Free access
Applied and Industrial Microbiology
Research Article
11 June 2021

Effective Generation of Glucosylpiericidins with Selective Cytotoxicities and Insights into Their Biosynthesis


Exploring unknown glycosyltransferases (GTs) is important for compound structural glycodiversification during the search for drug candidates. Piericidin glycosides have been reported to have potent bioactivities; however, the GT responsible for piericidin glucosylation remains unknown. Herein, BmmGT1, a macrolide GT with broad substrate selectivity and isolated from Bacillus methylotrophicus B-9987, was found to be able to glucosylate piericidin A1 in vitro. Next, the codon-optimized GT gene sbmGT1, which was designed based on BmmGT1, was heterologously expressed in the piericidin producer Streptomyces youssoufiensis OUC6819. Piericidin glycosides thus significantly accumulated, leading to the identification of four new glucopiericidins (compounds 3, 4, 6, and 7). Furthermore, using BmmGT1 as the probe, GT1507 was identified in the genome of S. youssoufiensis OUC6819 and demonstrated to be associated with piericidin glucosylation; the overexpression of this gene led to the identification of another new piericidin glycoside, N-acetylglucosamine-piericidin (compound 8). Compounds 4, 7, and 8 displayed cytotoxic selectivity toward A549, A375, HCT-116, and HT-29 solid cancer cell lines compared to the THP-1 lymphoma cell line. Moreover, database mining of GT1507 homologs revealed their wide distribution in bacteria, mainly in those belonging to the high-GC Gram-positive and Firmicutes clades, thus representing the potential for identification of novel tool enzymes for compound glycodiversification.
IMPORTANCE Numerous bioactive natural products are appended with sugar moieties and are often critical for their bioactivities. Glycosyltransferases (GTs) are powerful tools for the glycodiversification of natural products. Although piericidin glycosides display potent bioactivities, the GT involved in glucosylation is unclear. In this study, five new piericidin glycosides (compounds 3, 4, 6, 7, and 8) were generated following the overexpression of GT-coding genes in a piericidin producer. Three of them (compounds 4, 7, and 8) displayed cytotoxic selectivity. Notably, GT1507 was demonstrated to be related to piericidin glucosylation in vivo. Furthermore, mining of GT1507 homologs from the GenBank database revealed their wide distribution across numerous bacteria. Our findings would greatly facilitate the exploration of GTs to glycodiversify small molecules in the search for drug candidates.


Numerous bioactive natural products contain sugar moieties, which have profound impacts on the various characteristics of the products, such as solubility, bioavailability, stability, mode of action, and pharmacokinetics (14). Glycosyltransferases (GTs) are usually responsible for the attachment of sugar moieties to aglycone scaffolds. GTs catalyze the transfer of sugar moieties from an activated donor to specific acceptor molecules by forming glycosidic bonds (14). The reversibility of the reaction and substrate promiscuity of GTs provide them the ability of glycodiversification of small molecules, which may help in the search of drug candidates (58). Therefore, expanding the application of GTs as well as exploring unknown GTs is critical for drug development.
Piericidins belong to a family of α-pyridone antibiotics produced by various Streptomyces species, which feature a highly substituted pyridone core with a lipophilic side chain (9) (Fig. 1). Because of their structural resemblance to coenzyme Q10, piericidins have been demonstrated to be potent NADH-ubiquinone oxidoreductase (complex I) inhibitors in the mitochondrial electron transport chain (913). Different piericidin glycosides have been simultaneously isolated with aglycone piericidin A1 but usually in small amounts (12, 1416), and their bioactivities are strongly influenced by the type and position of the sugar unit (14). For example, 10-glucopiericidin A1 (GPA) and 4′-glucopiericidin A1 display stronger antimicrobial activities than piericidin A1 and other glycosides, such as 3′-rhamnopiericidin A1 (17), 3′-deoxytalopiericidin A1 (18) and 7-demethyl-3′-rhamnopiericidin A1 (19). While the piericidin aglycone is biosynthesized via a type I polyketide synthase (PKS) machinery, no GT gene has been observed within the biosynthetic gene clusters of piericidins reported to date (2022).
FIG 1 Chemical structures of piericidins (compounds 1 to 8). Compounds 3, 4, and 6 to 8 are new glycosylated piericidins generated in this study.
We previously identified a macrolide GT gene, bmmGT1, from Bacillus methylotrophicus B-9987 and demonstrated that it was responsible for the O-glucosylation of macrolactins (24-membered macrolide) and bacillaenes (polyunsaturated enamines); however, the gene is located within none of the biosynthetic gene clusters. Notably, BmmGT1 displays broad substrate flexibility, especially toward sugar acceptors, transferring glucose to macrolactins, bacillaenes, and phenolic compounds (2325). Given the potent bioactivities but limited yields of glucopiericidins, we aimed to produce glucopiericidins efficiently and understand their biosynthesis. The substrate promiscuity of BmmGT1 led us to probe its ability to recognize piericidins, which may further contribute toward understanding the biosynthesis of glucopiericidins. In this study, we investigated the ability of BmmGT1 to glucosylate piericidin A1. Using BmmGT1 as the probe, another GT, GT1507, was identified in a piericidin producer, marine-derived Streptomyces youssoufiensis OUC6819 (26), and was found to be probably related to the biosynthesis of glucosylated piericidin.


BmmGT1 is able to modify piericidin A1 in vitro through glucosylation.

To detect if BmmGT1 can transfer the glucosyl group onto piericidin A1 (compound 1) (see Table S1 and Fig. S1 in the supplemental material), BmmGT1 was overexpressed and purified as previously described (23) (see Fig. S2). A reaction was set up using compound 1 and UDP-d-glucose as the sugar acceptor and donor, respectively. As shown in Fig. 2, piericidin A1 (compound 1) was transformed into compound 2 (at retention time of 28.0 min) and small amounts of other glucosylated derivatives (panels i and ii). High-resolution mass spectroscopy (HRMS) analysis was used to measure the molecular weight of compound 2 as C31H47NO9 ([M+H]+ at m/z 578.3350, calculated [calcd] 578.3324), which was 162 mass units more than that of compound 1 ([M+H]+ at m/z 416.282, calcd 416.2795) (Fig. 2), suggesting the attachment of a glucosyl moiety onto compound 1. Compound 2 was subsequently isolated from a large-scale reaction mixture and was identified as GPA based on the one-dimensional (1D) and two-dimensional (2D) nuclear magnetic resonance (NMR) spectral data (Table S1; Fig. S3). The coupling constant of the anomeric proton H-1′′ (δH 4.15, d, J = 7.8 Hz) indicated that the glycosyl linkage had a β-configuration, and the 10-O-d-glucoside linkage was confirmed by the downfield shift of C-10 (δC 94.4) and the heteronuclear multiple-bond correlation (HMBC) from H-1′′ to C-10, which are consistent with the results of a previous study (12). This result demonstrated that BmmGT1 can transfer a glucosyl group onto 10-OH of compound 1. In addition, the molecular weights for the minor products are 740 ([M+H]+) (see Fig. S4), which was 162 mass units more than that of compound 2, indicating the generation of diglucopiericidin compounds.
FIG 2 In vitro assay of BmmGT1 using piericidin A (compound 1) as a sugar acceptor. (A) (i) Compound 1 plus UDP-d-glucose plus heated BmmGT1; (ii) 500 μM compound 1 plus 2 mM UDP-d-glucose plus 2 μM BmmGT1. Reaction mixtures were incubated at 30°C for 2 h. (B) HRMS data for compounds 1 and 2.

Heterologous expression of the codon-optimized sbmGT1 in OUC6819.

