TY - JOUR
AU - Russinova, Eugenia
AB - Abstract Protein aggregation is determined by short (5–15 amino acids) aggregation-prone regions (APRs) of the polypeptide sequence that self-associate in a specific manner to form β-structured inclusions. Here, we demonstrate that the sequence specificity of APRs can be exploited to selectively knock down proteins with different localization and function in plants. Synthetic aggregation-prone peptides derived from the APRs of either the negative regulators of the brassinosteroid (BR) signaling, the glycogen synthase kinase 3/Arabidopsis SHAGGY-like kinases (GSK3/ASKs), or the starch-degrading enzyme α-glucan water dikinase were designed. Stable expression of the APRs in Arabidopsis (Arabidopsis thaliana) and maize (Zea mays) induced aggregation of the target proteins, giving rise to plants displaying constitutive BR responses and increased starch content, respectively. Overall, we show that the sequence specificity of APRs can be harnessed to generate aggregation-associated phenotypes in a targeted manner in different subcellular compartments. This study points toward the potential application of induced targeted aggregation as a useful tool to knock down protein functions in plants and, especially, to generate beneficial traits in crops. In order to function properly, proteins must fold into their native structure, but protein folding is often challenged by protein misfolding and aggregation (Tyedmers et al., 2010). Although protein aggregation has long been considered as a disordered process mediated by nonspecific hydrophobic interactions, it is now understood to be a sequence-specific self-association process (Mitraki, 2010; Tyedmers et al., 2010). Indeed, both in bacterial (Sabaté et al., 2010) and mammalian systems (Rajan et al., 2001), aggregation of nonhomologous proteins has been shown to occur preferentially in distinct inclusion bodies. In vitro aggregation of protein solutions can be accelerated by seeding with preformed aggregates, and this process efficiency depends critically on the sequence homology between seed and target protein (Krebs et al., 2004; O’Nuallain et al., 2004). Self-seeding is generally several orders of magnitude more efficient than cross-seeding (Ganesan et al., 2015; Surmacz-Chwedoruk et al., 2014). Aggregation-associated human diseases, such as Alzheimer’s or Parkinson’s disease, are in line with this notion because the processes underlying these diseases are highly specific and characterized by the aggregation of one or a few proteins in particular tissues and cell types (Jucker and Walker, 2013). The elucidation of the structure of amyloid-forming peptides and protein fragments has shed light on the molecular origin of the sequence specificity of protein aggregation. The amyloid structure consists of the formation of a so-called cross-β conformation, whereby the peptide backbone of the aggregate creates hydrogen bond-mediated β-strand interactions, whereas the side chains contribute to the stability of these β-strands by aligning with, and closely packing to, the identical sequence of the neighboring strand (Sawaya et al., 2007; Makin et al., 2005). The registered stacking of side chains explains the aggregation sequence specificity. Indeed, backbone interactions contribute comparatively more to the amyloid structure than to the globular protein structure (Fitzpatrick et al., 2011). The portions of a protein sequence that are susceptible to associate into aggregates by β-strand-mediated interactions are limited to short segments, defined as aggregation-prone regions (APRs). The APRs consist of 5 to 15 amino acids in length (Rousseau et al., 2006; Goldschmidt et al., 2010) and can be identified by prediction algorithms (Fernandez-Escamilla et al., 2004). The determining role of APRs has been demonstrated by “aggregation-grafting” experiments, in which insertion of an APR of an aggregating protein into the sequence of a nonaggregating protein results in a protein with aggregation propensity and morphology similar to those of the original protein (Ventura et al., 2004). Application of the prediction algorithm TANGO (Fernandez-Escamilla et al., 2004) to the Arabidopsis (Arabidopsis thaliana) proteome revealed that 80% of the proteins contain APRs, implying that, similar to other eukaryotes, plant proteomes are also susceptible to protein aggregation (Rousseau et al., 2006). As most of the Arabidopsis proteins harbor aggregation-prone sequence segments within their primary structure and as aggregation is sequence specific, it should, in principle, be possible to induce aggregation and, subsequently, functional depletion of a protein by exposing it to a short target-specific aggregating peptide in plants. First, we tested this hypothesis by targeting proteins with kinase activity in Arabidopsis plants. We selected the cytosolic glycogen synthase kinase 3/Arabidopsis SHAGGY-like kinases (GSK3/ASKs) and the chloroplast-localized α-glucan water dikinase (GWD). Arabidopsis possesses 10 ASKs grouped into four clades (Youn and Kim, 2015) that share a 50% overall sequence identity across the whole protein. Among the ASKs, BRASSINOSTEROID INSENSITIVE2 (BIN2) was characterized as a negative regulator of BR signaling (Li and Nam, 2002; Vert and Chory, 2006; Yan et al., 2009). In addition to BIN2 and its two close homologs, BIN2-LIKE1 (BIL1) and BIL2 (clade II), at least four other ASKs redundantly convey BR signals via a mechanism similar to that of BIN2 (De Rybel et al., 2009; Kim et al., 2009; Rozhon et al., 2010). The GWD enzyme catalyzes the phosphorylation of starch in the chloroplasts by transferring β-ATP phosphate to either the C6 or the C3 position of the glycosyl residue of amylopectin and, thus, plays an essential role in starch metabolism (Mitsui et al., 2010). The phosphate groups influence the susceptibility of the starch granules to degrading enzymes, such as β-amylases. As a result, the starch breakdown is impaired in GWD-deficient plants. In GWD-antisense potato (Solanum tuberosum) plants (Lorberth et al., 1998), as well as in the GWD-deficient starch excess1 (sex1) mutants of Arabidopsis (Yu et al., 2001), the foliar starch content is significantly higher than that of the respective wild-type plants. In addition to the model plant Arabidopsis, we applied the APR-mediated aggregation by targeting the GWD enzyme in maize (Zea mays). Our work demonstrates that overexpression of different APRs, derived from a single protein or protein family, fused to a fluorescent carrier, results in specific knockdowns similar to previously described genetic mutants. We show that direct interactions between the APRs and the target proteins caused the loss of function of the proteins. Moreover, specific subcellular targeting of the synthetic APRs can be achieved in both model and crop plant species. Hence, the APR expression approach presented here can be used as an innovative knockdown method to inactivate proteins by specific in vivo pull-down in defined subcellular compartments of plants. In addition, the results also underline that, at least in plants, protein aggregation is not cytotoxic per se, but rather that the functional effect of the aggregates observed here appear to be dominated by sequence-specific cross-seeding of the aggregation of cellular APR-sharing proteins. RESULTS Design of the Aggregation Constructs To simultaneously knock out the function of all 10 ASKs in Arabidopsis by inducing specifically their misfolding and inactivation, we applied the aggregation prediction algorithm TANGO (Fernandez-Escamilla et al., 2004) to BIN2 in order to identify overlapping aggregation-prone peptides in the 10 target proteins. One APR of nine amino acids with a TANGO aggregation score greater than 50 (out of a maximum of 100) and coding for the sequence 249QLVEIIKVL257 in BIN2 was detected (hereafter referred to as BIN2249-257; Fig. 1A; Supplemental Table S1). The BIN2249-257 APR was situated in the kinase domain preceding the highly conserved TREE domain that plays a key role in the BIN2 function (Choe et al., 2002). Comparison of the BIN2249-257 amino acid sequence with the Arabidopsis proteome revealed that this APR was identical to the APRs identified in 8 of 10 ASKs and differed with only one amino acid (Val-256 to Ile-256) in the APRs of the remaining two ASKs (Fig. 1A; Supplemental Table S2). For modulation of the APR aggregation properties, different synthetic aggregating blocks (SABs) were designed. Each SAB was C-terminally fused to GFP for visualization and solubility increase (Fig. 1C; Supplemental Table S3). To stimulate aggregation, the BIN2 APR was combined with a modified version of the unnatural amyloid-forming booster (B) sequence STVIIE (López De La Paz et al., 2002; BIN2249-257B). In contrast, to slow down the aggregation of the synthetic booster by charge repulsion (Chiti et al., 2003), an Arg (R) was included on both APR flanks, thus modifying the BIN2249-257B into BIN2249-257RB (Supplemental Table S3). As charged residues are enriched at the flanks of APRs to decrease aggregation and function as natural “aggregation gatekeepers” (De Baets et al., 2014), five to six naturally flanking (NF) amino acids were added to the BIN2249-257 APR and expressed, including the first six amino acids of the BIN2 protein (MADDKE) in a single copy (BIN2249-257NF) or in tandem (T) (BIN2249-257NFT) (Supplemental Table S3). The aim of the tandem constructs was to amplify the aggregation potential by mimicking the repeating patterns of APRs that are observed in naturally occurring functional amyloids, such as the yeast prions (Bednarska et al., 2016). Figure 1. Open in new tabDownload slide Selected APRs and aggregation constructs design. A, Multiple alignment of ASK amino acid sequences with the aggregating peptide BIN2249-257. B, Multiple alignment of the APRs targeting the GWD protein in Arabidopsis (AtGWD534-541, AtGWD821-829, and