilization of Neocallimastix patriciarum xylanase on artificial oil bodies peptides. Peptides 31, 1421–1425. doi: 10.1016/j.peptides.2010.04.032 and statistical optimization of enzyme activity. Bioresour. Technol. 99, 8662–8666. Takagi, H., Hiroi, T., Yang, L., Tada, Y., Yuki, Y., Takamura, K., et al. (2005a). doi: 10.1016/j.biortech.2008.04.017 A rice-based edible vaccine expressing multiple T cell epitopes induces oral Ibl, V., and Stoger, E. (2012). The formation, function and fate of protein tolerance for inhibition of Th2-mediated IgE responses. Proc. Natl. Acad. storage compartments in seeds. Protoplasma 249, 379–392. doi: 10.1007/ Sci. USA 102, 17525–17530. doi: 10.1073/pnas.0503428102 s00709-011-0288-z Takagi, H., Hiroi, T., Yang, L., Takamura, K., Ishimitsu, R., Kawauchi, H., Khan, I., Twyman, R. M., Arcalis, E., and Stoger, E. (2012). Using storage et al. (2008). Efficient induction of oral tolerance by fusing cholera toxin organelles for the accumulation and encapsulation of recombinant proteins. B subunit with allergen-specific T-cell epitopes accumulated in rice seed. Biotechnol. J. 7, 1099–1108. doi: 10.1002/biot.201100089 Vaccine 26, 6027–6030. doi: 10.1016/j.vaccine.2008.09.019 Liu, J. H., Selinger, L. B., Cheng, K. J., Beauchemin, K. A., and Moloney, M. M. Takagi, H., Saito, S., Yang, L., Nagasaka, S., Nishizawa, N., and Takaiwa, F. (1997). Plant seed oil-bodies as an immobilization matrix for a recombinant (2005b). Oral immunotherapy against a pollen allergy using a seed- xylanase from the rumen fungus Neocallimastix patriciarum. Mol. Breed. 3, based peptide vaccine. Plant Biotechnol. J. 3, 521–533. doi: 10.1111/ 463–470. doi: 10.1023/A:1009604119618 j.1467-7652.2005.00143.x López-García, B., Harries, E., Carmona, L., Campos-Soriano, L., López, J. J., Takaiwa, F., Wakasa, Y., Hayashi, S., and Kawakatsu, T. (2017). An overview Manzanares, P., et al. (2015). Concatemerization increases the inhibitory on the strategies to exploit rice endosperm as production platform for activity of short, cell-penetrating, cationic and tryptophan-rich antifungal biopharmaceuticals. Plant Sci. 263, 201–209. doi: 10.1016/j.plantsci.2017.07.016 peptides. Appl. Microbiol. Biotechnol. 99, 8012–8021. doi: 10.1007/ Tseng, C., Liao, C., Sun, Y., Peng, C., Tzen, J. T. C., Guo, R., et al. (2014). s00253-015-6541-1 Immobilization of Clostridium cellulolyticum. J. Agric. Food Chem. 62, López-García, B., Pérez-Payá, E., and Marcos, J. F. (2002). Identification of 6771–6776. doi: 10.1021/jf502022w novel hexapeptides bioactive against phytopathogenic fungi through screening Twyman, R. M., Stoger, E., Schillberg, S., Christou, P., and Fischer, R. (2003). of a synthetic peptide combinatorial library. Appl. Environ. Microbiol. 68, Molecular farming in plants: host systems and expression technology. Trends 2453–2460. doi: 10.1128/AEM.68.5.2453-2460.2002 Biotechnol. 21, 570–578. doi: 10.1016/j.tibtech.2003.10.002 Montesinos, L., Bundó, M., Badosa, E., San Segundo, B., Coca, M., and van Rooijen, G. J., and Moloney, M. M. (1995). Plant seed oil-bodies as carriers Montesinos, E. (2017). Production of BP178, a derivative of the synthetic for foreign proteins. Biotechnology 13, 72–77. antibacterial peptide BP100, in the rice seed endosperm. BMC Plant Biol. Wakasa, Y., Zhao, H., Hirose, S., Yamauchi, D., Yamada, Y., Yang, L., et al. 17:63. doi: 10.1186/s12870-017-1011-9 (2011). Antihypertensive activity of transgenic rice seed containing an Montesinos, L., Bundó, M., Izquierdo, E., Campo, S., and Badosa, E. (2016). 18-repeat novokinin peptide localized in the nucleolus of endosperm cells. Production of biologically active cecropin A peptide in rice seed oil bodies. Plant Biotechnol. J. 9, 729–735. doi: 10.1111/j.1467-7652.2010.00576.x PLoS One 11:e0146919. doi: 10.1371/journal.pone.0146919 Wang, S., Takahashi, H., Kajiura, H., Kawakatsu, T., Fujiyama, K., and Takaiwa, F. Muñoz, A. M. U., Read, N. D., and Marcos, J. F. (2013). Understanding the (2013). Transgenic rice seeds accumulating recombinant hypoallergenic birch mechanism of action of cell-penetrating antifungal peptides using the rationally pollen allergen bet v 1 generate giant protein bodies. Plant Cell Physiol. designed hexapeptide PAF26 as a model. Fungal Biol. Rev. 26, 146–155. 54, 917–933. doi: 10.1093/pcp/pct043 doi: 10.1016/j.fbr.2012.10.003 Yang, L., Ding, J., Zhang, C., Jia, J., Weng, H., Liu, W., et al. (2005). Estimating Müntz, K. (1998). Deposition of storage proteins. Plant Mol. Biol. 38, 77–99. the copy number of transgenes in transformed rice by real-time quantitative doi: 10.1023/A:1006020208380 PCR. Plant Cell Rep. 23, 759–763. doi: 10.1007/s00299-004-0881-0 Frontiers in Plant Science | www.frontiersin.org 12 June 2019 | Volume 10 | Article 731 Bundó et al. Rice OBs for PAF Production Yang, D., Guo, F., Liu, B., Huang, N., and Watkins, S. C. (2003). Expression Conflict of Interest Statement: The authors declare that the research was conducted and localization of human lysozyme in the endosperm of transgenic rice. in the absence of any commercial or financial relationships that could be construed Planta 216, 597–603. doi: 10.1007/s00425-002-0919-x as a potential conflict of interest. Yasuda, H., Tada, Y., Hayashi, Y., Jomori, T., and Takaiwa, F. (2005). Expression of the small peptide GLP-1 in transgenic plants. Transgenic Res. 14, 677–684. Copyright © 2019 Bundó, Shi, Vernet, Marcos, López-García and Coca. This is an open- doi: 10.1007/s11248-005-6631-4 access article distributed under the terms of the Creative Commons Attribution License Zasloff, M. (2002). Antimicrobial peptides of multicellular organisms. Nature (CC BY). The use, distribution or reproduction in other forums is permitted, provided 415, 389–395. doi: 10.1038/415389a the original author(s) and the copyright owner(s) are credited and that the original Zhang, L., and Gallo, R. L. (2016). Antimicrobial peptides. Curr. Biol. 26, publication in this journal is cited, in accordance with accepted academic practice. No R14–R19. doi: 10.1016/j.cub.2015.11.017 use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Plant Science | www.frontiersin.org 13 June 2019 | Volume 10 | Article 731