Considering the presence of intracellular UDP-d-glucose, we next aimed to heterologously express BmmGT1 in the piericidin producer S. youssoufiensis OUC6819 to effectively produce glycosylpiericidins. Given that the bmmGT1 gene is obtained from the low-GC-content B. methylotrophicus B-9987, to obtain efficient expression, a codon-optimized gene, sbmGT1, was designed based on the high-GC-content criterion for codon usage in Streptomycetes, as described in Materials and Methods (see Table S2). The modified gene sbmGT1 was put under the control of a strong constitutive promoter Pgapdh (27), and the resulting expression plasmid pWLI901 was then introduced into OUC6819 to yield wild type (WT)/pWLI901. High-pressure liquid chromatography (HPLC) analysis of the fermentation products showed the presence of compound 2 in OUC6819 WT/pWLI901 (Fig. 3A, iii), but it was not detected in the controls (Fig. 3A, i and ii), suggesting that compound 1 was glycosylated when sbmGT1 was overexpressed.
FIG 3 HPLC traces of the fermentation products from OUC6819 strains. (A) Heterologous expression of the codon-optimized sbmGT1 in OUC6819 strains: (i) wild-type strain (WT); (ii) WT/pSET152C (empty vector); (iii) WT/pWLI901 (harboring sbmGT1 under the control of the constitutive promoter Pgapdh); (iv) ΔrdmF mutant (reedsmycins-deficient strain); (v) OUC6819 ΔrdmF/pSET152C; (vi) OUC6819 ΔrdmF/pWLI901. The compound peaks at retention times of 24.0 to 26.5 min are reedsmycins (panels i to iii). (B) Overexpression of GT1507 and SY7155 in OUC6819 strains: (i) WT; (ii) WT/pWLI902 (harboring GT1507 under the control of Pgapdh); (iii) WT/pWLI903 (harboring GT7155 under the control of Pgapdh); (iv) ΔrdmF mutant; (v) OUC6819 ΔrdmF/pWL902; (vi) OUC6819 ΔrdmF/pWLI903.
Further liquid chromatography-mass spectrometry (LC-MS) analysis indicated the generation of diglucopiericidins in OUC6819 WT/pWLI901 (see Fig. S5). Unfortunately, they had similar retention times (24.0 to 26.5 min) to those of polyene antibiotics reedsmycins produced by OUC6819 (Fig. 3A, iii). Because reedsmycins are also assembled through a PKS pathway using malonyl coenzyme A (malonyl-CoA) and methylmalonyl-CoA as building blocks (26, 27), reedsmycins would probably compete for the biosynthetic precursors with piericidins. To eliminate these detection interferences and biosynthetic competitors, pWLI901 was introduced into the reedsmycins-deficient ΔrdmF mutant strain (27). As expected, the production of compound 1 in the ΔrdmF mutant increased by ∼2.5-fold (Fig. 3A, iv) compared to that in the wild-type strain (Fig. 3A, i), while the production of compound 2 in OUC6819 ΔrdmF/pWLI901 (Fig. 3A, vi) was enhanced by ∼1.2-fold compared to that in OUC6819 WT/pWLI901 (Fig. 3A, iii). In addition, few potential glucosylated piericidins with a UV spectrum similar to that of compound 2 were detected (see Fig. S6), indicating the occurrence of glucosylation with different patterns.

Isolation and identification of glucosylated piericidins.