AtGWD1227-1234) and in maize (ZmGWD599-610, ZmGWD889-897, and ZmGWD1082-1088) with their target protein, respectively. Identical residues are underlined with asterisks and indicated in red for Arabidopsis and in green for maize. The TREE kinase domain, conserved within the amino acid sequences of ASKs, is indicated in blue. Alignments were done with Clustal Omega. C, Schematic representation of constructs expressing different APR variants, indicated as SABs. SABs are fused to eGFP or cYFP at their C terminus. Only the GWD APRs were targeted to the chloroplast by adding a transit peptide at their N terminus. p35S, CaMV 35S promoter; pPePC, PepC promoter; eGFP, enhanced GFP; cYFP, citrine YFP; KanR/BastaR, kanamycin/Basta resistance gene; RB, right border; LB, left border. Figure 1. Open in new tabDownload slide Selected APRs and aggregation constructs design. A, Multiple alignment of ASK amino acid sequences with the aggregating peptide BIN2249-257. B, Multiple alignment of the APRs targeting the GWD protein in Arabidopsis (AtGWD534-541, AtGWD821-829, and AtGWD1227-1234) and in maize (ZmGWD599-610, ZmGWD889-897, and ZmGWD1082-1088) with their target protein, respectively. Identical residues are underlined with asterisks and indicated in red for Arabidopsis and in green for maize. The TREE kinase domain, conserved within the amino acid sequences of ASKs, is indicated in blue. Alignments were done with Clustal Omega. C, Schematic representation of constructs expressing different APR variants, indicated as SABs. SABs are fused to eGFP or cYFP at their C terminus. Only the GWD APRs were targeted to the chloroplast by adding a transit peptide at their N terminus. p35S, CaMV 35S promoter; pPePC, PepC promoter; eGFP, enhanced GFP; cYFP, citrine YFP; KanR/BastaR, kanamycin/Basta resistance gene; RB, right border; LB, left border. The GWD protein is encoded by a single gene in Arabidopsis and in maize. Similarly, the prediction algorithm TANGO (Fernandez-Escamilla et al., 2004) was used to identify APRs in orthologous proteins of Arabidopsis (AtGWD) and maize (ZmGWD; Supplemental Table S1). Three different APRs with a TANGO score higher than 50 were identified for each target protein in Arabidopsis and maize (Fig. 1B; Supplemental Table S1). The Arabidopsis APRs AtGWD534-541 and AtGWD821-829 were identical to the maize ZmGWD599-610 and ZmGWD889-897, respectively, whereas the APRs AtGWD1227-1234 and ZmGWD1082-1088 were specific for each species (Fig. 1B). Searches of the Arabidopsis proteome with the three selected GWD APRs did not reveal proteins containing identical APRs or APRs with a single mismatch (Supplemental Table S2). However, the ZmGWD1082-1088 APR was similar to five unrelated proteins in the maize proteome (one mismatch) and was excluded from further studies. The SABs for the GWD proteins consisted of tandem APRs flanked by the NFs of BIN2 and fused at their N terminus to a chloroplast transit peptide signal sequence for specific targeting to the chloroplasts (Supplemental Table S3) and at their C terminus to GFP or YFP when expressed in Arabidopsis and maize, respectively (Fig. 1C). The SABs AtGWD534-541NFT, AtGWD821-829NFT, and AtGWD1227-1234NFT were expressed in Arabidopsis, whereas ZmGWD599-610NFT and ZmGWD889-897NFT were introduced into maize after codon usage optimization. In Vivo Aggregation Induced by Expression of Selected APRs in Plant Cells All constructs, including controls, such as free GFP and booster (B)-GFP (Supplemental Table S3) were expressed transiently in leaf epidermis of tobacco (Nicotiana benthamiana) and stably in Arabidopsis with the constitutive cauliflower mosaic virus (CaMV) 35S promoter. Confocal fluorescence microscopy was used to evaluate the aggregation formation in the tobacco leaf epidermal cells 4 d after infiltration (Fig. 2) and in epidermal cells of roots, hypocotyls, and cotyledons of 7-d-old Arabidopsis plants grown in vitro (Supplemental Fig. S1). Consistently in both expression systems, the SABs containing either the booster sequence in combination with the Arg-flanked APR (BIN2249-257RB) or the tandemly repeated APRs (BIN2249-257NFT) were the most effective in perinuclear accumulation of GFP-labeled aggregates (Fig. 2A; Supplemental Fig. S1A). In contrast, the free GFP control accumulated throughout the cytosol and inside the nucleus. The B-GFP and BIN2249-257B-GFP fusions accumulated mainly in inclusion bodies, indicating low solubility of the aggregates, whereas the BIN2249-257NF-GFP fusion was found predominantly in the cytosol, implying a reduced aggregation tendency (Fig. 2A; Supplemental Fig. S1A). Figure 2. Open in new tabDownload slide Subcellular localization of the APRs. A, Tobacco leaf epidermis cells transiently expressing BIN2249-257B, BIN2249-257RB, BIN2249-257NF, BIN2249-257NFT, synthetic booster (B), and free GFP coinfiltrated with a nuclear localization signal-red fluorescent protein marker (NLS-RFP). Nuclei (N) are visible in the red channel. B, Subcellular localization of AtGWD534-541NFT-GFP, AtGWD821-829NFT-GFP, AtGWD1227-1234NFT-GFP, ZmGWD599-610NFT-YFP, ZmGWD889-897NFT-YFP, and ZmGWD1082-1088NFT-YFP. The chlorophyll (Ch) autofluorescence is visualized in the red channel. Arrows point to aggregates labeled with GFP or YFP. Bars = 20 μm (A) and 10 μm (B). Figure 2. Open in new tabDownload slide Subcellular localization of the APRs. A, Tobacco leaf epidermis cells transiently expressing BIN2249-257B, BIN2249-257RB, BIN2249-257NF, BIN2249-257NFT, synthetic booster (B), and free GFP coinfiltrated with a nuclear localization signal-red fluorescent protein marker (NLS-RFP). Nuclei (N) are visible in the red channel. B, Subcellular localization of AtGWD534-541NFT-GFP, AtGWD821-829NFT-GFP, AtGWD1227-1234NFT-GFP, ZmGWD599-610NFT-YFP, ZmGWD889-897NFT-YFP, and ZmGWD1082-1088NFT-YFP. The chlorophyll (Ch) autofluorescence is visualized in the red channel. Arrows point to aggregates labeled with GFP or YFP. Bars = 20 μm (A) and 10 μm (B). Transient expression of all GWD SABs in tobacco leaf epidermis resulted in aggregate formation inside the plastids, except for ZmGWD1082-1088NFT, in which the YFP fluorescence was visible as a diffused signal in the chloroplasts (Fig. 2B). Consistently, when stably transformed in Arabidopsis plants, the AtGWD534-541NFT and AtGWD821-829NFT SABs, but not AtGWD1227-1234NFT-GFP, caused the formation of GFP-labeled aggregates inside the chloroplasts in leaf epidermal and palisade cells (Supplemental Fig. S1B). The ZmGWD599-610NFT and ZmGWD889-897NFT SABs were stably expressed in maize under control of the maize mesophyll-specific phosphoenolpyruvate carboxylase (PepC) promoter, previously used to down-regulate GWD only in leaves, where the enzyme is the most abundant (Sattarzadeh et al., 2010). Confocal microscopy revealed that in mesophyll cells of the third or fourth leaves of 14-d-old transgenic maize seedlings, ZmGWD889-897NFT and ZmGWD599-610NFT induced pronounced aggregate formation (Supplemental Fig. S1C) in the chloroplasts, as seen in the tobacco experiments. Next, we analyzed cell extracts of transgenic Arabidopsis plants (Fig. 3, A and B) and tobacco leaves transiently expressing the GFP-tagged SABs (Supplemental Fig. S2, A and B) under nondenaturing conditions by means of Blue Native (BN)-PAGE to verify the Mr of the induced aggregates and whether the target proteins had presumably acquired a different electrophoretic mobility when a specific APR was overexpressed. The BN-PAGE analysis revealed the occurrence of high-M r protein complexes for all SABs. Notably, the BIN2 SABs containing the booster sequence induced the highest M r protein complexes and the APRs in tandem induced the formation of aggregates more than twice the M r of a single APRs (Fig. 3, A and B; Supplemental Fig. S2A). Interestingly, aggregates seemed to be less abundantly induced by the expression of YFP-tagged SABs than by that of the respective GFP fusions, suggesting that the choice of the fluorescent carrier protein most probably influences protein aggregation (Supplemental Fig. S2B). Figure 3. Open in new tabDownload slide Expression of BIN2 and GWD APRs leading to aggregate formation. A and B, BN-PAGE and immunoblots with anti-GFP antibodies of protein aggregates induced by expression of different BIN2 (A) and GWD (B) SABs in Arabidopsis, in respect to the wild type (Col-0), booster (b)-GFP, and free GFP controls. BN-PAGE was run under nondenaturing conditions to determine the native masses of protein complexes. C, FTIR analysis of the high-Mr aggregates immunoprecipitated with anti-GFP antibodies from lysates prepared from Arabidopsis transgenic seedlings that expressed different versions of BIN2 SABs. D to F, TEM immunolocalization with anti-GFP antibodies of BIN2249-257RB-GFP and BIN2249-257NFT-GFP in hypocotyls (D) and in cortical cells of the root elongation zone (E and F). Ultrathin sections were poststained in uranyl acetate and lead citrate and grids were viewed with a JEM-1010 TEM (Jeol) operating at 80 kV. Bars = 0.1 (inset) and 0.5 μm in D and F, and 0.1 μm in E. Seven-day-old T3 Arabidopsis seedlings were used in all experiments. Arrows point to GFP-labeled protein aggregates. Figure 3. Open in new tabDownload slide Expression of BIN2 and GWD APRs leading to aggregate formation. A and B, BN-PAGE and immunoblots with anti-GFP antibodies of protein aggregates induced by expression of different BIN2 (A) and GWD (B) SABs in Arabidopsis, in respect to the wild type (Col-0), booster (b)-GFP, and free GFP controls. BN-PAGE was run under nondenaturing conditions to determine the native masses of protein complexes. C, FTIR analysis of the high-Mr aggregates immunoprecipitated with anti-GFP antibodies from lysates prepared from Arabidopsis transgenic seedlings that expressed different versions of BIN2 SABs. D to F, TEM immunolocalization with anti-GFP antibodies of BIN2249-257RB-GFP and