Skip to main content
NCBI home page
Glycobiology logoLink to Glycobiology
. 2024 Jul 12;34(9):cwae050. doi: 10.1093/glycob/cwae050

Enzymatic glycosylation of aloesone performed by plant UDP-dependent glycosyltransferases

Natalia Putkaradze 1,✉, Laura Dato 2, Onur Kırtel 3, Jørgen Hansen 4, Ditte Hededam Welner 5,✉
PMCID: PMC11273223  PMID: 38995933

Abstract

Aloesone is a bioactive natural product and biosynthetic precursor of rare glucosides found in rhubarb and some aloe plants including Aloe vera. This study aimed to investigate biocatalytic aloesone glycosylation and more than 400 uridine diphosphate-dependent glycosyltransferase (UGT) candidates, including multifunctional and promiscuous enzymes from a variety of plant species were assayed. As a result, 137 selective aloesone UGTs were discovered, including four from the natural producer rhubarb. Rhubarb UGT72B49 was further studied and its catalytic constants (kcat = 0.00092 ± 0.00003 s−1, KM = 30 ± 2.5 μM) as well as temperature and pH optima (50 °C and pH 7, respectively) were determined. We further aimed to find an efficient aloesone glycosylating enzyme with potential application for biocatalytic production of the glucoside. We discovered UGT71C1 from Arabidopsis thaliana as an efficient aloesone UGT showing a 167-fold higher catalytic efficiency compared to that of UGT72B49. Interestingly, sequence analysis of all the 137 newly identified aloesone UGTs showed that they belong to different phylogenetic groups, with the highest representation in groups B, D, E, F and L. Finally, our study indicates that aloesone C-glycosylation is highly specific and rare, since it was not possible to achieve in an efficient manner with any of the 422 UGTs assayed, including multifunctional GTs and 28 known C-UGTs.

Keywords: aloesin, aloesone, aloesone-O-glucoside, glycosyltransferase, rhubarb

Introduction

Aloesone (2-acetonyl-7-hydroxy-5-methylchromone) is a natural product from rhubarb (Rheum palmatum) and Aloe species. It is one of the bioactive constituents of the medicinal plant Aloe vera exhibiting anti-pigmentation, anti-inflammatory and anti-epileptic effects (Lucini et al. 2015; Wang et al. 2022; Wang et al. 2023). The 8-C-glycoside of aloesone, called aloesin, is a valuable ingredient in cosmetic creams due to its powerful anti-pigmentation and wound-healing activities (Jones et al. 2002). Besides that, since glycosylation (a covalent attachment of sugar moieties to molecules through a glycosidic linkage) is a convenient way to change properties of chemicals (e. g. solubility, stability and bioactivity (Gantt et al. 2011; Nidetzky et al. 2018)), there is an increasing interest in glycosides for cosmetic and other industrial applications. In this context, aloesone glycosides represent interesting chemicals for production as well as further studies.

Aloesone glycosides have been found in Nature by isolation from plant extracts (Haynes et al. 1970; Kashiwada et al. 1990). It is believed that in planta aloesone is formed by the action of a type III polyketide synthase (Abe et al. 2004; Mizuuchi et al. 2009) and structurally modified via an enzymatic O-/C-glucosylation reaction leading to the respective glucoside in rhubarb and aloe (Fig. 1). Little is known about the enzymatic glycosylation step in these plants, and the corresponding enzymes have not been identified so far. In a transcriptome-based study of A. vera a highly promiscuous C-/O-glycosyltransferase has been characterized, though no activity on aloesone was detected (Xie et al. 2020). Moreover, even though aloesone can be synthesized chemically (Gramatica et al. 1986; Kim et al. 2017), its glycosylation has not been achieved in vitro neither by chemical nor biocatalytic means. Given that biocatalytic processes might be more sustainable than their chemical counterparts, identifying enzymes for aloesone glycosylation is of considerable importance.

Fig. 1.

Fig. 1

Open in a new tab

Proposed aloesone glycoside pathway in planta. RpPKS: Aloesone synthase from R. Palmatum, AaPKS: Aloesone synthase from A. vera.

Plant UGTs from the GT1 family are obvious candidates for glycosylation of low molecular weight natural products such as aloesone (Bowles et al. 2005; Lombard et al. 2014). These enzymes use a UDP-activated sugar, predominantly UDP-glucose, as their donor substrate and can transfer the sugar residue to an acceptor molecule to produce natural product glycosides. In contrast to their high specificity for the glycosyl donors, UGTs in many cases show low specificity for the acceptors by glycosylating diverse substrates and are, therefore, attractive tools for glycodiversification of natural products from different origins (Bowles et al. 2005; Louveau and Osbourn 2019; Zhang et al. 2022). Despite displaying a high degree of sequence variability, all UGTs adopt the GT-B fold consisting of two Rossman-like domains (Breton et al. 2006; Lairson et al. 2008). The substrate binding pocket is located between the two domains and contains highly – although not completely – conserved His-Asp catalytic dyad (Lairson et al. 2008; Osmani et al. 2009). It is established that UGTs use a SN2-like substitution mechanism and a histidine as a catalytic base (Breton et al. 2006; Brazier-Hicks et al. 2007). The vast majority of UGTs are known to be O-glycosyltransferases, meaning that they glycosylate an oxygen atom. However, C-, N- and S-glycosyltransferases also exist, using either aromatic carbon, nitrogen or sulfur as an acceptor site (Nidetzky et al. 2018; Tegl and Nidetzky 2020; Putkaradze et al. 2021). Moreover, multifunctional UGTs displaying different types of glycosylation activity towards a broad range of acceptors have also been described (Putkaradze et al. 2021).