Next, a large-scale fermentation of OUC6819 ΔrdmF/pWLI901 was performed. Compounds 3 to 7 were isolated, and their structures were determined by 1D (1H and 13C) and 2D (correlation spectroscopy [COSY], heteronuclear single quantum coherence [HSQC], HMBC, and nuclear Overhauser effect spectroscopy [NOESY]) NMR (Table 1; see also Fig. S7 to S11). Compounds 3, 4, and 5 had the same molecular formula as C37H57NO14 ([M+H]+ at m/z 740.3885 for compound 3, 740.3882 for compound 4, and 740.3894 for compound 5, calcd 740.3852), which was 162 mass units greater than that of compound 2, indicating the presence of another glucosyl substituent. The coupling constants of the anomeric protons H-1′′ (δH 4.28, d, J = 7.8 Hz) and H-1′′′ (δH 4.54, d, J = 7.8 Hz) indicated β-configured glycosyl linkages, and the HMBCs from H-1′′ to C-10 (δC 92.4) and H-1′′′ to C-3′′ (δC 86.9) in compound 3 unambiguously demonstrated that one β-glucopyranosyl moiety was positioned at 10-OH and another β-glucopyranosyl moiety was positioned at C-3′′ of the first one. The coupling constants of the anomeric protons H-1′′ (δH 5.20, d, J = 7.2 Hz) and H-1′′′ (δH 4.24, d, J = 7.7 Hz) suggested that the glycosyl linkages exhibited β-configuration, and HMBC from H-1′′ to C-4′ (δC 155.9) and H-1′′′ to C-6′′ (δC 69.6) in compound 4 revealed that the first β-glucopyranosyl moiety was attached to 4′-OH, while the second β-glucopyranosyl moiety was positioned at C-6′′ of the first one. The coupling constants of the anomeric protons H-1′′ (δH 5.25, d, J = 7.5 Hz) and H-1′′′ (δH 4.23, d, J = 7.9 Hz) indicated β-configured glycosyl linkages, and the HMBC from H-1′′ to C-4′ (δC 154.6) and H-1′′′ to C-10 (δC 92.4) in compound 5 revealed that the β-glucopyranosyl moieties were attached to 4′-OH and 10-OH of compound 5, which is consistent with previously reported results (16).
TABLE 1 1H (600 MHz) and 13C (150 MHz) NMR data of piericidin compounds 3 to 8 in CD3OD
δH (J in Hz)δC, typeδH (J in Hz)δC, typeδH (J in Hz)δC, typeδH (J in Hz)δC, typeδH (J in Hz)δC, typeδH (J in Hz)δC, type
13.38 (d, 6.1)33.9, CH23.38 (m)35.5, CH23.41 (m)34.1, CH23.45 (m)34.2, CH23.45 (d, 6.9)34.1, CH23.37 (m)33.6, CH2
25.34 (m)121.9, CH5.35 (m)123.4, CH5.37 (m)121.8, CH5.32 (m)121.6, CH5.30 (m)121.5, CH5.34 (m)122.1, CH
3 134.5, C 135.9, C 134.5, C 135.2, C 135.1, C 134.5, C
42.78 (d, 7.0)42.6, CH22.77 (m)44.1, CH22.78 (m)42.7, CH22.79 (d, 7.3)42.6, CH22.78 (d, 7.0)42.5, CH22.79 (m)42.6, CH2
55.55 (dt, 15.0, 7.0)124.7, CH5.53(dt, 15.2, 7.2)126.6, CH5.56 (dt, 15.5, 7.0)125.1, CH5.54 (dt, 14.8, 7.0)124.8, CH5.54 (dt, 15.6, 7.0)124.9, CH5.55 (dt, 15.5, 7.0)124.8, CH
66.10 (d, 15.4)136.2, CH6.08 (d, 15.5)137.8, CH6.12 (d, 15.5)136.1, CH6.09 (d, 15.5)136.4, CH6.10 (d, 15.5)136.2, CH6.07 (d, 15.5)136.5, CH
7 135.9, C 134.2, C 136.4, C 133.5, C 133.2, C 133.1, C
85.41 (d, 9.2)134.5, CH5.28 (d, 9.4)135.9, CH5.42 (d, 9.2)134.6, CH5.30 (d, 9.2)134.3, CH5.40 (d, 9.3)134.7, CH5.26 (d, 9.2)134.6, CH
92.79 (m)35.3, CH2.68 (m)37.5, CH2.80 (m)35.4, CH2.69 (m)36.3, CH2.80 (m)35.2, CH2.73 (m)35.2, CH
103.77 (m)*92.4, CH3.69 (d, 7.9)83.9, CH3.74 (d, 8.6)92.4, CH3.71 (d, 7.7)82.4, CH3.74 (d, 8.5)92.4, CH3.61 (d, 8.3)93.1, CH
11 135.4, C 138.0, C 135.7, C 134.4, C 134.7, C 136.3, C
125.50 (m)123.3, CH5.44 (m)122.9, CH5.49 (m)123.0, CH5.40 (m)121.4, CH5.48 (m)123.0, CH5.45 (m)122.5, CH
131.65 (s)*11.8, CH31.61 (s)13.1, CH31.65 (s)11.1, CH31.64 (s)11.7, CH31.64 (s)10.9, CH31.63 (s)11.8, CH3
141.64 (s)*10.3, CH31.60 (s)11.0, CH31.63 (s)10.4, CH31.62 (s)9.6, CH31.63 (s)10.4, CH31.62 (s)10.6, CH3
150.84 (d, 6.9)16.4, CH30.78 (d, 6.9)18.1, CH30.84 (d, 6.9)16.4, CH30.82 (d, 7.0)16.7, CH30.84 (d, 6.9)16.3, CH30.82 (d, 6.9)16.5, CH3
161.76 (s)11.7, CH31.74 (s)*13.1, CH31.76 (s)*11.7, CH31.76 (s)11.9, CH31.75 (s)11.5, CH31.72 (s)11.7, CH3
171.77 (s)15.3, CH31.74 (s)*16.7, CH31.77 (s)*15.37, CH31.77 (s)15.2, CH31.77 (s)15.1, CH31.77 (s)15.3, CH3
2′ 154.4, C 157.1, C 155.7, C 163.6, C 163.2, C 154.6, C
3′ 128.5, C 130.2, C 133.2, C6.48 (s)92.1, CH6.40 (s)91.6, CH 128.7, C
4′ 155.3, C 155.9, C 154.6, C 164.2, C 164.5, C 155.2, C
5′ 113.1, C 119.9, C 118.74, C 113.7, C 113.8, C 113.4, C
6′ 150.8, C 151.8, C 150.5, C 156.9, C 157.1, C 150.1, C
7′3.90 (s)52.1, CH33.9253.7, CH33.93 (s)52.1, CH33.89 (s)52.6, CH33.87 (s)52.6, CH33.92 (s)52.3, CH3
8′3.75 (s)59.3, CH33.81 (s)61.2, CH33.83 (s)59.8, CH3    3.76 (s)59.4, CH3
9′2.07 (s)9.4, CH32.13 (s)11.9, CH32.17 (s)10.3, CH32.13 (s)9.1, CH32.13 (s)9.0, CH32.08 (s)9.4, CH3
(10-) 1′′4.28 (d, 7.8),102.1, CH        4.40 (d, 8.4)102.3, CH
2′′3.36 (m)76.5, CH        3.58 (m)56.3, CH
3′′3.50 (m)86.9, CH        3.41 (m)75.2, CH
4′′3.42 (m)68.5, CH        3.34 (m)70.7, CH
5′′3.17 (m)76.1, CH        3.10 (m)76.2, CH
6′′3.65 (m)61.2, CH2        3.65 (m)61.3, CH2
 3.77 (m)*         3.75 (m) 
7′′           172.2, C
8′′          1.91 (s)22.1, CH3
(10-) 1′′′4.54 (d, 7.8)103.8, CH  4.23 (d, 7.9)102.7, CH  4.23 (d, 7.8)102.7, CH  
2′′′3.28 (m)70.1, CH  3.16 (m)74.3, CH  3.16 (m)74.2, CH  
3′′′3.17 (m)*76.1, CH  3.13 (m)*76.4, CH  3.12 (m)76.2, CH  
4′′′3.42 (m)*68.5, CH  3.43 (m)*70.1, CH  3.42 (m)69.9, CH  
5′′′3.17 (m)*76.1, CH  3.30 (m)*76.9, CH  3.30 (m)77.0, CH  
6′′′3.65 (m)*61.2, CH2  3.66 (m)*61.3, CH2  3.73 (m)61.0, CH2  
 3.77 (m)*   3.78 (m)*   3.93 (m)   
(4′-) 1′′  5.20 (d, 7.2)104.2, CH5.25 (d, 7.5)102.8, CH5.03 (d, 7.1)99.6, CH5.03 (d, 7.2)99.8, CH  
2′′  3.46 (m)75.7, CH3.47 (m)74.2, CH3.53 (m)73.2, CH3.53 (m)73.2, CH  
3′′  3.44 (m)78.0, CH3.13 (m)76.4, CH3.50 (m)76.6, CH3.31 (m)76.8, CH  
4′′  3.64 (m)71.4, CH3.43 (m)70.1, CH3.42 (m)70.0, CH3.42 (m)69.9, CH  
5′′  3.13 (m)*75.2, CH3.30 (m)76.9, CH3.79 (m)75.8, CH3.50 (m)77.0, CH  
6′′  3.77 (m)69.6, CH23.66 (m)61.3, CH23.84 (m)69.0, CH23.73 (m)61.0, CH2  
   4.07 (m) 3.79 (m) 4.17 (m) 3.93 (m)   
(4′-) 1′′′  4.24 (d, 7.7)104.7, CH  4.36 (d, 7.8)103.9, CH    
2′′′  3.14 (m)*75.2, CH  3.25 (m)73.7, CH    
3′′′  3.25 (m)*77.9, CH  3.26 (m)76.6, CH    
4′′′  3.45 (m)71.72, CH  3.32 (m)70.2, CH    
5′′′  3.25 (m)*77.8, CH  3.28 (m)76.6, CH    
6′′′  3.76 (m)63.2, CH2  3.69 (m)61.30, CH2    
   3.83 (m)   3.88 (m)     
*, signals are overlapped.
Compounds 6 and 7 had the molecular formula of C36H55NO13 ([M+H]+ at m/z 710.3771 for compound 6 and 710.3781 for compound 7, calcd 710.3746), which was 30 mass units less than that of compounds 3, 4, and 5, indicating the loss of one methoxy group. Eventually, the structures of compounds 6 and 7 were elucidated based on extensive analyses of NMR data (1H, 13C, COSY, HSQC, HMBC, and NOESY) (Table 1; Fig. S10 and S11). The NMR spectroscopic data of compounds 6 and 7 were similar to those of compounds 4 and 5, respectively, except for the desubstitution of a methoxy group at C-3′ (δH 6.48 and 6.40 for H-3′ of compounds 6 and 7, respectively), suggesting that BmmGT1 can accept desmethoxy piericidin, which is an intermediate formed during compound 1 biosynthesis (22).

Insights into glucosylpiericidin biosynthesis.