BIN2249-257NFT-GFP in hypocotyls (D) and in cortical cells of the root elongation zone (E and F). Ultrathin sections were poststained in uranyl acetate and lead citrate and grids were viewed with a JEM-1010 TEM (Jeol) operating at 80 kV. Bars = 0.1 (inset) and 0.5 μm in D and F, and 0.1 μm in E. Seven-day-old T3 Arabidopsis seedlings were used in all experiments. Arrows point to GFP-labeled protein aggregates. To investigate the biochemical nature of the aggregates formed in plant cells, we analyzed by means of Fourier transform infrared (FTIR) spectroscopy the high-M r aggregates that had been immunoprecipitated with anti-GFP antibodies from lysates prepared from Arabidopsis transgenic seedlings that expressed different BIN2 SABs. For all constructs, IR absorption peaks 1620, 1635, and 1690 cm−1 were detected (Fig. 3C). This result supports the formation of an amyloid-like β-structure because absorbance of IR light at these wave numbers is a characteristic feature of β-structures. Transmission electron microscopy (TEM) combined with immunogold labeling of different Arabidopsis tissues expressing the BIN2 SABs with the strongest aggregation properties, namely, BIN2249-257RB and BIN2249-257NFT, revealed that aggregates localized in the cytosol. In the BIN2249-257RB-producing plants, the aggregated proteins accumulated either as amorphous clusters or as more ordered and elongated fibril-like structures (Fig. 3, D and E), whereas in the BIN2249-257NFT plants, these ordered structures were absent and free cytosolic proteins occurred most frequently (Fig. 3F). These variations in aggregate morphology reflected the differences in construct design and the impact of the strongly aggregating amyloid-forming booster sequence. The APRs Interacted Specifically with Their Targeted Proteins in Vivo To evaluate the specificity of the induced protein aggregations, we carried out colocalization experiments in leaf epidermal cells of tobacco between GFP-tagged ASKs (BIN2, BIL1, ASKα, ASKγ, and ASKθ) and the RFP-tagged BIN2249-257NFT, expressed from CaMV 35S and estradiol-inducible promoters, respectively. Four days after transfection and following a 24-h induction of the BIN2249-257NFT-RFP expression, a strong colocalization was observed between the APR and all target proteins (Supplemental Fig. S3). The direct interaction between the BIN2249-257 APR and each of the 10 target ASKs was confirmed by a bimolecular fluorescence complementation (BiFC) assay. Coexpression of each of the ASKs tagged with the N-terminal GFP fragment (nGFP) and the BIN2249-257NFT tagged with the C-terminal part of GFP (cGFP) in tobacco leaves resulted in a fluorescent signal (Fig. 4A; Supplemental Table S4). The self-interaction property of the BIN2249-257 APR was assessed by coexpressing the BIN2249-257NFT-nGFP and BIN2249-257NFT-cGFP constructs (Fig. 4A). In all cases, despite the observed GFP signal, no GFP-labeled aggregates were formed, probably due to the slow reassociation of the two GFP fragments that could have slowed down the aggregate formation. Figure 4. Open in new tabDownload slide Specific in vivo interaction between BIN2 and the BIN2249-257 APR. A, BiFC assay of BIN2249-257NFT-cGFP coexpressed with different nGFP-tagged ASKs and BIN2249-257NFT-nGFP in tobacco leaves, 3 d after infiltration. In the last panel, the interaction between BIN2249-257NFT-cGFP and MUTE is also shown as a negative control. Bars = 50 μm. B, Coimmunoprecipitation in tobacco leaves of BIN2-HA with different BIN2249-257 SABs after coexpression for 3 d. Booster (B)-GFP, free GFP, BIN2-HA, mock (not infiltrated leaf), and BIN2P249-257NFT-GFP are included as negative controls. Proteins were detected with anti-HA and anti-GFP antibodies. C, Coimmunoprecipitation of BIN2249-257NFT-GFP with MUTE-GS or GS-MUTE proteins coproduced for 3 d as in B. GFP, MUTE-GS, GS-MUTE, and mock are included as negative controls; the GS tag (consisting of a protein G tag and a streptavidin-binding peptide) reacts with the antiperoxidase (PAP) antibody. Anti-PAP and anti-GFP antibodies were used for protein detection. Figure 4. Open in new tabDownload slide Specific in vivo interaction between BIN2 and the BIN2249-257 APR. A, BiFC assay of BIN2249-257NFT-cGFP coexpressed with different nGFP-tagged ASKs and BIN2249-257NFT-nGFP in tobacco leaves, 3 d after infiltration. In the last panel, the interaction between BIN2249-257NFT-cGFP and MUTE is also shown as a negative control. Bars = 50 μm. B, Coimmunoprecipitation in tobacco leaves of BIN2-HA with different BIN2249-257 SABs after coexpression for 3 d. Booster (B)-GFP, free GFP, BIN2-HA, mock (not infiltrated leaf), and BIN2P249-257NFT-GFP are included as negative controls. Proteins were detected with anti-HA and anti-GFP antibodies. C, Coimmunoprecipitation of BIN2249-257NFT-GFP with MUTE-GS or GS-MUTE proteins coproduced for 3 d as in B. GFP, MUTE-GS, GS-MUTE, and mock are included as negative controls; the GS tag (consisting of a protein G tag and a streptavidin-binding peptide) reacts with the antiperoxidase (PAP) antibody. Anti-PAP and anti-GFP antibodies were used for protein detection. Next, the hemagglutinin (HA)-tagged BIN2 was coimmunoprecipitated in all samples after transient coexpression with each BIN2249-257 APR-containing construct in tobacco leaf epidermal cells, validating the direct interaction between BIN2249-257 and its target protein in vivo (Fig. 4B). Remarkably, substitution of Val-251 and Ile-254 by two prolines (P) that lowers the aggregation propensity (Richardson and Richardson, 2002) in the BIN2249-257 APR, completely abolished its aggregation capacity (Supplemental Tables S1–S3) and the interaction with its target protein BIN2 (Fig. 4B). To additionally assess the specificity of the BIN2249-257 APR for binding random proteins containing similar APR sequences, we tested the interactions between the BIN2249-257NFT SAB and the Arabidopsis basic helix-loop-helix transcription factor MUTE (Pillitteri et al., 2007). MUTE had been identified as the only Arabidopsis protein containing an APR that differed from BIN2249-257 by two amino acids (Glu-252 into Lys-252 and Lys-255 into Ser-255) and had a TANGO score of 46 (Supplemental Table S2). Both BiFC and coimmunoprecipitation experiments did not reveal interactions between the BIN2249-257 APR and the MUTE protein (Fig. 4, A and C; Supplemental Table S4). The BIN2249-257 APR also did not interact with randomly selected, nonhomologous, and overexpressed proteins, such as the clathrin light chain (CLC) (Supplemental Fig. S4). In addition, we detected the GWD protein by means of a GWD-specific antibody after BN-PAGE of protein extracts from tobacco leaves and after immunoprecipitation with anti-GFP antibodies (Supplemental Fig. S2C). Altogether, this evidence supports the high APR specificity in target interactions. Arabidopsis Plants Expressing an APR That Targeted the 10 ASKs Showed Weak Constitutive BR Responses To assess whether the expression of the BIN2249-257 APR induced loss of function of the targeted ASK proteins, we analyzed the growth and developmental phenotypes of 7-d-old in vitro- and light-grown Arabidopsis plants overexpressing the SABs with strong aggregation properties (BIN2249-257RB and BIN2249-257NFT). The ASK promoter-GUS studies revealed that ASKα, ASKγ, and BIN2 are the most abundantly expressed ASKs at this developmental stage (Supplemental Fig. S5) and, thus, most probably targeted ASKs by the BIN2249-257 APR. Representative T3 homozygous transgenic lines overexpressing each BIN2249-257RB and BIN2249-257NFT (Supplemental Fig. S6A) displayed longer hypocotyls and roots than the wild-type control Columbia-0 (Col-0; Fig. 5, A–C) and 1-month-old plants from the same transgenic lines grown in soil had larger rosettes than Col-0 (Fig. 5D). All tested transgenic lines were partially resistant to the specific BR biosynthesis inhibitor brassinazole (BRZ; Asami et al., 2000; Fig. 5E) and showed down-regulation of the BR-biosynthetic genes CONSTITUTIVE PHOTOMORPHOGENIC DWARF (CPD; Szekeres et al., 1996) and DWARF4 (DWF4; Choe et al., 1998; Fig. 5F) and up-regulation of the transcription factor BRASSINAZOLE RESISTANT1 (BZR1; Wang et al., 2002). Furthermore, overexpression of the BIN2249-257RB in a weak mutant allele of the BR receptor bri1-5 partially rescued the mutant phenotype (Fig. 5G). These phenotypes were in line with the anticipated enhanced BR signaling due to the inactivation of the BR-negative regulators, as shown previously for the bin2-3 knockout mutant (Yan et al., 2009). Figure 5. Open in new tabDownload slide Constitutive BR responses in Arabidopsis plants expressing the BIN2 APR. A, Seven-day-old in vitro-grown Arabidopsis wild-type (Col-0) and T3 transgenic plants expressing BIN2249-257RB and BIN2249-257NFT. B and C, Hypocotyl and root length measurements of the plants shown in A (n > 15). D, Rosettes and rosette area quantification of the lines shown in A and grown in soil for 30 d (n = 8). E, Hypocotyl lengths, relative to Col-0, of BIN2249-257RB, BIN2249-257NFT, and bin2/bil1/bil2 seedlings grown in the dark for 5 d on medium containing DMSO or 1 μM BRZ (n > 15). F, Relative expression of DWF4, CPD, and BZR1 genes in 7-d-old Col-0, BIN2249-257RB, BIN2249-257NFT, and bin2/bil1/bil2 seedlings. G, Phenotypes and rosette area quantification of 30-d-old bri1-5 and two independent bri1-5/BIN2249-257RB T3 line 5 and line 8 (n = 8). Error bars represent sd, *P<0.05, **P<0.001, and ***P<0.0001 (Student's t-test). N, number of plants analyzed. Figure 5. Open in new tabDownload slide Constitutive BR responses in Arabidopsis plants expressing the BIN2 APR. A, Seven-day-old in vitro-grown Arabidopsis wild-type (Col-0) and T3 transgenic plants expressing BIN2249-257RB and BIN2249-257NFT. B and C, Hypocotyl and root length measurements of the plants shown in A (n > 15). D, Rosettes and rosette area quantification of the lines shown in A and