Introduction

Top Abstract Introduction Methods Results Discussion References

Biochemical studies have shown that the conjugation of indole-3-acetic acid (IAA) to many kinds of molecules is part of a regulatory mechanism for controlling IAA levels through sequestration and reuse, or as an entry into catabolism (1). In addition, some longer-term responses, such as resistance to higher temperatures (2), are mediated by the conjugated form of the phytohormone. IAA in plants occurs in both conjugated and free forms, and there is increasing evidence that the ratio of free to conjugated IAA is controlled by tissue-specific and developmentally regulated processes (3). To date, a majority of the information on IAA conjugation has focused on IAA linked to a single amino acid, such as aspartate (4), or to a mono or disaccharide (5). Studies of these lower molecular mass conjugates have shown that they are part of a biochemical system for the homeostatic control of IAA levels in higher plants (4). The use of conjugation by plants to regulate IAA levels appears to have become increasingly more important as plants evolved from liverworts to mosses and tracheophytes (6). The presence of a small 3.6-kDa solvent-extractable peptide from bean that accounted for the majority of conjugated IAA present in acetone extracts of bean seeds (Phaseolus vulgaris) has been reported (7). Subsequent studies, however, have shown that the solvent extractable IAA peptides account for only a small fraction of the total IAA present (8). We now report the cloning of the gene for a member of the family of proteins that are modified by the covalent attachment of IAA and describe its expression during seed development and germination.

	Methods

Top Abstract Introduction Methods Results Discussion References

Protein Purification and Immunoblot Analysis. Total proteins from dried bean seeds (Phaseolus vulgaris) were separated on a 10-20% denaturing polyacrylamide gel. The proteins were blotted for 75 min at 0.8 mA·cm² onto nitrocellulose (Schleicher & Schüll, Dassel, Germany) membranes. The membrane was blocked with 5% (wt/vol) nonfat dried milk in PBS buffer followed by incubation with antibody Ab3.6K raised against the bean 3.6 kDa IAA peptide or preimmune serum (9) at 1:1,000 dilution. Detection was done by chemiluminescence (Western-Star, Tropix, Bedford, MA). For purification of IAP1, dried bean seeds were ground with 10 volumes of 10 mM Tris·HCl, 0.1% thioglycerol, 25% of ammonium sulfate, and the mixture was heat treated at 121°C for 20 min. After cooling to 4°C, the crude extract was centrifuged (10,000 × g for 60 min) and the resulting supernatant brought to 50% saturation with ammonium sulfate. The mixture was stirred for 1 h and centrifuged (10,000 × g for 60 min). The resulting pellet was resuspended with 20 mM Tris·HCl and 0.1% thioglycerol and dialyzed. The sample was partially purified by sequential chromatography on High Q anion exchange columns (Bio-Rad) eluted by 10 mM Mops at pH 7.0 and 30 mM NaCl and on hydroxyapatite columns (Bio-Rad) eluted by sodium phosphate buffer linear gradient from 10 mM to 150 mM.

Analysis of IAA Conjugation to IAP1. The IAP1 protein band was determined by locating the zone of immuno reactivity on the membrane blot. The band was cut out of the membrane, hydrolyzed in 7 N NaOH, and purified on a Baker C₁₈ solid-phase extraction column before GC-MS analysis. [¹³C₆]IAA was used as internal standard. Analysis was performed by GC-MS (Hewlett-Packard 6890 GC/5973 MSD). The molecular and quinolinium ions for methyl-IAA at m/z 189/195 and 130/136, respectively, were monitored (ions deriving from the methyl esters of endogenous and [¹³C₆]IAA, respectively; refs. 10 and 11). The amount of IAA released by alkaline hydrolysis of the protein was calculated by the isotope dilution equation (10, 12).

Molecular Mass Determination of IAP1. The mass spectrum of purified IAP1 protein (10 µg) was obtained using a Biflex III time-of-flight mass spectrometer (Bruker) equipped with an N₂ laser for ionization. Sinapinic acid was used as the matrix, and BSA was used for calibration. The spectrum represents the sum of consecutive laser shots smoothed. The iap1 cDNA was cloned in the pTrcHis2 vector, and its expression in E. coli TOP10 cells (Invitrogen) induced with isopropyl -D-thiogalactoside. The newly synthesized gene product was detected using anti-his (C-term) antibodies (Invitrogen). The iap1 cDNA was cloned into a pGEM-T vector and was used for coupled in vitro transcription and translation using the TnT Quick Coupled System (Promega) with [³⁵S]methionine. GeneRacer Kit (Invitrogen) was used for full-length, RNA ligase-mediated rapid amplification of 5' ends using RNA isolated 24 days after flowering. 5' RACE System Version 2.0 (Life Technologies) was used for conformation of the results.