Since minor amounts of piericidin glycosides are usually isolated along with piericidin A1 (12, 1416), we further inspected the LC-MS data for the fermentation products of wild-type OUC6819. The molecular weight of GPA ([M+H]+ at m/z 578.3325) was detected, albeit at a very low intensity (see Fig. S12). This result suggested that a gene outside the piericidin biosynthetic gene cluster was probably involved in the glucosyl transfer reaction. Considering the biochemical reaction (Fig. 2) and in vivo overexpression results (Fig. 3), we searched for the homologs of BmmGT1 against the genome of OUC6819, revealing two GTs (GT1507 and SY7155) with identities >30% to BmmGT1 (Table 2). Given the bare production of GPA (compound 2) in the wild-type strain (Fig. 3A, i and ii) and the ΔrdmF mutant (Fig. 3A, iv and v), to probe their functions, each gene was cloned under the control of the constitutive promoter Pgapdh, followed by overexpression in OUC6819 strains. As shown in Fig. 3B, overexpression of GT1507 in the wild-type strain and the ΔrdmF mutant led to the production of compound 2 and diglucopiericidin (compound 5) (Fig. 3B, ii and v), and noticeably, a new compound 8 accumulated (Fig. 3B, ii and v); conversely, the overexpression of SY7155 had no impact on the metabolic profiles (Fig. 3B, iii and vi), suggesting that GT1507 might be related to the glucosylation of compound 1 in OUC6819 in vivo.
TABLE 2 Proposed function of the BmmGT1 homologs from S. youssoufiensis OUC6819
ProteinSize (aa)aProposed functionHomolog
Protein/organismAccession no. (% identity/% similarity)Alignment with BmmGT1 (% identity/% similarity)
GT1507408Macrolide-inactivating glycosyl transferaseBKD26_18935/Streptomyces sp. CB03238WP_084899436.1 (72/83)32/47
SY7155384Glycosyl transferaseB4N89_13400/Streptomyces scabrisporus NF3OPC85028.1 (78/86)30/47
aa, amino acids.
Subsequently, compound 8 was isolated, and its structure was determined. According to the HRMS data (see Fig. S14), the molecular formula of compound 8 was established as C33H50N2O9 ([M+H]+ at m/z 619.3605, calcd 619.3589), which was 203 mass units greater than that of compound 1. A full set of 1D and 2D NMR spectra of compound 8 was acquired, thereby allowing us to assign its 1H and 13C NMR chemical shifts (Table 1). The spectroscopic data for compound 8 were similar to those for compound 2, except for the presence of an N-acetylglucosamine group at C-10, which was determined by the HMBC from H-1′′ (δH 4.40, d, J = 8.4 Hz) to C-10 (δC 93.1) and from H3-8′′ (δH 1.91, s) and H-2′′ (δH 3.85, m) to C-7′′ (δC 172.2) (Table 1; Fig. S14).
To characterize the biochemical functions of GT1507 in vitro, GT1507 was overexpressed in Escherichia coli BL21(DE3) and purified to near homogeneity (>95%) (Fig. S2). As shown in Fig. 4, GT1507 transferred glucose or N-acetylglucosamine onto compound 1, generating compounds 2 and 5 (Fig. 4, ii) or 8 (Fig. 4, iii), respectively. We then examined the catalytic activities of GT1507 at different pH values, temperatures, and cations and obtained the following optimized reaction condition: 50 mM Tris-HCl buffer (pH 8.5) with 10 mM MgCl2 at 30°C (Fig. S14). Under the optimized condition, the steady-state kinetic characterizations of GT1507 were performed. As shown in Table 3 and in Fig. S15, GT1507 displayed Km values of 6.9 and 207.4 μM for compound 1 and UDP-d-glucose, respectively, and showed Km values of 10.2 and 927.2 μM for compound 1 and UDP-d-N-acetylglucosamine, respectively; kcat/Km for UDP-d-glucose (1.0 × 103 s−1 M−1) was approximately 9-fold higher than that for UDP-d-N-acetylglucosamine (1.1 × 102 s−1 M−1). These results demonstrated that UDP-d-glucose was clearly favored over UDP-d-N-acetylglucosamine for GT1507. In addition, the kinetic parameters of BmmGT1 toward UDP-d-glucose and compound 1 were measured (Table 3 and Fig. S15). The Km and kcat/Km values of BmmGT1 were similar to those of GT1507, indicating the comparable catalytic activities of these two enzymes.
FIG 4 In vitro assay of GT1507 using piericidin A (compound 1) as sugar acceptor. (i) Compound 1 with heated GT1507; (ii) 500 μM compound 1 plus 2 mM UDP-d-glucose + 2 μM GT1507; (iii) 500 μM compound 1 plus 2 mM UDP-d-N-acetylglucosamine plus 2 μM GT1507; (iv) compound 2 standard; (v) compound 5 standard; (vi) compound 8 standard. Reaction mixtures were incubated at 30°C for 2 h.
TABLE 3 Kinetic parameters of GT1507 for different substratesa
ProteinSubstrateKm (μM)Vmax (μM s−1 × 10−2)Kcat (s−1 × 10−2)Kcat/Km (s−1 M−1)
GT1507Donor, UDP-d-glucose207.4 ± 19.811.0 ± 0.321.9 ± 0.61.0 × 103
Donor, UDP-d-N-acetylglucosamine927.2 ± 136.010.6 ± 0.610.6 ± 0.61.1 × 102
Acceptor, 1b6.9 ± 1.43.3 ± 0.232.7 ± 2.34.8 × 104
Acceptor, 1c10.2 ± 1.53.1 ± 0.215.6 ± 0.81.5 × 104
BmmGT1Donor, UDP-d-glucose234.0 ± 32.310.7 ± 0.421.4 ± 0.89.2 × 102
Acceptor, 1b7.1 ± 0.63.1 ± 0.831.2 ± 0.84.4 × 104
Values are the means from three independent assays reported with standard deviations.
Kinetics determined with saturating concentration (10 mM) of UDP-d-glucose.
Kinetics determined with saturating concentration (10 mM) of UDP-d-N-acetylglucosamine.

Bioactivities of compounds 1 to 8.

The cytotoxicities of compounds 1 to 8 were then evaluated against five human cancer cell lines (lung cancer cell line, A549; melanoma cell line, A375; colon cancer cell lines, HCT-116 and HT-29; and lymphoma cell line, THP-1), as described in Materials and Methods. As shown in Table 4, aglycone 1 was inactive against all tested cancer cell lines up to 25 μM; compound 2 with a glucosyl group at 10-OH exhibited strong cytotoxicities against A549, A375, HCT-116, and HT-29 cell lines, with 50% inhibitory concentration (IC50) values ranging from 0.16 to 2.42 μM, and compound 8 with a N-acetylglucosamine group at 10-OH displayed moderate cytotoxicities against these cell lines. Although further glucosylation at 4′-OH (compound 5) led to comparable cytotoxicities to those of compound 2, diglucopiericidin compound 4 (with a disugar unit at 4′-OH) exhibited moderate cytotoxicity against these cell lines, with IC50 values ranging from 2.07 to 9.57 μM, and diglucopiericidin 3 (with a disugar unit at 10-OH) was completely inactive. In addition, the cytotoxicities of compounds 4 and 5 were stronger than those of their corresponding demethoxyl derivative compounds 6 and 7 (Table 4), indicating the importance of the methoxyl group at C-3′ for the cytotoxic activities. Notably, all compounds were inactive toward the THP-1 cell line, suggesting that compounds 2, 4, 5, 7, and 8 exhibit selectivity toward solid cancer cell lines—A549, A375, HCT-116, and HT-29—versus the lymphoma cell line THP-1.
TABLE 4 Cytotoxicities of compounds 1 to 8 against five human cancer cell lines
CompoundIC50 (μM)
22.42 ± 0.010.80 ± 0.010.16 ± 0.010.97 ± 0.03>25
49.57 ± 0.014.30 ± 0.012.07 ± 0.085.94 ± 0.02>25
51.12 ± 0.010.74 ± 0.030.27 ± 0.032.34 ± 0.01>25
624.8 ± 0.06>255.03 ± 0.5213.93 ± 0.03>25
719.45 ± 0.162.73 ± 0.012.55 ± 0.266.45 ± 0.03>25
815.72 ± 0.023.11 ± 0.011.47 ± 0.037.43 ± 0.03>25

Sequence similarity networks of the GT1507 homologs and their distribution.