grown in soil for 30 d (n = 8). E, Hypocotyl lengths, relative to Col-0, of BIN2249-257RB, BIN2249-257NFT, and bin2/bil1/bil2 seedlings grown in the dark for 5 d on medium containing DMSO or 1 μM BRZ (n > 15). F, Relative expression of DWF4, CPD, and BZR1 genes in 7-d-old Col-0, BIN2249-257RB, BIN2249-257NFT, and bin2/bil1/bil2 seedlings. G, Phenotypes and rosette area quantification of 30-d-old bri1-5 and two independent bri1-5/BIN2249-257RB T3 line 5 and line 8 (n = 8). Error bars represent sd, *P<0.05, **P<0.001, and ***P<0.0001 (Student's t-test). N, number of plants analyzed. Comparison of different tissues of wild-type and transgenic Arabidopsis plants expressing BIN2249-257RB and BIN2249-257NFT at the ultrastructural level indicated that the analyzed subcellular organelles, such as mitochondria and chloroplasts, were similar in terms of shape, size, and localization with those of the wild-type plants grown in vitro (Supplemental Fig. S7). In accordance, a genome-wide expression analysis revealed that only a few genes were differentially regulated (Supplemental Table S5), of which 17 were down-regulated (<0.5-fold change) and 33 were up-regulated (>1.5-fold change) in the BIN2249-257NFT-expressing line when compared to the wild type. The changes in expression were subtle with a median increase below 2 (1.7) and gene ontology searches associated the affected genes with stress responses, hormone signaling, or chaperones. Interestingly, a 5-fold increase in the expression of the heat shock protein 70 (HSP70) was detected and later confirmed by quantitative reverse-transcription PCR (qRT-PCR) experiments (Supplemental Fig. S6B), implying that the observed aggregate formation had triggered the chaperone machinery to minimize protein aggregation. Altogether, the beneficial phenotypic traits, the plant tissue morphology, and the transcriptome data suggest that Arabidopsis plants are able to accommodate the constitutive expression of APRs without cytotoxic side effects, thus allowing the expression of aggregation-induced knockdown phenotypes. GWD-Targeted Aggregation in Arabidopsis and Maize In order to prove that the targeted protein aggregation is usable for proteins with different functions and subcellular localizations, Arabidopsis plants expressing AtGWD534-541NFT and AtGWD821-829NFT SABs fused to GFP and designed to target the AtGWD protein (Fig. 1B; Supplemental Table S3) were evaluated for loss-of-function phenotypes and compared with the known Arabidopsis GWD mutant sex1-5 (Yu et al., 2001). The AtGWD1227-1234NFT-expressing plants were not analyzed because of lack of GFP fluorescence. T3 homozygous transgenic lines, each overexpressing AtGWD534-541NFT or AtGWD821-829NFT (Supplemental Fig. S6C), were grown in soil for 6 weeks. The rosettes of AtGWD821-829NFT-expressing plants were smaller than those of the wild type and comparable to those of the sex mutants (Fig. 6, A and B). In agreement with the phenotypic observations, a significant increase in starch content in the fourth and fifth leaves of 6-week-old Arabidopsis plants was detected by means of an iodine staining only in the line overexpressing the AtGWD821-829NFT-GFP construct (Fig. 6, C and D). Figure 6. Open in new tabDownload slide GWD-targeted aggregation in Arabidopsis. A and B, Phenotypes and rosette area quantification (n = 10) of 6-week-old Arabidopsis T3 transgenic plants expressing AtGWD534-541NFT-GFP and AtGWD821-829NFT-GFP, respectively. The wild type (Col-0) and sex1-5 mutant were used as controls (n = 10). C and D, Lugol staining and intensity color quantification of the 4th and 5th leaves from 6-week-old Arabidopsis T3 transgenic plants shown in A. The Lugol staining intensities are shown as grey values in pixels, with the values 0 and 250 pixels for black and white for 8-bit images, respectively (n = 4). Error bars represent sd, ***P<0.0001 (Student's t-test). n, Number of plants analyzed. Figure 6. Open in new tabDownload slide GWD-targeted aggregation in Arabidopsis. A and B, Phenotypes and rosette area quantification (n = 10) of 6-week-old Arabidopsis T3 transgenic plants expressing AtGWD534-541NFT-GFP and AtGWD821-829NFT-GFP, respectively. The wild type (Col-0) and sex1-5 mutant were used as controls (n = 10). C and D, Lugol staining and intensity color quantification of the 4th and 5th leaves from 6-week-old Arabidopsis T3 transgenic plants shown in A. The Lugol staining intensities are shown as grey values in pixels, with the values 0 and 250 pixels for black and white for 8-bit images, respectively (n = 4). Error bars represent sd, ***P<0.0001 (Student's t-test). n, Number of plants analyzed. T2 transgenic maize plants (Fig. 7) overexpressing ZmGWD599-610NFT-YFP and ZmGWD889-897NFT-YFP proteins (Supplemental Fig. S6D) that had been predicted to aggregate the maize GWD ortholog were evaluated for growth phenotypes and starch content. At least two segregating T2 transgenic lines per construct were analyzed. Eight-week-old mature plants from both lines showed mild growth retardation phenotypes in comparison to the B104 wild-type control when grown in the greenhouse (Fig. 7). Iodine staining of 10 leaf disks from the mature zone (Nelissen et al., 2012) of leaf 7 to leaf 10, collected at approximately 3 cm distance from each other, revealed a 10% starch increase in leaf 8 and leaf 9 of plants expressing ZmGWD599-610NFT-YFP and approximately 8% in leaf 9 and 10 of plants expressing ZmGWD889-897NFT-YFP (Fig. 7B; Supplemental Fig. S8). Figure 7. Open in new tabDownload slide GWD-targeted aggregation in maize. A, Phenotypes of 80-d-old T2 segregating maize lines expressing ZmGWD599-610NFT-YFP or ZmGWD889-897NFT-YFP from the mesophyll-specific promoter (pPePC) and compared with the wild-type control (B104). B, Lugol staining quantifications of mean gray values of leaf 7 to10 in plants shown in Supplemental Fig. S8, A and B. The Lugol staining intensities are presented as relative grey values. Error bars represent sd, *P<0.05 (Student's t-test). Figure 7. Open in new tabDownload slide GWD-targeted aggregation in maize. A, Phenotypes of 80-d-old T2 segregating maize lines expressing ZmGWD599-610NFT-YFP or ZmGWD889-897NFT-YFP from the mesophyll-specific promoter (pPePC) and compared with the wild-type control (B104). B, Lugol staining quantifications of mean gray values of leaf 7 to10 in plants shown in Supplemental Fig. S8, A and B. The Lugol staining intensities are presented as relative grey values. Error bars represent sd, *P<0.05 (Student's t-test). DISCUSSION Here, we demonstrated the potential of targeted aggregation to specifically down-regulate a protein function in plants without affecting the cellular viability and the overall plant fitness. The proposed method is based on the fact that protein aggregation is often mediated by short aggregation-prone segments of polypeptide chains that become exposed upon misfolding, leading to their assembly into intermolecular aggregates. Overall, this self-assembly process has been shown to be remarkably specific because most proteins are unable to coaggregate and protein deposits in patients affected by neurodegenerative diseases are highly enriched in one particular protein (Rajan et al., 2001; Ren et al., 2009). Although aggregate formation has been observed in bacteria, fungi, insects, invertebrates, and humans, in which it is usually associated with numerous diseases, this process is most probably ubiquitous across all the kingdoms, including plants. The aggregation propensity of the complete Arabidopsis proteome, analyzed by the protein aggregation prediction algorithm TANGO, was similar to that of other eukaryotes: 12% of the Arabidopsis proteome possesses a significant aggregation tendency versus 11.3% of the human proteome (Rousseau et al., 2006), suggesting that plants can be an attractive model to study protein aggregation. We show that it is possible to induce targeted aggregation of selected proteins in different locations in the plant cell and in different plant species. Our results indicate that the overexpression of APRs with a predicted high aggregation potential can trigger misfolding and subsequent aggregation of the endogenous proteins (e.g. ASKs and GWD), resulting in a conditional loss of their activity and leading to constitutive BR responses and increased amounts of starch, respectively. Thus, our data demonstrate that the coaggregation of polypeptides in plant cells also depends on the involvement of a short aggregation-nucleating region. The process of targeted aggregation shares similarities with the functional regulation of yeast prions, although important differences should be noted. Both yeast prion formation and targeted aggregation of ASKs or GWD appear to be governed by protein-specific aggregation without overall cell toxicity. This notion is supported by the observed phenotypes with an increased plant biomass and lack of defects at the ultrastructural level in various subcellular organelles, such as mitochondria and chloroplasts, for the ASK plants that overexpress APRs. In addition, a genome-wide expression study of these plants revealed very subtle changes in gene expression. The 5-fold increase of the HSP70 expression is most probably an adaptive change to the proteostatic network rather than a strong stress response. Although previous expression profile analyses of the Arabidopsis HSP70 genes have shown that these chaperones are indeed overexpressed in response to environmental stresses, such as heat, drought, and chemical treatments, the amplitude of the HSP70 up-regulation in these experiments ranged between a 15- to 20-fold change (Sugio et al., 2009). In addition, a high overexpression of the major cytosolic HSP70 in Arabidopsis had negative consequences for plant growth and viability (Sung and Guy, 2003). In contrast, no changes were observed in root growth. The profile of the transcriptional features of the responses