Abstract

Two auxin amidohydrolases, BrIAR3 and BrILL2, from Chinese cabbage [Brassica rapa L. ssp. pekinensis (Lour.) Hanelt] were produced by heterologous expression in Escherichia coli, purified, and screened for activity towards N-(indol-3-ylacetyl)–l-alanine (IAA-Ala) and the long-chain auxin–amino acid conjugates, N-[3-(indol-3-yl)propionyl]–l-alanine (IPA-Ala) and N-[4-(indol-3-yl)butyryl]–l-alanine (IBA-Ala). IPA-Ala was shown to be the favored substrate of both enzymes, but BrILL2 was approximately 15 times more active than BrIAR3. Both enzymes cleaved IBA-Ala and IAA-Ala to a lesser extent. The enzyme kinetics were measured for BrILL2 and the obtained parameters suggested similar binding affinities for the long-chain auxin–amino acid conjugates (IPA-Ala and IBA-Ala). The velocity of the hydrolyzing reaction decreased in the order IPA-Ala > IBA-Ala > IAA-Ala. In a root growth bioassay, higher growth inhibition was caused by IPA-Ala and IBA-Ala in comparison with IAA-Ala. Neither these conjugates nor the corresponding free auxins affected the expression of the BrILL2 gene. A modeling study revealed several possible modes of IPA-Ala binding to BrILL2. Based on these results, two possible scenarios for substrate hydrolysis are proposed. In one the metal binding water is activated by the carboxyl group of the substrate itself, and in the other by a glutamate residue from the active site of the enzyme.

Introduction

Auxins, mostly arylalkanoic acids, are multifunctional plant hormones which have been studied from Cziesielki’s and Darwin’s times since the 1880s, with unusually slow progress. An aspect of their metabolism which is still incompletely understood is how their concentrations at points of physiological action are regulated. In part, this is accomplished by reversible conjugation with amino acids (Normanly et al. 2004). Such a conjugate, N-(indol-3- ylacetyl)–l-aspartic acid (IAA-Asp), was first detected in 1955 in pea seedlings (Andreae and Good 1955). This work has been followed by a large number of subsequent studies. For instance, IAA-Asp has been identified in Scots pine (Anderson and Sandberg 1982) and Douglas fir (Chiwocha and von Aderkas 2002); IAA-Glu and IAA-Asp in cucumber (Sonner and Purves 1985) and soybean (Cohen 1982, Epstein et al. 1986), and IAA-Ala in spruce (Östin et al. 1992). Additionally, IAA-Ala, IAA-Asp, IAA-Leu and IAA-Glu have been found in Arabidopsis (Tam et al. 2000, Kowalczyk and Sandberg 2001). Based on the identification of the oxidative metabolites, 6-OH-IAA-Val and 6-OH-IAA-Phe, in Arabidopsis, Kai et al. (2007) suggested that the conjugates of IAA with phenylalanine and valine, which so far had not been found in this species, can also be formed endogenously in that plant. Just recently, IAA-Val was identified, after treatment with IAA, in the moss Physcomitrella patens (Ludwig-Müller et al. 2009). Amide conjugates of the long-chain auxin 4-(indol-3-yl)butyric acid (IBA) (particularly IBA-Asp) were reported to be found in pea tissue (Nordström et al. 1991) and petunia cell suspension culture (Epstein et al. 1993) after feeding with IBA, and in maize roots during arbuscular mycorrhiza formation (Fitze et al. 2005). Conjugates of 3-(indol-3-yl)propionic acid (IPA) have so far not been reported in plants.

The enzymology involved in the metabolism of amide conjugates is just beginning to be understood. The biosynthesis appears to include auxin–AMP mixed anhydrides as critical intermediates (Staswick et al. 2002). The hydrolysis of IAA–amide conjugates results in free IAA and is mediated by auxin amidohydrolases, metalloenzymes belonging to the M20 family of peptidases (reviewed by Bartel et al. 2001, Woodward and Bartel 2005). Such hydrolases were first cloned from Arabidopsis thaliana (Davies et al. 1999, LeClere et al. 2002) and tested for activity towards conjugates of the most common auxin, IAA. A homologous amidohydrolase from wheat (Triticum aestivum L. cv. Caledonia) was more active towards conjugates of IBA and IPA (Campanella et al. 2004), two known (Segal and Wightman 1982, Schneider et al. 1985, Epstein and Ludwig-Müller 1993, Ludwig-Müller 2000), but so far neglected, endogenous auxins. Exogenous IBA conjugates have been successfully used for the rooting of cuttings (Epstein and Wiesman 1987, Wiesman et al. 1989), but only sporadic effort has been devoted to looking for their endogenous occurrence. Here we screened the hydrolytic specificity of two amidohydrolases from Chinese cabbage (Brassica rapa), and discovered that they, in addition to IAA–amino acid conjugates (Ludwig-Müller et al. 1996, Schuller and Ludwig-Müller 2006), preferentially cleave amino acid conjugates of long-chain auxins (IPA-Ala and IBA-Ala). The physiological activity of the screened substrates was examined in vivo by root growth bioassay. In addition, using homology modeling, we attempted to explain the mode of binding of the preferred substrate, IPA-Ala, into the binding pocket of BrILL2 amidohydrolase.

Results

Enzyme activity

Herein we produced the Brassica hydrolases, BrIAR3 and BrILL2, by heterologous expression in Escherichia coli, purified them to approximately 95%, and used them in tests of enzyme activity towards auxin–amino acid conjugates. To define the optimal activity conditions for these enzymes, we screened potential metal cofactors and the requirement for reducing agents. Mn2+ was found to be the only metal ion with cofactor activity. Reducing agents [dithiothreitol (DTT) or β-mercaptoethanol] were necessary for activity of both enzymes towards the analyzed substrates. Substrate screening revealed activity towards IPA-Ala and IBA-Ala at much higher rates than towards IAA-Ala. Hydrolysis rates for BrIAR3 and BrILL2 were determined and compared at the point when approximately 40–50% of substrate was hydrolyzed (Table 1). Maximum activity of both examined enzymes was observed towards IPA-Ala: approximately 200 and 3,000 nmol mg–1 of protein min–1 for BrIAR3 and BrILL2, respectively. Activities of BrILL2 towards IBA-Ala and IAA-Ala were lower by almost one and two orders of magnitude, and those of BrIAR3 by 2.7- and 47-fold, respectively. There was no detectable activity towards IPA-β-alanine (Bal) and IPA-Asp, with or without the presence of reducing agent (Table 1). Fig. 1 shows a representative HPLC chromatogram of BrILL2 hydrolysates of IAA-Ala, IBA-Ala and IPA-Ala. The products of hydrolysis, IAA, IPA and IBA, were detected according to their retention times which were 16, 19 and 23 min, respectively. The enzyme BrILL2 showed in general about a 10 times higher hydrolysis rate in comparison with BrIAR3. Thus we proceeded to determine the kinetics parameters of BrILL2 for the substrates IPA-Ala, IBA-Ala and IAA-Ala (Table 2). Fig. 2 shows representative kinetics curves (velocity vs. substrate) obtained for the examined substrates. As can be seen from Table 2, the Km constant for the longest chain substrate, IBA-Ala, was found to be the lowest, about 1.1 mM. Km values for IPA-Ala and IAA-Ala were about twice (2.0 mM) and about six times (6.5 mM) higher in comparison with that for IBA-Ala, respectively. The kcat value for IPA-Ala was about four times that measured for IBA-Ala, and about 30 times that for IAA-Ala. The catalytic efficiency (kcat/Km) of BrILL2 for IPA-Ala was found to be 2.6 times higher than that for IBA-Ala and about 100 times higher than that for IAA-Ala.

Table 1

Enzyme activity expressed as nmol auxin released mg protein–1 min–1 of two Brassica rapa auxin amidohydrolases measured at the time when 40–50% of free auxin had been released

Substrate Enzyme activity
 
 BrIAR3 BrILL2 
IAA-Ala 4.1 ± 1.1 36 ± 7.3 
IBA-Ala 72 ± 10 415 ± 35 
IPA-Ala 191 ± 14 2970 ± 124 
IPA-Bal n.d. n.d. 
IPA-Asp n.d. n.d. 
Substrate Enzyme activity
 
 BrIAR3 BrILL2 
IAA-Ala 4.1 ± 1.1 36 ± 7.3 
IBA-Ala 72 ± 10 415 ± 35 
IPA-Ala 191 ± 14 2970 ± 124 
IPA-Bal n.d. n.d. 
IPA-Asp n.d. n.d. 

The initial concentration of the substrate in the reaction mixture was 0.5 mM.

n.d., not detectable

Fig. 1

HPLC chromatograms of the BrILL2 hydrolyzing reaction with the substrates IPA-Ala after 10 min, IBA-Ala after 45 min and IAA-Ala after 3 h. Reaction mixtures were as defined in Materials and Method containing 2.13 × 10–7 M of enzyme and 0.5 mM substrate. 61 × 46 mm (600 × 600 dpi).

Fig. 1

HPLC chromatograms of the BrILL2 hydrolyzing reaction with the substrates IPA-Ala after 10 min, IBA-Ala after 45 min and IAA-Ala after 3 h. Reaction mixtures were as defined in Materials and Method containing 2.13 × 10–7 M of enzyme and 0.5 mM substrate. 61 × 46 mm (600 × 600 dpi).

Table 2

Michaelis–Menten kinetics parameters of the enzyme BrILL2 calculated by using the program PSI-Plot

Substrate Vmax (mM/min) Km (mM) kcat (min–1kcat/Km (M–1 min–1
IPA-Ala 0.143 ± 0.004 2.03 ± 0.39 668.75 ± 21.15 328,785 
IBA-Ala 0.061 ± 0.002 1.14 ± 0.05 143.7 ± 5.05 125,942 
IAA-Ala 0.046 ± 0.006 6.52 ± 1.31 21.36 ± 2.84 3,274 
Substrate Vmax (mM/min) Km (mM) kcat (min–1kcat/Km (M–1 min–1
IPA-Ala 0.143 ± 0.004 2.03 ± 0.39 668.75 ± 21.15 328,785 
IBA-Ala 0.061 ± 0.002 1.14 ± 0.05 143.7 ± 5.05 125,942 
IAA-Ala 0.046 ± 0.006 6.52 ± 1.31 21.36 ± 2.84 3,274 

Protein concentrations used were: 2.13 × 10–7 M for IPA-Ala, 4.26 × 10–7 M for IBA-Ala and 2.13 × 10–6 M for IAA-Ala.

Fig. 2

Representative kinetics curves of BrILL2. Hydrolyzing rates were determined at varying concentrations of IPA-Ala (circles), IBA-Ala (triangles) and IAA-Ala (squares) with enyzme concentrations of 2.13 × 10–7 M, 4.26 × 10–7 M and 2.13 × 10–6 M, respectively. The curves were generated using the program SigmaPlot 10.0. 80 × 62 mm (600 × 600 dpi).

Fig. 2

Representative kinetics curves of BrILL2. Hydrolyzing rates were determined at varying concentrations of IPA-Ala (circles), IBA-Ala (triangles) and IAA-Ala (squares) with enyzme concentrations of 2.13 × 10–7 M, 4.26 × 10–7 M and 2.13 × 10–6 M, respectively. The curves were generated using the program SigmaPlot 10.0. 80 × 62 mm (600 × 600 dpi).

Modeling

The 3D structure of BrILL2 was derived by homology modeling using the X-ray structure of the IAA–amino acid hydrolase from Arabidopsis thaliana, AtILL2 (pdbs: 1XMB, and 2Q43), as a template. The amino acid sequence identity between these two hydrolases, AtILL2 and BrILL2, is 76%.

Since amidohydrolases of the M20 family possess two metal cofactors, two Mn2+ ions were placed into the putative binding site (see Materials and Methods for details), and the obtained model was energy optimized and subjected to short 300 ps molecular dynamics (MD) simulations using explicit water molecules (Fig. 3).

Fig. 3

The 3D structure of BrILL2 obtained by comparative modeling with the secondary structure elements displayed. The positions of the two Mn2+ ions are presented as violet balls. The substrate binds to the Mn2+ ions. Coloring is according to the secondary structure: α-helix = red, β-sheet = yellow, coil = green. 80 × 59 mm (300 × 300 dpi).

Fig. 3

The 3D structure of BrILL2 obtained by comparative modeling with the secondary structure elements displayed. The positions of the two Mn2+ ions are presented as violet balls. The substrate binds to the Mn2+ ions. Coloring is according to the secondary structure: α-helix = red, β-sheet = yellow, coil = green. 80 × 59 mm (300 × 300 dpi).

The equilibrated protein structure was used to build complexes with IPA-Ala since it is the most favorable substrate for BrILL2. Docking was performed by the AutoDock3.05 program, and the most populated binding sites were selected as the most probable ones. We selected 15 different BrILL2–IPA-Ala complexes altogether. They were solvated, energy-minimized and, in order to learn more about the complex stability and the substrate-binding site, subjected to 300 ps of MD simulations. The resulting structures were analyzed and grouped according to their similarity. Finally, we could distinguish four main binding modes (Fig. 4).

Fig. 4

Different binding modes found for IPA-Ala: BM1 (A), BM2 (B), BM3 (C) and BM4 (D). The amino acid residues in the close vicinity of the metal cations and the substrate are displayed. For clarity, the orientation of the substrate differs in the different modes. The Mn2+ ions are colored violet and water molecules are shown as red and white sticks.

Fig. 4

Different binding modes found for IPA-Ala: BM1 (A), BM2 (B), BM3 (C) and BM4 (D). The amino acid residues in the close vicinity of the metal cations and the substrate are displayed. For clarity, the orientation of the substrate differs in the different modes. The Mn2+ ions are colored violet and water molecules are shown as red and white sticks.

In the first binding mode, BM1 (Fig. 4A), IPA-Ala is coordinating one of the Mn2+ ions with its carbonyl oxygen. This manganese ion (Mn12+) is also coordinated with Cys139, His399, Glu175 [bidentately (B)] and a water molecule which is, at the same time, H-bonded to both the carbonyl and the carboxyl oxygens of the ligand. The other Mn2+ (Mn22+) is coordinated by Asp114 [monodentately (M)], Glu370 (B), Glu174 (B) and a water molecule.

The indole ring is accommodated between Thr201, Arg203 and Phe385. Interaction between the imidazole nitrogen and the Thr201 hydroxyl group is established by water-mediated H-bonds. The substrate carboxyl group is H-bonded to NH of Gly369 and the alanine methyl group is hydrophobically interacting with Ile366 and the alkyl part of the Glu370 side chain. If Ala was replaced by Asp in this binding mode, its carboxyl group would be in an unfavorable steric and electrostatic interaction with the Glu370 carboxyl group.

In the second binding mode, BM2 (Fig. 4B), IPA-Ala is also coordinating one of the Mn2+ ions with its carbonyl oxygen, but the orientation of the substrate in the binding site differs from that in BM1. The coordination of the manganese ions is also slightly different. The ion coordinated by the substrate carbonyl oxygen is, instead of bidentately by Glu175 and one water molecule, coordinated monodentately by Glu175 and by two water molecules and, as in BM1, by Cys139 and His399. The other Mn2+ is coordinated by His199, instead of by a water molecule.

In the third binding mode, BM3 (Fig. 4C), IPA-Ala is coordinating both Mn2+ ions with its carboxyl oxygens while the carbonyl oxygen weakly electrostatically interacts with NH of Gly369 and with CH2 of Gly368. Ala398 (through its CH3 group) hydrophobically interacts with CH3 of the substrate side chain. The indole ring is stabilized by stacking interaction with Phe385, by hydrophobic interactions with Ile366 and Arg203, and by weak CH–O interaction with Ser210.

In the fourth binding mode, BM4 (Fig. 4D), the substrate coordinates the metal ions by both the carbonyl oxygen (Mn12+) and one carboxyl oxygen (Mn12+ and Mn22+). However, during 1 ns of MD simulation no water molecule entered the metal ions’ cordination spheres.

Conformational analysis of IPA-Ala and IPA-Bal revealed a highly folded conformation for IPA-Bal, in which both carboxyl oxygens were involved in NH–O hydrogen bonds (Supplementary Fig. S1). Such a conformation is energetically very stable and hinders proper accommodation of the substrate in the binding site.

Root growth bioassay

The amino acid conjugates, IAA-Ala, IBA-Ala, IPA-Ala, IPA-Bal and IPA-Asp, as well as the free auxins, IAA, IBA and IPA, in a concentration range of 0.1 μM to 0.1 mM in 1% agar plates, were tested for inhibitory effects on the growth of roots of Chinese cabbage seedlings (B. rapa L.). Auxin-free agar plates were used as a control. The resulting dose– response curves are shown in Fig. 5. In the group of free auxins (Fig. 5A), IAA showed the steepest curve causing stronger inhibition of root growth, at lower concentrations, than IBA and IPA. However, at higher auxin concentrations (0.01 and 0.1 mM), the inhibitory effects of IPA and IBA on root growth were the same as for IAA. Among the amino acid conjugates, IPA-Ala and IBA-Ala had the most pronounced inhibitory effect on root growth, while IAA-Ala, IPA-Asp and IPA-Bal were significantly less effective (Fig. 5B). The root growth inhibition by free auxins, at a concentration of 0.1 mM, with respect to a corresponding control, was about 91% for all three free auxins. Comparing auxin conjugates, IPA-Ala caused the highest root growth inhibition (86.1%), followed by IBA-Ala (71.0%), IAA-Ala (46.3%) and IPA-Asp (45.1%). IPA-Bal showed the weakest effect on root growth and caused only 16.1% inhibition at the compared concentration (0.1 mM). The IC50 value for each auxin used in the root growth bioassay was calculated from non-linear regression curves (Table 3, figures not shown). As can be seen, the IC50 for IAA was the lowest (6.3 × 10–8 M), while IPA and IBA had approximately 10 times higher values. Considering the auxin conjugates, IBA-Ala showed a 2.5 times lower IC50 than IPA-Ala, while those for IAA-Ala, IPA-Bal and IPA-Asp were two orders of magnitude higher.

Fig. 5

Root growth curves of Brassica rapa L. seedlings upon treatment with the free auxins IAA, IPA and IBA (A), and auxin–amino acid conjugates (B). Each point represents the mean ± SE, n = 30. The experiments were performed as three sets of bioassays with the corresponding controls. 80 × 107 mm (600 × 600 dpi).

Fig. 5

Root growth curves of Brassica rapa L. seedlings upon treatment with the free auxins IAA, IPA and IBA (A), and auxin–amino acid conjugates (B). Each point represents the mean ± SE, n = 30. The experiments were performed as three sets of bioassays with the corresponding controls. 80 × 107 mm (600 × 600 dpi).

Table 3

Concentrations of auxins causing 50% root growth inhibition (IC50)

Auxin IC50 (M) 
IAA 6.30 × 10–8 
IBA 6.48 × 10–7 
IPA 4.98 × 10–7 
IAA-Ala 2.85 × 10–4 
IBA-Ala 3.67 × 10–6 
IPA-Ala 9.53 × 10–6 
IPA-Bal 3.20 × 10–4 
IPA-Asp 1.21 × 10–4 
Auxin IC50 (M) 
IAA 6.30 × 10–8 
IBA 6.48 × 10–7 
IPA 4.98 × 10–7 
IAA-Ala 2.85 × 10–4 
IBA-Ala 3.67 × 10–6 
IPA-Ala 9.53 × 10–6 
IPA-Bal 3.20 × 10–4 
IPA-Asp 1.21 × 10–4 

The values were calculated from non-linear regression curves (modified hyperbolae)generated by the program SigmaPlot 10.0.

BrILL2 gene expression analysis

To check the influence of the examined substrates and the products of the BrILL2-catalyzed hydrolysis at the gene regulation level, we analyzed the expression of the BrILL2 gene in seedlings treated for 24 h with free auxins (IAA, IBA and IPA) and their amino acid conjugates (IAA-Ala, IBA-Ala, IPA-Ala, IPA-Asp and IPA- Bal), at a concentration of 0.1 mM. Untreated seedlings of the same age were used as a control. Fig. 6 shows the average results of three independent experiments. As can be seen, none of the free auxins nor their amino acid conjugates significantly affected the transcription of the BrILL2 gene cumulatively in the whole seedling.

Fig. 6

Analysis of BrILL2 gene expression. The ratio of BrILL2 transcript and geometric mean of cytochrome oxidase (cox) and 18S rRNA in qRT-PCR of each auxin-treated sample was normalized against the average ratio in untreated samples and plotted using arbitrary units. Each bar represents the mean ± SE of three biological replicates of the experiment. Seedlings were treated with 0.1 mM of the following auxins: control, no treatment (1), IAA (2), IBA (3), IPA (4), IAA-Ala (5), IBA-Ala (6), IPA-Ala (7), IPA-Asp (8), IPA-Bal (9). 80 × 64 mm (600 × 600 dpi).

Fig. 6

Analysis of BrILL2 gene expression. The ratio of BrILL2 transcript and geometric mean of cytochrome oxidase (cox) and 18S rRNA in qRT-PCR of each auxin-treated sample was normalized against the average ratio in untreated samples and plotted using arbitrary units. Each bar represents the mean ± SE of three biological replicates of the experiment. Seedlings were treated with 0.1 mM of the following auxins: control, no treatment (1), IAA (2), IBA (3), IPA (4), IAA-Ala (5), IBA-Ala (6), IPA-Ala (7), IPA-Asp (8), IPA-Bal (9). 80 × 64 mm (600 × 600 dpi).

Discussion

Brassica auxin amidohydrolases: substrate specificity

Enzymes capable of cleaving the auxin–amino acid linkage belong to the ILR1-like IAA amidohydrolase gene family. They were originally characterized in the model plant A. thaliana (Bartel and Fink 1995, Davies et al. 1999, LeClere et al. 2002). Furthermore, Campanella et al. (2003a, 2003b) isolated and characterized sILR1 amidohydrolases from the related species Arabidopsis suecica. The amidohydrolases from the two Arabidopsis species are shown to be specific for IAA–amino acid conjugates, with overlapping substrate specificities (ILR1 cleaves IAA-Leu and IAA-Phe preferentially; ILL1, ILL2 and IAR3 prefer IAA-Ala; sILR1 shows specificity for IAA-Gly and IAA-Ala). Investigation of the amidohydrolases of B. rapa L. revealed four full-length cDNAs that showed high homology to the IAA-amidohydrolases of Arabidopsis (Ludwig-Müller et al. 1996, Schuller and Ludwig-Müller 2006). Two of them, BrIAR3 and BrILL2, were previously reported to cleave IAA-Ala, while BrILL2 can additionally hydrolyze IAA-Val and IAA-Leu to a lesser extent. Herein, we examined the specificity of these two enzymes towards amino acid conjugates of long-chain auxins (IBA and IPA). BrIAR3 and BrILL2 from Chinese cabbage preferably cleaved IPA-Ala, with hydrolyzing rates of about 200 and 3,000 nmol of released free IPA mg protein–1 min–1, respectively. To our knowledge, this is the first report on preferential cleavage of an IPA–amino acid conjugate by a plant amidohydrolase. Hydrolyzing activity towards long-chain auxin–amino acid conjugates was first reported for wheat auxin amidohydrolase (TaIAR3) (Campanella et al. 2004), its activity decreasing in the order IBA-Ala > IPA-Ala > IAA-Ala. Just recently, five homologous enzymes from Medicago truncatula were found to cleave IBA-Ala to an extent of 10–680 pmol of released auxin min–1 ml–1E. coli extract (Campanella et al. 2008). Wheat and Medicago hydrolase activity tests were performed with crude bacterial lysate containing the recombinant plant proteins. Herein, enriched enzymes enabled us to obtain results directly attributable to the plant hydrolases, (without possible impact of other components from the bacterial extract). To learn more details about the catalytic specificity, we determined the enzyme kinetics of BrILL2 [which cleaved IPA-Ala about 15 times faster than BrIAR3 (Table 1)] towards the substrates IAA-Ala, IPA-Ala and IBA-Ala (Table 2). The Km values for the alanine conjugates of IPA and IBA were similar, with a slightly higher value for IPA-Ala, while the Km value for IAA-Ala was approximately six times higher. This finding may imply higher enzyme affinity for the long-chain substrates, IBA-Ala and IPA-Ala, in comparison with IAA-Ala. The values of kcat suggested that the velocity of hydrolysis decreases in the order IPA-Ala > IBA-Ala > IAA-Ala. Based on values of the catalytic efficiency (kcat/Km) BrILL2 is about 2.6 times more efficient in hydrolyzing IPA-Ala than IBA-Ala and about 100 times less efficient for IAA-Ala. To date, kinetic data are available for the recombinant Arabidopsis homologs towards IAA conjugates with the amino acids Ala and Leu (LeClere et al. 2002). The results show, for IAA-Ala, a 100 times lower Km value for AtILL2 (0.052 mM) than we found for BrILL2 (6.5 mM), and significantly higher catalytic efficiency of the Arabidopsis enzyme in comparison with the Brassica homolog (about 700 times).

Modeling

Based on our results and the previously reported results of LeClere et al. (2002), it is obvious that both components of the substrate, the auxin moiety as well as the amino acid residue, are important for enzyme specificity and activity. Which parts of the auxin–amino acid conjugate molecule, and which amino acid moieties of the protein, are involved in substrate binding and the hydrolyzing reaction are so far unclear.

Based on our experimental data, and previous reports on metal cofactors involved in hydrolyzing reactions catalyzed by auxin amidohydrolases (LeClere et al. 2002), we used Mn2+ as a metal cofactor. Its binding site predicted for BrILL2, by comparative modeling of the derived 3D structure, is similar to the one found for the Bacillus subtilis YXEP protein (Minasov et al. 2005) as well as to the one recently predicted for AtILL2 (Bitto et al. 2009). Starting from this structure we found several possible modes of IPA-Ala binding to BrILL2 (Fig. 4) and could propose two possible scenarios for hydrolysis. In the first one (BM1), the water which is coordinated by the metal ion and, at the same time, H-bonded to both the carbonyl and the carboxyl groups of the substrate might be the water to be activated (H2O→OH) in order to accomplish the nucleophilic attack at the carbonyl C. According to the theory of Mock and Zhang (1991), and in agreement with the stability of the arrangement found, the substrate carboxylate may be the proton transfer agent in the hydrolytic cycle if the carboxy group is properly positioned. In BM1 the IPA-Ala carboxyl group is H-bonded to the Gly369 amide (NH) which assures its appropriate position.

In the second scenario (BM2), the Glu175 would activate the water molecule which would in turn accomplish the nucleophilic attack at the carbonyl C. This is the water molecule which is, at the same time, coordinated by a manganese ion and H-bonded to Glu175 and the substrate carbonyl. In this scenario, His399 may have an important role in the proton shuffling process.

The majority of amino acid residues in and close to the enzyme active site appear to be conserved among members of the M20 family; however, there are also a few differences which may be speculated to explain the different specificities of these enzymes. Comparing three auxin hydrolases with different substrate specificities (AtILL2 specific for IAA-Ala, BrILL2 most specific for IPA-Ala, and TaIAR3 most specific for IBA-Ala) ( Fig. 7), some key amino acid moieties were noted. For example, instead of Ile366 and Phe385, which in BrILL2 appear to play important roles in selectivity towards the amino acid part of the substrates and indole ring stabilization, respectively, in AtILL2 there are Val364 and Leu383. In TaIAR3, the situation regarding these two amino acids is more similar to that in BrILL2: there is Phe379 as in BrILL2, while isoleucine is replaced by leucine (Leu359). Bitto et al. (2009) assumed Phe214 to be important for indole ring stabilization in AtILL2. In BrILL2 it is replaced by methionine (Met216), and does not interact directly with the substrate, but interacts with Ile366 through a backbone atom-mediated H-bond, and stabilizes the Phe385 side chain in the proper orientation by C–H–π interactions. Both these residues take part in the stabilization of the IPA-Ala indole ring. TaIAR3 has isoleucine at that position which is again more similar to BrILL2 than to AtILL2.

Fig. 7

Alignment of amino acid sequences of auxin amidohydrolases from Arabidopsis thaliana (AtILL2, NCBI: NP_200477.1) (1), Brassica rapa (BrILL2, GenBank: ABB60092.1) (2) and Triticum aestivum (TaIAR3, GenBank: AAU06081.1) (3). Identity between 1 and 2 is 76%, between 1 and 3 is 54%, and between 2 and 3 is 53%. The alignment was generated by ClustalW and visualized by BioEdit version 7.0.9.0. Identical amino acids are shaded black, and similar amino acids are shaded gray. Amino acids discussed in the text are marked by asterisks. 99 × 82 mm (600 × 600 dpi).

Fig. 7

Alignment of amino acid sequences of auxin amidohydrolases from Arabidopsis thaliana (AtILL2, NCBI: NP_200477.1) (1), Brassica rapa (BrILL2, GenBank: ABB60092.1) (2) and Triticum aestivum (TaIAR3, GenBank: AAU06081.1) (3). Identity between 1 and 2 is 76%, between 1 and 3 is 54%, and between 2 and 3 is 53%. The alignment was generated by ClustalW and visualized by BioEdit version 7.0.9.0. Identical amino acids are shaded black, and similar amino acids are shaded gray. Amino acids discussed in the text are marked by asterisks. 99 × 82 mm (600 × 600 dpi).

Bitto et al. (2009) assumed Leu175 to be responsible for selectivity against IAA–amino acid conjugates with amino acid chains bulkier than alanine or serine. This residue is preserved in BrILL2, but regarding the binding modes that we found it seems a bit too far for such a role. Phe383 creates a hydrophobic environment for the alanine methyl group and it appears more appropriate for such a role in selectivity for the amino acid part of the substrate. This residue is preserved in AtILL2 (Phe381), and in TaIAR3 it is replaced by tyrosine. Mutagenesis experiments will be required to confirm the importance and role of the above-discussed amino acids.

Root growth assay

The biological effects of free auxins and their amino acid conjugates were tested on root growth of Brassica seedlings. The results showed inhibitory effects of the free auxins, IAA, IBA and IPA, which are in accordance with those obtained for Arabidopsis (Woodward and Bartel 2005) and water cress (Jerchel and Staab-Müller 1954), both of which are close relatives to Chinese cabbage. In standard auxin bioassays, in addition to IAA and IBA, which were examined in more detail, IPA showed moderate auxin activity (Fawcett et al. 1960, Katekar 1979). Amino acid conjugates of IAA were shown to exhibit auxin activity in numerous bioassays (for review, see Bartel et al. 2001). Among the amino acid conjugates examined herein, IPA-Ala and IBA-Ala caused strong inhibition of root growth in Brassica seedlings with IC50 values of 9.53 × 10–6 and 3.67 × 10–6 M. As these conjugates are also the preferred substrates of BrILL2 and BrIAR3, their activity suggests that at least part of their inhibitory effect is due to hydrolysis releasing the respective free auxin. Indeed, in bean stem sections, the biological activity of amino acid conjugates was found to be directly correlated to the amount of free auxin resulting from their hydrolysis (Bialek et al. 1983).

Regulation of BrILL2 gene transcription

It was questionable whether the presence of amino acid conjugates which are substrates of BrILL2 may have an impact on the expression of the BrILL2 gene. The regulation of auxin amidohydrolase genes by potential substrates, or by the reaction products, has not been examined previously. The results of our experiments implied that none of the examined auxin–amino acid conjugates, nor the corresponding free auxins as products of the hydrolyzing reaction, has significant effects on the cumulative accumulation of BrILL2 transcripts in young seedlings. It is well known from published data that gene expression of different hydrolases is spatially and temporally controlled, as observed in Medicago (Campanella et al. 2008) and Arabidopsis species (Campanella et al. 2003b, Rampey et al. 2004). Furthermore, processes such as the colonization with an arbuscular mycorrhizal fungus or the induction of rhizobial nodulation in Medicago (Campanella et al. 2008) and the infection of Brassicaceae with Plasmodiophora brassicae (Schuller and Ludwig-Müller 2006) may mediate the expression of some hydrolase genes. The regulation of gene expression of auxin amidohydrolases is obviously a complex process affected by many factors and will be examined in future experiments.

Physiological role of IPA

The physiological role of IPA and its conjugates in planta is still incompletely understood. Based on retention values in thin-layer and gas chromatographies, Linser et al. (1954), Bayer (1969) and Aung (1972) suggested IPA’s occurrence in different varieties of Brassica oleracea L., in leaves of Nicotiana tabacum L. and in Lycopersicon esculentum Mill. cotyledons. IPA was first identified by GC-MS in the hypocotyls of Cucurbita pepo L. (Segal and Wightman, 1982), then in the roots of Pisum sativum L. seedlings at a quite low level (5 pmol g FW–1) (Schneider et al. 1985). On the other hand, microorganisms such as Clostridium species (Elsden et al. 1976) and ruminal bacteria (Mohammed et al. 2003) are able to produce IPA from tryptophan; others like Pseudomonas solanacearum and Ralstonia solanacearum are strongly sensitive to IPA and its derivatives (Matsuda et al. 1993, Matsuda et al. 1998), which makes those indoles possibly interesting as antimicrobial agents. Finally, some bacteria such as Bacillus megatherium developed resistance to IPA through conjugation with amino acids and sugars (Tabone and Tabone 1953, Tabone 1958). Walker et al. (2003) reported IPA as one of the secondary metabolites found in the root exudate of Arabidopsis upon treatment with salicylic acid, which is usually produced by plants as a response to microbial attack. It is still questionable if the so far detected IPA in the above-mentioned plants results from bacterial contamination, represents a response to some kind of stress or has an, as yet unidentified, endogenous role. To our knowledge, IPA conjugates have not yet been identified in plants. It is thus an open question if IPA-Ala is a natural substrate of the two herein examined Brassica hydrolases. Additional experiments on the possible identification of IPA and IPA conjugates in B. rapa, which are in progress, may lead to more details and conclusions about their physiological role in vivo.

Materials and Methods

General

Commercial, analytical grade chemicals and solvents were used if not stated otherwise. Thin-laye chromatography (TLC) was performed on glass plates coated with silica gel GF254 (Merck, Darmstadt, Germany). The chromatograms were developed with 2-propanol : ethyl acetate : concentrated aqueous ammonia (35 : 45 : 20, by vol.) or dichloromethane : methanol : acetic acid (90 : 10 : 1, by vol.). Indolic compounds were detected by UV absorbance or by spraying with Ehrlich’s reagent (1% dimethylaminobenzaldehyde in ethanol : HCl, 1 : 1).

HPLC was performed using a reversed-phase (Nucleosil C18; Macherey-Nagel, Düren, Germany) HPLC column (25 cm × 4.6 mm i.d., 5 μm particle size), connected to a refillable guard column (20 × 2.0 mm i.d.) packed with ‘porous C18’ silica gel, particle size 35–70 μm (Alltech, Deerfield, IL, USA) and kept at a temperature of 22°C. The eluent was mixed from components A (methanol) and B (100 mM formic acid in 5% aqueous methanol). The elution cycle started with 25% A in B (1 min), continued with a series of linear gradients to reach 50% A at 11 min, 60% A at 26 min and 100% A at 36 min, then returned to the initial conditions (25% A at 50 min), and was completed by 15 min of stabilization at 25% A. The flow rate was 1 ml min–1. The effluent was monitored for absorbance at 284 nm using a diode array detector (K-2800, Knauer, Berlin, Germany). Chromatograms were analyzed using the software ChromGate 2.8 (Knauer).

Preparation of the substrates

N-(indol-3-yl)acetyl–l-alanine (IAA-Ala) (Wieland and Hörlein 1955) and N-[3-(indol-3-yl)]propionyl–l-alanine (IPA-Ala) (Campanella et al. 2004) were synthesized as published previously.

N-[4-(indol-3-yl)]butyryl–l-alanine (IBA-Ala). IBA (1.016 g, 5 mmol) and N-hydroxysuccinimide (604 mg, 5.25 mmol) were dissolved in a stirred mixture of dry, peroxide-free, dioxane (6 ml) and dry ethyl acetate (4 ml), and dicyclohexylcarbodiimide (1.135 g, 5.5 mmol) was added gradually, during 45 min, at –10°C. The mixture was stirred overnight at 4°C. The precipitated dicyclohexylurea was removed by filtration. Evaporation of the filtrate afforded the crude N-hydroxysuccinimide ester of IBA as a yellowish oil. The latter was dissolved in peroxide-free dioxane (20 ml) and a solution of l-Ala (445 mg, 5 mmol) in 10% aqueous NaHCO3 was added, with vigorous stirring, during 15 min, at room temperature. After 90 min, the resulting suspension was diluted with water (30 ml) and extracted with ether (2 × 30 ml) to remove by-products. The aqueous phase was acidified to pH 1.5 and partitioned against ethyl acetate (4 × 50 ml). The organic phase was evaporated. The residue was redissolved in 2-propanol : water (1 : 1; 5 ml) and passed through a column (48 × 2.5 cm) of Sephadex LH-20, eluted with the same solvent, to yield the title compound (957 mg, 70%) as a yellowish, amorphous, solid. Purity by HPLC: 95% (retention time 17.8 min; conditions as described above); contained no free IBA. The mass spectrum and nuclear magnetic resonance (NMR) spectrum details are given in the Supplementary data.

N-[3-(indol-3-yl)propionyl]–l-alanine (IPA-Bal). To a cooled (–10°C) solution of 3-(indol-3-yl)propionic acid (2 g, 10.57 mmol) and N-hydroxysuccinimide (1.28 g, 11.10 mmol) in a mixture of dry ethyl acetate (8 ml) and dry peroxide- free dioxane (12 ml), dicyclohexylcarbodiimide (2.40 g, 11.63 mmol) was added in small portions, during 30 min, with stirring. Stirring was continued at 0°C for 30 min, and at 4°C overnight. The dicyclohexylurea formed was removed by filtration, and the filtrate was evaporated. Recrystallization of the residue from 2-propanol (100 ml) afforded IPA-N-hydroxysuccinimide ester (2.65 g, 87.4%) as off-white crystals, melting point 139–140°C. Purity by HPLC: 97.2 % (retention time 21.3 min).

To an aliquot (1,145 mg, 4 mmol) of IPA-N-hydroxysuccinimide ester, dissolved in peroxide-free dioxane (17 ml), was added a solution of Bal (392 mg, 4.4 mmol) in 10% aqueous NaHCO3 (17 ml), with stirring, during 5 min. Stirring was continued for a further 15 min, when TLC indicated that the reaction was complete. The reaction mixture was diluted with water (10 ml). Colored impurities were extracted with ether (2 × 10 ml). The aqueous phase was acidified to pH 1.5 and extracted with ethyl acetate (4 × 50 ml). The organic phase was extracted with brine (2 × 5 ml), dried over anhydrous Na2SO4 and evaporated to yield the crude title compound as a yellow foam. Repeated crystallization from 50% aqueous 2-propanol afforded white crystals (679 mg), melting point 145–146°C, purity by HPLC: 99.2% (retention time 15.7 min). The combined mother liquors were chromatographed on a column (50 × 2.5 cm) of Sephadex LH-20 and eluted with 2-propanol : water (1 : 1) to yield a further crop of the title compound (213 mg), melting point 142–143°C, purity by HPLC: 99.0%. Overall yield: 95.6%. Mass spectrum and NMR spectrum details are given in the Supplementary data.

N-[3-(indol-3-yl)propionyl]–dl-aspartic acid (IPA-Asp). To a stirred solution of IPA-N-hydroxysuccinimide ester (1 g, 3.49 mmol), prepared as described above, in dioxane (15 ml) was added a solution of dl-aspartic acid (511 mg, 3.84 mmol) in 10% aqueous NaHCO3 (15 ml), during 5 min. Stirring was continued until TLC indicated that the reaction was complete (90 min). Work-up as described for IPA-Bal afforded the crude title compound (1.24 g). The latter was dissolved in 2-propanol : H2O (1 : 1; 5 ml) and passed through a column (50 × 2.5 cm) of Sephadex LH-20, eluting with 2-propanol : H2O (1 : 1), to yield the main fraction of the title compound (830 mg), purity by HPLC 98.3% (retention time 13.8 min), as a pink solid foam, in addition to 105 mg of less pure (94.1%) material. Mass spectrum and NMR spectrum details are given in the Supplementary data.

Protein expression and purification

For heterologous expression in E. coli, the previously cloned cDNAs corresponding to the BrIAR3 24 and BrILL2 proteins (Schuller and Ludwig-Müller 2006) inserted into the expression vector pTrcHis2-Topo (appends a His-tag to the expressed protein) were used. A few colonies of freshly transformed cells of the E. coli strain BL21(DE3)RIL+ (Stratagene, La Jolla, CA, USA) were grown in 10 ml of Luria–Bertani (LB) broth containing 100 μg ml–1 ampicillin, overnight, at 37°C. One liter of LB medium supplemented with ampicillin was inoculated with the overnight culture, incubated at 37°C, with vigorous shaking until reaching an A600 between 0.6 and 0.8, supplemented with isopropyl-β-d-thiogalactopyranoside (IPTG; 0.5 mM) to induce protein expression, and incubated at 20°C for 3 h. The bacteria were then harvested by centrifugation at 5,000 × g for 20 min at 4°C, and stored at –20°C until use.

To isolate recombinant proteins, the bacterial pellet was resuspended in lysis buffer [50 mM Tris–HCl, 300 mM NaCl, 10 mM imidazole, 50 μg ml–1 phenylmethylsulfonyl fluoride (PMSF), 1 mg ml–1 lysozyme, 0.1% Triton X-100, pH 8.0], 10 ml g–1 of pellet and, following incubation on ice for 30 min, frozen (liquid nitrogen) and thawed three times. The resulting lysate was supplemented with 10 μg of DNase I (Boehringer Mannheim GmbH, Germany), left at room temperature for 20 min and centrifuged (16,000 × g, 40 min, 4°C). For purification of the His-tagged protein, the supernatant was mixed with 50 μl of a 50% slurry of Ni-NTA–agarose (Qiagen, Valencia, CA, USA) and incubated for 1 h at 4°C under rotation. The Ni-NTA agarose was then washed three times with washing buffer (50 mM Tris–HCl, 300 mM NaCl, 20 mM imidazole, pH 8.0). The bound proteins were eluted with 3 × 100 μl of elution buffer (50 mM Tris–HCl, 300 mM NaCl, 250 mM imidazole, pH 8.0). The proteins recovered were analyzed by SDS–PAGE (12.5%) and quantified according to Bradford (1976).

Enzyme assay

Enzyme reactions were run in 100 mM Tris–HCl, pH 8.0, in the presence of 0.2 mM MnCl2 and 1 mM DTT, at 37°C. In addition to Mn2+, the two-valent metal ions Zn2+, Mg2+, Ni2+, Ca2+, Cu2+ and Co2+ were tested as metal cofactors in concentrations of 0.1–1 mM. The effect of the reducing agents DTT and β-mercaptoethanol was also examined. Amino acid conjugates of the auxins, IAA, IPA and IBA were screened as potential substrates.

Enzyme kinetics were determined for BrILL2 and the alanine conjugates of IPA, IBA and IAA calculating initial reaction rates (v0) in the concentration ranges 0.1–4.0, 0.2–4.5 and 0.2–9 mM, respectively. Protein concentrations used were 2.13 × 10–7 M for IPA-Ala, 4.26 × 10–7 M for IBA-Ala and 2.13 × 10–6 M for IAA-Ala. For each concentration, measurements were taken at 5, 10 and 15 min. The reactions were stopped by lowering the pH to 2 by adding 1 M HCl and the mixtures were extracted with ethyl acetate. The organic phase was collected, evaporated to dryness in a stream of nitrogen and re-evaporated twice with 200 μl of hexane to remove traces of acid. The residue was suspended in 60 μl of methanol. The suspension was passed through a membrane filter (Millex-LH, pore size 0.45 μm) and analyzed by HPLC (Knauer).

The amount of free auxin (IAA, IPA or IBA) formed was calculated from the integrated area of the respective HPLC peak. The kinetic constants Vmax, Km and kcat were determined from Michaelis–Menten curves. All calculations were performed using the program PSI-Plot (Poly Software International, Pearl River, NY, USA). Reaction mixtures containing boiled or no enzyme were run for each substrate as controls.

Tests of root growth inhibition

Seeds of Chinese cabbage [B. rapa L. ssp. pekinensis (Lour.) Hanelt cv. Cantonner Witkrop] were purchased from ISP International Seed Processing GmbH, Quedlinburg, Germany. Seeds were washed, sterilized in 3% NaClO, rinsed in sterile distilled water and incubated on 1% agar plates at 4°C for 24 h. The plates were then placed under continuous light, at 22°C, for the next 24 h, for seed germination. One-day-old seedlings, with roots of about 10 mm in length, were placed on 1% agar plates supplemented with the auxins IBA, IPA, IAA or their amino acid conjugates (IBA-Ala, IPA-Ala, IPA-Bal, IPA-Asp and IAA-Ala) at concentrations of 0.1, 1, 10 and 100 μM. Control experiments were performed on 1% agar plates without any auxin. Following incubation of the seedlings under continuous light for a further 24 h at 22°C, the increase of root lengths was monitored.

BrILL2 gene expression analysis

To observe the effect of auxin and auxin conjugates on expression of the BrILL2 gene, three sets of 1-day-old Brassica seedlings were treated with different auxin compounds at 0.1 mM concentration for 24 h (growing conditions and treatment are described above for the bioassay of root growth inhibition). Total RNA was extracted from 100 mg of plant tissue using an RNeasy Mini kit (Qiagen). DNase-treated (Invitrogen) total RNA (1–2 μg) was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Samples were analyzed in the set-up for quantitative real-time PCRs (qRT-PCR) using SYBRGreen I chemistry for the target gene BrILL2 (Schuller and Ludwig-Müller 2006) and TaqMan chemistry for endogenous controls cytochrome oxidase (Weller et al. 2000) and 18S rRNA (Eukaryotic 18S rRNA TaqMan endogenous control, Applied Biosystems, Carlsbad, CA, USA).

All qRT-PCRs were performed on an ABI PRISM® 7900 HT Sequence Detection System (Applied Biosystems) as described earlier (Hren et al. 2009). The relative quantification approach was used to quantify the presense of BrILL2 transcript as described in Pfaffl (2001) with the additional quality control system in place (Hren et al. 2009) using cytochrome oxidase and 18S rRNA as endogenous controls required for the normalization process. The final results on BrILL2 gene expression were calculated as an average of all three biological replicates of the experiment and standard error taking into account correlation coefficients of independent BrILL2 and endogenous controls measurements.

Modeling

System preparation. The 3D structure of BrILL2 was modeled by the program Modeller9v2 (Sali and Blundell 1993) using the X-ray structures of IAA-amino acid hydrolase from A. thaliana (PDB_ids 1XMB and 2Q43) as templates (Wesenberg et al. 2004, Levin et al. 2007).

We assumed the presence of two Mn2+ ions within the protein, and their positions were determined by combining the grid-based method for sodium cation positioning available in the Leap modul of the AMBER9 package (http://ambermd.org/; Case et al. 2005) with the position of the two Ni2+ ions in the Bacillus subtilis YXEP protein (pdbid 1YSJ) (Minasov et al. 2005). Sequence identity between 1YSJ and BrILL2 is about 36%.

Substrate docking was accomplished by the AutoDock3.05 program (Morris et al. 1999), and the most populated poses were selected. (Details of substrate preparation and docking are given in the Supplementary data.) The complexes were completed by hydrogen atom addition. According to the experimental conditions used for the kinetic measurements, all Asp and Glu were negatively charged and all Lys and Arg positively charged. Histidines were neutral and their protonation site (either Nδ or Nε) was determined according to their neighborhood and probability of H-bond formation.

The parameters for the Mn2+ ions were derived combining the results of quantum mechanical calculations and the data determined by a PDB survey (Dokmanić et al. 2008, Yang et al. 2008) (for details see Supplementary data).

Simulations. Energy minimization and MD simulations were performed with the SANDER modules of the AMBER9 and AMBER10 packages (Case et al. 2008) using the ff03 force field (Duan et al. 2003) (simulation details are given in the Supplementary data).

All calculations were performed on Dual Core Opteron under the Linux (SuSE) operating system.

Supplementary data

Supplementary data are available at PCP online.

Funding

The Croatian Ministry of Science, Education and Sports (grants No. 098-0982913-2829 and 098-1191344-2860); Slovenian research agency program P4-0165; the Joint Board for Scientific and Technological Cooperation between the Republic of Croatia and the Republic of Slovenia.

Acknowledgments

We are particularly indebted to Dr. Marija Abramić for critical reading of the manuscript and constructive discussions. We thank Vladimir Vraneša for technical assistance.

Abbreviations

    Abbreviations
  • Bal

    β-alanine (3-aminopropionic acid)

  • DTT

    dithiothreitol

  • IBA

    indolebutyric acid [4-(indol-3-yl)butyric acid]

  • IPA

    indolepropionic acid [3-(indol-3-yl)propionic acid]

  • IPTG

    isopropyl β-d-thiogalactopyranoside

  • LB

    Luria–Bertani

  • MD

    molecular dynamics

  • NTA agarose

    nitrilotriacetic acid agarose

  • PMSF

    phenylmethylsulfonyl fluoride

  • qRT-PCR; quantitative real-time PCR; TLC

    thin-layer chromatography.

References

Andersson
B
Sandberg
G
Identification of endogenous N-(3-indolacetyl)aspartic acid in Scots pine (Pinus sylvestris L.) by combined gas chromatography–mass spectrometry, using high-performance liquid chromatography for quantification
J. Chromatogr.
 , 
1982
, vol. 
238
 (pg. 
151
-
156
)
Andreae
WA
Good
NE
The formation of indoleacetylaspartic acid in pea seedlings.
Plant Physiol.
 , 
1955
, vol. 
30
 (pg. 
380
-
382
)
Aung
LH
The nature of root-promoting substances in Lycopersicon esculentum seedlings.
Physiol. Plant.
 , 
1972
, vol. 
26
 (pg. 
306
-
309
)
Bartel
B
Fink
G
ILR1, an amidohydrolase that releases active indole-3-acetic acid from conjugates
Science
 , 
1995
, vol. 
268
 (pg. 
1745
-
1748
)
Bartel
B
LeClere
S
Magidin
M
Zolman
BK
Inputs to the active indole-3-acetic acid pool: de novo synthesis, conjugate hydrolysis, and indole-3-butyric acid β-oxidation
J. Plant Growth Regul.
 , 
2001
, vol. 
20
 (pg. 
198
-
216
)
Bayer
MH
Gas chromatographic analysis of acidic indole auxins in Nicotiana.
Plant Physiol.
 , 
1969
, vol. 
44
 (pg. 
267
-
271
)
Bialek
K
Meudt
WJ
Cohen
JD
Indole-3-acetic acid (IAA) and IAA conjugates applied to bean stem sections.
Plant Physiol.
 , 
1983
, vol. 
73
 (pg. 
130
-
134
)
Bitto
E
Bingman
CA
Bittova
L
Houston
NL
Boston
RS
Fox
BG
, et al.  . 
X-ray structure of ILL2, an auxin-conjugate amidohydrolase from Arabidopsis thaliana
Proteins
 , 
2009
, vol. 
74
 (pg. 
61
-
71
)
Bradford
MM
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding.
Anal. Biochem.
 , 
1976
, vol. 
72
 (pg. 
248
-
254
)
Campanella
JJ
Bakllamaja
V
Restieri
T
Vomacka
M
Herron
J
Petterson
M
, et al.  . 
Isolation of an ILR1 auxin conjugate hydrolase homolog from Arabidopsis suecica.
Plant Growth Regul.
 , 
2003
, vol. 
39
 (pg. 
175
-
181
)
Campanella
JJ
Ludwig-Müller
J
Bakllamaja
V
Sharma
V
Cartier
A
ILR1 and sILR1 IAA amidohydrolase homologs differ in expression pattern and substrate specificity.
Plant Growth Regul.
 , 
2003
, vol. 
41
 (pg. 
215
-
223
)
Campanella
JJ
Olajide
AF
Magnus
V
Ludwig-Müller
J
A novel conjugate hydrolase from wheat with substrate specificity for longer side-chain auxin amide conjugates.
Plant Physiol.
 , 
2004
, vol. 
135
 (pg. 
2230
-
2240
)
Campanella
JJ
Smith
SM
Leibu
D
Wexler
S
Ludwig-Müller
J
The auxin conjugate hydrolase family of Medicago truncatula and their expression during the interaction with two symbionts.
J. Plant Growth Regul.
 , 
2008
, vol. 
27
 (pg. 
26
-
38
)
Case
DA
Cheatham
TE
III
Darden
TA
Gohlke
H
Luo
R
Merz
KM
Jr
, et al.  . 
The Amber biomolecular simulation programs.
J. Computat. Chem.
 , 
2005
, vol. 
26
 (pg. 
1668
-
1688
)
Case
DA
Darden
TA
Cheatham
TE
III
Simmerling
CL
Wang
J
Duke
RE
, et al.  . 
AMBER 10
2008
San Francisco
University of California
Chiwocha
S
von Aderkas
P
Endogenous levels of free and conjugated forms of auxin, cytokinins and abscisic acid during seed development in Douglas fir
Plant Growth Regul.
 , 
2002
, vol. 
36
 (pg. 
191
-
200
)
Cohen
JD
Identification and quantitative analyses of indole-3-acetyl-l-aspartate from seeds of Glycine max L.
Plant Physiol.
 , 
1982
, vol. 
70
 (pg. 
749
-
753
)
Davies
RT
Goetz
DH
Lasswell
J
Anderson
MN
Bartel
B
IAR3 encodes an auxin conjugate hydrolase from Arabidopsis.
Plant Cell
 , 
1999
, vol. 
11
 (pg. 
365
-
367
)
Dokmanić
I
Šikić
M
Tomić
S
Correlation between the metal ion type, coordination number and the amino acid residues involved in the coordination
Acta Crystallogr. D
 , 
2008
, vol. 
64
 (pg. 
257
-
263
)
Duan
Y
Wu
C
Chowdhury
S
Lee
MC
Xiong
G
Zhang
W
, et al.  . 
A point-charge force field for molecular mechanics simulations of proteins.
J. Computat. Chem.
 , 
2003
, vol. 
24
 (pg. 
1999
-
2012
)
Elsden
SR
Hilton
MG
Waller
JM
The end products of the metabolism of aromatic amino acids by Clostridia.
Arch. Microbiol.
 , 
1976
, vol. 
107
 (pg. 
283
-
288
)
Epstein
E
Baldi
BG
Cohen
JD
Identification of indole-3-acetylglutamate from seeds of Glycine max L.
Plant Physiol.
 , 
1986
, vol. 
80
 (pg. 
256
-
258
)
Epstein
E
Ludwig-Müller
J
Indole-3-butyric acid in plants: occurrence, synthesis, metabolism, and transport
Physiol. Plant.
 , 
1993
, vol. 
88
 (pg. 
382
-
389
)
Epstein
E
Sagee
O
Zelcer
A
Uptake and metabolism of indole-3-butyric acid and indole-3-acetic acid by Petunia cell suspension culture.
Plant Growth Regul.
 , 
1993
, vol. 
13
 (pg. 
31
-
40
)
Epstein
E
Wiesman
Z
Improved vegetative propagation of olive cultivars with IBA-alanine.
Olea
 , 
1987
, vol. 
18
 (pg. 
35
-
38
)
Fawcett
CH
Wain
RI
Wightman
F
The metabolism of 3-indolylalkanecarboxylic acids, and their amides, nitriles and methyl esters in plant tissues
Proc. R. Soc. B: Biol. Sci
 , 
1960
, vol. 
152
 (pg. 
231
-
254
)
Fitze
D
Wiepning
A
Kaldorf
M
Ludwig-Müller
J
Auxins in the development of an arbuscular mycorrhizal symbiosis in maize.
J. Plant Physiol.
 , 
2005
, vol. 
162
 (pg. 
1210
-
1219
)
Hren
M
Ravnikar
M
Brzin
J
Ermacora
P
Carraro
L
Bianco
PA
, et al.  . 
Induced expression of sucrose synthase and alcohol dehydrogenase I genes in phytoplasma-infected grapevine plants grown in the field.
Plant Pathol.
 , 
2009
, vol. 
58
 (pg. 
170
-
180
)
Jerchel
VD
Staab-Müller
R
Analytische kennzeichen und wuchsstoffwirkung von homologen und peptiden der indol-3-essigsäure.
Z. Naturforsch.
 , 
1954
, vol. 
9b
 (pg. 
411
-
415
)
Kai
K
Horita
J
Wakasa
K
Miyagawa
H
Three oxidative metabolites of indole-3-acetic acid from Arabidopsis thaliana.
Phytochemistry
 , 
2007
, vol. 
68
 (pg. 
1651
-
1663
)
Katekar
G
Auxins: on the nature of the receptor site and molecular requirements for auxin activity.
Phytochemistry
 , 
1979
, vol. 
18
 (pg. 
223
-
233
)
Kowalczyk
M
Sandberg
G
Quantitative analysis of indole-3-acetic acid metabolites of Arabidopsis.
Plant Physiol.
 , 
2001
, vol. 
127
 (pg. 
1845
-
1853
)
LeClere
S
Tellez
R
Rampey
RA
Matsuda
SPT
Bartel
B
Characterization of a family of IAA–amino acid conjugate hydrolases from Arabidopsis.
J. Biol. Chem.
 , 
2002
, vol. 
277
 (pg. 
20446
-
20452
)
Levin
EJ
Kondrashov
DA
Wesenberg
GE
Phillips
GN
Ensemble refinement of protein crystal structures: validation and application.
Structure
 , 
2007
, vol. 
15
 (pg. 
1040
-
1052
)
Linser
H
Mayr
H
Maschek
F
Papierchromatographie von zellstreckend wirksamen indolkörpern aus Brassica—Arten.
Planta
 , 
1954
, vol. 
44
 (pg. 
103
-
120
)
Ludwig-Müller
J
Indole-3-butyric acid in plant growth and development.
Plant Growth Regul.
 , 
2000
, vol. 
32
 (pg. 
219
-
230
)
Ludwig-Müller
J
Epstein
E
Hilgenberg
W
Auxin-conjugate hydrolysis in Chinese cabbage: characterization of an amidohydrolase and its role during infection with clubroot disease.
Physiol. Plant.
 , 
1996
, vol. 
97
 (pg. 
627
-
634
)
Ludwig-Müller
J
Jülke
S
Bierfreund
NM
Decker
EL
Reski
R
Moss (Physcomitrella patens) GH3 proteins act in auxin homeostasis.
New Phytol.
 , 
2009
, vol. 
181
 (pg. 
323
-
338
)
Matsuda
K
Toyoda
H
Yokoyama
K
Wakita
K
Nishio
H
Nishida
T
, et al.  . 
Growth inhibition of Pseudomonas solanacearum by substituted 3-indolepropionic acids and related compounds.
Biosci. Biotech. Biochem.
 , 
1993
, vol. 
57
 (pg. 
1766
-
1767
)
Matsuda
K
Toyoda
H
Nishio
H
Nishida
T
Bingo
M
Matsuda
Y
, et al.  . 
Control of the bacterial wilt of tomato plants by a derivative of 3-indolepropionic acid based on selective actions on Ralstonia solanacearum.
J. Agric. Food Chem.
 , 
1998
, vol. 
46
 (pg. 
4416
-
4419
)
Minasov
G
Shuvalova
L
Brunzelle
JS
Collart
FR
Anderson
WF
Structure of Bacillus subtilis YXEP protein, a dinuclear metal binding peptidase from M20 family.
2005
 
Protein Data Bank(www.rcsborg/pdb/home) ID 1ysj, DOI: 10.2210/pdb2211ysj/pdb
Mock
WL
Zhang
JZ
Mechanistically significant diastereoselection in the sulfoximine inhibition of carboxypeptidase A
J. Biol. Chem.
 , 
1991
, vol. 
266
 (pg. 
6393
-
6400
)
Mohammed
N
Onodera
R
Or-Rashid
MM
Degradation of tryptophan and related indolic compounds by ruminal bacteria, protozoa and their mixture in vitro
Amino Acids
 , 
2003
, vol. 
24
 (pg. 
73
-
80
)
Morris
G
Goodsell
D
Halliday
R
Huey
R
Hart
W
Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function.
J. Computat. Chem.
 , 
1999
, vol. 
19
 (pg. 
1639
-
1662
)
Nordström
AC
Jacobs
FA
Eliasson
L
Effect of exogenous indole-3-acetic acid and indole-3-butyric acid on internal levels of the respective auxins and their conjugation with aspartic acid during adventitious root formation in pea cuttings.
Plant Physiol.
 , 
1991
, vol. 
96
 (pg. 
856
-
861
)
Normanly
J
Slovin
JP
Cohen
JD
Davies
PJ
Auxin biosynthesis and metabolism.
Plant Hormones. Biosynthesis, Signal Transduction, Action
 , 
2004
Dordrecht
Kluwer Academic Publishers
(pg. 
36
-
62
)
Östin
A
Moritz
T
Sandberg
G
Liquid chromatography/mass spectrometry of conjugates and oxidative metabolites of indole-3-acetic acid.
Biol. Mass Spectrom.
 , 
1992
, vol. 
21
 (pg. 
292
-
298
)
Pfaffl
MW
A new mathematical model for relative quantification in real-time RT–PCR.
Nucleic Acids Res.
 , 
2001
, vol. 
29
 (pg. 
2002
-
2007
)
Rampey
RA
LeClere
S
Kowalczyk
M
Ljung
K
Sandberg
G
Bartel
B
A family of auxin-conjugate hydrolases that contribute to free indole-3-acetic acid levels during Arabidopsis germination.
Plant Physiol.
 , 
2004
, vol. 
135
 (pg. 
978
-
988
)
Sali
A
Blundell
TL
Comparative protein modelling by satisfaction of spatial restraints.
J. Mol. Biol.
 , 
1993
, vol. 
234
 (pg. 
779
-
815
)
Schneider
EA
Kazakoff
CW
Wightman
F
Gas chromatography–mass spectrometry evidence for several endogenous auxins in pea seedling organs.
Planta
 , 
1985
, vol. 
165
 (pg. 
232
-
241
)
Schuller
A
Ludwig-Müller
J
A family of auxin conjugate hydrolases from Brassica rapa: characterization and expression during clubroot disease.
New Phytol.
 , 
2006
, vol. 
171
 (pg. 
1
-
13
)
Segal
LM
Wightman
F
Gas chromatographic and GC–MS evidence for the occurrence of 3-indolylpropionic acid and 3-indolylacetic acid in seedlings of Cucurbita pepo.
Physiol. Plant.
 , 
1982
, vol. 
56
 (pg. 
367
-
370
)
Sonner
JM
Purves
WK
Natural occurrence of indole-3-acetyl aspartate and indole-3-acetyl glutamate in cucumber shoot tissue.
Plant Physiol.
 , 
1985
, vol. 
77
 (pg. 
784
-
785
)
Staswick
PE
Tiryaki
I
Rowe
ML
Jasmonate response locus JAR1 and several related Arabidopsis genes encode enzymes of the firefly luciferase superfamily that show activity on jasmonic, salicylic, and indole-3-acetic acids in an assay for adenylation
Plant Cell
 , 
2002
, vol. 
14
 (pg. 
1405
-
1415
)
Tabone
J
Tabone
D
Bio-esterification du glucose.
V. Bio-synthèse par Bacillus megatherium de l’ester β glucosidique de l’acide indolpropionique. C. R. Hebd. Seances Acad. Sci.
 , 
1953
, vol. 
237
 (pg. 
943
-
944
)
Tabone
D
Biosynthèse par B.
megatherium de combinaisons de l’acide indol propionique avec certains acides aminés. Bull. Soc. Chim. Biol.
 , 
1958
, vol. 
40
 (pg. 
5
-
6
)
Tam
YY
Epstein
E
Normanly
J
Characterization of auxin conjugates in Arabidopsis.
Low steady-state levels of indole-3-acetyl-aspartate, indole-3-acetyl-glutamate, and indole-3-acetyl-glucose. Plant Physiol.
 , 
2000
, vol. 
123
 (pg. 
589
-
595
)
Walker
TS
Bais
HP
Halligan
KM
Stermitz
FR
Vivanco
JM
Metabolic profiling of root exudates of Arabidopsis thaliana.
J. Agric. Food Chem.
 , 
2003
, vol. 
51
 (pg. 
2548
-
2554
)
Weller
SA
Elphinstone
JG
Smith
NC
Boonham
N
Stead
DE
Detection of Ralstonia solanacearum strains with a quantitative, multiplex, real-time, fluorogenic PCR (TaqMan) assay
Appl. Environ. Microb.
 , 
2000
, vol. 
66
 (pg. 
2853
-
2858
)
Wesenberg
GE
Smith
DW
Phillips
GN
Jr
Bitto
E
Bingman
CA
Allard
STM
Center for Eukaryotic Structural Genomics:
X-ray structure of IAA-amino acid hydrolase from Arabidopsis thaliana gene AT5G56660.
2004
 
DOI 10.2210/pdb1xmb/pdb
Wiesman
Z
Riov
J
Epstein
E
Characterization and rooting ability of indole-3-butyric acid conjugates formed during rooting of mung bean cuttings.
Plant Physiol.
 , 
1989
, vol. 
91
 (pg. 
1080
-
1084
)
Wieland
T
Hörlein
G
Synthesen einiger β-Indolacetylaminosäuren und -peptide, Liebigs Ann
Chem.
 , 
1955
, vol. 
591
 (pg. 
192
-
199
)
Woodward
AW
Bartel
B
Auxin: regulation, action, and interaction
Ann. Bot.
 , 
2005
, vol. 
95
 (pg. 
707
-
735
)
Yang
T-Y
Dudev
T
Lim
C
Mononuclear versus binuclear metal-binding sites: metal-binding affinity and selectivity from PDB survey and DFT/CDM calculations.
J. Am. Chem. Soc.
 , 
2008
, vol. 
130
 (pg. 
3844
-
3852
)

Author notes

4This author performed the modeling and any correspondence regarding this should be addressed to him: E-mail, sanja.tomic@irb.hr.