Given that BmmGT1 homologs are conserved in the Bacillus subtilis clade (24), we further analyzed the distribution of GT1507 homologs. We performed sequence similarity network (SSN) analysis of 1,573 GT homologs from the nonredundant (nr) protein sequence database with a sequence identity of 30% at a cutoff of 10−150. As indicated in Fig. 5, although the homologs are distributed across bacteria, they can be divided into two major clades, high-GC Gram-positive and Firmicutes. The Firmicutes clade mainly has three paralogous groups, BmmGT1/YjiC, BmmGT2/YdhE, and BmmGT3/YojK, which are consistent with the previous result (24). The BmmGT1/YjiC group is associated with the high-GC Gram-positive clade, suggesting their close relationship during evolution. Noticeably, GTs from Deltaproteobacteria (18 of 23) and a GT from Alphaproteobacteria (FKO01_23780, WP_150831670.1 from Mesorhizobium sp. B2-3-3) are clustered within the GT1507/OleD/LobA2/Kcn28 group in the high-GC Gram-positive clade, suggesting possible horizontal transfer of the GT1507 homologs across the bacterial world. Except the GTs with indicated names, all other GTs are unknowns, thus representing the potential for identification of novel enzyme tools for compound glycodiversification.
FIG 5 Distribution of GT1507 homologs in bacteria. SSN analysis of the GT1507 homologs in diverse bacteria is based on a cutoff E value of 10−85 and visualized by Cytoscape (34). Colors in the nodes show GTs from different classes of taxonomy according to the legend. Edges (lines) indicate that pairwise alignments of nodes are better than the cutoff value. GT1507, OleD, LobA2, Kcn28, CalG4, CalG2, EspG2, BmmGT1/YjiC, BmmGT2/YdhE, and BmmGT3/YojK are shown in black circles. WP_150831670.1 (Mesorhizobium sp. B2-3-3) and WP_033205283.1 (Patulibacter americanus) are shown in black circles. High-GC Gram-positive clade mainly includes Amycolatopsis, Kitasatospora, Micromonospora, Nocardiopsis, Rhodococcus, and Streptomyces. Firmicutes-clade mainly includes Bacillus, Paenibacillus, and Faecalibacterium.


Glycosylation plays an essential role in the bioactivity, specificity, stability, and pharmacokinetics of natural products (28). To date, more than 60 piericidins, including 24 piericidin glycosides, have been isolated (13, 15, 16, 29). In this study, we efficiently generated glycosylated piericidins by introducing sbmGT1 into OUC6819, from which another four piericidin glycosides were identified. In addition, GT1507 was demonstrated to be probably associated with the glucosylation of piericidins in OUC6819. Analysis of GT1507 homologs in the genome database revealed their wide distribution in bacteria, which could considerably expand the members of potential GTs for compound structural diversification.
BmmGT1 exhibits considerable substrate flexibility regarding both aglycones and sugar donors (2325). The glucosylation ability of BmmGT1 toward piericidin A1 (Fig. 2) further supported the potential of BmmGT1 as an enzymatic tool for compound glycodiversification. Taking advantage of the presence of sugar as well as piericidin analogs, glycosylated piericidins, which were barely detectable in the wild-type strain (Fig. 3A, i and ii), were effectively generated via constitutively expressing codon-optimized sbmGT1 (Fig. 3A, iii and vi). This strategy could contribute to the large-scale production of compound 2 as well as the structural diversification of piericidins. The relative yields of each piericidin glycoside might be ascribed to the availability of the aglycones and catalytic selectivity of BmmGT1.
Using BmmGT1 as the probe, we successfully identified the GT gene which is probably associated with piericidin glucosylation in OUC6819. Similar to bmmGT1, GT1507 is also located outside the biosynthetic gene cluster. However, glycosylated piericidins were barely detected in the wild-type strain (see Fig. S12 in the supplemental material), which could be explained by the following reasons: (i) the expression levels of GT1507 are low and (ii) GT1507 may function on other substrates in addition to piericidins. BmmGT1, the GT1507 homolog, is indeed involved in glucosylation of macrolactins as well as bacillaenes in vivo (23, 24). Moreover, glycosylation confers host cell immunity from endogenous and exogenous agents (30). Therefore, in addition to piericidin glucosylation, GT1507 could probably have other biological function(s). Interestingly, when GT1507 was overexpressed in OUC6819, in addition to compounds 2 and 5, piericidin glycoside compound 8 carrying an N-acetylglucosamine group also accumulated (Fig. 3), indicating the different selectivity of GT1507 and BmmGT1 toward sugar donors. Further experiments are required to elucidate the underling molecular mechanism.
With compound 2 being the prominent piericidin glycoside, compounds 3 to 7 are diglucosylated piericidin derivatives, interestingly, with different glucosylation patterns. Although O-diglucosylation at C-4′ was obtained (compound 4) (Fig. 3), single O-glucosylation at C-4′ was not observed either in vitro (Fig. 2) or in vivo (Fig. 3). Nevertheless, BmmGT1 prefers to transfer the glucosyl group onto C-10-OH over C-4′-OH. Noticeably, O-diglucosylation at C-10 and C-4′ occurs by 1,3- (compound 3) and 1,6-linkages (compounds 4 and 6), respectively, which is possibly ascribed to the substrate binding manners of the sugar receptors. Mutagenesis and crystallographic studies will be helpful to understand the molecular basis of the BmmGT1-catalyzed O-diglucosylation reactions.
Compared with compound 1, monoglucosylated compounds 2 and 8 exhibited stronger cytotoxicities toward solid cancer cell lines A549, A375, HCT-116, and HT-29. Diglucopiericidins 3 to 5 displayed varied cytotoxicities against these cell lines (Table 4). These results indicated that glycosyl groups and their patterns are important for the bioactivities, consistent with previously reported results (14). Notably, diglucosylated compound 5 exhibited comparable cytotoxic activities to those of compound 2 and might serve as a potential candidate for further drug development.
The wide distribution of the GT1507 homologs across diverse bacteria indicates the importance of these GTs, which might have served as the key manner for administration of endogenous chemical diversification and/or detoxification of xenobiotics for adaptation to environmental changes. From this viewpoint, these GTs might display broad substrate flexibility, recognizing diverse sugar acceptors and donors. Thus, these GTs might be potential enzymatic tools for compound glycodiversification.


In this study, five new piericidin glycosides (compounds 3, 4, 6, 7, and 8) were effectively generated by the overexpression of GT genes in a piericidin producer. Three of these compounds (compounds 4, 7, and 8) displayed cytotoxic selectivity. Notably, GT1507 was demonstrated to be associated with glucopiericidin biosynthesis. Mining of GT1507 homologs from the GenBank database revealed their wide distribution across various bacteria. Our findings will facilitate the exploration of GTs that can glycodiversify small molecules in the search for drug candidates.


Bacterial strains, plasmids, and culture conditions.