to misfolded protein accumulation due to heat stress in the cytosol has revealed that the overall number of genes affected is much higher (2696 genes; Sugio et al., 2009) than that in our study (39 genes), hence excluding any proteotoxic effect generated by the overexpression of the APRs. Moreover, threshold effects were detected for the functional knockout of the selected proteins. Plants that displayed obvious ASK knockdown phenotypes accumulated perinuclear aggregates marked by GFP fluorescence (Fig. 2A; BIN2249-257RB and BIN2249-257NFT), as previously shown in yeast and mammalian cells (Kaganovich et al., 2008). When GFP was visible in inclusion bodies (Fig. 2A; BIN2249-257B), indicative of low aggregate solubility, no phenotypes were observed, similarly to the cytoplasmic APRs (Fig. 2A; BIN2249-257NF). In the case of the GWD protein, different APRs were tested in Arabidopsis and in maize. Likewise to the BIN2 knockdown, GWD loss-of-function phenotypes were generated with the APRs capable to form GFP-labeled aggregates inside the chloroplasts. Interestingly, the level of overexpression of the APRs and their protein knockdown capacities did not correlate. Overall, the APR-generated phenotypes were weaker than those of the known genetic mutants. When applied to multiple gene families, the efficiency of the targeted aggregation is most probably limited by the expression pattern of each gene member, the protein turnover, and the successful delivery of the APR to the respective targets. Therefore, tissue- and cell-specific promoters and specific sequences are essential to target the APRs to cellular organelles, as is the case for GWD. We demonstrated that the target-specific APRs can be used to selectively knockdown proteins in plants. Our results show that APRs with similar TANGO scores (49–52) and high sequence identity (89%, one mismatch) can bind the target proteins, whereas an APR with a TANGO score of 46, but bearing two mismatches (78% sequence identity), displayed a interaction loss. A complete lack of interaction was also observed when Val-251 and Ile-254 in the BIN2 APR were substituted by Pro that drastically reduces the aggregation propensity by breaking the β-strand structure (Richardson and Richardson, 2002) and, thus, lowers the TANGO score to 0, even with maintenance of the sequence identity (78%). In general, our data are in agreement with previous studies in which the specificity of APR-mediated interactions is determined by the combination of aggregation propensity and sequence matching (Ganesan et al., 2015). Besides its use as protein function-suppressing method, our approach emerges as a powerful tool to study protein aggregation mechanisms. Aggregation is often examined in the context of human diseases, in which aggregation of particular proteins is generally linked to lethal phenotypes. Virtually nothing is known about the mechanisms that control the self-assembly of proteins into aggregates in plants. The FTIR spectroscopy supports the hypothesis that β-sheet-containing aggregates are induced by the overexpression of the selected APRs. In addition, TEM combined with immunogold APR labeling in different Arabidopsis tissues has revealed that aggregated proteins accumulate both as amorphous clusters and as more ordered and elongated fibril-like structures, suggesting that aggregate formation in plants might differ from the known amyloid formation in mammalian cells. Overall, our data show that endogenous production or artificial introduction into a cell of small peptides with the APRs of a targeted protein will provide the opportunity to generate highly specific protein knockdowns posttranslationally in different plant species. This method has several potential advantages in respect to known knockdown approaches that act at the genomic or transcriptional level, such as T-DNA or transposon insertions and RNA-mediated gene suppression (RNA interference [RNAi], artificial microRNA, and antisense RNA) that can often suffer from significant drawbacks, such as off-target effects or systemic silencing. Additionally, sensitivity to environmental and developmental stresses and the observed trait instability also affect the efficiency of the RNA-silencing technology (Small, 2007; Frizzi and Huang, 2010). The APR peptides can also be expressed over several generations without silencing, therefore overcoming the phenotypic instability of the RNAi technology. As the synthetic APR-containing peptides can be targeted to different cellular compartments or be secreted in the apoplast, protein knockdowns with high selectivity can be obtained. The latter might generate the development of a knockdown strategy with applications in crop protection, when RNAi has failed to induce resistance against a number of pathogens (Price and Gatehouse, 2008). In comparison to the antibody-based technology, in which the produced recombinant proteins can have a low activity due to incorrect folding and often have low product yields and recovery problems (Ahmad et al., 2012), the intrinsic nature of the APRs to form β-sheet structures assures their structural stability when overexpressed in cells. MATERIALS AND METHODS In Silico Analysis Multiple sequences of ASKs and BIN2249-257 APR were aligned with the Clustal Omega program (Sievers et al., 2011), as well as AtGWD534-541, AtGWD821-829, AtGWD1227-1234, ZmGWD599-610, ZmGWD889-897, and ZmGWD1082-1088 APRs against their corresponding target protein fragments in AtGWD and ZmGWD sequences. The GWD gene of maize (Zea mays; GRMZM2G412611, UniProtKB annotation, A0A096TN87), orthologous to that of Arabidopsis (Arabidopsis thaliana; At1g10760) was identified with the PLAZA2.5 bioinformatics platform (Proost et al., 2009). The aggregation propensity of the GSK3/ASK proteins was calculated with the algorithm TANGO (Fernandez-Escamilla et al., 2004) that predicts aggregation-nucleating sequences in proteins. To ensure the discovery of all sequences matching a given APR within a certain number of mutations, we used an exhaustive algorithm that compares the APR sequence to all possible fragments of the same size in the proteome (Ganesan et al., 2015). Plant Material, Growth Conditions, and Plasmid Engineering Arabidopsis accessions Col-0 or Wassilewskija (Ws-2) and maize B104 inbred line (Hallauer et al., 1997) were used for transformation and 4-week-old Nicotiana benthamiana plants for leaf infiltration experiments. The bin2/bil1/bil2, bri1-5, and sex1-5 mutant lines had been described previously (Vert and Chory, 2006; Noguchi et al., 1999; Yu et al., 2001). The ASK, MUTE, and CLC genes and the BIN2249-257, BIN2P249-257, AtGWD534-541, AtGWD821-829, AtGWD1227-1234, ZmGWD599-610, and ZmGWD889- 897 sequences (Supplemental Table S6) were cloned in the pDONR221 vector (Invitrogen) and ASKs and PepC promoters in pDONR-P4P1 via Gateway cloning (Invitrogen). The PepC promoter sequence was derived from the pPTN512 vector (Sattarzadeh et al., 2010). The CaMV 35S promoter containing pEN-L4-2-R1 (Karimi et al., 2007) was also used, whereas the pEN-R2-F-L3 (Karimi et al., 2007), pEN-R2-citrineYFP-L3 (kind gift from Hilde Nelissen), and pDONR-P2R-TagRFP-P3 (Merzlyak et al., 2007) entry clones were used to generate translational fusions to GFP, YFP, or RFP in the pK7FWG2, pK7m34GW, pBb7m34GW, or pH7m34GW destination vectors (Karimi et al., 2007). For MUTE-GS fusions, the pDONR221-expressing MUTE (At3g06120) was fused N-terminally rather than C-terminally to the GS Rhino tag. For the N-terminal fusions, pDONR221-MUTE was cloned into the pkNGSrhino destination vector (Van Leene et al., 2015). For the C-terminal fusions, pDONR221-MUTE was recombined with pDONR-P2R-GSRhino-P3 and pDONR-P4-35S-P1 donor vectors (Van Leene et al., 2015). For BIN2-HA fusions, a pKm43GW destination vector (Karimi et al., 2007) was used overexpressing (CaMV 35S promoter) the BIN2 gene fused to 3× HA tag. The estradiol-inducible BIN2249-257NFT-RFP construct was engineered with the pMDC7-m13GW destination vector (Curtis and Grossniklaus, 2003). The sequence of the chloroplast transit peptide signal from the small ribulose-1,5-biphosphate carboxylase/oxygenase subunit of pea (Pisum sativum; Bowler et al., 1991; Supplemental Table S6) was included at the 5′ of the AtGWD534-541, AtGWD821-829, AtGWD1227-1234, ZmGWD599-610, and ZmGWD889-897 sequences. The ASK promoter sequences were recombined into the pMK7S*NFm14GW vector (Karimi et al., 2007) to generate transcriptional fusions to a nuclear localization signal (NLS)-GFP-GUS, of which the pBIN2::NLSGFP-GUS construct had been described previously (Gudesblat et al., 2012). For the BiFC experiments, BIN2249-257NFT, ASKs, and MUTE were fused to the N or C terminus of GFP fragments (nGFP or cGFP) as described (Boruc et al., 2010). As negative controls, constructs overexpressing nGFP or cGFP were used. Codon usage was optimized for the expression in Arabidopsis and maize. The resulting expression clones were transformed into Agrobacterium tumefaciens for plant transformation. For transient expression experiments, the abaxial sides of 4-week-old tobacco leaves were infiltrated with A. tumefaciens strains cultivated with the virulence gene activator acetosyringone as described (Boruc et al., 2010). For estradiol inductions, tobacco leaves were reinfiltrated with 20 μm estradiol 3 d after injection and imaged 24 h after induction with the ImageJ software (http://rsb.info.nih.gov/ij/). Arabidopsis seeds were stratified in the dark at 4°C for 2 d and germinated on half-strength Murashige and Skoog (MS) medium (1% [w/v] Suc) under long-day (16 h light/8 h dark) conditions at 20 to 22°C before transfer to soil. BRZ and estradiol were purchased from TCI Europe and Sigma-Aldrich, respectively. For the GUS activity analysis, 7-d-old Arabidopsis seedlings were processed as reported (Zhiponova et al., 2013). For maize transformation, immature embryos of the B104 inbred line