Immuno-Localization. Hand sections (80-100 µm thick) were cut through the seed with a razor blade, and the tissue was rinsed with distilled water and PBS, pH 7.4. Tissues were fixed by incubation in 5% (vol/vol) formaldehyde for 7 min at 4°C, rinsed with PBS, and then blocked in PBS containing 5% (wt/vol) dry milk. Sections then were incubated with preimmune serum or with Ab3.6K at 1:1,000 dilution followed by incubation in secondary antibody, peroxidase-conjugated anti-rabbit IgG, at 1:9,000 dilution. The sections then were stained with 0.01% 3,3'-diaminobenzidine in 50 mM Tris·HCl buffer (pH 7.6) containing 0.012% hydrogen peroxide.

Amino Acid Analysis and Microsequencing of IAP1. Amino acid analysis of protein from two spots obtained from two-dimensional PAGE (1 pmol and 5 pmol, respectively) was performed by hydrolyzing the samples with 6N HCl containing 1% phenol in the vapor phase at 150°C for 1 h; the resultant amino acids were determined on a Perkin-Elmer/Applied Biosystems 420A amino acid analyzer. Microsequencing using 15 pmol of protein was performed by microcapillary reverse-phase HPLC nano-electrospray tandem mass spectrometry on a Finnigan LCQ quadrupole ion trap mass spectrometer.

cDNA Library Synthesis and Screening of Libraries. Total RNA from bean seed at 21 days postanthesis was isolated (13) and used to synthesize a cDNA library in Lambda ZAP II (custom synthesis by Stratagene). Primers (5'AAGGATTATACCGCTGAGAA and 5'GTAATACGACTCACTATAGGGC) were derived based on the microsequencing results, and the most common bean codon usage (14) was used to PCR screen the cDNA library. The Phaseolus genomic library was obtained from David Mok (Oregon State Univ.). A full-length ³²P-labeled iap1 cDNA clone was used to screen the library.

DNA and RNA Blotting. A biotinylated full-length cDNA probe was synthesized and used as described (15). A chemiluminescence detection system was used according to manufacturer instructions (Tropix).

	Results

Top Abstract Introduction Methods Results Discussion References

Even though earlier work by Bandurski and Schulze (16) had shown that bean seeds contain a significant amount of amide-linked IAA, we were unable previously to isolate simple IAA amino acid conjugates from bean seeds by using conventional methods (7). Instead, we isolated a 3.6-kDa peptide from bean seeds to which IAA was conjugated (7). Antibodies (Ab3.6K) raised against the bean 3.6 kDa IAA peptide (9) detected several other polypeptides in bean seeds of varying molecular masses from 17-60 kDa (Fig. 1A). The most abundant protein (IAP1) with an apparent M_r on SDS/PAGE of 42 kDa was found to have IAA covalently attached (Fig. 2). This protein was purified to near homogeneity (Fig. 1B), and quantitative GC-MS analysis of the purified protein (10-12) showed a protein to IAA ratio of 2:1. Silver staining of the purified protein fractions analyzed by two-dimensional PAGE revealed two polypeptides of different pIs (Fig. 3A). Ab3.6K detected both polypeptides (Fig. 3B). The amino acid composition of two spots, cut from several blots and subjected to amino acid analysis, were essentially identical (data not shown). The more acidic spot was then subjected to microsequencing. Based on these results, specific primers were used to PCR screen a cDNA library made from 3-week-old bean seeds. A cDNA clone encoding a 35-kDa protein was isolated, sequenced, and used to screen a genomic DNA library. Analysis of the genomic clone revealed the presence of a single intron and two potential methionine start sites, one of which resulted in a 42-kDa protein, the other encoded a protein of 35 kDa. Matrix-assisted laser desorption ionization (MALDI-TOF) mass spectral analysis of the purified protein from bean (Fig. 4A) indicated an average molecular mass for IAP1 of 35,214 Da. Heterologous expression studies in two different systems (Fig. 4 B and C) confirmed that the cDNA clone for iap1 was full length, but that the protein product runs anomalously on SDS/PAGE at an apparent molecular mass of 42 kDa. In addition, full-length RNA ligase-mediated 5'-RACE confirmed that the full-length cDNA encoded a 35-kDa protein.