Bacterial strains and plasmids used in this study are listed in Table 5. Escherichia coli strains were cultivated at 37°C in lysogeny broth (LB) liquid medium or on LB agar. S. youssoufiensis OUC6819 strains (26) were cultured at 30°C in TSBY (yeast extract, 5 g liter−1; tryptic soy broth, 30 g liter−1; sucrose, 103 g liter−1) for DNA extraction and on International Streptomyces Project synthetic salts-starch medium (ISP4) agar containing 0.5% glycine for genetic manipulation. When necessary, the medium was supplemented with 50 μg ml−1 of apramycin, 100 μg ml−1 of kanamycin, 20 μg ml−1 of nalidixic acid for Streptomyces, and 50 μg ml−1 of apramycin, 100 μg ml−1 of kanamycin, 25 μg ml−1 of chloramphenicol, 25 μg ml−1 of tetracycline, and 100 μg ml−1 of ampicillin for E. coli.
TABLE 5 Bacteria and plasmids used in this study
Strain or plasmidDescriptionReference or source
    E. coli DH5αHost strain for general cloningStratagene
    E. coli ET12567/pUZ8002Host strain for conjugation35
    S. youssoufiensisStrain producing reedsmycins and piericidins26
    OUC6819 ΔrdmFrdmF inactivation strain of S. youssoufiensis OUC6819, which is abolished in reedsmycins production27
    OUC6819 WT/pWLI901OUC6819 wild-type strain with pWLI901This study
    OUC6819 WT/pWLI902OUC6819 wild-type strain with pWLI902This study
    OUC6819 WT/pWLI903OUC6819 wild-type strain with pWLI903This study
    OUC6819 ΔrdmF/pWLI901OUC6819 ΔrdmF with pWLI901This study
    OUC6819 ΔrdmF/pWLI902OUC6819 ΔrdmF with pWLI902This study
    OUC6819 ΔrdmF/pWLI903OUC6819 ΔrdmF with pWLI903This study
    pWLI206pET28a carrying bmmGT123
    pSET152CpSET152 derivative, with insertion of the neo gene from SuperCos1 at the sites of ApaI and SgrAIOur laboratory
    pWLI901pSET152C derivative harboring sbmGT1 under the control of PgapdhThis study
    pWLI902pSET152C derivative harboring GT1507 under the control of PgapdhThis study
    pWLI903pSET152C derivative harboring SY7155 under the control of PgapdhThis study
    pWLI904pET28a carrying GT1507This study

Codon optimization of bmmGT1 and DNA manipulation.

The gene sequence of bmmGT1 was codon optimized per the codon usage of Streptomyces coelicolor A3(2) using JCat software (31). The codon adaptation index (CAI) value was improved from 0.64 (bmmGT1) to 1.00 (sbmGT1) with the GC content increasing from 48% (bmmGT1) to 66% (sbmGT1). The resulting gene sequence named as sbmGT1 (see Table S2 in the supplemental material) was synthesized by Sunny Biotech Company (Shanghai, China). Plasmid extractions and DNA purifications were carried out using commercial kits (OMEGA, Bio-Tek, Guangzhou, China). PCRs were carried out using Pfu DNA polymerase. Restriction endonucleases and T4 DNA ligase were purchased from Fermentas (Shenzhen, China). Oligonucleotide synthesis and DNA sequencing were performed by Sunny Biotech Company (Shanghai, China).

In vitro assays.

GT1507 was amplified from the S. youssoufiensis genome by PCR using Pfu DNA polymerase with the primer pairs GT1507-PFP/PRP (Table 6). The resulting PCR products were digested with NdeI and XhoI, purified, and cloned into the same sites of pET28a to generate expression vector pWLI904. After confirmation by sequencing, pWLI904 was introduced into E. coli BL21(DE3). Expression of GT1507 was induced at an optical density at 600 nm (OD600) of approximately 0.6 by addition of isopropyl-β-d-thiogalactopyranoside (IPTG; 0.2 mM final concentration), and cultivation was continued for an additional 4 h at 30°C. Cells were harvested by centrifugation, washed twice, and resuspended in 50 mM Tris-HCl buffer (pH 7.5). The resuspended cells were lysed by sonication in an ice-water bath with an ultrasonic processor VCX750 (Sonics & Materials Inc., PA, USA) and centrifuged at 11,000 × g for 30 min at 4°C. The supernatant was applied to a His-Trap HP column (1 ml, GE Healthcare) and the His6-tagged GT1507 was eluted with a linear gradient of imidazole (10 to 500 mM) in the binding buffer using an ÄKTA purifier system. The purified protein was desalted using Ultra free VR 24 centrifugal filter unit (Millipore, Bedford, MA, USA), and stored in 50 mM Tris-HCl (pH 8.0) buffer at −80°C until use. The protein concentration was determined by the Bradford method using bovine serum albumin (BSA) as the standard.
TABLE 6 Primers used in this study
NameSequence (5′→3′)a
Underlined letters represent restriction sites. The primer pair GT1507-PFP/PRP was used for GT1507 cloning. The primer pair SGT1-FP/SGT1-RP was used for sbmGT1 cloning. The primer GT1507-FP/RP was used for GT1507 cloning. The primer of SY7155-FP/RP was used for SY7155 cloning. The 3′-OH of pGRP was phosphorylated. pGFP/RP was used for amplification of strong constitutive promoter Pgapdh.
A typical reaction was conducted in a 30-μl system consisting of 500 μM compound 1, 2 μM enzyme, 2 mM UDP-d-glucose (or UDP-d-N-acetylglucosamine), and 10 mM MgCl2 in 50 mM Tris-HCl buffer (pH 8.0). The reaction mixtures were incubated at 30°C for 2 h, the reactions were quenched by the addition of 30 μl CH3CN, and the denatured proteins were removed by centrifugation. The assays were monitored by HPLC analysis, using a C18 YMC pack ODS-AQ column (5 μm, 150 mm by 4.6 mm) with UV detection at 260 nm under the following program: solvent system (phase A, 0.1% formic acid in H2O; phase B, 0.1% formic acid in CH3CN); 20% B (0 to 5 min), 20% to 100% B (5 to 45 min), 100% B (45 to 55 min), at a flow rate of 1 ml min−1.
To determine the kinetic parameters of GT1507 and BmmGT1 toward different substrates, the reactions were performed in 30-μl systems containing 50 mM Tris-HCl buffer (pH 8.0), 10 mM MgCl2, GT1507/BmmGT1, and substrate at various concentrations (see the legend to Fig. S15). After incubation at 30°C for 5 min, the reactions were quenched by the addition of 30 μl CH3CN, and the denatured proteins were removed by centrifugation. The assays were monitored by HPLC analysis as mentioned above. The consumption of compound 1 in each reaction was quantified using standard curves, and the reaction rates were calculated. The Vmax and Km values were determined by plotting the reaction rates against the substrate concentrations. The data fitting was conducted by hyperbolic regression using Origin software (OriginLab Corporation, USA). The values represent the means from three independent reactions.

Gene overexpression.

For gene overexpression, sbmGT1, GT1507, and SY7155 were each put under the control of the constitutive promoter Pgapdh from S. youssoufiensis OUC6819 (Table 6). Construction was performed as follows: taking the sbmGT1 expression plasmid as an example, Pgapdh was amplified using the primer pair pGFP/3′-OH phosphorylated pGRP (Table 6) and was digested with EcoRI; the sbmGT1 fragment was amplified with the primer pair SGT1-FP/RR (Table 6) and was digested with XbaI. Then, these two digested fragments were ligated and cloned into the EcoRI and XbaI sites of pSET152C to give pWLI901. Similarly, the fragments of GT1507 and SY7155 were cloned into the same sites of pSET152C to obtain pWLI902 and pWLI903, respectively (Table 5). After confirmation by sequencing, the resulting plasmids were transformed into E. coli ET12567/pUZ8002 and then introduced into the wild-type S. youssoufiensis OUC6819 and the ΔrdmF mutant strain via conjugation.