were cocultivated for 3 d with A. tumefaciens EH101 containing the ZmGWD constructs in plasmid pBb7m34GW (Coussens et al., 2012). Cocultivated embryos were cultured in the dark for 1 week on nonselective medium and transformed embryogenic callus was selected for 10 weeks on phosphinothricin-containing medium. Transgenic rooted (T0) plantlets were induced in light on regeneration medium. The presence of the transgene was confirmed by PCR and a commercial phosphinothricin activity assay (TraitChek Crop and Grain Test Kit; Strategic Diagnostic) was used to test the selection marker activity. Transgenic T0 plants were grown to maturity in the greenhouse, back-crossed (BC) to the wild type B104. BC progenies were harvested and analyzed (T1). BC of T0 plants to the wild type B104 in a reciprocal way was done to secure sufficient transgenic T1 seed production. T1 plants were grown to maturity and self-fertilized. The resulting T2 seeds were also germinated and analyzed. The primers used are presented in Supplemental Table S4. Phenotype Analysis Root and hypocotyl lengths of vertically grown 7-d-old seedlings were measured with the ImageJ software (http://rsb.info.nih.gov/ij/). Rosette leaf areas were calculated with ImageJ on 5- to 6-week-old Arabidopsis plants grown in soil. Means and standard deviations were calculated with the Excel 2010 software and the statistical significance by the P values of a two-tailed Student's t test. For maize, the pictures of approximately 80-d-old T2 plants were taken during anthesis and compared to untransformed plants from each transformation event. For qualitative starch analyses, nonsenescent foliar tissues were boiled in 80% ethanol to remove chlorophyll and subsequently stained with Lugol iodine solution (Sigma-Aldrich). For Arabidopsis, the fourth and the fifth leaves of 6-week-old plants were taken, whereas for maize, 10 leaf disks per leaf, in leaves 7 to 10 of 8-week-old T2 segregating plants, were collected starting approximately 45 cm from the tip (punch number 1) until about 12 cm from the leaf base, keeping a distance of approximately 3 cm between each punch. For Lugol staining quantifications of starch, mean gray values in pixels were measured in 8-bit RGB images with the Image J software. A fixed area was measured for each leaf sample, setting the scale in pixels and to 0. In 8-bit images, the grayscale goes from a minimum value of 0 pixel (black) to a maximum value of 250 pixel (white). qRT-PCR and Microarray For qRT-PCR analyses, cDNA was prepared from 1 μg of total RNA extracted in technical triplicates from 7-d-old Arabidopsis seedlings or from 30-d-old maize leaf material from the mature zone (Nelissen et al., 2012) with the RNeasy Kit (Qiagen); qRT-PCR was run on a LightCycler 480 apparatus (Roche Diagnostics) with the SYBR Green I Master kit (Roche Diagnostics) or on a MyIQ cycler with the TaqMan master mix (Bio-Rad). Targets were quantified with specific primer pairs (Supplemental Table S6). Data were analyzed with the Biogazelle qBASEplus software (Hellemans et al., 2007) with the translation initiation factor elongation factor 1-α (EF1A), cyclin-dependent kinase A (CDKA;1), ubiquitin (UBQ), and heat shock factor 1 (HSF1) as reference genes. For maize, 18S rRNA (18S) was used as reference gene. For microarray analyses, 7-d-old seedlings of Arabidopsis Col-0 and expressing BIN2249-257NFT-GFP were grown vertically on half-strength MS medium. Total RNA was extracted from shoot material with TRIzol (Invitrogen) and further purified with the RNeasy Kit (Qiagen). Per array, 200 μg was used to hybridize the Arabidopsis ATH1 GeneChips (Affymetrix) at the VIB Nucleomics Core Facility (Leuven, Belgium; www.nucleomics.be) according to the manufacturer’s instructions. Raw data were processed with the RMA algorithm (Irizarry et al., 2003) within BioConductor with the ATH1-121501 chip definition file (www.bioconductor.org) to assign probes to genes, followed by a one-way ANOVA on all genes in parallel. P values were calculated with GenStat (Payne, 2012) and subsequently transformed into false discovery rates (Storey and Tibshirani, 2003) to identify differentially expressed genes. Microscopy Images of GUS-stained seedlings were taken with a MZ16 binocular microscope (Leica) and a Nikon 198 camera. Seven-day-old Arabidopsis seedlings and tobacco leaves were analyzed 3 to 4 d after injections with a FluoView1000 (Olympus) inverted confocal microscope equipped with a 63× water-corrected objective. Images were captured at 488- and 559-nm laser excitation and 500- to 550-nm and 570- to 670-nm long-pass emission filters for GFP and RFP, respectively. Emission fluorescence was captured in the frame-scanning mode and images were analyzed with the FluoView FV1000 software (Olympus). Intensity correlation analysis and Manders’ overlap coefficient calculations were done as described (Scacchi et al., 2009) by means of an ImageJ plug-in (http://wwwfacilities.uhnresearch.ca/wcif/imagej/colour_analysis.htm). For the BiFC experiments, the autofluorescence background level measured in tobacco leaves coexpressing the cGFP and nGFP constructs was used to set the GFP signal threshold. Combinations were scored as positive interactions when the GFP signal was higher than the threshold. For the morphological studies with TEM, fragments (1–2 mm2) of cotyledons, hypocotyls, and roots of 7-d-old BIN2249-257NFT-GFP and BIN2249-257RB-GFP, plants were embedded in Spurr’s resin as described (Betti et al., 2012). For immunocytochemical detection, the tissue fragments were infiltrated at 4°C in LR-White hard grade (London Resin) and immunolabeled with an anti-GFP antibody (AbCam) and secondary colloidal gold-protein A conjugates, PAG10nm (Cell Biology Department, Utrecht University) as described (Betti et al., 2012). Ultrathin sections were poststained at 20°C for 40 min in uranyl acetate and 10 min in lead citrate with an automatic contrasting system (EM AC20; Leica). Grids were viewed with a JEM-1010 TEM (JEOL) operating at 80 kV. Protein Extraction, Pull-Down, and Immunoblots To extract proteins for SDS-PAGE, flash-frozen 7-d-old Arabidopsis seedlings or tobacco leaves were ground and homogenized in ice-cold extraction buffer (50 mm Tris-HCl, pH 7.5, 150 mm NaCl, 1% [v/v] NP-40, 0.1% [v/v] SDS, 0.1% Na-deoxycholate, 5 mm DTT, and Complete protease inhibitor [Roche Diagnostics]). The homogenate was centrifuged at 14,000g twice for 20 min at 4°C and protein concentration was determined with Quick Start Bradford 1× dye reagent (Bio-Rad). Approximately 60 μg of total protein was separated on a 12% SDS-PAGE gel and transferred to polyvinylidene fluoride membranes (GE Healthcare). For pull-down experiments, proteins were extracted from flash-frozen tobacco leaves and immunoprecipitated with GFPTrap-A beads (Chromotek) as described (Roux et al., 2011). For BN-PAGE, total proteins were extracted from flash-frozen 7-d-old Arabidopsis seedlings or tobacco leaves, separated on Novex gels (Invitrogen), and transferred to polyvinylidene fluoride membranes as described (Xu et al., 2011). For immunodetection, mouse anti-GFP (JL-8, Living Colors; Clontech), rat anti-HA (Roche Diagnostics), mouse antiperoxidase (PAP; ab21867; AbCam), or rabbit anti-GWD antibodies (kind gift of Prof. Jeorg Fettke) were used as primary antibodies at 1:5,000 or 1:1,000 dilutions. Secondary anti-mouse, anti-rat, or anti-rabbit antibodies (GE Healthcare) were used at 1:10,000 dilutions. The proteins were detected by ECL (Perkin-Elmer). FTIR Spectroscopy FTIR was done on a Tensor 37 FT-IR spectrometer equipped with a BioATR II cell (Bruker) as described (Xu et al., 2011). Accession Numbers Sequence data from this article can be found in the GenBank/EMBL data libraries, in the Arabidopsis information Resource or in the Maize Genetics and Genomics databases under accession numbers: At4g18710 (BIN2/ASKη), At5g26751 (ASKα), At3g05840 (ASKγ), At5g14640 (ASKε), At2g30980 (BIL1/ASKζ), At1g06390 (BIL2/ASKι), At4g00720 (ASKθ), At3g61160 (ASKβ), At1g09840 (ASKκ), At1g57870 (ASKδ), At1g10760, (GWD), At3g06120 (MUTE), At2g2060 (CLC) and GRMZM2G412611 (ZmGWD). Supplemental Data The following materials are available. Supplemental Figure S1. Subcellular localization of the aggregates. Supplemental Figure S2. Biochemical analysis of the aggregates. Supplemental Figure S3. Colocalization analysis. Supplemental Figure S4. Absence of CLC binding by the BIN2 APR. Supplemental Figure S5. ASK promoter-GUS expression patterns. Supplemental Figure S6. Expression analysis of the transgenic lines. Supplemental Figure S7. TEM analysis. Supplemental Figure S8. Starch content analysis of the transgenic maize lines. Supplemental Figure S9. Full scans of blots. Supplemental Table S1. TANGO analysis. Supplemental Table S2. PepMatch results for selected APRs. Supplemental Table S3. Aggregation constructs design. Supplemental Table S4. Protein-protein interactions tested by bimolecular fluorescence complementation (BiFC). Supplemental Table S5. Microarray gene expression analysis. Supplemental Table S6. Sequence information. ACKNOWLEDGMENTS We thank Carina Braeckman for Arabidopsis transformation, Griet Coussens and Leen Vercruysse for maize transformation, Wilson Ardiles for sequencing support, Thao Bui Puong, Marieke Lippens, and Anaxi Houbaert for technical help, Jacinte Beerten for assistance with TEM, Mansour Karimi for cloning the pPepC sequence in the Gateway entry vector, Hilde Nelissen for the pEN-R2-citrineYFP-L3 vector, Daniël Van Damme for the 35S::CLC-HA construct, the VIB Nucleomics Core facility for microarray experiments, Prof. Joerg Fettke for the anti-GWD antibody, and Martine De Cock for help in preparing the manuscript Glossary APR aggregation-prone region GWD α-glucan water dikinase BR brassinosteroid SAB synthetic aggregating block CaMV cauliflower mosaic virus BN Blue Native FTIR Fourier transform infrared TEM transmission