A Novel Thioredoxin h Is Secreted in Nicotiana alata and Reduces S-RNase in Vitro*

Thioredoxins type h are classified into three subgroups. The subgroup II includes thioredoxins containing an N-terminal extension, the role of which is still unclear. Although thioredoxin secretion has been observed in animal cells, there is no evidence suggesting that any thioredoxin h is secreted in plants. In this study, we report that a thioredoxin h, subgroup II, from Nicotiana alata (NaTrxh) is secreted into the extracellular matrix of the stylar transmitting tract tissue. Fractionation studies showed that NaTrxh is extracted along with well characterized secretion proteins such as S-RNases and NaTTS (N. alata transmitting tissue-specific protein). Moreover, an NaTrxh-green fluorescent fusion protein transiently expressed in Nicotiana benthamiana and Arabidopsis thaliana leaves was also secreted, showing that NaTrxh has the required information for its secretion. We performed reduction assays in vitro to identify potential extracellular targets of NaTrxh. We found that S-RNase is one of the several potential substrates of the NaTrxh in the extracellular matrix. In addition, we proved by affinity chromatography that NaTrxh specifically interacts with S-RNase. Our findings showed that NaTrxh is a new thioredoxin h in Nicotiana that is secreted as well as in animal systems. Because NaTrxh is localized in the extracellular matrix of the stylar transmitting tract and its specific interaction with S-RNase to reduce it in vitro, we suggest that this thioredoxin h may be involved either in general pollenpistil interaction processes or particularly in S-RNase-based self-incompatibility.

Thioredoxins type h are classified into three subgroups. The subgroup II includes thioredoxins containing an N-terminal extension, the role of which is still unclear. Although thioredoxin secretion has been observed in animal cells, there is no evidence suggesting that any thioredoxin h is secreted in plants. In this study, we report that a thioredoxin h, subgroup II, from Nicotiana alata (NaTrxh) is secreted into the extracellular matrix of the stylar transmitting tract tissue. Fractionation studies showed that NaTrxh is extracted along with well characterized secretion proteins such as S-RNases and NaTTS (N. alata transmitting tissue-specific protein). Moreover, an NaTrxh-green fluorescent fusion protein transiently expressed in Nicotiana benthamiana and Arabidopsis thaliana leaves was also secreted, showing that NaTrxh has the required information for its secretion. We performed reduction assays in vitro to identify potential extracellular targets of NaTrxh. We found that S-RNase is one of the several potential substrates of the NaTrxh in the extracellular matrix. In addition, we proved by affinity chromatography that NaTrxh specifically interacts with S-RNase. Our findings showed that NaTrxh is a new thioredoxin h in Nicotiana that is secreted as well as in animal systems. Because NaTrxh is localized in the extracellular matrix of the stylar transmitting tract and its specific interaction with S-RNase to reduce it in vitro, we suggest that this thioredoxin h may be involved either in general pollenpistil interaction processes or particularly in S-RNase-based self-incompatibility.
Thioredoxins (Trxs) 2 are small conserved proteins that play an important role in cellular redox regulation. When the Trx active site WCGPC is reduced, it is able to reduce the disulfide bonds of target proteins. Trxs are widely distributed in nature from prokaryotes to eukaryotes. In photosynthetic organisms, Trxs have been shown to be highly polymorphic and to participate in several central cellular processes (1,2). The diversity of physiological roles in which Trxs participate depends entirely on the target proteins (3)(4)(5). The 20 genes encoding Trxs in the Arabidopsis thaliana genome (6) and the proteomic analysis (2) of their targeted proteins by Trx h3 in A. thaliana reinforce the wide range of functions in which these proteins are involved.
Although Trxs h have traditionally been considered to be cytoplasmic proteins, recent analyses (13) revealed that they can be further differentiated into three subgroups. Subgroup II includes Trxs h with N-terminal extensions. The biochemical function of these extensions remains unclear, and the available algorithms do not predict any targeting signal. Gelhaye et al. (14) reported on a Trx h subgroup II from poplar (Populus tremula), PtTrxh2, the N-terminal extension of which is necessary to target this Trx h to mitochondria. In addition, RPP13-1, a Trx h present in the rice (Oryza sativa) phloem sap, mediates its own transport from cell to cell through plasmodesmata only if its N-terminal extension is present (15,16).