Production, analyses, and purification of piericidins.

Spores of Streptomyces strains were inoculated into 50 ml medium in a 250-ml flask for production analysis or into 200 ml in a 1-liter flask for isolation and were incubated on a rotatory shaker at 30°C with 220 rpm for 7 days. The fermentation cultures were harvested by centrifugation, and the supernatant was extracted twice with an equal volume of ethyl acetate to afford residue A. The pelleted mycelia were extracted with acetone, followed by evaporation in vacuo to yield residue B. The combined residues were dissolved in methanol, filtered through a 0.22-μm filter, and subjected to HPLC analysis. The HPLC system consisted of Agilent 1260 Infinity quaternary pumps and a 1260 Infinity diode-array detector. Analytical HPLC was performed on an Eclipse C18 column (5 μl, 4.6 mm by 150 mm) with a linear gradient from 20% to 100% acetonitrile (ACN)/H2O in 40 min followed by an additional 10 min at 100% ACN at flow rate of 1 ml·min−1 and UV detection at 260 nm.
A total volume of 20 liters (OUC6819 ΔrdmF/pWLI901) or 15 liters (OUC6819 ΔrdmF/pWLI902) of fermentation broths were harvested and treated as described above with ethyl acetate (EtOAc) three times to get the crude extracts. The crude extracts were partitioned between 90% methanol (MeOH) and n-hexane to remove nonpolar components. Then, the MeOH layer was subjected to a stepped-gradient open column (ODS-A, 120 Å, S-30/50 mesh) eluting with 20 to 100% MeOH to yield 14 fractions. Compounds 7 (5 mg) and 6 (3 mg) were obtained by further purification of fraction 7 of OUC6819 ΔrdmF/pWLI901 on reversed-phase HPLC (YMC-Pack ODS-A column 250 mm by 10 mm; inside diameter [i.d.], 5 μm; wavelength, 260 nm) eluting with 48% CH3CN plus 0.1% HCOOH (vol/vol) (1.5 ml·min−1). Compounds 5 (3 mg), 4 (2 mg), and 3 (6 mg) were obtained from fraction 8 of OUC6819 ΔrdmF/pWLI901 eluting with 48% CH3CN plus 0.1% HCOOH (vol/vol) (1.5 ml·min−1). Compound 2 (80 mg) was obtained from fraction 9 of OUC6819 ΔrdmF/pWLI901 eluting with 58% CH3CN plus 0.1% HCOOH (vol/vol) (1.5 ml·min−1). Compound 1 (10 mg) was obtained from fraction 11 of OUC6819 ΔrdmF/pWLI901 eluting with 75% CH3CN plus 0.1% HCOOH (vol/vol) (1.5 ml·min−1). Compound 8 (12 mg) was obtained from fraction 9 of OUC6819 ΔrdmF/pWLI902 eluting with 58% CH3CN plus 0.1% HCOOH (vol/vol) (1.5 ml·min−1). The identities of these compounds were elucidated by high-resolution electrospray ionization mass spectrometry (HRESIMS) and NMR analysis. HRESIMS was carried out on a Thermo LTQ-XL mass spectrometer. NMR data were recorded with an Agilent-DD2500 spectrometer.

Sequence similarity networks of the GT1507 homologs.

The EFI-Enzyme Similarity Tool (EFI-EST) is used to generate sequence similarity networks (SSNs) (32). For distribution of GT1507 homologs, the GT1507 sequence was used as the query for searching nonredundant protein sequence database using PSI-BLAST (position-specific iterated BLAST) in NCBI. Partially, 1,573 GT1507 homologs were used to build the GT sequence similarity networks. Network analysis was performed by BLASTP searches, comparing each sequence against each of the other sequences. The 10−85 SSNs were generated by applying an E value cutoff of 10−85 to the full network. Each node in the network represents a single sequence, and each edge represents the pairwise connection between two sequences for which the BLASTP E value was lower than the cutoff value. SSNs were visualized and colored by Cytoscape (v3.7.2) (33).

Cytotoxicity assay.

The cytotoxicities of compounds 1 to 8 were evaluated against A549, A375, HCT-116, HT-29 and THP-1 tumor cell lines by using the sulforhodamine B (SRB) method (34). A549, A375 HCT-116, HT-29, and THP-1 cells were seeded in 96-well plates at 5,000 cells/well. After 24 h, test samples at different concentrations were added, using 1 μM doxorubicin as the positive control and corresponding medium as the blank control. Three replicates were set for each concentration. After incubation for 72 h, 50% (wt/vol) cold trichloroacetic acid (TCA) was added to each well to fix the cells, followed by staining with SRB. After the addition of 150 μl Tris solution to each well, OD540 values were measured using a microplate reader (SpectraMax I3; Molecular Devices, USA). The inhibition rate of tumor cell growth was calculated according to the following formula: inhibition rate (%) = [(OD540 (control well) − OD540 (test well))/OD540 (control well)] × 100. The IC50 values were calculated using GraphPad Prism 5.

Data availability.

The sequences of GT1507 and SY7155 have been deposited in GenBank under the accession numbers MN953422 and MN953423, respectively.


This work was supported by grants from the National Key R&D Program of China (2019YFC0312501), the National Natural Science Foundation of China (U1706206, 32070054, 31570032, 31711530219, and 31171201), the Taishan Scholars Program (tsqn201909170), and the Science and Technology Project of Qingdao under grant 17-3-3-44-nsh.
We declare no conflict of interest.
Z.L. performed all the experiments. F.X., T.W., C.L. and Y.J. assisted with in vitro assays. H.L. was involved in NMR analysis. S.C. and X.W. performed the cytotoxicity assays. Q.C., T.Z., and D.L. provided the wild-type S. youssoufiensis OUC6819. Z.L. and F.X. were involved in the draft manuscript writing. W.L. supervised the whole work and wrote the manuscript. All authors read and approved the final manuscript.

Supplemental Material

File (aem.00294-21-s0001.pdf)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.