We hypothesized aloesone to be a promising target for selective glycosylation based on its structural features, such as O- and C-glycosylation sites and a chromone core quite similar to well-known UGT acceptors such as flavones (Putkaradze et al. 2021). Thus, we aimed to study aloesone glycosylation by a variety of plant UGTs, to learn how widespread this activity is, to discover novel aloesone UGTs, and gain first insights into aloesone-glycosylating enzymes (their glycosylation types, catalytic properties, and specificities). To accomplish these goals, we initially tested UGTs from one of the two natural producers of aloesone glycosides, rhubarb. Limited data is available in the literature about rhubarb UGTs and their specificities, and to the best of our knowledge, only one rhubarb UGT (RpUGT1, UGT73BE14) has been characterized so far (Yamada et al. 2020). RpUGT1 was identified as an emodin-6-O-glucosyltransferase and a role in the biosynthesis of anthraquinone and stilbene glucosides in planta was proposed (Yamada et al. 2020). Besides that, the enzyme showed broad substrate specificity by glycosylating various flavonoids, including flavones (Yamada et al. 2020). For this reason, we included RpUGT1 as a particularly promising candidate for aloesone glycosylation in our assay together with other non-characterized rhubarb UGTs (Palcic 2011). In addition to these, we initially performed a targeted screening including 35 plant UGTs which either: I) are evolutionary related to the rhubarb counterparts (e.g. sequence identity >75%, same phylogenetic group); II) have a broad range of known acceptor substrates, including flavones; or III) have proven C-glycosylation activity. In addition, the aloesone UGT candidate space was significantly extended by screening a large UGT library consisting of 380 randomly selected enzymes from most phylogenetic groups. Novel aloesone UGTs were identified, and analysis of all data was employed to investigate if any phylogenetic bias for aloesone glucosylation exists. The most interesting aloesone UGTs were further characterized.

Results

To discover an enzyme capable of glycosylating aloesone, UGTs from different plants were heterologously produced, partially purified and assayed in vitro. We first screened the 7 UGTs from rhubarb (UGT73BE14, UGT71AQ1, UGT92A1, UGT71U24, UGT71U1, UGT76F1 and UGT72B49, (Table S1)) for aloesone glucosylation activity, and found that 4 enzymes were active: UGT72B49, UGT92A1, UGT71U24, and UGT71U1 (Fig. S1). They were all able to transform aloesone into the same product with a retention time of 2.63 min (Fig. 2A) which was well separated from the C-glucoside aloesin (Fig. 2B and E) via HPLC and confirmed to be aloesone 7-O-glucoside by subsequent MS/MS analysis (Fig. 2G and H).

Fig. 2.

Fig. 2

Open in a new tab

HPLC (A–C) and LC–MS/MS data (D–I) showing UGT72B49 reaction with aloesone as well as authentic standards of aloesone and aloesin. The HPLC chromatograms of the reaction with UGT72B49 (A), aloesin standard (B) and aloesone standard (C). The XIC of the aloesone O-glucoside [M + H]+ in the reaction with UGT72B49 (D). The XIC of the aloesone O-glucoside and aloesin [M + H]+ from the reaction in panel D spiked with aloesin (E). Combined XIC of aloesin [M + H]+ and aloesone [M + H]+ from standard solutions (F). MS/MS spectrum of the aloesone O-glucoside-containing reaction in panel D (G), aloesin (H) and aloesone (I). P: Reaction product.

Low protein production titers of UGT92A1, UGT71U24 and UGT71U1 prevented us from performing a detailed comparison and enzymatic characterization of these enzymes. UGT72B49, however, showed sufficient production titers (1.8 mg protein per L culture after one-step IMAC purification) for its further enzymatic characterization. For the following experiments, UGT72B49 was expressed in 6 L of culture volume and purified by using two-step IMAC purification strategy as described in the Materials and Methods section. SDS-PAGE (Fig. S2) confirmed adequate purity of the final sample. Firstly, aloesone glycosylation by UGT72B49 was monitored at four different temperatures (Fig. S3A). Interestingly, the enzyme showed the highest aloesone O-glucoside production at 50 °C for both time points. Therefore, for the following experiment we also included 60 °C and measured initial rates at 5 different temperatures (Fig. 3A). The glycosylation rate of UGT72B49 was found to peak at 50 °C and was >60% of the maximum level of activity at 40 °C. In contrast, it showed only residual activity at 60 °C. Further, we determined the pH optimum of UGT72B49 by measuring initial rates of the reaction at pH 5–10. Optimal pH was estimated to be 7, with ca. 6% decrease of the maximum activity at pH 8, <10% activity at pH 9 and no activity at pH 10 (Fig. 3B). The catalytic constants kcat and KM for aloesone glycosylation by UGT72B49 were determined in a HEPES buffer at pH 7.4 and 37 °C and was found to be 0.00092 ± 0.00003 s−1 (0.055 ± 0.002 min−1) and 30 ± 2.5 μM, respectively (Fig. 3C).

Fig. 3.

Fig. 3

Open in a new tab

Enzymatic characterization of UGT72B49. Temperature optimum (A), pH optimum (B) and Michaelis–Menten curve (C). Mean values were calculated from three independent experiments and standard deviations are shown as error bars.

Furthermore, we investigated if aloesone glycosylation is a widespread capability even in plant species that do not produce aloesone. To this end, we first selected 2 UGTs that share the highest sequence identity with UGT72B49 (UGT72B19 from Fagopyrum esculentum and UGT72B30 from Polygonum tinctorium with 80% and 77% sequence identity, respectively). Since UGT72B49 belongs to phylogenetic group E (Yamada et al. 2020), we also produced and assayed 5 additional UGT enzymes from this phylogenetic group (Table S1). We further expanded the aloesone UGT candidates’ sequence space by adding UGTs with different substrate scope and glycosylation types such as C-glycosyltransferases from the family 708 (Ito et al. 2017; Putkaradze et al. 2023) and multifunctional UGTs showing different types of glycosylation reaction (O-/C-/N-/S-) on a wide range of plant natural products, such as TcCGT1 from Trolius chinensis (He et al. 2019). We hypothesized that UGTs with reported C-glycosylation activity towards other structurally similar substrates were particularly promising candidates. In total 35 UGTs (Table S1) were produced, purified, and tested for activity towards aloesone. As a result, several additional aloesone UGTs were identified (Table S2). Interestingly, all 7 selected UGTs from phylogenetic group E were active on aloesone whereas only 4 enzymes out of 28 C-glycosyltransferases showed low activities (conversion ≤10%). UGT71C1 and UGT72B30 were capable of full conversion of 50 μM aloesone whereas UGT72B19, the UGT with the highest (80%) similarity to UGT72B49, showed <20% conversion after overnight incubation. Interestingly, no UGT could C-glycosylate aloesone in an efficient manner, with only TcCGT1 being capable of forming a trace amount of the aloesone C-glycoside, namely aloesin (<0.5%, data not shown), which was confirmed by HPLC and an authentic standard of aloesin. These results indicate that aloesone is a common substrate for O-glycosylation by UGTs from the phylogenetic group E and that aloesone C-glycosylation by UGTs is rare and requires specific enzymatic properties. To further test this, we then screened a large random UGT library consisting of 380 enzymes from 17 phylogenetic groups (GLY-it library). 122 of these UGTs glucosylated aloesone, among which 28 fully converted it in the overnight reaction (Fig. S6). Surprisingly, UGTs from no less than 13 phylogenetic groups were identified as aloesone UGTs, with the highest representation in groups B, D, E, F, and L. Of these, UGTs from group B came out with the highest percentage of active UGTs among those tested (62%). Figure 4 shows a phylogenetic analysis of all the identified aloesone UGTs.

Fig. 4.

Fig. 4

Open in a new tab

Phylogenetic analysis of aloesone UGTs identified in our study. Each node represents an individual enzyme. UGTs from the GLY-it library are shown as black squares. The tree was constructed by neighbor-joining method. The bootstrap analysis was performed with 100 replicates. The scale bar represents distance between nodes in terms of expected number of substitutions per amino acid.

From the 30 newly discovered aloesone UGTs capable of complete conversion of aloesone overnight, we selected UGT72B30 and UGT71C1 for further characterization since both enzymes also showed relatively high protein production titers in contrast to the vast majority of the investigated UGTs. Conversion % was determined for both enzymes at different temperatures and two timepoints (Fig. S3B and C). UGT72B30 showed 49% lower conversion than UGT71C1 at 30 °C, and therefore only the latter was chosen for further kinetic characterization. UGT71C1 was found to be a more efficient aloesone O-glycosyltransferase, showing 293- and 167-fold higher kcat and kcat/KM than the rhubarb UGT72B49 (Fig. 5, Table 1).

Fig. 5.

Fig. 5

Open in a new tab

Michaelis–Menten curve of UGT71C1 for aloesone glycosylation. The values represent the mean of three independent experiments and standard deviations are shown as error bars.

Table 1.

Kinetic parameters of aloesone O-GTs determined in this study.

name k cat  [s−1] K M  [μM] k cat /K M  [s−1 M−1]
UGT72B49 0.00092 ± 0.00003 30.04 ± 2.56 0.03 × 103
UGT71C1 0.27 ± 0.01 50.54 ± 7.29 0.05 × 105
Open in a new tab

In the following experiment we could achieve the production of 1 mg of aloesone O-glucoside (as quantified with an aloesin calibration curve) by UGT71C1 in 45 mL of total reaction volume (Fig. S5). For this 96% of 100 μM aloesone was glycosylated in 4 h. It is noteworthy that the reaction took place much faster since there was >80% conversion even in the t0 sample taken after a few seconds of the reaction (data not shown).

Discussion

Glycosylated phytochemicals attract interest due their valuable properties and applications as therapeutics, food additives, colorants, and cosmetic ingredients (Griffith et al. 2005; Wang et al. 2008; Ceunen and Geuns 2013; Esatbeyoglu et al. 2015; Louveau and Osbourn 2019). UGTs enable the introduction of sugar moieties into a plethora of small molecules and are, therefore, attractive catalysts for the synthesis of glycosylated natural product with industrial applications. Although many UGTs have been characterized, their catalytic properties are mostly far from being industrially relevant and for many glycosylation reactions the respective glycosyltransferase remains to be discovered. Aloesone glycosides are bioactive chemicals that are known to be synthesized in rhubarb and aloe plants, however, an aloesone glycosyltransferase has never been characterized. Therefore, this study aimed to identify, firstly, aloesone UGTs, and secondly, acquire knowledge enabling the development of efficient biocatalysts for aloesone glycoside synthesis in the industry.

For the first activity assay, we selected 7 rhubarb enzymes as particularly promising natural aloesone UGT candidates. UGT72B49, UGT92A1 and UGT71U24 were identified as aloesone UGTs in our study, but none showed complete conversion of 50 μM of aloesone after overnight incubation. We cannot rule out that specific UGT-mediated glycoside cleavage might have given rise to limited final conversion values. All 3 enzymes converted aloesone to one product, aloesone O-glucoside. UGT72B49 was selected since it was the most efficient rhubarb aloesone UGT and further characterized by estimating the temperature and pH optima as well as Michaelis–Menten constants. Interestingly, in contrast to most characterized wildtype UGTs, the temperature optimum was estimated to be at 50 °C. The pH optimum and KM (7–8 and 30 μM, respectively) are within a typical range of those for other plant UGTs towards flavonoid acceptors, kcat is significantly lower (0.00092 s−1). While the identified aloesone UGTs might display different catalytic properties in planta, UGT72B49 seems to be an interesting target for protein engineering to improve activity in vitro. Besides that, future studies using transcriptome data might lead to the identification of efficient aloesone UGTs from this plant (Liu et al. 2020).

In the following assays we significantly extended the number of aloesone UGT candidates, as we hypothesized aloesone to be a common glycosyl acceptor for characterized UGTs from various plants. For this part of the studies, we selected 7 UGTs that were from the same phylogenetic group as UGT72B49 and included characterized multifunctional UGTs as well as C-glycosyltransferases from the literature. The latter enzymes were deemed promising candidates, particularly for the C-glycosylation of aloesone. It is noteworthy that outside aloe plants, a synthesis of aloesone C-glucoside (aloesin) has been reported in engineered E. coli strain containing a C-glycosyltransferase (GtUF6CGT1) variant although the productivity was very low (Yang et al. 2021). Interestingly, neither any tested C-UGTs, nor multifunctional UGTs showed a sufficient glucoside formation under the assay conditions. As an exception, both C-and O-glucosides were detected in the reaction with TcCGT1 but amounts of both were low. TcCGT1 is a promiscuous and multifunctional (O-/C-/N-/S) UGT with a spacious active site as seen in its crystal structure (6JTD) (He et al. 2019). Our results suggest that aloesone might bind with a low affinity to TcCGT1 and it might adopt multiple binding modes, enabling both C- and O-glycosylation, as well as catalytically incompetent binding modes probably. In contrast, 6 out of 7 selected UGTs from the phylogenetic group E were identified as aloesone O-GTs, and among these, UGT72B30 and UGT71C1 achieved full conversion. Notably, UGT71C1 was identified as a superior aloesone UGT compared to UGT72B49 (293- and 167-fold higher kcat and kcat/KM, respectively). UGT71C1 is known as a quercetin glycosyltransferase (Lim et al. 2004), but it was revealed in our study that it is also a selective aloesone O-glycosyltransferase. Interestingly, while UGT72B49 shows significantly higher substrate affinity for aloesone than UGT71C1, it is catalytically 167-times less efficient and seems to be an interesting candidate for directed evolution efforts to improve the turnover rate. However, we believe that slight/fewer modifications in the active site of UGT71C1 might be more promising to improve binding affinity/catalytic efficiency for aloesone and this UGT is an attractive candidate for rational engineering.

Given that aloesone UGTs from different phylogenetic groups were identified in our study, we wanted to find out if aloesone glycosylation could be accomplished by other UGTs with a high sequence variability, especially in the N-terminal domain, and if we could identify any phylogenetic bias. For this, a GLY-it screen consisting of 380 UGTs were assayed and 122 UGTs were identified as active towards aloesone. Our results in combination with primary sequence analysis suggested that aloesone glycosylation activity is distributed among at least 13 different phylogenetic groups of UGTs. Aloesone seems to be a common UGT acceptor capable of productive binding to many UGTs with remarkable differences in the active site. The highest representation of aloesone UGTs were found in groups B, D, E, F, and L, but since the representation of other groups was pretty low, we cannot identify any group as the most promising one for aloesone glycosylation. It is noteworthy to indicate that the highest number of aloesone UGTs are from group E; however, group B (which was also relatively well represented in the screen) showed the highest percentage of aloesone UGTs internally (aloesone UGTs/total tested UGTs) among all groups.

In conclusion, this work identified the first aloesone UGTs from plants and characterized them as O-glycosyltransferases. Moreover, we showed that many UGTs from different phylogenetic groups are aloesone glycosyltransferases. UGT71C1 was characterized as an efficient UGT biocatalyst for the synthesis of aloesone O-glucoside. In contrast, C-glycosylation of aloesone seems to be very specific since no enzyme could show any significant activity. Aloesone O-GTs identified in this study represent interesting targets for future design of aloesone C-glycosyltransferases for aloesin production as well as further optimize O-glycosyltransferases towards industrial applications.

Materials and methods

Chemicals

Most of the chemicals were purchased from Sigma-Aldrich and were of the highest purity available. Aloesone was synthesized by Aurora Fine Chemicals and was of >95% purity. Aloesin was purchased from Santa Cruz Biotechnology and was of ≥98% purity.

Selection of aloesone GT candidates

Full-length UGT sequences from R. palmatum were obtained from CAZy database (http://www.cazy.org). UGT73BE14, UGT71AQ1 and UGT72B49 were previously described as RpUGT1, RpUGT2 and RpUGT3, respectively (Yamada et al. 2020). UGT72B19 and UGT72B30 selected based on their high sequence similarity to UGT72B49 (80 and 77%, respectively). UGT708A6, UGT72B10, UGT71E5, UGT71D1 and UGT71C1 were randomly selected from our in-house UGT sequence library (Della Gala and Welner 2023) and they shared 32–64% sequence identity with UGT72B49. C-GT and multifunctional UGT sequences were selected from the literature (Chen et al. 2015; Sasaki et al. 2015; Ito et al. 2017; Wang et al. 2017; He et al. 2019; Mashima et al. 2019; Zhang et al. 2020; Putkaradze et al. 2023). Ready-to-use UGT library containing 380 UGTs from 17 phylogenetic groups (GLY-it screen) were obtained from River Stone Biotech.

Heterologous protein production and purification

Sequences encoding the selected UGTs except for UGT72B30 were synthesized and cloned in the pET28a(+) vector (Novagen) between the NcoI and XhoI restriction sites by Biomatik (Canada). The protein coding sequences were lacking the start methionine and contained an N-terminal sequence consisting of a 6xHis-tag and a TEV cleavage site (HHHHHHDYDIPTTENLYFQGS). Expression strains were created by heat shock transformation of respective plasmids into One Shot™ BL21 Star™ (DE3) chemically competent E. coli cells. Transformed cells were chosen on LB plates supplemented with 50 mg/L kanamycin after overnight incubation at 37 °C. The subsequent precultures were grown overnight in LB medium supplemented with 50 mg/L kanamycin at 37 °C. For the heterologous expression of UGT72B30, E. coli Rosetta (DE3) cells (Novagen) were transformed with pTMH307 vector (Hsu et al. 2018) and media with 100 mg/L ampicillin were used. Main expression cultures were prepared by adding 5 mL (1:100 v/v) of overnight culture to 500 mL of 2×YT medium in 2 L baffled Erlenmeyer flasks with the same final concentration of antibioticum and were grown at 37 °C and 220 rpm until OD600 = 0.5–0.7 were reached. The cultures were induced with 0.25–0.5 mM IPTG and incubated in a rotary shaker at 20 °C and 180 rpm for the following 20–22 h. The cells were harvested by centrifugation at 4 °C and resuspended in lysis buffer (50 mM HEPES, 300 mM NaCl, pH 7.6, with 20 mM imidazole) supplemented with DNAse I (0.6 mg DNAse I for each 500 mL culture pellet). The cell suspensions were subjected to cell disruption by sonication (70% nominal power, 20 s on/20 s off for 10 min) on ice using Sonics VCX-130 Ultrasonic Processor (Medline Scientific). The lysate was clarified by centrifugation at 14,500 × g and 4 °C for 50 min. The supernatants after filtration through 0.45 μM syringe filters were loaded into equilibrated 1 mL prepacked HisTrap FF column (Cytiva). After protein binding, the column was washed with 18–22 column volumes of washing buffer (50 mM HEPES, 300 mM NaCl, pH 7.6, with 35 mM imidazole). Proteins were eluted with an imidazole buffer gradient from 35 to 500 mM. The elution peak fractions were concentrated and buffer-exchanged against a storage buffer (25 mM HEPES, 150 mM NaCl, pH 7.5) using Amicon® Ultra-15 centrifugal filters with 30 kDa cutoff. The total concentration of partially purified proteins was measured by spectrophotometric measurements at 280 nm in a NanoDrop Instrument (ThermoFisher Scientific, Wilmington, USA). UGT72B49, UGT71C1 and UGT72B30 enzymes for enzymatic characterization were produced in the same way but in larger culture volumes (6 L for UGT72B49, 1 L for UGT71C1 and UGT72B30) and purified by combining two purification steps with HisTrap FF column. After the first purification step as described above, the protein samples were transferred to 25 mM HEPES buffer (150 mM NaCl, pH 7.6, with 0.5 mM EDTA and 1 mM DTT) and treated with a recombinant Tobacco Etch Virus (TEV) protease (10 μL of 5 mg/mL TEV protease per 10 mg target protein) to remove the 6xHis-tag and loaded into 1 mL prepacked HisTrap FF equilibrated with 25 mM HEPES buffer (150 mM NaCl, pH 7.6). The flowthrough fractions were pooled, concentrated, and transferred to the storage buffer. All UGTs in 10, 20 and 50 uL aliquots were snap frozen using liquid nitrogen and stored at −70 °C until further use.

UGT activity assays and reaction scale-up

Activity assays were carried out in 25 μL reaction volume using either PCR tubes or 96-Well PCR Plates. Each reaction contained 50–250 μg/mL of partially purified UGT sample, 50 μM aloesone and 200 μM UDP-Glc in 50 mM HEPES buffer (50 mM NaCl, pH 7.4). For the assays with RSB UGTs, 100 mM Tris–HCl buffer (pH 7.4 with 500 μM MgCl2, 100 μM KCl and 0.1 U FastAP thermosensitive alkaline phosphatase) and 2 μL of each enzyme were used. All reactions were incubated overnight at 30 °C. The UGT reactions were quenched either by adding of 25 μL of 40% acetonitrile (in water) or incubating at 95 °C for 2–3 min.

Reaction with UGT71C1 were scaled up to 45 mL of total volume (9 × 5 mL in 15-mL Falcon tubes) in 50 mM HEPES buffer (pH 7.4, with 50 mM NaCl). The reaction ran at 30 °C and 300 rpm in an orbital shaker contained 100 μM aloesone (total amount 1.05 mg), 400 μM UDP-glucose and 8 mg of UGT71C1, and were quenched by heating after 4 h.

HPLC analysis

All reactions were analyzed by reverse-phase HPLC using an Ultimate 3000 Series apparatus (Thermo Fisher) and either a Kinetex C18 (2.6 μm, 100 Å, 100 × 4.6 mm, Phenomenex) or Zorbax RR Eclipse Plus C18 (3.5 μm, 95 Å, 150 × 4.6 mm, Agilent) analytical columns. The column temperature was kept at 30 °C. Water and acetonitrile containing 0.1% formic acid with a flow rate of 1 mL/min were used for mobile phase. A gradient from 5% to 25% acetonitrile for 1.5 min followed by gradients from 25% to 80% for 2 min and from 80% to 100% for 1.5 min were used for the analysis of aloesone and control reactions of previously characterised C-GTs with phloretin, apigenin and genistein. The detection wavelength was either 270 nm (genistein) or 290 nm (aloesone, phloretin) or 320 nm (apigenin). For the quantification of HPLC data a Chromeleon 7 software (Thermo Fisher) was used.

Enzymatic characterization of aloesone UGTs

Temperature preference of UGT72B49, UGT71C1 and UGT72B30 were assayed in 100 μL reaction volume with 50 mM HEPES buffer (pH 7.4, with 50 mM NaCl), 50 μM aloesone, 200 μM UDP-Glc and different amounts of purified enzyme (40 μg UGT72B49, 43 μg UGT72B30 and 1.36 μg UGT71C1) at 22, 30, 40 and 50 °C. Aloesone reactions with UGT72B49 and UGT72B30 were stopped after 30 and 90 min, with UGT71C1 after 5 and 15 min. To determine temperature optimum of UGT72B49 initial rates at 22, 30, 40, 50 and 60 °C were measured. For this, 220 μL of each reaction in the same buffer containing 100 μM aloesone, 400 μM UDP-Glc and 25 μg UGT72B49 were incubated in a thermocycler at different temperatures. 48 μL samples were taken after 10, 20, 30, 40 min and reactions were quenched by thermal denaturation at 95 °C. After the HPLC measurement product formation rates were calculated by linear regression using at least three time-points. The relative initial activity was calculated as a percentage of the maximum observed rate. The experiment was performed in triplicates.

To determine the pH optimum reactions were performed in 100 mM citrate–phosphate (pH 5, 6) HEPES (7, 8) and glycine (9, 10) buffers. Each reaction was in 220 μL volume and contained 100 μM aloesone, 400 μM UDP-Glc and 48 μg UGT72B49. 48 μL samples were taken after 10, 20, 30 and 40 min and reactions were quenched by thermal denaturation at 95 °C. After the HPLC measurement product formation rates were calculated by linear regression using at least three time-points. The relative initial activity was calculated as a percentage of the maximum observed rate. The experiment was performed in triplicates.

Michaelis–Menten kinetics of UGT72B49 and UGT71C1 were measured in 50 mM HEPES buffer (50 mM NaCl, pH 7.4) consisting of 0.8 mM UDP-Glc, varying concentrations of aloesone (6.25 to 300 μM) and either 0.44 μM UGT72B49 or 0.011 μM UGT71C1. The reactions were started by adding the UGT, 50-μL samples were taken at different time-points (4, 8, 12 and 16 min for UGT72B49; 2, 4, 8 and 12 min for UGT71C1) and quenched by thermal denaturation at 95 °C. The linear increase of the product peak area (Fig. S4) within either 16 min (UGT72B49) or 12 min (UGT71C1) was measured via HPLC and Chromeleon 7 software (Thermo Scientific). Product formation rates for the different substrate concentrations were calculated by linear regression using at least three time-points and an aloesin calibration curve. Data were fit to the Michaelis–Menten equation by non-linear regression using OriginPro software (OriginLab) to obtain KM and kcat. Mean values and standard deviations were calculated from the three separate reactions.

LC–MS/MS analysis

UGT72B49 and UGT71C1 reaction samples as well as aloesone and aloesin standards were analyzed by the LC–MS/MS. The UGT reactions with 50 μM aloesone were prepared as described for HPLC but in duplicates. After thermal deactivation one replicate was spiked with 15 μM aloesin. The LC–MS/MS analysis was performed on a Vanquish Duo UHPLC binary system (Thermo Fisher Scientific) coupled to an IDX-Orbitrap mass spectrometer (Thermo Fisher Scientific). The chromatographic separation was achieved using a Waters ACQUITY BEH C18 (10 cm × 2.1 mm, 1.7 μm) equipped with an ACQUITY BEH C18 guard column kept at 40 °C. The mobile phase consisted of water with 0.1% formic acid (A) and acetonitrile with 0.1% formic acid (B). The gradient composition was as follow: initial composition was 2%B held for 0.8 min, followed by a linear gradient till 5% in 3.3 min, and afterwards, 100%B was reached in 10 min and held for 1 min before going back to initial conditions. Re-equilibration time was 2.7 min. The flow rate was 0.35 mL/min. The MS measurement was done in positive—heated electrospray ionization (HESI) mode with a voltage of 3,500 V acquiring in full MS/MS spectra (Data dependent Acquisition-driven MS/MS) in the mass range of 70–1,000 Da.

Phylogenetic analysis of aloesone UGTs

The analysis was carried out with the CLC software (QIAGEN), version 20.0.4. The multiple sequence alignment of all the amino acid sequences from glycosyltransferase enzymes that were found active on aloesone was performed with the following settings: alignment mode = very accurate (progressive alignment-based algorithm); gap open cost = 10.0; gap extension cost = 1.0; end gap cost = free. From this alignment, a maximum likelihood based phylogenetic tree was constructed using the following parameters: tree construction method = Neighbor Joining; protein substitution model = WAG; Transition / transversion ratio = 2.0; Number of substitution rate categories = 4; Gamma distribution parameter = 1.0; bootstrap analysis performed with 100 replicates.

Supplementary Material

SI_Aloesone_O-glc_revised_to_submit_FINAL_cwae050
si_aloesone_o-glc_revised_to_submit_final_cwae050.pdf (475.1KB, pdf)

Acknowledgments

The authors thank Dr. Michael Court from UGT Nomenclature Committee (Washington State University) for assigning a name to UGT71U24 and Daniela Rago from analytics department of The Novo Nordisk Foundation Center for Biosustainability for LC–MS measurement.

Contributor Information

Natalia Putkaradze, The Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Søltofts Plads 220, Lyngby DK-2800, Denmark.

Laura Dato, River Stone Biotech ISG, Fruebjergvej 3, Copenhagen DK-2100, Denmark.

Onur Kırtel, The Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Søltofts Plads 220, Lyngby DK-2800, Denmark.

Jørgen Hansen, River Stone Biotech ISG, Fruebjergvej 3, Copenhagen DK-2100, Denmark.

Ditte Hededam Welner, The Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Søltofts Plads 220, Lyngby DK-2800, Denmark.

Author contributions

Natalia Putkaradze (Conceptualization [equal], Data curation [lead], Formal analysis [lead], Funding acquisition [lead], Investigation [lead], Methodology [equal], Project administration [equal], Supervision [lead], Validation [equal], Visualization [lead], Writing—original draft [lead], Writing—review & editing [lead]), Laura Dato (Data curation [equal], Investigation [equal], Validation [equal], Visualization [equal], Writing—review & editing [equal]), Onur Kirtel (Data curation [equal], Investigation [equal], Writing—review & editing [equal]), Jørgen Hansen (Investigation [equal], Writing—review & editing [equal]), and Ditte Welner (Conceptualization [equal], Investigation [lead], Methodology [equal], Project administration [equal], Supervision [lead], Writing—review & editing [lead])

Funding

This work was supported by the Novo Nordisk Foundation through the grant NNF21OC0071624.

Conflict of interest statement

L.D. and J. H. have financial interests in GLY-it. The remaining authors declare no competing interests.

Data availability

The GLY-it sequence data do not contain novel protein sequences. The protein composition of the GLY-it screen is a third-party property and can be shared by permission of River Stone Biotech ISG (laurad@gly-it.com).

References

  1. Abe  I, Utsumi  Y, Oguro  S, Noguchi  H. The first plant type III polyketide synthase that catalyzes formation of aromatic heptaketide. FEBS Lett. 2004:562(1–3):171–176. [DOI] [PubMed] [Google Scholar]
  2. Bowles  D, Isayenkova  J, Lim  E, Poppenberger  B. Glycosyltransferases: managers of small molecules. Curr Opin Plant Biol. 2005:8(3):254–263. [DOI] [PubMed] [Google Scholar]
  3. Brazier-Hicks  M, Offen  WA, Gershater  MC, Revett  TJ, Lim  EK, Bowles  DJ, Davies  GJ, Edwards  R. Characterization and engineering of the bifunctional N- and O-glucosyltransferase involved in xenobiotic metabolism in plants. Proc Natl Acad Sci USA. 2007:104(51):20238–20243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Breton  C, Šnajdrová  L, Jeanneau  C, Koča  J, Imberty  A. Structures and mechanisms of glycosyltransferases. Glycobiology. 2006:16(2):29–37. [DOI] [PubMed] [Google Scholar]
  5. Ceunen  S, Geuns  JMC. Steviol glycosides: chemical diversity, metabolism, and function. Nat Prod Rep. 2013:76(6):1201–1228. [DOI] [PubMed] [Google Scholar]
  6. Chen  D, Chen  R, Wang  R, Li  J, Xie  K, Bian  C, Sun  L, Zhang  X, Liu  J, Yang  L, et al.  Probing the catalytic promiscuity of a Regio- and stereospecific C-glycosyltransferase from Mangifera indica. Angew Chem. 2015:127(43):12869–12873. [DOI] [PubMed] [Google Scholar]
  7. Della Gala  V, Welner  DH. Identification and functional characterization of novel plant UDP-glycosyltransferase (LbUGT72B10) for the bioremediation of 3,4-dichloroaniline. RSC Sustainability. 2023:1(8):2024–2032. [Google Scholar]
  8. Esatbeyoglu  T, Wagner  AE, Schini-Kerth  VB, Rimbach  G. Betanin-a food colorant with biological activity. Mol Nutr Food Res. 2015:59(1):36–47. [DOI] [PubMed] [Google Scholar]
  9. Gantt  RW, Peltier-Pain  P, Thorson  JS. Enzymatic methods for glyco(diversification/randomization) of drugs and small molecules. Nat Prod Rep. 2011:28(11):1811–1853. [DOI] [PubMed] [Google Scholar]
  10. Gramatica  P, Gianotti  M, Speranza  G, Manitto  P. Synthesis of naturally occurring 2,5-Dialkylchromones. Part 1. Synthesis of Aloesone and Aloesol. Heterocycles. 1986:24:743–750. [Google Scholar]
  11. Griffith  BR, Langenhan  JM, Thorson  JS. “Sweetening” natural products via glycorandomization. Curr Opin Biotechnol. 2005:16(6):622–630. [DOI] [PubMed] [Google Scholar]
  12. Haynes  LJ, Holdsworth  DK, Russell  R. C-Glycosyl compounds. Part VI. Aloesin, a C-glucosylchromone from aloe sp. J Chem Soc C Org. 1970:18:2581–2586. [Google Scholar]
  13. He  JB, Zhao  P, Hu  ZM, Liu  S, Kuang  Y, Zhang  M, Li  B, Yun  CH, Qiao  X, Ye  M. Molecular and structural characterization of a promiscuous C-glycosyltransferase from Trollius chinensis. Angew Chem Int Ed Engl. 2019:58(33):11513–11520. [DOI] [PubMed] [Google Scholar]
  14. Hsu  TM, Welner  DH, Russ  ZN, Cervantes  B, Prathuri  RL, Adams  PD, Dueber  JE. Employing a biochemical protecting group for a sustainable indigo dyeing strategy. Nat Chem Biol. 2018:14(3):256–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ito  T, Fujimoto  S, Suito  F, Shimosaka  M, Taguchi  G. C-glycosyltransferases catalyzing the formation of di-C-glucosyl flavonoids in citrus plants. Plant J. 2017:91(2):187–198. [DOI] [PubMed] [Google Scholar]
  16. Jones  K, Hughes  J, Hong  M, Jia  Q, Orndorff  S. Modulation of melanogenesis by aloesin: a competitive inhibitor of tyrosinase. Pigment Cell Res. 2002:15(5):335–340. [DOI] [PubMed] [Google Scholar]
  17. Kashiwada  Y, Nonaka  G-I, Nishioka  I. Chromone glucosides from rhubarb. Phytochemistry. 1990:29(3):1007–1009. [Google Scholar]
  18. Kim  HJ, Parveen  S, Lee  J, Hassan  AHE, Jin  C, Saleem  M, Lee  YS. Concise synthesis of Aloesone and Aloesol natural products. Bull Korean Chem Soc. 2017:38(10):1121–1122. [Google Scholar]
  19. Lairson  LL, Henrissat  B, Davies  GJ, Withers  SG. Glycosyltransferases: structures, functions, and mechanisms. Annu Rev Biochem. 2008:77(1):521–555. [DOI] [PubMed] [Google Scholar]
  20. Lim  EK, Ashford  DA, Hou  B, Jackson  RG, Bowles  DJ. Arabidopsis glycosyltransferases as biocatalysts in fermentation for regioselective synthesis of diverse quercetin glucosides. Biotechnol Bioeng. 2004:87(5):623–631. [DOI] [PubMed] [Google Scholar]
  21. Liu  J, Leng  L, Liu  Y, Gao  H, Yang  W, Chen  S, Liu  A. Identification and quantification of target metabolites combined with transcriptome of two rheum species focused on anthraquinone and flavonoids biosynthesis. Sci Rep. 2020:10(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lombard  V, Golaconda Ramulu  H, Drula  E, Coutinho  PM, Henrissat  B. The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res. 2014:42(D1):490–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Louveau  T, Osbourn  A. The sweet side of plant-specialized metabolism. Cold Spring Harb Perspect Biol. 2019:11(12):1–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lucini  L, Pellizzoni  M, Pellegrino  R, Molinari  GP, Colla  G. Phytochemical constituents and in vitro radical scavenging activity of different aloe species. Food Chem. 2015:170:501–507. [DOI] [PubMed] [Google Scholar]
  25. Mashima  K, Hatano  M, Suzuki  H, Shimosaka  M, Taguchi  G. Identification and characterization of Apigenin 6-C-glucosyltransferase involved in biosynthesis of Isosaponarin in wasabi (Eutrema japonicum). Plant Cell Physiol. 2019:60(12):2733–2743. [DOI] [PubMed] [Google Scholar]
  26. Mizuuchi  Y, Shi  SP, Wanibuchi  K, Kojima  A, Morita  H, Noguchi  H, Abe  I. Novel type III polyketide synthases from Aloe arborescens. FEBS J. 2009:276(8):2391–2401. [DOI] [PubMed] [Google Scholar]
  27. Nidetzky  B, Gutmann  A, Zhong  C. Leloir glycosyltransferases as biocatalysts for chemical production. ACS Catal. 2018:8(7):6283–6300. [Google Scholar]
  28. Osmani  SA, Bak  S, Møller  BL. Substrate specificity of plant UDP-dependent glycosyltransferases predicted from crystal structures and homology modeling. Phytochemistry. 2009:70(3):325–347. [DOI] [PubMed] [Google Scholar]
  29. Palcic  MM. Glycosyltransferases as biocatalysts. Curr Opin Chem Biol. 2011:15(2):226–233. [DOI] [PubMed] [Google Scholar]
  30. Putkaradze  N, Teze  D, Fredslund  F, Welner  DH. Natural product: C-glycosyltransferases-a scarcely characterised enzymatic activity with biotechnological potential. Nat Prod Rep. 2021:38(3):432–443. [DOI] [PubMed] [Google Scholar]
  31. Putkaradze  N, Gala  VD, Vaitkus  D, Teze  D, Welner  DH. Sequence mining yields 18 phloretin C-glycosyltransferases from plants for the efficient biocatalytic synthesis of nothofagin and phloretin-di-C-glycoside. Biotechnol J. 2023:18(6):1–10. [DOI] [PubMed] [Google Scholar]
  32. Sasaki  N, Nishizaki  Y, Yamada  E, Tatsuzawa  F, Nakatsuka  T, Takahashi  H, Nishihara  M. Identification of the glucosyltransferase that mediates direct flavone C-glucosylation in Gentiana triflora. FEBS Lett. 2015:589(1):182–187. [DOI] [PubMed] [Google Scholar]
  33. Tegl  G, Nidetzky  B. Leloir glycosyltransferases of natural product C-glycosylation: structure, mechanism and specificity. Biochem Soc Trans. 2020:48(4):1583–1598. [DOI] [PubMed] [Google Scholar]
  34. Wang  Z, Li  X, Yang  Z, He  X, Tu  J, Zhang  TV. Effects of aloesin on melanogenesis in pigmented skin equivalents. Int J Cosmet Sci. 2008:30(2):121–130. [DOI] [PubMed] [Google Scholar]
  35. Wang  X, Li  C, Zhou  C, Li  J, Zhang  Y. Molecular characterization of the C-glucosylation for puerarin biosynthesis in Pueraria lobata. Plant J. 2017:90(3):535–546. [DOI] [PubMed] [Google Scholar]
  36. Wang  Y, Li  C, Xiong  Z, Chen  N, Wang  X, Xu  J, Yuemei  W, Liu  L, Wu  H, Huang  C, et al.  Up-and-coming anti-epileptic effect of aloesone in Aloe vera: evidenced by integrating network pharmacological analysis, in vitro, and in vivo models. Front Pharmacol. 2022:13:1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang  X, Li  Q, Li  Y. Multiple beneficial effects of Aloesone from Aloe vera on. Molecules. 2023:28(4):1617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Xie  K, Zhang  X, Sui  S, Ye  F, Dai  J. Exploring and applying the substrate promiscuity of a C-glycosyltransferase in the chemo-enzymatic synthesis of bioactive C-glycosides. Nat Commun. 2020:11(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yamada  A, Kondo-Kaneko  M, Ishiuchi  K, Makino  T, Terasaka  K. Isolation and characterization of a novel glucosyltransferase involved in production of emodin-6-O-glucoside and rhaponticin in rheum palmatum. Plant Biotechnol. 2020:37(3):293–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Yang  D, Jang  WD, Lee  SY. Production of Carminic acid by metabolically engineered Escherichia coli. J Am Chem Soc. 2021:143(14):5364–5377. [DOI] [PubMed] [Google Scholar]
  41. Zhang  M, Li  F-D, Li  K, Wang  Z-L, Wang  Y-X, He  J-B, Su  H-F, Zhong-Yi  Z, Chi  C-B, Shi  X-M, et al.  Functional characterization and structural basis of an efficient di- C -glycosyltransferase from Glycyrrhiza glabra. J Am Chem Soc. 2020:142(7):3506–3512. [DOI] [PubMed] [Google Scholar]
  42. Zhang  LJ, Wang  DG, Zhang  P, Wu  C, Li  YZ. Promiscuity characteristics of versatile plant glycosyltransferases for natural product Glycodiversification. ACS Synth Biol. 2022:11(2):812–819. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI_Aloesone_O-glc_revised_to_submit_FINAL_cwae050
si_aloesone_o-glc_revised_to_submit_final_cwae050.pdf (475.1KB, pdf)

Data Availability Statement

The GLY-it sequence data do not contain novel protein sequences. The protein composition of the GLY-it screen is a third-party property and can be shared by permission of River Stone Biotech ISG (laurad@gly-it.com).

Articles from Glycobiology are provided here courtesy of Oxford University Press

RESOURCES