Here, we report an analysis of a Trx h that belongs to the subgroup II, from Nicotiana alata, called NaTrxh. The N-terminal extension of NaTrxh was not recognizable as a secretion signal, and yet we demonstrated that NaTrxh accumulates in the extracellular matrix (ECM) of the stylar transmitting tract (TT) in N. alata. Moreover, fusion of NaTrxh to the green fluorescent protein (GFP) directed secretion of this fusion protein in two heterologous systems. We used two separate biochemical approaches, monobromobimane (mBBr) labeling and 4-aminophenylarsine oxide (PAO) chromatography, to test potential substrates of NaTrxh. S-RNase was clearly among the proteins reduced by NaTrxh, raising the possibility that it is involved in pollination, particularly in self-incompatibility (SI).
ana benthamiana and A. thaliana Columbia (Col-0) ecotype plants were grown under greenhouse conditions using a 16:8 photoperiod at 22°C.
NaTrxh-GFP Fusion Constructs and Transient Expression-An NaTrxh cDNA with BamHI and NcoI sites was produced by PCR and fused in-frame to the N-terminal of the GFP in the pHBT vector (20). Primers used are as follows: forward, 5Ј-CGCGCGGATCCATGG-GATCGTATCTTTCAA-3Ј; reverse, 5Ј-AGCCATGGTTGGACATG-3Ј. The NaTrxh-GFP fusion was cloned into pBIN19 (21) under control of the 35S cauliflower mosaic virus (CaMV35S) promoter.
After Agrobacterium tumefaciens pGV2260 (22) transformation, an agroinfiltration of N. benthamiana leaves was carried out, as described previously (23,24). The protein expression was observed by confocal microscopy after 3 days. A. thaliana Col-0 transformation was carried out by bombarding (25) 2-week-old plantlets (grown aseptically in 0.5ϫ Murashige and Skoog with 1% sucrose) with DNA-coated tungsten particles (Tungsten M-17, Bio-Rad) containing the NaTrxh-GFP construct cloned into the pHBT vector (20). The protein expression was observed by confocal microscopy after 1 week.
Protein Assay-Protein concentrations were determined as described previously (27), using bovine serum albumin (BSA) as standard.
Reductase Activity Assay-Recombinant NaTrxh (NaTrxh rec ) and recombinant Escherichia coli thioredoxin (Sigma) were evaluated for their ability to reduce insulin disulfide bonds as described previously (28). The ability of the NADPH/NTR system to reduce NaTrxh rec was evaluated as follows: NaTrxh rec (2.5 g) and E. coli Trx (2.5 g) in 50 mM Tris⅐HCl, pH 7.9, were incubated at 37°C with 2.0 g of recombinant E. coli NTR and 0.125 mol of NADPH in a final volume of 100 l. After incubation, 0.2 mol of mBBr in 10 l of acetonitrile was added, and the samples were incubated for 20 min at room temperature. Proteins were boiled for 5 min in SDS sample buffer free of reducing agents, separated in 10 -20% gradient SDS-PAGE, and visualized under UV light (Fluor-S, Fuji Corp Saddle Brook, NJ).
NaTrxh-Affi-Gel Affinity Column and Purification of Anti-NaTrxh Antibody-NaTrxh rec (16 mg) was immobilized to Affi-Gel-10 (Bio-Rad), as recommended by the manufacturer. The anti-NaTrxh rec serum was precipitated with 50% saturated ammonium sulfate. After desalting, this IgG fraction was passed over the NaTrxh-Affi-Gel affinity column.
Specific antibodies against NaTrxh rec were eluted with 50 mM glycine, 50 mM NaCl, pH 2.6. The samples were neutralized by the addition of 1 M Tris.
Phylogenetic Analysis-Amino acid sequences of plant Trxs were aligned by Clustal X (29). Based on the alignment, a phylogenetic tree was constructed by the neighbor-joining method (30). A phylogenetic test was done based on 1,000 bootstrap replicates. Phylogenetic analysis was conducted using MEGA version 2.1 (31). The GenBank TM accession numbers used for this analysis are as follows: Trx type f, A. thaliana Protein Gel Blot Analysis and Immunostaining-Proteins were fractionated in a 12.5% SDS-PAGE, blotted onto nitrocellulose, and immunostained with anti-S 105 -RNase (1:10,000 dilution), anti-NaTTS (1:10,000 dilution) (32), or anti-NaTrxh rec (1:1,000 dilution).
In Vitro Identification of NaTrxh Target Proteins-Reduction of the disulfide bonds of target proteins was determined using the two-dimensional SDS-PAGE system (33) using the low salt buffer extract without Na 2 S 2 O 4 . Before the first dimension, proteins are labeled with mBBr after non-reducing (control) or reducing (with dithiothreitol (DTT) or NaTrxh rec plus NADPH and NTR) conditions. The second dimension gel was observed under UV light (Fluor-S, Fuji Corp., Saddle Brook, NJ).
Low salt protein extracts from N. alata S 105 S 105 in 50 mM Tris⅐HCl, pH 7.9, were passed over PAO resin (Invitrogen) after reduction by NaTrxh rec , 20 mM 2-mercaptoethanol, or without reduction. After washing, the unbound fraction was collected. Bound proteins were eluted by sequentially washing the resin twice with 1 mM 2-mercaptoethanol containing buffer, five times with 5 mM 2-mercaptoethanol, and three times with 500 mM 2-mercaptoethanol buffer. All the fractions were concentrated by lyophilization and were analyzed by 12.5% SDS-PAGE. Gels were blotted onto nitrocellulose and immunostained with anti-S 105 -RNase antibody.
In Vitro Protein-Protein Interactions by Affinity Chromatography-Protein extracts (1 mg) from N. alata S 105 S 105 styles were obtained with binding buffer (BB, 50 mM Tris⅐HCl, pH 7.9) and passed over the NaTrxh-Affi-Gel affinity column. After recovering the unbound fraction, 10 bed volume washes were done with BB, and then the column was sequentially washed as follows: (a) BB plus 1% Tween 20, (b) BB plus 0.1 M NaCl, and (c) BB plus 0.2 M NaCl. Stronger interacting proteins were eluted with elution buffer (50 mM glycine, 50 mM NaCl, pH 2.6). The samples were neutralized by the addition of 1 M Tris. Fractions were concentrated by cold acetone precipitation and were analyzed by 12.5% SDS-PAGE. Gels were blotted onto nitrocellulose and immunostained with anti-S 105 -RNase antibody.

RESULTS
Isolation of the Trx h cDNA-We isolated a cDNA (AFLP25F), which is differentially expressed between N. alata cv Breakthrough, an SC mutant plant that does not express the S-RNase, and SC N. plumbaginifolia. The full-length cDNA sequence of this transcript (Fig. 1A) showed extensive sequence similarity with Trx genes from plants. The predicted open reading frame contains the sequence WCGPC, described as the conserved Trx active site (3,4).
To identify the type of Trx encoded by the AFLP25F cDNA (i.e. m, f, x, or h), we performed a phylogenetic analysis using the deduced AFLP25F amino acid sequence and other plant Trxs. The phylogenetic tree in Fig. 1B displays five major groups. Four of those groups correspond to organellar thioredoxins (f, m, x, and o types). The AFLP25F protein sequence clearly clusters with Trxs h, subgroup II. The AFLP25F product was named NaTrxh.
To evaluate the disulfide reductase activity of NaTrxh, we expressed this gene in E. coli as a GST fusion protein. Fig. 2A shows the recombinant NaTrxh (NaTrxh rec ) purified after cleavage from GST. Its ability to reduce insulin disulfide bonds using DTT as an electron donor (28) is shown in Fig. 2B. The results show that NaTrxh rec was able to reduce insulin disulfide bridges with a similar qualitative activity profile as the recombinant E. coli Trx (34). Furthermore, when NaTrxh rec was incubated with NADPH and recombinant E. coli NTR, the thiol groups of the reduced NaTrxh rec were labeled with mBBr. Fig. 2C shows NaTrxh rec and recombinant E. coli Trx sulfhydryl labeling in a time-dependent manner only when NADPH and NTR were present. Thus, similar to other Trxs h, oxidized NaTrxh rec can be regenerated by the addition of NADPH and NTR (12,35). NaTrxh Expression Pattern in N. alata-To analyze the expression pattern of NaTrxh in plants, we prepared specific affinity-purified anti-NaTrxh antibodies. Fig. 3A (right panel) shows that this antibody detects specifically NaTrxh rec with no cross reaction to E. coli Trx. This antibody was used to analyze the presence of the NaTrxh in different tissues of N. alata. NaTrxh protein is detected in all the tissues analyzed (Fig. 3B, right panel). This protein is particularly abundant in floral tissues including petals, ovaries, and styles, and it is present in lower levels in anthers, sepals, and leaves.
Biochemical fractionation studies suggest an extracellular location for NaTrxh. SI N. alata S 105 S 105 styles were hand-bisected, and the secreted proteins that accumulate in this tissue were differentially extracted using sequential washes with low salt and high salt buffers, as described (26). Using this procedure, soluble ECM proteins are eluted by a low salt buffer, whereas more tightly bound proteins are released only after the high salt buffer wash. The extraction profile obtained for the NaTrxh protein was similar to S 105 -RNase and to the transmitting tissue-specific protein of N. alata (NaTTS) (Fig. 4). It is known that these two proteins are secreted and that they accumulate in the stylar TT ECM. NaTrxh is eluted with the higher molecular mass (i.e. 55-110-kDa) NaTTS isoforms (Fig. 4C, LS lane) that are thought to be present in the ECM (26), suggesting that NaTrxh is also secreted into the ECM of the stylar TT.
NaTrxh Is Secreted into the Extracellular Matrix of the Transmitting Tissue-To further investigate NaTrxh localization and its secretion to the stylar TT ECM, we performed an immunohistochemical analysis using three-dimensional confocal microscopy. Cross sections of SI N. alata S C10 S C10 styles were simultaneously immunolabeled with a mouse anti-S C10 -RNase antibody (32) and with the affinity-purified polyclonal anti-NaTrxh antibody. Fig. 5, A2, A4 and A6, show that both NaTrxh and S C10 -RNase colocalized outside of the TT cells, indicated by the yellow signal generated from the mixture of both green (S C10 -RNase) and magenta (NaTrxh) fluorochrome signals. To show that the anti-NaTrxh antibody is specifically reacting with NaTrxh in the TT cells, we pretreated the antibody prior to apply to the tissue with NaTrxh rec . As shown in Fig. 5, A3 and A5, after pretreatment, only the green signal of S C10 -RNase was detected.
The experiments shown in Figs. 4 and 5A demonstrate that NaTrxh is secreted into the ECM. However, NaTrxh does not possess a canonical secretory signal peptide, and different secretion signal algorithms give conflicting results. For example, the Bendtsen neural network algorithm (36) does not predict any signal peptide, whereas the hidden Markov model algorithm predicts a signal peptide with a probability of 0.953 with a maximal cleavage site probability of 0.593 between amino acid residues 16 and 17 (Fig. 1A). Likewise, the Secretome 1.0 predictor (37) predicts that NaTrxh (NN-score 0.874) is a non-classical secreted protein.
NaTrxh Sequence Contains the Information for Its Secretion-To further corroborate that NaTrxh is indeed secreted in plants, an NaTrxh-GFP fusion protein expressed from the CaMV35S promoter (20) was constructed. This construct was used to analyze GFP expression pattern in transient assays in N. benthamiana and A. thaliana leaves (Fig. 5B). When the NaTrxh-GFP fusion protein is expressed, most of the GFP protein accumulates in the cell wall in either N. benthamiana (Fig. 5, B2 and B4) or A. thaliana (Fig. 5, B10 and B12). As expected, the GFP protein without fusion, used as control, was mainly localized in the cytoplasm (Fig. 5, B6 and B8). Thus, NaTrxh is sufficient to cause secretion of GFP, supporting the idea that this protein is targeted outside of the cell in N. alata.
S-RNase Is Reduced by NaTrxh in Vitro-To investigate possible substrates for NaTrxh, we performed in vitro reduction reactions with stylar proteins and NaTrxh rec plus NADPH and NTR. Reduced sulfhydryls were labeled with the fluorescent probe mBBr and visualized after separation in a two-dimensional SDS-PAGE system in which non-reducing conditions were used for the first dimension and reducing ones were used for the second dimension (33). For these experiments, we used the low salt extractable ECM stylar protein fraction from SI N. alata S 105 S 105 prepared without a reducing agent in the extraction buffer. As shown in Fig. 6, A and B, there is a small amount of mBBr labeling where S 105 -RNase is the most prominent labeled protein observed (Fig. 6, B and G). Proteins reduced after the first dimension show altered mobility and shift off the diagonal in the second dimension. If reduction is incomplete, two spots are visible, as seen for S 105 -RNase (Fig. 6, A and  B). Proteins fully reduced with DTT prior to mBBr labeling show brighter fluorescence and appear as a single spot after two-dimensional SDS-PAGE (Fig. 6, C and D). As expected, many proteins are reduced by DTT, and mBBr labeling is visible along the entire diagonal. Treatment with purified NaTrxh rec , NADPH, and NTR caused a specific increase in mBBr labeling of S 105 -RNase similar to the one obtained by DTT (Fig. 6,   FIGURE 2. NaTrxh exhibits disulfide reductase activity, and its redox state is regenerated by NADPH and NTR. A, SDS-PAGE showing NaTrxh and GST after thrombin digestion. B, NaTrxh rec exhibits Trx activity. NaTrxh (squares) and E. coli Trx (triangles) were incubated with insulin with (filled) or without DTT (open) as electron donor. Insulin was reduced with DTT as control (asterisks). C, Trx reduction with NADPH/NTR. NaTrxh rec and E. coli Trx (Ec Trx) were incubated with NADPH and recombinant E. coli NTR (Ec NTR) and labeled with mBBr, which indicates reduction of NaTrxh rec and E. coli Trx.   FEBRUARY 10, 2006 • VOLUME 281 • NUMBER 6 JOURNAL OF BIOLOGICAL CHEMISTRY 3421 E and F). S 105 -RNase changed its mobility after reduction with 2-mercaptoethanol for the second dimension (i.e. it appears as two spots in the second dimension), suggesting that NaTrxh only partially reduced S 105 -RNase before the first dimension. The specificity of NaTrxh reduction is apparent from comparison of the large number of proteins that shift mobility and appear off the diagonal after treatment with NaTrxh but not with the nonspecific reduction by DTT (compare Fig. 6, C and E).

Thioredoxin Type h Secretion
PAO affinity chromatography provided further evidence that S 105 -RNase is a substrate for NaTrxh. PAO matrices specifically bind proteins with vicinal thiols that can reversibly form a covalent bond with the resin (38). Low salt extracts from SI N. alata S 105 S 105 either were applied directly to the PAO matrix or were first treated with 20 mM 2-mercaptoethanol or NaTrxh rec plus NADPH and E. coli NTR. Fig. 7 shows that little or no S 105 -RNase bound to the PAO matrix unless it was first subjected to reducing conditions. Some S 105 -RNase was weakly bound under all conditions and eluted with 1 mM or 5 mM 2-mercaptoethanol (Fig. 7A). Large amounts of S 105 -RNase, however, bound tightly to this column after reduction by 2-mercaptoethanol (Fig. 7B) or NaTrxh rec (Fig. 7C). The tightly bound S 105 -RNase eluted with 500 mM 2-mercaptoethanol (Fig. 7, B and C).
S 105 -RNase Interacts with NaTrxh Independently of Disulfide Bridges-We performed affinity chromatography experiments to prove that S 105 -RNase specifically interacts with NaTrxh. An NaTrxh-Affi-Gel affinity column was prepared and used to pass over a stylar protein extract from SI N. alata S 105 S 105 obtained without reducing agents. Fig. 8A shows that S 105 -RNase is specifically retained in the NaTrxh-Affi-Gel matrix because S 105 -RNase is not released from the column when this was washed under harsh conditions (1% Tween 20 and 0.2 M NaCl), but it is eluted only with a low pH buffer (Fig. 8A, B lane). Therefore, we conclude that S 105 -RNase retention in the column is due to its specific interaction with NaTrxh because it was not retained when the BSA-Affi-Gel (Fig. 8B) or the Gly-Affi-Gel columns (Fig. 8C) were used.

DISCUSSION
In the present work, we identified a 16.8-kDa Trx h (NaTrxh). Sequence analysis of this protein showed that it belongs to the Trx h subgroup II (13), and it also showed that it is a novel thioredoxin in Nicotiana since only two different Trxs h have been reported in N. tabacum (39), which are clustered in the subgroup I (Fig. 1B). Trx h subgroup II is characterized by the presence of an N-terminal extension thought to be important for targeting (16,14). Despite the fact that this domain characterizes all the proteins of this subgroup, its function remains unclear. Scattered information has linked this domain to a particular function; the ability to move from cell to cell of a rice phloem sap Trx h has been associated with its N-terminal extension (16). Similarly, Gelhaye et al. (14) reported that poplar Ptrxh2 is associated with mitochondria, proving that its N-terminal extension functions as a mitochondrial transit peptide.
Using different approaches in this work, we demonstrated that NaTrxh is secreted from the cell in different plant species. Moreover, the biochemical data showed that NaTrxh behaves as a soluble component in the ECM and that its fusion to GFP is sufficient to cause GFP secretion. To our knowledge, accumulation of Trx h in the extracellular space has not been previously reported. However, the mechanism by which this N-terminal extension promotes secretion is still unclear. This sequence might function as a non-canonical secretion signal for processing through the endoplasmic reticulum and Golgi apparatus for eventual secretion. It is also possible that NaTrxh secretion occurs through an alternative pathway. In mammalian cells, it has been reported that a Trx is secreted through a pathway that does not involve the Golgi apparatus, the redox status of the cell, or its own Trx redox state (40).
Trxs modify the structure and activity of target proteins by reducing disulfide bonds. They have been implicated in diverse physiological processes. For example, cytoplasmic Trxs are involved in the Brassica sporophytic SI system. Bower et al. (41) identified two Trxs h, THL-1 and THL-2, that specifically interact with the protein kinase domain of the S-locus receptor (SRK), the female determinant of S-specificity in Brassica (42). Cabrillac et al. (43) suggested that THL-1 binds to SRK and inhibits its auto-phosphorylation. When SRK binds SCR (S-locus cysteine-rich protein), the male determinant (44), it appears to release THL-1 concomitant with activation of kinase activity. Our results show that, similarly to other plant Trxs (THL-1, THL-2) (5, 41), NaTrxh is ubiquitously expressed in N. alata. This result suggests that NaTrxh may function in the regulation of the redox state of extracellular proteins. In the particular case of the style, the ECM in the TT is especially rich in secreted proteins including S-RNases and arabinogalactan proteins such as NaTTS (26). As NaTrxh appears to be able to regulate these proteins, it might explain its higher abundance in pistils when compared with other tissues. We have also observed that NaTrxh is expressed at approximately a 6-fold higher level in N. alata than in N. plumbaginifolia (data not shown), and this is consistent with its higher level of expression of secreted proteins in N. alata. The abundance of secreted proteins in the TT ECM afforded an opportunity to directly test whether they are substrates for NaTrxh reduction.
If NaTrxh functions to reduce ECM proteins, then the maintenance of its reduced active state would require NTR in the ECM as well. There is not yet evidence for a secreted NTR in plants. In animal systems, however, an NTR is secreted together with a Trx in monocytes and also in malignant leukemia or melanoma cell lines (45). Another possibility is that NaTrxh functions after uptake into the pollen tube cytoplasm, where NADPH and NTR are more likely to be present. In N. alata, NaTrxh could be taken up by the pollen tubes along with other stylar proteins, such as the S-RNase (46) and 120K (47). 3 Using mBBr labeling and PAO chromatography, we showed that S-RNase is clearly a substrate for NaTrxh. S-RNase was preferentially labeled by mBBr even in the presence of other potential substrates such as NaTTS, 120K, and the class III pistil-specific extensin-like proteins (48).
We detected a stable interaction between S-RNase and NaTrxh, which is specific and independent of any disulfide bridge. It is known that the interactions based on disulfide bridges between a Trx and its target protein are ephemeral since it only depends on the intermolecular disulfide bridge that is formed when the first cysteine residue of the target protein is reduced by the N-terminal cysteine, and it is quickly attacked by the second cysteine of the Trx, releasing the reduced target protein (49,50). Therefore, it seems that S 105 -RNase reduction by 3 B. A. McClure, unpublished data. FIGURE 6. A two-dimensional SDS-PAGE system for analysis of protein disulfide reduction. A, C, and E, Coomassie Blue staining. B, D, and F, mBBr staining. A and B, control TT ECM proteins extracted from SI N. alata S 105 S 105 without reducing agent and not subjected to reduction before running the first dimension. C and D, as in A and B but reduced with DTT before the first dimension. E and F, as in A and B but reduced with NaTrxh plus NADPH/NTR before the first dimension. G, S 105 -RNase immunostaining gel blot treated as in E. FIGURE 7. Vicinal thiol groups are produced in S 105 -RNase after NaTrxh reduction. A, soluble extracellular proteins treated without reducing agent (Non-reduced) were bound to a PAO matrix. The column was washed sequentially with 1 and 5 mM 2-mercaptoethanol, as indicated. Proteins with reduced disulfide bonds bind to PAO matrix and elute with 500 mM 2-mercaptoethanol, as indicated. Fractions were analyzed by anti-S 105 -RNase antibody immunostaining. B and C, same procedure as in A, but the proteins were previously treated with 20 mM 2-mercaptoethanol (2-ME (20 mM)) or with NaTrxh rec plus NADPH/NTR (NADPH/NTR/NaTrxh), respectively. FIGURE 8. S 105 -RNase specifically interacts with NaTrxh. A, stylar proteins from N. alata S 105 S 105 obtained without reducing agents (crude extract, CE) were passed over the NaTrxh-Affi-Gel affinity column. After recovering the unbound fraction (UB), the column was washed with 10 bed volumes of binding buffer (only the 1st, 5th, and 10th washes are shown: W1, W5, and W10) and then with 1% Tween 20 buffer (Tw), 0.1 M NaCl, and 0.2 M NaCl buffers, as indicated. The proteins that strongly bind to NaTrxh were eluted with elution buffer (B lane). Fractions were analyzed by anti-S 105 -RNase antibody immunostaining. B and C, same procedure as in A, but the stylar proteins were passed over the BSA-Affi-Gel or the Gly-Affi-Gel columns, respectively.
NaTrxh is specific and depends on previous substrate recognition by NaTrxh.
These observations may have significance for SI. The specific interaction of S-RNase with NaTrxh and its consequent reduction could favor its interaction with SLF (S-locus F-box protein), the pollen determinant of S-specificity (51). Alternatively, reduction of S-RNase may alter its conformation and facilitate its uptake. Oxley and Bacic (52) have observed changes in S-RNase conformation after reduction. Thus, because of its remarkable secretion into the ECM, NaTrxh may play a role in pollen rejection. Further studies of this possibility are warranted.