Functional and structural characterization of IdnL7, an adenylation enzyme involved in incednine biosynthesis

The adenylation enzyme IdnL7, which is involved in the biosynthesis of incednine, has been biochemically and structurally characterized. Biochemical analysis showed that IdnL7 selects and activates several small amino acids. The structure of IdnL7 in complex with an l-alanyl-adenylate intermediate mimic explains the broad substrate specificity of IdnL7 towards small l-amino acids.

Abstract

Adenylation enzymes play an important role in the selective incorporation of the cognate carboxylate substrates in natural product biosynthesis. Here, the biochemical and structural characterization of the adenylation enzyme IdnL7, which is involved in the biosynthesis of the macrolactam polyketide antibiotic incednine, is reported. Biochemical analysis showed that IdnL7 selects and activates several small amino acids. The structure of IdnL7 in complex with an l-alanyl-adenylate intermediate mimic, 5′-O-[N-(l-alanyl)sulfamoyl]adenosine, was determined at 2.1 Å resolution. The structure of IdnL7 explains the broad substrate specificity of IdnL7 towards small l-amino acids.

1. Introduction  

Adenylation enzymes are involved in the biosynthesis of nonribosomal peptides and related natural products (Gulick, 2009 ▸). Canonical adenylation enzymes, such as PheA in gramicidin biosynthesis (Conti et al., 1997 ▸), first activate a carboxylate substrate by catalyzing its reaction with ATP to form an acyl-AMP intermediate (Fig. 1 ▸ a). This intermediate then undergoes nucleophilic substitution with the thiol group of the 4′-phosphopantetheine arm of a carrier protein (CP) to give a thioester product, which is used for the condensation reaction in nonribosomal peptide synthetase (NRPS) systems. The adenylation enzyme functions as a gatekeeper in the selective incorporation of an appropriate carboxylate substrate in natural product biosynthesis. The carboxylate substrates for adenylation enzymes include nonproteinogenic amino acids such as β-amino acids in addition to standard proteinogenic α-amino acids, leading to the structural diversifi­cation of natural products. The substrate specificity of adenylation enzymes is dictated by a group of ten residues, referred to as the specificity-conferring code, which are located in the substrate-binding pocket (Khayatt et al., 2013 ▸; Rausch et al., 2005 ▸; Stachelhaus et al., 1999 ▸; Kudo et al., 2019 ▸). Bioinformatic tools have been developed for predicting the substrate specificity of adenylation enzymes based on the specificity-conferring code, although it is still difficult to predict the substrate specificities of some adenylation enzymes from the specificity-conferring codes only.

Figure 1.

Reactions of adenylate-forming enzymes. (a) Typical reactions of the adenylation enzyme PheA with carrier protein (CP). (b) Reactions of VinM-type enzymes with aminoacyl-CP.

β-Amino-acid-containing macrolactam antibiotics are an important class of macrocyclic polyketides (Kudo et al., 2014 ▸; Miyanaga, Kudo et al., 2016 ▸). They contain various β-amino-acid starter units in their polyketide skeletons: vicenistatin, produced by Streptomyces halstedii HC34, contains a 3-aminoisobutyrate unit, and incednine, produced by Streptomyces sp. ML694-90F3, contains a 3-aminobutyrate unit. The mechanism of incorporation of the β-amino-acid unit was first elucidated in the vicenistatin-biosynthetic pathway (Shinohara et al., 2011 ▸). In vicenistatin biosynthesis, an adenylation enzyme, VinN, transfers (2S,3S)-3-methylaspartate to a standalone CP, VinL, to give 3-methylaspartyl-VinL (Supplementary Fig. S1a). After the decarboxylation of 3-methylaspartyl-VinL, another adenylation enzyme, VinM, catalyzes the amino­acylation of 3-aminoisobutyryl-VinL with l-alanine to give l-alanyl-3-aminoisobutyryl-VinL. Thus, VinM catalyzes amide-bond formation with aminoacyl-VinL instead of the typical thioesterification reaction with CP (Fig. 1 ▸ b). VinM first adenylates l-alanine and then uses the resulting l-alanyl-adenylate intermediate for amide-bond formation with 3-aminoisobutyryl-VinL. The dipeptidyl group is then transferred to the CP domain of polyketide synthase by an acyltransferase, VinK, for polyketide-chain elongation. The terminal l-alanyl group remains attached during the poly­ketide elongation reaction and probably functions as a biosynthetic protecting group (Shinohara et al., 2011 ▸; Miyanaga, Kudo et al., 2016 ▸). After polyketide-chain elongation, the terminal l-alanyl group is removed by the amidohydrolase VinJ. The homologous genes for the starter-related enzymes, VinN, VinL, VinM, VinK and VinJ, are fully conserved in the incednine-biosynthetic gene cluster, suggesting that the (S)-3-aminobutyrate unit is incorporated using the same mechanism in incednine biosynthesis (Takaishi et al., 2013 ▸). In the proposed incednine-biosynthetic pathway (Supplementary Fig. S1b), the VinN-type adenylation enzyme IdnL1 activates (S)-3-aminobutyric acid, which is biosynthetically derived from l-glutamate (Takaishi et al., 2012 ▸), and subsequently transfers it to the standalone CP IdnL6. Next, the VinM-type adenylation enzyme IdnL7 catalyzes the aminoacylation of (S)-3-aminobutyryl-IdnL6 with a small amino acid, such as l-alanine, to form dipeptidyl-IdnL6.

We have recently determined the crystal structures of VinN-type adenylation enzymes, including IdnL1, and these provided mechanistic insights into the recognition of β-amino acids (Miyanaga et al., 2014 ▸; Cieślak et al., 2017 ▸). However, the substrate-recognition mechanism of the VinM-type adenyl­ation enzyme remained elusive because of a lack of structural information. Here, we report a substrate-recognition analysis and the crystal structure of IdnL7. The structure allows the visualization of the active site of IdnL7 and explains the recognition modes of a variety of small amino acids by the VinM-type family of adenylation enzymes.

2. Materials and methods  

2.1. Cloning, expression and purification of His₆-IdnL7 protein  

The idnL7 gene was amplified by PCR from a cosmid obtained in a previous experiment (Takaishi et al., 2013 ▸) with the primers IdnL7-N, 5′-CGCATATGGACGGCATCTTG-3′ (the NdeI site is underlined), and IdnL1-C, 5′-CCCCTCGAGACCCCTAAAGA-3′ (the XhoI site is underlined). The PCR fragment was ligated into the pT7Blue T-Vector (Merck Millipore, Billerica, Massachusetts, USA). After sequence confirmation, the resulting plasmid was digested with NdeI and XhoI, and the obtained fragment was subsequently inserted into the corresponding site of the pColdI vector (Takara Biochemicals, Ohtsu, Japan) to prepare N-terminally His₆-tagged recombinant IdnL7 protein. Escherichia coli Rosetta 2 (DE3) cells harboring pColdI-idnL7 were cultivated at 37°C in Luria–Bertani broth containing ampicillin (50 µg ml⁻¹) and chloramphenicol (30 µg ml⁻¹) until an optical density of 0.5 at 600 nm was reached. The cultures were shifted to a cold water bath and protein expression was induced by the addition of 0.1 mM isopropyl β-d-1-thio­galactopyranoside. The cells were grown at 15°C for a further 20 h. The harvested cell pellets were resuspended in buffer (50 mM Tris–HCl pH 7.5, 10% glycerol) and lysed by sonication. The supernatant was directly applied onto a His60 Ni Superflow affinity column (Clontech, Mountain View, California, USA). Bound recombinant IdnL7 protein was eluted with elution buffer (50 mM Tris–HCl pH 7.5, 300 mM imidazole, 10% glycerol), desalted on a PD-10 column (GE Healthcare, Buckinghamshire, England) with exchange into 10 mM Tris–HCl pH 7.5, 1 mM TCEP and concentrated using an Amicon Ultra centrifugal filter (Merck Millipore). The purity of the recombinant protein fraction was verified by SDS–PAGE analysis. Table 1 ▸ summarizes information regarding the production of the macromolecule.

Table 1. Macromolecule-production information.

Source organism	Streptomyces sp. ML694-90F3
Forward primer	CGCATATGGACGGCATCTTG
Reverse primer	CCCCTCGAGACCCCTAAAGA
Cloning vector	pT7Blue
Expression vector	pColdI
Expression host	E. coli Rosetta 2 (DE3)
Complete amino-acid sequence of the construct produced	MNHKVHHHHHHIEGRHMDGILDHGLHARFLRGLSAAPDGAAVRIGTTSVSYRHLHRTALLWAGALTAAGARSVGVLAGKSATGYAGILAALYAGAAVVPLRPDFPAARTREVLRASDADVLIADRAGLPVLAGALAGDGAADVPVLAPDALDGELPEGVARLVPRPELSLSEPARCKPADPAYLLFTSGSTGRPKGVVITHGATGHYFDVMERRYDFGASDVFSQAFDLNFDCAVFDLFCAWGAGATVVPVPPPAYRDLPGFITAQGITVWFSTPSVIDLTRRLGALDGPRMPGLRWSLFAGEALKCRDAADWRAAAPGATLENLYGPTELTITVAAHRWDDEESPRAAVNGLAPIGAVNDGHDHLLLGPDGDPSPDEGELWVTGPQLAAGYLDPADERGRFAERDGRRWYRTGDRVRRAPGGDLVYVGRLDSQLQVHGWRVEPAEVEHAVRACGADDAVVVGVDTPGGTELVAFYTGIPVEPRELVRRLREVVPDGVLPRHFRHLDAFPLNANRKTDRLRLTTMAADGYGPRSGPAL

2.2. Kinetics  

Enzyme-activity and kinetic parameters were determined according to a continuous spectrophotometric assay that measures the release of inorganic pyrophosphate (Webb, 1992 ▸). The following commercially available amino acids were tested: glycine, l-alanine, l-cysteine, l-valine, l-aspartic acid (all from Kanto Chemical Co. Inc., Tokyo, Japan), d-alanine, l-serine, l-threonine, l-proline (all from TCI, Tokyo, Japan) and l-asparagine (from FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan). Reaction mixtures (150 µl total volume) consisted of 50 mM Tris–HCl pH 7.5, 10% glycerol, 1 mM ATP, 1 mM MgCl₂, 0.1 mM 2-amino-6-mercapto-7-methylpurine ribonucleoside, 1 U ml⁻¹ inorganic pyrophos­phatase, 1 U ml⁻¹ purine nucleoside phosphorylase and 1 mM amino acid. Reactions were initiated by the addition of 4 µM IdnL7 and were then incubated at 25°C. The increase in the absorbance at 360 nm, which is attributed to the formation of 2-amino-6-mercapto-7-methylpurine (∊_360 nm = 11 000), was monitored using a UV-2450 spectrophotometer (Shimadzu, Tokyo, Japan). For kinetic analysis, the concentration of l-alanine was varied between 0.01 and 2 mM. K _m and k _cat values were determined using the Michaelis–Menten equation. All assays were carried out in triplicate.

2.3. Crystallization, data collection and structure determination of IdnL7  

IdnL7 protein (at 10 mg ml⁻¹) in 10 mM Tris–HCl pH 7.5, 1 mM TCEP was incubated at 28°C for 10 min in the presence of 0.5 mM 5′-O-[N-(l-alanyl)sulfamoyl]adenosine (l-Ala-SA) and 0.5 mM MgCl₂ before being mixed with reservoir solution. Crystals of IdnL7 were grown by sitting-drop vapor diffusion by mixing the protein solution with an equal volume of reservoir solution [0.1 M Tris–HCl pH 7.5, 30%(w/v) PEG 4000, 0.2 M MgCl₂] at 26°C. The detailed crystallization conditions are summarized in Table 2 ▸. Crystals were soaked in reservoir solution containing 25%(v/v) glycerol as a cryoprotectant and flash-cooled in a liquid-nitrogen stream prior to X-ray data collection. Diffraction data were collected on the AR-NW12A beamline at the Photon Factory, Tsukuba, Japan and were processed using the HKL-2000 program suite (Otwinowski & Minor, 1997 ▸). Data-collection and processing statistics are summarized in Table 3 ▸. The initial phase of IdnL7 was determined by molecular replacement with MOLREP (Vagin & Teplyakov, 2010 ▸) using the PheA structure (PDB entry 1amu; Conti et al., 1997 ▸) as a search model. Protein model building of IdnL7 was carried out automatically with ARP/wARP (Morris et al., 2002 ▸) and subsequently inspected using Coot (Emsley & Cowtan, 2004 ▸). REFMAC (Murshudov et al., 2011 ▸) was used to refine the structure. The final refinement statistics are shown in Table 4 ▸. Structural representations were prepared with PyMOL (Schrödinger). The geometries of the final structure were evaluated using MolProbity (Chen et al., 2010 ▸). The atomic model and structure factors have been deposited in the Protein Data Bank (PDB entry 6akd).