View larger version (85K): [in this window] [in a new window]   Fig. 1.   (A) Immunoblot of IAA proteins from bean seeds. (B) Fractions were analyzed by SDS/PAGE and proteins detected with Coomasie blue. 1, crude extract; 2, ammonium sulfate pellet; 3, High Q; 4, hydroxyapatite. IAP1 containing fractions from each column step were detected with Ab3.6K before analysis by SDS/PAGE.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   (A) The IAP1 protein band was blotted onto nitrocellulose and analyzed by GC-MS after alkaline hydrolysis. The presence of IAA in the protein hydrolysate is shown by the peak at 10 min in the ion trace (C) and by the presence of the m/z 130 and 189 peaks in the selected ion mass spectrum (B).

View larger version (63K): [in this window] [in a new window]   Fig. 3.   The partially purified IAP1 protein was analyzed by two-dimensional SDS/PAGE and silver staining (A) and by immunoblotting with Ab3.6K (B). The more basic and more acidic proteins are labeled 1 and 2, respectively.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   (A) The mass spectrum of purified IAP1 protein (10 µg) was obtained using a MALDI-TOF instrument. (B) The iap1 cDNA was expressed in E. coli. A protein with a predicted molecular mass of 38 kDa (35 kDa plus the his tags that contributed 3 kDa to the protein mass) was obtained, and it migrated on SDS/PAGE as expected for a protein of mass 45 kDa. (C) The iap1 cDNA in a pGEM-T vector was used for coupled in vitro transcription and translation. A distinct band of the largest protein at 42 kDa was produced by this system, even though the largest possible transcript encodes a 35-KDa product. Together, these results demonstrate that secondary structure, or other attributes, of this protein affect the migration and thus the apparent molecular mass of the protein on SDS/PAGE.

Expression of iap1 in E. coli yielded a his-tagged protein product unmodified by IAA. This microbial protein was detected by Ab3.6K (data not shown), suggesting that the antibody recognizes a domain common to proteins modified by IAA rather than the IAA moiety itself. Nevertheless, all six plant-derived proteins detected by this antibody [five from bean and one from Arabidopsis (Fig. 2 and see Fig. 10; ref. 17)] have been shown by GC-MS analysis to be modified by IAA.

IAP1 appears to be encoded by a single copy gene (Fig. 5). The gene has 65.5% similarity at the nucleotide level, interspersed throughout the sequence, and 64.7% similarity at the amino acid level to a 35-kDa late seed maturation protein from soybean (GenBank accession no. AF009953) (http://pbil.univ-lyon1.fr/pbil.html; http://www.expasy.ch/tools/sim-prot.html). As shown in Fig. 6, the gene also has close similarity (55.7% at the nucleotide level and 44.1% at the amino acid level) to a gene for a late embryogenesis abundant protein from Arabidopsis (GenBank accession no. 7629988). The gene does not contain known signal sequences nor does it appear to encode the smaller 3.6-kDa bean IAA peptide (18). RNA blot analysis showed that the iap1 transcripts accumulate in significant amounts in seeds late in their development (Fig. 7A). No iap1 mRNA was found in other bean tissues (Fig. 7B), suggesting that IAP1 accumulates during seed maturation to provide a mechanism for hormonal regulation during germination. This suggestion is supported by the fact that amide IAA accumulates throughout bean seed development (Fig. 7A). GC-MS analysis of IAP1 isolated from seeds 24 days post anthesis (the time at which its presence was first clearly detectable) showed that IAA was attached to 20% of the IAP1 protein (data not shown), and this raises to 50% modification of IAP1 by IAA in mature seeds. By the time the seed is mature, less than a third of the total IAA occurs as the free acid (19). Immunoblots with Ab3.6K detected protein accumulating late during seed development (data not shown) showed that levels decrease dramatically in axes and cotyledons germinating in the dark in the course of a single day after imbibition (Fig. 8). Unlike maize (20, 21), bean seedlings begin de novo IAA biosynthesis within the first hours after imbibition in the dark (22).

View larger version (58K): [in this window] [in a new window]   Fig. 5.   Genomic DNA blot analysis of iap1 in Phaseolus vulgaris. Genomic DNA was digested with HindIII, EcoRI, and BamHI. HindIII cuts within the genomic sequence of iap1, whereas EcoRI cuts only 16 bp into the sequence and no BamHI sites are present. These results indicate that iap1 is encoded by a single copy gene. A full-length iap1 cDNA probe was used for hybridization and the mobility of the DNA size standards are indicated in kilobases.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Amino acid sequence alignment of IAP1 (GenBank accession no. AF293023) and its homologues in Glycine max (GenBank accession no. AF009953) and Arabidopsis thaliana (GenBank accession no. 7629988). The underlined amino acids are the sequences obtained by microsequence analysis of peptide fragments. Amino acids indicated in black are unique to that protein; those in blue are present at that position in two of the proteins, and those in red are conserved in all three sequences.

View larger version (54K): [in this window] [in a new window]   Fig. 7.   RNA blot analysis of iap1 in bean. (A) Bean seeds at different stages of development. (B) Bean seeds in different tissues of mature bean plants. The levels of total IAA (the sum of free, ester, and amide forms) and of amide-linked IAA during bean-seed development are shown below A (data from ref. 19).

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Levels of IAP1 decrease during seed germination, as determined by immunoblotting with Ab3.6K or, as a control, preimmune serum (Pi). Arrows indicate the expected mobility for IAP1.

Previous studies have shown that the model dicot Arabidopsis has both ester- and amide-conjugated IAA (23), but in total, the smaller molecular mass conjugates previously identified did not account for the bulk of the conjugate pool. Ab3.6K was used for immunostaining of different Arabidopsis tissues. Crossreacting proteins were localized to the root meristem and outer cell regions of the cotyledons and radicle of Arabidopsis seeds (Fig. 9). Ab3.6K crossreacting proteins were also found in the outer cell layers of cantaloupe and tomato fruit (24), but the individual proteins were never isolated. Immunoblot analysis also indicates that Ab3.6K cross reacts with proteins in many different plant species (Fig. 10). The Arabidopsis seed 35-kDa immunostaining protein was partially purified and subjected to alkaline hydrolysis (10-12). GC-MS analysis confirmed the presence of IAA covalently bound to the protein. These results establish the presence of IAA proteins in dicots other than bean.

View larger version (62K): [in this window] [in a new window]   Fig. 9.   Detection of Ab3.6K cross-reactive proteins in cross sections of Arabidopsis seeds. Preimmne serum (A) or with Abe.6K incubation (B). Cotyledons and radicle are labeled c and r, respectively. Positive staining is dark brown to black, depending on the intensity.

View larger version (35K): [in this window] [in a new window]   Fig. 10.   Presence of Ab3.6K cross-reacting proteins in different plant seeds as determined by immunoblotting. Plant species examined were: bean (Phaseolus vulgaris cv. Bush Burpee Stringless), lima bean (Phaseolus limensis cv. Fordhook 242), pea (Pisum sativium cv. Thomas Laxton), lettuce (Lactuca sativa cv. Grand Rapids), arabidopsis (Arabidopsis thaliana ecotype Columbia), tomato (Lycopersicon esculentum cv. Ailsa Craig), cabbage (Brassica campestris cv. Granat), maize (Zea mays cv. Silver Queen), carrot (Daucus carota cv. Danvers). Several proteins that crossreact with this antibody are detected in all plant species so far examined. Several smaller proteins from bean and the major cross-reacting protein from Arabidopsis at 35 kDa were partially purified and shown by GC-MS analysis to release IAA following strong alkaline hydrolysis (data not shown).

	Discussion

Top Abstract Introduction Methods Results Discussion References

Plants contain phytohormones in conjugated forms (5), and the existence of conjugated auxin has been postulated since the studies by Cholodny in the 1930s (25). A comparatively large amount of information is now available on the biosynthesis and hydrolysis of the lower molecular mass conjugates of IAA in higher plants (5). These conjugates are thought to be involved in a variety of hormonally related processes: in the transport of IAA within the plant; the storage and subsequent reuse of IAA; protection of IAA from enzymatic destruction; as components of a homeostatic mechanism for control of IAA levels (26); and as an entry route into the subsequent catabolism of IAA (27).