electron microscopy BiFC bimolecular fluorescence complementation BRZ brassinazole qRT-PCR quantitative reverse-transcription PCR LITERATURE CITED Ahmad P , Ashraf M, Younis M, Hu X, Kumar A, Akram NA, Al-Qurainy F ( 2012 ) Role of transgenic plants in agriculture and biopharming . Biotechnol Adv 30 : 524 – 540 Google Scholar Crossref Search ADS PubMed WorldCat Asami T , Min YK, Nagata N, Yamagishi K, Takatsuto S, Fujioka S, Murofushi N, Yamaguchi I, Yoshida S ( 2000 ) Characterization of brassinazole, a triazole-type brassinosteroid biosynthesis inhibitor . Plant Physiol 123 : 93 – 100 Google Scholar Crossref Search ADS PubMed WorldCat Bednarska NG , van Eldere J, Gallardo R, Ganesan A, Ramakers M, Vogel I, Baatsen P, Staes A, Goethals M, Hammarström P, et al. ( 2016 ) Protein aggregation as an antibiotic design strategy . Mol Microbiol 99 : 849 – 865 Google Scholar Crossref Search ADS PubMed WorldCat Betti C , Lico C, Maffi D, D’Angeli S, Altamura MM, Benvenuto E, Faoro F, Baschieri S ( 2012 ) Potato virus X movement in Nicotiana benthamiana: new details revealed by chimeric coat protein variants . Mol Plant Pathol 13 : 198 – 203 Google Scholar Crossref Search ADS PubMed WorldCat Boruc J , Mylle E, Duda M, De Clercq R, Rombauts S, Geelen D, Hilson P, Inzé D, Van Damme D, Russinova E ( 2010 ) Systematic localization of the Arabidopsis core cell cycle proteins reveals novel cell division complexes . Plant Physiol 152 : 553 – 565 Google Scholar Crossref Search ADS PubMed WorldCat Bowler C , Slooten L, Vandenbranden S, De Rycke R, Botterman J, Sybesma C, Van Montagu M, Inzé D ( 1991 ) Manganese superoxide dismutase can reduce cellular damage mediated by oxygen radicals in transgenic plants . EMBO J 10 : 1723 – 1732 Google Scholar Crossref Search ADS PubMed WorldCat Chiti F , Stefani M, Taddei N, Ramponi G, Dobson CM ( 2003 ) Rationalization of the effects of mutations on peptide and protein aggregation rates . Nature 424 : 805 – 808 Google Scholar Crossref Search ADS PubMed WorldCat Choe S , Dilkes BP, Fujioka S, Takatsuto S, Sakurai A, Feldmann KA ( 1998 ) The DWF4 gene of Arabidopsis encodes a cytochrome P450 that mediates multiple 22α-hydroxylation steps in brassinosteroid biosynthesis . Plant Cell 10 : 231 – 243 Google Scholar PubMed OpenURL Placeholder Text WorldCat Choe S , Schmitz RJ, Fujioka S, Takatsuto S, Lee M-O, Yoshida S, Feldmann KA, Tax FE ( 2002 ) Arabidopsis brassinosteroid-insensitive dwarf12 mutants are semidominant and defective in a glycogen synthase kinase 3β-like kinase . Plant Physiol 130 : 1506 – 1515 Google Scholar Crossref Search ADS PubMed WorldCat Coussens G , Aesaert S, Verelst W, Demeulenaere M, De Buck S, Njuguna E, Inzé D, Van Lijsebettens M ( 2012 ) Brachypodium distachyon promoters as efficient building blocks for transgenic research in maize . J Exp Bot 63 : 4263 – 4273 Google Scholar Crossref Search ADS PubMed WorldCat Curtis MD , Grossniklaus U ( 2003 ) A gateway cloning vector set for high-throughput functional analysis of genes in planta . Plant Physiol 133 : 462 – 469 Google Scholar Crossref Search ADS PubMed WorldCat De Baets G , Van Durme J, Rousseau F, Schymkowitz J ( 2014 ) A genome-wide sequence-structure analysis suggests aggregation gatekeepers constitute an evolutionary constrained functional class . J Mol Biol 426 : 2405 – 2412 Google Scholar Crossref Search ADS PubMed WorldCat De Rybel B , Audenaert D, Vert G, Rozhon W, Mayerhofer J, Peelman F, Coutuer S, Denayer T, Jansen L, Nguyen L, et al. ( 2009 ) Chemical inhibition of a subset of Arabidopsis thaliana GSK3-like kinases activates brassinosteroid signaling . Chem Biol 16 : 594 – 604 Google Scholar Crossref Search ADS PubMed WorldCat Fernandez-Escamilla A-M , Rousseau F, Schymkowitz J, Serrano L ( 2004 ) Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins . Nat Biotechnol 22 : 1302 – 1306 Google Scholar Crossref Search ADS PubMed WorldCat Fitzpatrick AW , Knowles TP, Waudby CA, Vendruscolo M, Dobson CM ( 2011 ) Inversion of the balance between hydrophobic and hydrogen bonding interactions in protein folding and aggregation . PLoS Comput Biol 7 : e1002169 Google Scholar Crossref Search ADS PubMed WorldCat Frizzi A , Huang S ( 2010 ) Tapping RNA silencing pathways for plant biotechnology . Plant Biotechnol J 8 : 655 – 677 Google Scholar Crossref Search ADS PubMed WorldCat Ganesan A , Debulpaep M, Wilkinson H, Van Durme J, De Baets G, Jonckheere W, Ramakers M, Ivarsson Y, Zimmermann P, Van Eldere J, Schymkowitz J, Rousseau F ( 2015 ) Selectivity of aggregation-determining interactions . J Mol Biol 427 : 236 – 247 Google Scholar Crossref Search ADS PubMed WorldCat Goldschmidt L , Teng PK, Riek R, Eisenberg D ( 2010 ) Identifying the amylome, proteins capable of forming amyloid-like fibrils . Proc Natl Acad Sci USA 107 : 3487 – 3492 Google Scholar Crossref Search ADS PubMed WorldCat Gudesblat GE , Schneider-Pizoń J, Betti C, Mayerhofer J, Vanhoutte I, van Dongen W, Boeren S, Zhiponova M, de Vries S, Jonak C, Russinova E ( 2012 ) SPEECHLESS integrates brassinosteroid and stomata signalling pathways . Nat Cell Biol 14 : 548 – 554 Google Scholar Crossref Search ADS PubMed WorldCat Hallauer AR , Lamkey KR, White PR ( 1997 ) Registration of five inbred lines of maize: B102, B103, B104, B105, and B106 . Crop Sci 37 : 1405 – 1406 Google Scholar Crossref Search ADS WorldCat Hellemans J , Mortier G, De Paepe A, Speleman F, Vandesompele J ( 2007 ) qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data . Genome Biol 8 : R19 Google Scholar Crossref Search ADS PubMed WorldCat Irizarry RA , Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP ( 2003 ) Exploration, normalization, and summaries of high density oligonucleotide array probe level data . Biostatistics 4 : 249 – 264 Google Scholar Crossref Search ADS PubMed WorldCat Jucker M , Walker LC ( 2013 ) Self-propagation of pathogenic protein aggregates in neurodegenerative diseases . Nature 501 : 45 – 51 Google Scholar Crossref Search ADS PubMed WorldCat Kaganovich D , Kopito R, Frydman J ( 2008 ) Misfolded proteins partition between two distinct quality control compartments . Nature 454 : 1088 – 1095 Google Scholar Crossref Search ADS PubMed WorldCat Karimi M , Bleys A, Vanderhaeghen R, Hilson P ( 2007 ) Building blocks for plant gene assembly . Plant Physiol 145 : 1183 – 1191 Google Scholar Crossref Search ADS PubMed WorldCat Kim T-W , Guan S, Sun Y, Deng Z, Tang W, Shang J-X, Sun Y, Burlingame AL, Wang Z-Y ( 2009 ) Brassinosteroid signal transduction from cell-surface receptor kinases to nuclear transcription factors . Nat Cell Biol 11 : 1254 – 1260 Google Scholar Crossref Search ADS PubMed WorldCat Krebs MR , Morozova-Roche LA, Daniel K, Robinson CV, Dobson CM ( 2004 ) Observation of sequence specificity in the seeding of protein amyloid fibrils . Protein Sci 13 : 1933 – 1938 Google Scholar Crossref Search ADS PubMed WorldCat Li J , Nam KH ( 2002 ) Regulation of brassinosteroid signaling by a GSK3/SHAGGY-like kinase . Science 295 : 1299 – 1301 Google Scholar PubMed OpenURL Placeholder Text WorldCat López De La Paz M , Goldie K, Zurdo J, Lacroix E, Dobson CM, Hoenger A, Serrano L ( 2002 ) De novo designed peptide-based amyloid fibrils . Proc Natl Acad Sci USA 99 : 16052 – 16057 Google Scholar Crossref Search ADS PubMed WorldCat Lorberth R , Ritte G, Willmitzer L, Kossmann J ( 1998 ) Inhibition of a starch-granule-bound protein leads to modified starch and repression of cold sweetening . Nat Biotechnol 16 : 473 – 477 Google Scholar Crossref Search ADS PubMed WorldCat Makin OS , Atkins E, Sikorski P, Johansson J, Serpell LC ( 2005 ) Molecular basis for amyloid fibril formation and stability . Proc Natl Acad Sci USA 102 : 315 – 320 Google Scholar Crossref Search ADS PubMed WorldCat Merzlyak EM , Goedhart J, Shcherbo D, Bulina ME, Shcheglov AS, Fradkov AF, Gaintzeva A, Lukyanov KA, Lukyanov S, Gadella TWJ, Chudakov DM ( 2007 ) Bright monomeric red fluorescent protein with an extended fluorescence lifetime . Nat Methods 4 : 555 – 557 Google Scholar Crossref Search ADS PubMed WorldCat Mitraki A ( 2010 ) Protein aggregation from inclusion bodies to amyloid and biomaterials . Adv Protein Chem Struct Biol 79 : 89 – 125 Google Scholar Crossref Search ADS PubMed WorldCat Mitsui T , Itoh K, Hori H, Ito H ( 2010 ) Biosynthesis and degradation of starch . Bull Facul Agric Niigata Univ 62 : 49 – 73 Google Scholar OpenURL Placeholder Text WorldCat Nelissen H , Rymen B, Jikumaru Y, Demuynck K, Van Lijsebettens M, Kamiya Y, Inzé D, Beemster GTS ( 2012 ) A local maximum in gibberellin levels regulates maize leaf growth by spatial control of cell division . Curr Biol 22 : 1183 – 1187 Google Scholar Crossref Search ADS PubMed WorldCat Nelson R , Sawaya MR, Balbirnie M, Madsen AØ, Riekel C, Grothe R, Eisenberg D ( 2005 ) Structure of the cross-β spine of amyloid-like fibrils . Nature 435 : 773 – 778 Google Scholar Crossref Search ADS PubMed WorldCat Noguchi T , Fujioka S, Choe S, Takatsuto S, Yoshida S, Yuan H, Feldmann KA, Tax FE ( 1999 ) Brassinosteroid-insensitive dwarf mutants of Arabidopsis accumulate brassinosteroids . Plant Physiol 121 : 743 – 752 Google Scholar Crossref Search ADS PubMed WorldCat O’Nuallain B , Williams AD, Westermark P, Wetzel R ( 2004 ) Seeding specificity in amyloid growth induced by heterologous fibrils . J Biol Chem 279 : 17490 – 17499 Google Scholar Crossref Search ADS PubMed WorldCat Payne RW ( 2012 ) GenStat Release 15 Reference Manual, Part 3 Procedure Library PL21. VSN International , Hemel Hempstead, UK . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pillitteri LJ , Sloan DB, Bogenschutz NL, Torii KU ( 2007 ) Termination of asymmetric cell division and differentiation of stomata . Nature 445 : 501 – 505 Google Scholar Crossref Search ADS PubMed WorldCat Price DRG , Gatehouse JA ( 2008 ) RNAi-mediated crop protection against insects . Trends Biotechnol 26 : 393 – 400 Google Scholar Crossref Search ADS PubMed WorldCat Proost S , Van Bel M, Sterck L, Billiau K, Van Parys T, Van de Peer Y, Vandepoele K ( 2009 ) PLAZA: a comparative genomics resource to study gene and genome evolution in plants . Plant Cell 21 : 3718 – 3731 Google Scholar Crossref Search ADS PubMed WorldCat Rajan RS , Illing ME, Bence NF, Kopito RR ( 2001 ) Specificity in intracellular protein aggregation and inclusion body formation . Proc Natl Acad Sci USA 98 : 13060 – 13065 Google Scholar Crossref Search ADS PubMed WorldCat Ren P-H , Lauckner JE, Kachirskaia I, Heuser JE, Melki R, Kopito RR ( 2009 ) Cytoplasmic penetration and persistent infection of mammalian cells by polyglutamine aggregates . Nat Cell Biol 11 : 219 – 225 Google Scholar Crossref Search ADS PubMed WorldCat Richardson JS , Richardson DC ( 2002 ) Natural β-sheet proteins use negative design to avoid edge-to-edge aggregation . Proc Natl Acad Sci USA 99 : 2754 – 2759 Google Scholar Crossref Search ADS PubMed WorldCat Rousseau F , Serrano L, Schymkowitz JWH ( 2006 ) How evolutionary pressure against protein aggregation shaped chaperone specificity . J Mol Biol 355 : 1037 – 1047 Google Scholar Crossref Search ADS PubMed WorldCat Roux M , Schwessinger B, Albrecht C, Chinchilla D, Jones A, Holton N, Malinovsky FG, Tör M, de Vries S, Zipfel C ( 2011 ) The Arabidopsis leucine-rich repeat receptor-like kinases BAK1/SERK3 and BKK1/SERK4 are required for innate immunity to hemibiotrophic and biotrophic pathogens . Plant Cell 23 : 2440 – 2455 Google Scholar Crossref Search ADS PubMed WorldCat Rozhon W , Mayerhofer J, Petutschnig E, Fujioka S, Jonak C ( 2010 ) ASKθ, a group-III Arabidopsis GSK3, functions in the brassinosteroid signalling pathway . Plant J 62 : 215 – 223 Google Scholar Crossref Search ADS PubMed WorldCat Sabaté R , Espargaró A, de Groot NS, Valle-Delgado JJ, Fernàndez-Busquets X, Ventura S ( 2010 ) The role of protein sequence and amino acid composition in amyloid formation: scrambling and backward reading of IAPP amyloid fibrils . J Mol Biol 404 : 337 – 352 Google Scholar Crossref Search ADS PubMed WorldCat Sattarzadeh A , Fuller J, Moguel S, Wostrikoff K, Sato S, Covshoff S, Clemente T, Hanson M, Stern DB ( 2010 ) Transgenic maize lines with cell-type specific expression of fluorescent proteins in plastids . Plant Biotechnol J 8 : 112 – 125 Google Scholar Crossref Search ADS PubMed WorldCat Sawaya MR , Sambashivan S, Nelson R, Ivanova MI, Sievers SA, Apostol MI, Thompson MJ, Balbirnie M, Wiltzius JJ, McFarlane HT, et al. ( 2007 ) Atomic structures of amyloid cross-β spines reveal varied steric zippers . Nature 447 : 453 – 457 Google Scholar Crossref Search ADS PubMed WorldCat Scacchi E , Osmont KS, Beuchat J, Salinas P, Navarrete-Gómez M, Trigueros M, Ferrándiz C, Hardtke CS ( 2009 ) Dynamic, auxin-responsive plasma membrane-to-nucleus movement of Arabidopsis BRX . Development 136 : 2059 – 2067 Google Scholar Crossref Search ADS PubMed WorldCat Sievers F , Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J, Thompson JD, Higgins DG ( 2011 ) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega . Mol Syst Biol 7 : 539 Google Scholar Crossref Search ADS PubMed WorldCat Small I ( 2007 ) RNAi for revealing and engineering plant gene functions . Curr Opin Biotechnol 18 : 148 – 153 Google Scholar Crossref Search ADS PubMed WorldCat Storey JD , Tibshirani R ( 2003 ) Statistical significance for genomewide studies . Proc Natl Acad Sci USA 100 : 9440 – 9445 Google Scholar Crossref Search ADS PubMed WorldCat Sugio A , Dreos R, Aparicio F, Maule AJ ( 2009 ) The cytosolic protein response as a subcomponent of the wider heat shock response in Arabidopsis. Plant Cell 21 : 642 – 654 Google Scholar Crossref Search ADS PubMed WorldCat Sung DY , Guy CL ( 2003 ) Physiological and molecular assessment of altered expression of Hsc70-1 in Arabidopsis. Evidence for pleiotropic consequences . Plant Physiol 132 : 979 – 987 Google Scholar Crossref Search ADS PubMed WorldCat Surmacz-Chwedoruk W , Babenko V, Dzwolak W ( 2014 ) Master and slave relationship between two types of self-propagating insulin amyloid fibrils . J Phys Chem B 118 : 13582 – 13589 Google Scholar Crossref Search ADS PubMed WorldCat Szekeres M , Németh K, Koncz-Kálmán Z, Mathur J, Kauschmann A, Altmann T, Rédei GP, Nagy F, Schell J, Koncz C ( 1996 ) Brassinosteroids rescue the deficiency of CYP90, a cytochrome P450, controlling cell elongation and de-etiolation in Arabidopsis . Cell 85 : 171 – 182 Google Scholar Crossref Search ADS PubMed WorldCat Tyedmers J , Mogk A, Bukau B ( 2010 ) Cellular strategies for controlling protein aggregation . Nat Rev Mol Cell Biol 11 : 777 – 788 Google Scholar Crossref Search ADS PubMed WorldCat Van Leene J , Eeckhout D, Cannoot B, De Winne N, Persiau G, Van De Slijke E, Vercruysse L, Dedecker M, Verkest A, Vandepoele K, et al. ( 2015 ) An improved toolbox to unravel the plant cellular machinery by tandem affinity purification of Arabidopsis protein complexes . Nat Protoc 10 : 169 – 187 Google Scholar Crossref Search ADS PubMed WorldCat Ventura S , Zurdo J, Narayanan S, Parreño M, Mangues R, Reif B, Chiti F, Giannoni E, Dobson CM, Aviles FX, Serrano L ( 2004 ) Short amino acid stretches can mediate amyloid formation in globular proteins: the Src homology 3 (SH3) case . Proc Natl Acad Sci USA 101 : 7258 – 7263 Google Scholar Crossref Search ADS PubMed WorldCat Vert G , Chory J ( 2006 ) Downstream nuclear events in brassinosteroid signalling . Nature 441 : 96 – 100 Google Scholar Crossref Search ADS PubMed WorldCat Wang ZY , Nakano T, Gendron J, He J, Chen M, Vafeados D, Yang Y, Fujioka S, Yoshida S, Asami T, Chory J ( 2002 ) Nuclear-localized BZR1 mediates brassinosteroid-induced growth and feedback suppression of brassinosteroid biosynthesis . Dev Cell 2 : 505 – 513 Google Scholar Crossref Search ADS PubMed WorldCat Xu J , Reumers J, Couceiro JR, De Smet F, Gallardo R, Rudyak S, Cornelis A, Rozenski J, Zwolinska A, Marine J-C, et al. ( 2011 ) Gain of function of mutant p53 by coaggregation with multiple tumor suppressors . Nat Chem Biol 7 : 285 – 295 Google Scholar Crossref Search ADS PubMed WorldCat Yan Z , Zhao J, Peng P, Chihara RK, Li J ( 2009 ) BIN2 functions redundantly with other Arabidopsis GSK3-like kinases to regulate brassinosteroid signaling . Plant Physiol 150 : 710 – 721 Google Scholar Crossref Search ADS PubMed WorldCat Youn JH , Kim TW ( 2015 ) Functional insights of plant GSK3-like kinases: multi-taskers in diverse cellular signal transduction pathways . Mol Plant 8 : 552 – 565 Google Scholar Crossref Search ADS PubMed WorldCat Yu T-S , Kofler H, Häusler RE, Hille D, Flügge U-I, Zeeman SC, Smith AM, Kossmann J, Lloyd J, Ritte G, et al. ( 2001 ) The Arabidopsis sex1 mutant is defective in the R1 protein, a general regulator of starch degradation in plants, and not in the chloroplast hexose transporter . Plant Cell 13 : 1907 – 1918 Google Scholar Crossref Search ADS PubMed WorldCat Zhiponova MK , Vanhoutte I, Boudolf V, Betti C, Dhondt S, Coppens F, Mylle E, Maes S, González-García M-P, Caño-Delgado AI, et al. ( 2013 ) Brassinosteroid production and signaling differentially control cell division and expansion in the leaf . New Phytol 197 : 490 – 502 Google Scholar Crossref Search ADS PubMed WorldCat Author notes 1 This work was supported by grants from the Agency for Innovation by Science and Technology (“Strategisch Basisonderzoek” project no. 60839), Ghent University (“Industrieel Onderzoeksfonds” F2014/IOF-StarrTT261 and Multidisciplinary Research Partnership “Biotechnology for a Sustainable Economy” no. 01MRB510W), the Research Foundation-Flanders (Joint Project Bulgarian Academy of Sciences VS.025.13N), the Interuniversity Attraction Poles Program (IUAP VII/29), initiated by the Belgian State, Science Policy Office, University of Leuven, and the European Research Council under the European Union's Horizon 2020 Framework Programme (ERC Grant agreement 647458). 2 Present address: Bekintex at Bekaert NV, 8000 Bruges, Belgium. 3 Present address: Institute of Plant Physiology and Genetics, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria. 4 Present address: Dana Farber Cancer Institute and the Broad Institute/MIT, Cambridge, MA, 02141. 5 Present address: State Key Laboratory for Oncogenes and Related Genes, Division of Gastroenterology and Hepatology, Renji Hospital, Shanghai Institute for Digestive Diseases, Shanghai Jiao-Tong University School of Medicine, Shanghai 200001, China. * Address correspondence to frederic.rousseau@switch.vib-kuleuven.be, joost.schymkowitz@switch.vib-kuleuven.be, or eugenia.russinova@psb.vib-ugent.be. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is Eugenia Russinova (eugenia.russinova@psb.vib-ugent.be). D.I., F.R., J.S., and E.R. conceived the research and designed and supervised the experiments; S.C. initiated the work; C.B. designed and performed most of the experiments; I.V. provided technical assistance to C.B.; R.D.R. performed the TEM work; K.M. and M.V. did the expression studies and data analysis; S.A. and M.V.L. made transgenic maize lines; S.A., D.R., M.V.L., F.V.B., and D.I. contributed to the maize analyses; R.G., F.D.S., and J.X. performed the FTIR and expression analysis; C.B., F.R., J.S., and E.R. wrote the article with contributions of all the authors. [OPEN] Articles can be viewed without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.16.00335 © 2016 American Society of Plant Biologists. All Rights Reserved. © The Author(s) 2016. Published by Oxford University Press on behalf of American Society of Plant Biologists. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
TI - Sequence-Specific Protein Aggregation Generates Defined Protein Knockdowns in Plants  
JF - Plant Physiology
DO - 10.1104/pp.16.00335
DA - 2016-06-10
UR - https://www.deepdyve.com/lp/oxford-university-press/sequence-specific-protein-aggregation-generates-defined-protein-Nda3beoI0l
SP - 773
EP - 787
VL - 171
IS - 2
DP - DeepDyve
ER -