Lairson LL, Henrissat B, Davies GJ, Withers SG. 2008. Glycosyltransferases: structures, functions, and mechanisms. Annu Rev Biochem 77:521–555.
Luzhetskyy A, Mendez C, Salas JA, Bechthold A. 2008. Glycosyltransferases, important tools for drug design. Curr Top Med Chem 8:680–709.
Thibodeaux CJ, Melancon CE, Liu HW. 2008. Biosynthese von naturstoffzuckern und enzymatische glycodiversifizierung. Angew Chem 120:9960–10007.
Williams GJ, Thorson JS. 2009. Natural product glycosyltransferases: properties and applications. Adv Enzymol Relat Areas Mol Biol 76:55–119.
Zhang C, Griffith B, Fu Q, Albermann C, Fu X, Lee I, Li L, Thorson J. 2006. Exploiting the reversibility of natural product glycosyltransferase-catalyzed reactions. Science 313:1291–1294.
Gantt RW, Peltier-Pain P, Cournoyer WJ, Thorson JS. 2011. Using simple donors to drive the equilibria of glycosyltransferase-catalyzed reactions. Nat Chem Biol 7:685–691.
Chen R, Zhang H, Zhang G, Li S, Zhang G, Zhu Y, Liu J, Zhang C. 2013. Characterizing amosamine biosynthesis in amicetin reveals amig as a reversible retaining glycosyltransferase. J Am Chem Soc 135:12152–12155.
Gantt RW, Peltier-Pain P, Singh S, Zhou M, Thorson JS. 2013. Broadening the scope of glycosyltransferase-catalyzed sugar nucleotide synthesis. Proc Natl Acad Sci U S A 110:7648–7653.
Takahashi N, Suzuki A, Tamura S. 1965. Structure of piericidin A. J Am Chem Soc 87:2066–2068.
Hall C, Wu M, Crane FL, Takahashi H, Tamura S, Folkers K. 1966. Piericidin A: a new inhibitor of mitochondrial electron transport. Biochem Biophys Res Commun 25:373–377.
Gutman M, Singer TP, Beinert H, Casida JE. 1970. Reaction sites of rotenone, piericidin A, and amytal in relation to the nonheme iron components of NADH dehydrogenase. Proc Natl Acad Sci U S A 65:763–770.
Matsumoto M, Mogi KI, Nagaoka K, Ishizeki S, Kawahara R, Nakashima T. 1987. New piericidin glucosides, glucopiericidins A and B. J Antibiot (Tokyo) 40:149–156.
Zhou X, Fenical W. 2016. The unique chemistry and biology of the piericidins. J Antibiot (Tokyo) 69:582–593.
Shaaban KA, Helmke E, Kelter G, Fiebig HH, Laatsch H. 2011. Glucopiericidin C: a cytotoxic piericidin glucoside antibiotic produced by a marine Streptomyces isolate. J Antibiot (Tokyo) 64:205–209.
Shang N, Zhang Z, Huang J, Wang L, Luo J, Yang J, Peng T, Yan Y, Ma Y, Huang S. 2018. Glycosylated piericidins from an endophytic streptomyces with cytotoxicity and antimicrobial activity. J Antibiot (Tokyo) 71:672–676.
Zhou X, Liang Z, Li K, Fang W, Tian Y, Luo X, Chen Y, Zhan Z, Zhang T, Liao S, Liu S, Liu Y, Fenical W, Tang L. 2019. Exploring the natural piericidins as anti-renal cell carcinoma agents targeting peroxiredoxin 1. J Med Chem 62:7058–7069.
Kimura K, Nakayama S, Nakajima N, Yoshihama M, Miyata N, Kawanishi G. 1990. A new piericidin rhamnoside, 3'-rhamnopiericidin A1. J Antibiot (Tokyo) 43:1341–1343.
Iwasaki H, Kamisango K, Kuboniwa H, Sasaki H, Matsubara S. 1991. 3'-Deoxytalopiericidin A1, a novel analog of antitumor antibiotics from oligotroph. J Antibiot (Tokyo) 44:451–452.
Kimura K, Takahashi H, Miyata N, Yoshihama M, Uramoto M. 1996. New piericidin antibiotics, 7-demethylpiericidin A1 and 7-demethyl-3'-rhamnopiericidin A1. J Antibiot (Tokyo) 49:697–699.
Tanabe M, Seto H. 1970. Biosynthetic studies with carbon 13. Piericidin A. J Org Chem 35:2087–2088.
Liu Q, Yao F, Chooi YH, Kang Q, Xu W, Li Y, Shao Y, Shi Y, Deng Z, Tang Y, You D. 2012. Elucidation of piericidin A1 biosynthetic locus revealed a thioesterase-dependent mechanism of α-pyridone ring formation. Chem Biol 19:243–253.
Chen Y, Zhang W, Zhu Y, Zhang Q, Tian X, Zhang S, Zhang C. 2014. Elucidating hydroxylation and methylation steps tailoring piericidin A1 biosynthesis. Org Lett 16:736–739.
Qin W, Liu Y, Ren P, Zhang J, Li H, Tian L, Li W. 2014. Uncovering a glycosyltransferase provides insights into the glycosylation step during macrolactin and bacillaene biosynthesis. Chembiochem 15:2747–2753.
Liu Y, Qin W, Liu Q, Zhang J, Li H, Xu S, Ren P, Tian L, Li W. 2016. Genome-wide identification and characterization of macrolide glycosyltransferases from a marine-derived Bacillus strain and their phylogenetic distribution. Environ Microbiol 18:4770–4781.
Liu Q, Ren P, Liu Y, Qin W, Li H, Li W. 2018. Exploration of the glycosyltransferase BmmGT1 from a marine-derived Bacillus strain as a potential enzyme tool for compound glycol-diversification. J Microbiol Biotechnol 28:931–937.
Che Q, Li T, Liu X, Yao T, Li J, Gu Q, Li D, Li W, Zhu T. 2015. Genome scanning inspired isolation of reedsmycins A-F, polyene-polyol macrolides from Streptomyces sp. CHQ-64. RSC Adv 5:22777–22782.
Yao T, Liu Z, Li T, Zhang H, Liu J, Li H, Che Q, Zhu T, Li D, Li W. 2018. Characterization of the biosynthetic gene cluster of the polyene macrolide antibiotic reedsmycins from a marine-derived Streptomyces strain. Microb Cell Fact 17:98.
Song MC, Kim E, Ban YH, Yoo YJ, Kim EJ, Park SR, Pandey RP, Sohng JK, Yoon YJ. 2013. Achievements and impacts of glycosylation reactions involved in natural product biosynthesis in prokaryotes. Appl Microbiol Biotechnol 97:5691–5704.
Han X, Liu Z, Zhang Z, Zhang X, Zhu T, Gu Q, Li W, Che Q, Li D. 2017. Geranylpyrrol A and Piericidin F from Streptomyces sp. CHQ-64 ΔrdmF. J Nat Prod 80:1684–1687.
Bolam DN, Roberts S, Proctor MR, Turkenburg JP, Dodson EJ, Martinez-Fleites C, Yang M, Davis BG, Davies GJ, Gilbert HJ. 2007. The crystal structure of two macrolide glycosyltransferases provides a blueprint for host cell antibiotic immunity. Proc Natl Acad Sci U S A 104:5336–5341.
Grote A, Hiller K, Scheer M, Munch R, Nortemann B, Hempel DC, Jahn D. 2005. JCat: a novel tool to adapt codon usage of a target gene to its potential expression host. Nucleic Acids Res 33:W526–W531.
Gerlt JA, Bouvier JT, Davidson DB, Imker HJ, Sadkhin B, Slater DR, Whalen KL. 2015. Enzyme Function Initiative-Enzyme Similarity Tool (EFI-EST): a web tool for generating protein sequence similarity networks. Biochim Biophys Acta 1854:1019–1037.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. 2003. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498–2504.
Vichai V, Kirtikara K. 2006. Sulforhodamine B colorimetric assay for cytotoxicity screening. Nat Protoc 1:1112–1116.
Gust B, Challis GL, Fowler K, Kieser T, Chater KF. 2003. PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin. Proc Natl Acad Sci U S A 100:1541–1546.

Information & Contributors


Published In

cover image Applied and Environmental Microbiology
Applied and Environmental Microbiology
Volume 87Number 1311 June 2021
eLocator: e00294-21
Editor: M. Julia Pettinari, University of Buenos Aires
PubMed: 33893110


Received: 9 February 2021
Accepted: 12 April 2021
Accepted manuscript posted online: 23 April 2021
Published online: 11 June 2021


Request permissions for this article.


  1. glycosyltransferase (GT)
  2. glucosylpiericidins
  3. glycosylation
  4. biosynthesis
  5. overexpression



Zengzhi Liu
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Fei Xiao
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Siqi Cai
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Chunni Liu
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Huayue Li
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
Ting Wu
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Yuechen Jiang
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Xin Wang
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
Qian Che
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
Tianjiao Zhu
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
Dehai Li
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China
Key Laboratory of Marine Drugs, Ministry of Education of China, School of Medicine and Pharmacy, Ocean University of China, Qingdao, China
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China


M. Julia Pettinari
University of Buenos Aires

Metrics & Citations


Note: There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.

Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy