UDP-glucosyltransferase UGT84B1 regulates the levels of indole-3-acetic acid and phenylacetic acid in Arabidopsis

Abstract

Auxin is a key plant growth regulator for diverse developmental processes in plants. Indole-3-acetic acid (IAA) is a primary plant auxin that regulates the formation of various organs. Plants also produce phenylacetic acid (PAA), another natural auxin, which occurs more abundantly than IAA in various plant species. Although it has been demonstrated that the two auxins have distinct transport characteristics, the metabolic pathways and physiological roles of PAA in plants remain unsolved. In this study, we investigated the role of Arabidopsis UDP-glucosyltransferase UGT84B1 in IAA and PAA metabolism. We demonstrated that UGT84B1, which converts IAA to IAA-glucoside (IAA-Glc), can also catalyze the conversion of PAA to PAA-glucoside (PAA-Glc), with a higher catalytic activity in vitro. Furthermore, we showed a significant increase in both the IAA and PAA levels in the ugt84b1 null mutants. However, no obvious developmental phenotypes were observed in the ugt84b1 mutants under laboratory growth conditions. Moreover, the overexpression of UGT84B1 resulted in auxin-deficient root phenotypes and changes in the IAA and PAA levels. Our results indicate that UGT84B1 plays an important role in IAA and PAA homeostasis in Arabidopsis.

1. Introduction

Auxin, a class of plant hormones, is essential for diverse developmental processes, including the formation of leaves, flowers, vasculature, adventitious roots, and lateral roots [1]. Plant organ development requires the formation of concentration gradients of indole-3-acetic acid (IAA), a primary natural auxin, in plant tissues. Endogenous IAA concentrations in plants are strictly regulated via multiple regulation steps: biosynthesis, inactivation, and polar auxin transport [2–4]. Previous studies concerning IAA inactivation in Arabidopsis revealed the occurrence of various auxin metabolic enzymes, such as GRETCHEN HAGEN 3 (GH3) auxin-amido synthetase, IAA CARBOXYL METHYLTRANSFERASE 1 (IAMT1), DIOXYGENASE FOR AUXIN OXIDATION (DAO), and the UDP-glucosyltransferase UGT84B1 (Fig. 1A) [5]. Members of the GH3 auxin-amido synthetases play important roles in the conversion of IAA to IAA-amino acid conjugates. For example, GH3.17/VAS2 forms IAA-glutamate (IAA-Glu) conjugate and plays an important role in hypocotyl elongation during the shade avoidance response [6]. IAMT1 catalyzes the conversion of IAA to IAA-methyl ester (MeIAA) and potentially regulates leaf development [7,8]. DAO-mediated oxidation of IAA to 2-oxindole-3-acetic acid (oxIAA) is essential for anther dehiscence, pollen fertility, and seed initiation in rice [9]. In Arabidopsis, oxIAA is further metabolized to oxIAA-glucoside (oxIAA-Glc) by UGT74D1 (Fig. 1A) [10]. IAA-glucoside (IAA-Glc) is one of the major IAA metabolites in angiosperms and gymnosperms [11–13]. Arabidopsis UGT84B1 catalyzes the glucosylation of IAA to IAA-Glc in vitro and in vivo, but its physiological role remains unclear [14,15]. The overexpression of UGT84B1 causes a remarkable increase in IAA-Glc levels and severe auxin-deficient phenotypes in Arabidopsis [15]. Interestingly, UGT84B1 overexpression also increases the levels of IAA [15], suggesting that certain hydrolases may actively hydrolyze IAA-Glc to IAA. UGT84B1 exhibits high catalytic activity towards IAA, but also uses various compounds, such as indole-3-butyric acid (IBA) and cis-cinnamic acid, as substrates in vitro [14]. These observations suggest that UGT84B1 may be implicated in the metabolism of various auxin-related compounds in Arabidopsis.

Fig. 1.

Enzyme activity of UGT84B1 proteins with IAA and PAA

(A) Schematic representation of the IAA metabolic pathways in plants. (B) Schematic representation of the PAA metabolic pathways in plants. (C) MS/MS chromatogram (m/z 336.0 > 130.1) of authentic IAA-Glc (upper), the reaction product of UGT84B1 with IAA (middle), and the control without UGT84B1 (lower). IAA-Glc is detected at 5.4 min. (D) The MS/MS chromatogram (m/z 297.1 > 91.1) of authentic PAA-Glc (upper), the reaction product of UGT84B1 with PAA (middle), and the control without UGT84B1 (lower). PAA-Glc is detected at 5.3 min.

Phenylacetic acid (PAA), a natural auxin, regulates the expression of various auxin responsive genes via the TRANSPORT INHIBITOR RESPONSE 1/AUXIN SIGNALING F-BOX (TIR1/AFB) pathway [16,17]. IAA directionally moves and forms concentration gradients, whereas PAA does not show these polar transport characteristics in plants [16]. Based on this evidence, we recently proposed a model where IAA and PAA coordinately regulate plant growth and development [18]. In PAA metabolism, GH3s can catalyze the conversion of PAA to its amino acid conjugates (Fig. 1B) [16,19]. More recently, we demonstrated that members of the GH3 family can modulate the ratio of IAA and PAA levels in Arabidopsis [20]. On the other hand, IAMT1 has been shown to exclusively metabolize IAA in Arabidopsis, despite it being able to convert both IAA and PAA to each carboxyl methyl ester in vitro [21]. A previous study on the role of UGT74B1 in glucosinolate biosynthesis demonstrated that it also catalyzes the glucosylation of PAA in vitro [22]. This suggests that UGT84B1 may play a crucial role in PAA metabolism in Arabidopsis.

In this study, we investigate the role of Arabidopsis UGT84B1 in the metabolism of IAA and PAA (Fig. 1A and B). We demonstrate that UGT84B1 expressed in E. coli can convert IAA to IAA-Glc and PAA to PAA-glucoside (PAA-Glc), with a higher catalytic activity towards PAA as a substrate. While no obvious developmental phenotypes were observed, the IAA and PAA levels significantly increased in the null ugt84b1 mutants compared to those in the wild type (WT) plants, suggesting that UGT84B1 metabolizes both auxins in Arabidopsis. The overexpression of UGT84B1 caused auxin-deficient phenotypes, but prompted a significant increase in the IAA and PAA levels. These results suggest that UGT84B1 metabolizes both IAA and PAA to the corresponding glucosides, while un-identified auxin-glucoside hydrolases revert these auxin metabolites back to active forms. Our results indicate that UGT84B1 plays a role in the regulation of IAA and PAA levels in Arabidopsis.

2. Materials and methods

2.1. Plant materials and growth conditions

The Arabidopsis thaliana ecotype Col-0 was used as the WT plants. Arabidopsis seeds were stratified at 4 °C for 2 days in the dark. Seedlings were grown on a Murashige and Skoog (MS) agar medium containing 1% sucrose at 23 °C under long-day light conditions with a 16 h light/8 h dark photoperiod. For ß-estradiol (ER) treatment, UGT84B1ox and pER8 plants [23] were grown vertically on an MS agar medium containing ER (2 μM) for 7 days.

2.2. Chemicals

All chemicals were purchased from Sigma-Aldrich unless otherwise stated. [phenyl-¹³C₆]IAA was purchased from Cambridge Isotope Laboratories. [phenyl-¹³C₆]oxIAA, [¹³C₄, ¹⁵N]IAA-Asp, [¹³C₅, ¹⁵N]IAA-Glu, [phenyl-¹³C₆]PAA, [¹³C₄, ¹⁵N]PAA-Asp, and [¹³C₅, ¹⁵N]PAA-Glu were previously synthesized [10,16,23]. IAA-Glc [1-O-(indol-3-ylacetyl)-ß-D-glucose] and PAA-Glc [1-O-(2-phenylacetyl)-ß-D-glucose] were synthesized according to the methods of Kai et al. [24]. Similarly, [2,2-D₂]IAA-Glc and [2,2-D₂]PAA-Glc were also synthesized as described in the supplementary data.

2.3. Preparation of recombinant UGT84B1 proteins

Total RNA was isolated from the WT plants using RNeasy (Qiagen, Venlo, Nether). The PrimeScript™ RT reagent kit with gDNA eraser (Takara Bio, Kusatsu, Japan) was used to generate first-strand cDNA. The cDNA of the UGT84B1 gene was cloned into a pCold-TF vector using the In-Fusion system (Takara Bio, Kusatsu, Japan). The E. coli strain BL21 star (DE3) was transformed using the pCold-TF:UGT84B1 plasmid. The transformants were incubated at 37 °C in a TB medium with 50 μg/mL carbenicillin. Protein expression was induced by adding isopropyl β-D-l-thiogalactopyranoside at a final concentration of 0.2 mM. After incubation at 15 °C for 48 h, the cells were collected by centrifugation. The purification of the trigger factor (TF)-fused UGT84B1 proteins and removal of the TF tag by the Factor Xa were performed as previously described [21].

2.4. Biochemical and steady-state kinetic analysis of UGT84B1

As for the glucosyltransferase assay, IAA-Glc and PAA-Glc formation was analyzed using LC-MS/MS. In a 50 μL reaction, the standard assay conditions included a HEPES buffer (50 mM, pH 7.5), UDP-glucose (4 mM), MgSO₄ (2.5 mM), β-mercaptoethanol (14.2 mM), and IAA or PAA. The kinetic analysis of IAA and PAA used varied auxin concentrations (0, 10, 20, 30, 40, 80, and 150 μM). The enzyme reaction was initiated by the addition of the UGT84B1 protein (4 μg). The reactions were conducted at 30 °C for 2 hours. The reactions were terminated by adding 50 μL of acetonitrile containing [2,2-D₂]IAA-Glc or [2,2-D₂]PAA-Glc. The reaction mixtures were evaporated to dryness using a Speed Vac (Thermo Fisher Scientific, Waltham, MA). The samples were re-dissolved with 20 μL of H₂O and quickly injected into the LC- MS/MS system for IAA-Glc and PAA-Glc analysis. The HPLC separation and MS/MS analysis conditions are shown in Table S1. Steady-state kinetic parameters were calculated from a Lineweaver–Burk plot.

2.5. Generation of ugt84b1 null mutants using CRISPR/Cas9 gene editing technology

A UGT84B1 fragment (approximately 1.1 kb) was to be removed from the Arabidopsis genome using our CRISPR/Cas9 gene editing technology, which employs two gRNAs and cuts the target gene twice [25,26]. One target sequence was CCCGTGATCTCCTCTCCACCGT, and the other was GATTGGCGTTGGCACCTGGTGG (the underlined bold sequences were PAM sites). The knockout mutants were screened using a PCR-based methodology involving the UGT84B1-GT1 and UGT84B1-GT2 primers listed in Table S2. The PCR reaction should generate 0.75 and 1.9 kb bands for mutant and WT DNA, respectively. To further differentiate zygosity, a third primer (UGT84B1-GT3) was used (Table S2). The PCR reaction with primers UGT84B1-GT2 and UGT84B1-GT3 would produce a 0.5 kb band for both heterozygous and WT plants, but no band was produced for ugt84b1 homozygous mutants. The two mutants used in this study had a deletion of 1159 and 1160 bp in UGT84B1, respectively.

2.6. Generation of UGT84B1ox plants

The full-length UGT84B1 cDNA was obtained from the RIKEN BioResource Research Center, Japan. The UGT84B1 ORF was amplified by adapter PCR from the RAFL cDNA clone using gene-specific (UGT84-F and UGT84-R) and adaptor attB primers (adaptor attB1 and attB2). The amplified UGT84B1 fragment was cloned into the pDONR221 vector via the BP clonase II reaction and then transferred into the pMDC7 vector [27] via the LR clonase II reaction (Invitrogen, Waltham, MA). Arabidopsis WT plants were transformed with the resulting construct, pMDC7:UGT84B1, using the floral dip method with the Agrobacterium tumefaciens GV3101 pMP90 strain.

2.7. qRT-PCR analysis

Total RNA was isolated from pER8 and UGT84B1ox plants using RNeasy (Qiagen, Venlo, Nether). The PrimeScript™ RT reagent kit with a gDNA eraser (Takara Bio, Kusatsu, Japan) was used to generate first-strand cDNA. Quantitative RT-PCR (qPCR) was performed on a G8830A AriaMx Real-time PCR system (Agilent Technology, Santa Clara, CA) using a THUNDERBIRD SYBR qPCR mix (Toyobo, Osaka, Japan), for which the specific primers are listed in Table S2.

2.8. LC-MS/MS analysis of auxin metabolites

IAA, oxIAA, IAA-Asp, IAA-Glu, PAA, PAA-Asp, and PAA-Glu were analyzed as previously reported [18]. LC-ESI-MS/MS analysis of IAA-Glc was performed as described in the supplementary data.

3. Results

3.1. UGT84B1 catalyzes the conversion of PAA to PAA-Glc in vitro

To elucidate the glucosyltransferase activity of UGT84B1 against PAA, we expressed the TF-fused UGT84B1 proteins in E. coli using the pCold-TF vector. After the removal of the TF region, UGT84B1 proteins were reacted with IAA or PAA in the presence of UDP-glucose. We analyzed IAA-Glc and PAA-Glc using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Authentic IAA-Glc was identified using a specific product ion at m/z 130.1, which was generated from a parent ion at m/z 336.0 (Fig. 1C). Similarly, authentic PAA-Glc was identified using a specific product ion at m/z 91.1, which was generated from a parent ion at m/z 297.1 (Fig. 1D). As shown in Fig. 1C, recombinant UGT84B1 proteins formed IAA-Glc from IAA and UDP-glucose, as previously described [14]. We found that UGT84B1 proteins also produced PAA-Glc from PAA and UDP-glucose (Fig. 1D). Assays in the absence of UGT84B1 proteins did not yield product ion peaks corresponding to IAA-Glc or PAA-Glc (Fig. 1C and D). Next, we performed a kinetics study on UGT84B1 with IAA and PAA as substrates. The determination of steady-state kinetic parameters for UGT84B1 highlighted the 6.4-fold higher catalytic efficiency (k_cat/K_m) for PAA than IAA (Table 1, Fig. S1). These results indicate that UGT84B1 may play a role in the metabolism of both IAA and PAA in Arabidopsis.

Table 1.

Kinetic analysis of UGT84B1 proteins with IAA and PAA

Steady-state kinetic parameters were determined for IAA and PAA as described in the Materials and Methods. The average values ± SD (n = 3) are shown.

Substrate	Km (μM)	Vmax (μM min−1)	kcat (min−1)	kcat /Km (mM−1 s−1)
IAA	23.07 ± 3.64	0.16 ±0.01	0.1 ± 0.01	0.07 ± 0.01
PAA	32.34 ± 2.64	1.46 ± 0.09	0.88 ± 0.05	0.45 ± 0.01

3.2. Both IAA and PAA levels increased in the ugt84b1 knockout mutants

The physiological role of UGT84B1 in Arabidopsis has not yet been clarified, mainly due to the lack of analysis concerning its loss-of-function mutants. We generated CRISPR/Cas9-based knockout mutants of ugt84b1 (ugt84b1-c1 and -c2) [25], which are deletion mutants of 1159 and 1160 bp, respectively (Fig. 2A, Fig. S2). We found that our ugt84b1-c1 and -c2 lines did not exhibit obvious developmental phenotypes compared with WT plants under our normal laboratory growth conditions (Fig. S3). However, the IAA levels were significantly increased in ugt84b1-c1 (1.3-fold) and -c2 (1.4-fold) lines compared with those of WT plants (Fig. 2B). In addition, the IAA-Glc levels were decreased by 28% and 27% in ugt84b1-c1 and -c2 lines, respectively, implying that other auxin UGTs may also function in Arabidopsis. Moreover, the levels of oxIAA, IAA-aspartate (IAA-Asp), and IAA-Glu were slightly increased in the ugt84b1-c1 and -c2 mutants compared with those of WT plants (Fig. 2B). These results suggest that a reduction in IAA glucosylation leads to an increase in IAA levels, which may promote other IAA inactivation pathways. Similar to IAA, the PAA levels were significantly increased by 1.2-fold in ugt84b1-c1 and -c2 mutants (Fig. 3C), suggesting that UGT84B1 also metabolizes PAA in Arabidopsis. These results suggest that UGT84B1 modulates the IAA and PAA levels in Arabidopsis through glucosylation.

Fig. 2.

Characterization of ugt84b1 knockout mutants

(A) Mutations in ugt84b1-c1 and -c2 mutants. Orange box represents the exon in the UGT84B1 gene. Underlines indicate PAM sites. Arrows indicate the gene-specific primers used for genotyping. (B) The levels of IAA and PAA metabolites in 10-day-old ugt84b1-c1 (84b1-c1) and -c2 (84b1-c2) mutants. Values are the means ± SD (n = 4). Differences between the WT plants and 84b1-c1/-c2 mutants are statistically significant at P < 0.05 (*P < 0.05, **P < 0.01, and ***P < 0.001, Student’s t-test).

Fig. 3.

Analysis of UGT84B1ox plants

(A) Phenotypes of pER8 and UGT84B1ox (84B1ox) seedlings treated with ER (2 μM) for 7 days. Scale bars indicate 1 cm. (B) Primary root length of pER8 and 84B1ox seedlings treated with ER for 7 days. Values are the means ± SD (n = 35). (C) Expression levels of UGT84B1 in pER8 and 84B1ox seedlings treated with ER for 7 days. Differences between pER8 and 84B1ox plants are statically significant at P < 0.01 (ANOVA, Tukey test). (D) The levels of IAA and PAA metabolites in pER8 and 84B1ox plants. Plants (6 days old) were treated with ER for 4 days. Values are the means ± SD (n = 4). Differences between pER8 and 84B1ox plants are statistically significant at P < 0.05 (*P < 0.05, **P < 0.01, and ***P < 0.001, Student’s t-test).

We attempted to analyze the PAA-Glc levels in Arabidopsis plants, but were unable to quantify this metabolite using LC-MS/MS.

3.3. UGT84B1 overexpression alters the IAA and PAA levels in Arabidopsis

To further elucidate the role of UGT84B1 in IAA and PAA metabolism in vivo, we generated transgenic Arabidopsis plants that overexpress UGT84B1 (UGT84B1ox) in an estradiol-inducible expression system using a pMDC7 vector. Two independent lines of UGT84B1ox (#10 and #31) displayed severe root growth defects compared to the empty-vector control pER8 plants (Fig. 3A, 3B), which is consistent with the previously reported phenotypes of 35S promoter-driven UGT84B1 overexpression plants [15]. Using quantitative RT-PCR (qPCR), we confirmed that the expression levels of UGT84B1 were elevated by 5.5-fold in UGT84B1ox #10 lines (Fig. 3C), which displayed severe root growth defects compared with those in pER8 plants. To investigate the impact of UGT84B1 overexpression on auxin homeostasis, we quantified the levels of auxin metabolites in the roots of pER8 and UGT84B1ox #10. Consistent with a previous study [15], the IAA-Glc levels were remarkably increased in UGT84B1ox (6.5-fold) (Fig. 3D). We also observed a significant increase in the IAA levels (12.7-fold) in UGT84B1ox, implying that accumulated IAA-Glc may be hydrolyzed to IAA by certain auxin-glucoside hydrolases. Moreover, the IAA-Asp and IAA-Glu levels were remarkably decreased in UGT84B1ox, suggesting the down-regulation of other IAA inactivation pathways for IAA homeostasis. Similar to IAA, we found that the PAA levels were increased by 1.9-fold in UGT84B1ox (Fig. 3D). Although the PAA-Glc levels in plants were not quantified, a significant increase in the PAA levels suggests that UGT84B1 also contributes to PAA metabolism in vivo.

4. Discussion

In this study, we provide evidence for the role that the auxin glucosyltransferase UGT84B1 plays in both IAA and PAA metabolism in Arabidopsis. First, we demonstrated that UGT84B1 proteins can catalyze the glucosylation of these auxins, with a higher substrate preference for PAA than IAA in vitro. Second, both IAA and PAA levels were significantly increased in Arabidopsis ugt84b1 knockout mutants. Finally, the overexpression of UGT84B1 caused the accumulation of both IAA and PAA in Arabidopsis, with remarkably increased IAA-Glc levels. Intriguingly, our CRISPR/Cas9-based knockout mutant of ugt84b1-c1 did not show obvious developmental phenotypes (Fig. S3), even though the IAA and PAA levels were significantly increased (Fig. 2B). The reason that the ugt84b1-c1 phenotype did not present the effects of increased auxin levels is unclear. However, we also found that ugt84b1-c1 still produces large amounts of IAA-Glc (Fig. 2B), suggesting that other UGTs play a compensatory role in ugt84b1-c1 mutants. As for the other Arabidopsis IAA glucosyltransferases, UGT74E2, an IBA glucosyltransferase involved in the modulation of plant architecture and water stress response, exhibited a weak glucosyltransferase activity against IAA in vitro [11]. Moreover, UGT74D1, which mainly catalyzes the glucosylation of oxIAA to oxIAA-Glc, can form IAA-Glc in vitro [10,28]. Further analysis of multiple knockout mutants of UGT84B1 and these UGTs will help to elucidate the metabolic regulation of IAA in Arabidopsis.

Similar to IAA accumulation in 35S promotor-driven UGT84B1 overexpression plants [15], a remarkable increase in the IAA levels was also observed in the estradiol-dependent UGT84B1ox plants, although the plants showed severe auxin-deficient root phenotypes (Fig. 3). Given that the IAA-Glc levels were remarkably increased in UGT84B1ox plants, these observations suggest that certain auxin-glucoside hydrolases play crucial roles in IAA-Glc homeostasis. Importantly, we observed an increase in the IAA (12.7-fold) and PAA levels (1.9-fold) in the UGT84B1ox plants compared to those in the pER8 control plants (Fig. 3D). Since UGT84B1 exhibits glucosylation activity against both IAA and PAA in vitro (Table 1), the higher accumulation of IAA suggests that certain auxin-glucoside hydrolases may strongly favor IAA-Glc as a substrate in Arabidopsis. A previous study demonstrated that the THOUSAND-GRAIN WEIGHT 6 (TGW6) gene, which encodes IAA-Glc hydrolase, plays important roles in the regulation of grain length and weight in rice [29]. In contrast, auxin-glucoside hydrolases in Arabidopsis still remain unknown. As such, it is crucial to explore auxin-glucoside hydrolases in Arabidopsis to further understand the mechanisms of IAA and PAA homeostasis.

Supplementary Material

1

Supplementary Table S1. HPLC gradient program and MS/MS conditions for IAA-Glc and PAA-Glc analysis

Supplementary Table S2. Primers used in this study

3

Supplementary Fig. S1. Lineweaver-Burk plot of UGT84B1 activity with IAA and PAA

Supplementary Fig. S2. Conformation of the ugt84b1-c mutations in Arabidopsis

Supplementary Fig. S3. Phenotypes of ugt84b1 knockout mutants

Highlights.

• Arabidopsis UGT84B1 proteins catalyzed the glucosylation of both IAA and PAA in vitro.

• Both the IAA and PAA levels increased in the ugt84b1 knockout mutants.

• The ugt84b1 null mutants still produced IAA-glucoside.

• UGT84B1 overexpression altered both the IAA and PAA levels in Arabidopsis.

• UGT84B1 homologs may also contribute to IAA and PAA homeostasis.

Abbreviations:

IAA

indole-3-acetic acid

PAA

phenylacetic acid

IAA-Glc

IAA-glucoside

PAA-Glc

PAA-glucoside

LC-MS/MS

liquid chromatography-tandem mass spectrometry

gFW

gram fresh weight