By contrast, much less information is known of the function of the higher molecular mass conjugates found in plants. The ester compounds identified so far are a glycoprotein from oat (28) and a cellulosic glucan from corn (29). Both of these compounds are the major forms of IAA in the tissue from which they were isolated. In corn, in which the glucan ester accounts for over 50% of the total IAA, the release of the attached IAA appears to occur during germination (30). This conclusion could be reached only by calculating the pool size and turnover rates for every major indole in the germinating kernel. The glycoprotein from oats is unstable, and it has not been possible to determine its structure other than that the IAA is conjugated by an ester linkage through the carbohydrate moiety (28).

In most studies done over the last 50 years, the amount of auxin obtained after solvent extraction did not account for the total IAA found following exhaustive extraction over time (31, 32). More recent studies showed a similar phenomenon, that the amount of total IAA (free plus conjugated) obtained by direct hydrolysis of the tissue was much greater than the amount readily extractable with solvents (12, 33, 34). This difficulty in accounting for the IAA present in plants has only recently been resolved by the isolation of proteins and peptides to which IAA was attached. The first studies that suggested IAA protein covalent complexes, however, left open the possibility that IAA attachment was a random event or incidental to the function of either the proteins or the phytohormone. We now show that a specific protein in bean, iap1, is covalently modified, that approximately half of that protein resident in the tissue is modified, and that the expression of the single copy gene encoding this gene is developmentally regulated.

IAP1, which is the major IAA-modified protein in bean seed, could be purified from crude protein extract using standard purification methods after taking advantage of the solubility of the protein following high temperature treatment. Since the total amount of IAA attached to protein in bean seeds is 600 ng/g (34), an upper limit at 240 µg/g can be calculated for the amount of IAP1 and other IAA-modified proteins present in bean. IAP1 therefore is less than 0.1% of the total seed protein. The solubility of IAP1 following heat treatment thus afforded an important early purification from the vast bulk of plant proteins that were denatured and precipitated following exposure to very high temperatures.

IAP1 is primarily found in seeds and seed-derived tissues. However, other bean tissues contain proteins that are detected by Ab3.6K (data not shown). Previous analytical results from our laboratory (22) showed that proteins with IAA attached are present in vegetative tissues of bean, but these earlier reports did not determine the characteristics of the protein moiety. The absence of iap1 mRNA in vegetative tissues (Fig. 7) indicates that the putative IAA proteins we have detected in such tissues are likely distinct from the IAP1 seed protein. Curiously, leaves contain a major Ab3.6K-detected protein that runs on PAGE coincident with IAP1 at an apparent 42 kDa (data not shown). Plants other than bean also contain Ab3.6K immunoreactive proteins (Fig. 10); however, the presence of IAA on these proteins has not been determined, although the attachment of minute amounts of IAA attached to maize zein has been described (35). To test the possibility that proteins detected in other species by AB3.6K might share the characteristic of an attached IAA prosthetic group, we partially purified the protein from Arabidopsis and found that it, indeed, also released IAA following strong base hydrolysis. The low number of IAP1 crossreacting proteins in seeds of Arabidopsis appeared to localize to the epidermal tissue of the cotyledons (Fig. 9).

The attachment of a plant hormone as a prosthetic group to a seed storage protein defines a novel class of sequestered IAA in plants. IAP1 accumulates during seed development and is utilized during the germination process, as would be expected for a storage protein. IAA-modified proteins, however, appear to be present in vegetative as well as seed tissues, suggesting other possible roles. Prosthetic groups are known to alter protein stability (36), and a recurrent theme in auxin signal transduction has been the involvement of protein metabolism in the process (37-39). The linkage between auxin activity and protein metabolism seems to be reinforced by a diverse array of experimental approaches, and the presence of specific proteins to which IAA is attached may have significance beyond the important regulatory roles already assigned to conjugates for modulation of IAA levels. Regardless of the broader consequences of protein modification, the description of IAA proteins should allow a more detailed picture of IAA metabolism, especially in plants such as bean and Arabidopsis where total IAA levels greatly exceed the IAA attributed to the free hormone and solvent-extractable conjugates (12, 23).