Table 2. Crystallization.

Method	Sitting-drop vapor diffusion
Plate type	96-well (Greiner)
Temperature (K)	299
Protein concentration (mg ml−1)	10
Buffer composition of protein solution	10 mM Tris–HCl pH 7.5, 0.5 mM magnesium chloride, 0.5 mM L-Ala-SA
Composition of reservoir solution	0.1 M Tris–HCl pH 8.5, 0.2 M magnesium chloride, 30% PEG 4000
Volume and ratio of drop	1 µl protein solution + 1 µl reservoir solution
Volume of reservoir (µl)	50

Table 3. Data-collection and processing statistics.

Values in parentheses are for the outer resolution shell.

Diffraction source	AR-NW12A, Photon Factory
Wavelength (Å)	1.00000
Temperature (K)	100
Detector	ADSC Quantum 270
Crystal-to-detector distance (mm)	244.0
Rotation range per image (°)	0.5
Total rotation range (°)	180
Exposure time per image (s)	1
Space group	P41212
a, b, c (Å)	68.53, 68.53, 225.17
α, β, γ (°)	90, 90, 90
Mosaicity (°)	0.47
Resolution range (Å)	50.00–2.10 (2.14–2.10)
Total No. of reflections	440969 (20168)
No. of unique reflections	32390 (1587)
Completeness (%)	99.7 (100.0)
Multiplicity	13.9 (14.2)
〈I/σ(I)〉	33.4 (2.6)
R meas	0.105 (0.990)
CC1/2	0.939 (0.706)
Overall B factor from Wilson plot (Å2)	39.6

Table 4. Structure refinement.

Resolution range (Å)	43.54–2.10
Completeness (%)	99.7
σ Cutoff	F > 0.00σ(F)
No. of reflections, working set	30667
No. of reflections, test set	1642
Final R cryst (%)	0.177
Final R free (%)	0.220
R.m.s. deviations
Bond lengths (Å)	0.018
Bond angles (°)	1.844
No. of chains in the asymmetric unit	1
No. of non-H atoms
Protein	3682
L-Ala-SA	28
Solvent	167
Average B factors (Å2)
Protein	43.2
L-Ala-SA	28.5
Solvent	43.9
Ramachandran plot
Favored region (%)	98.3
Allowed region (%)	1.5
Outlier region (%)	0.2

3. Results and discussion  

3.1. Substrate specificity of IdnL7  

The activity of IdnL7 with various amino acids was analyzed by measuring the release of pyrophosphate in an ATP-consuming reaction (Fig. 2 ▸). The results indicated that IdnL7 prefers small amino acids such as l-alanine (100% relative activity), l-serine (117%) and glycine (104%), consistent with the reported substrate specificity of VinM (Shinohara et al., 2011 ▸). The K _m and k _cat values of IdnL7 for l-alanine were 0.083 ± 0.020 mM and 0.40 ± 0.03 min⁻¹, respectively. IdnL7 showed moderate activity against l-cysteine (56%). On the other hand, IdnL7 showed very weak activity against larger amino acids such as l-threonine and l-valine. In addition, IdnL7 showed no activity against d-alanine, suggesting that IdnL7 is strictly an l-amino-acid adenylation enzyme.

Figure 2.

Substrate specificity of IdnL7. A relative enzyme activity of 100% corresponds to the specific activity of IdnL7 for l-Ala. Standard errors are calculated from three independent trials.

3.2. Overall structure  

To understand the structural basis for the substrate specificity of IdnL7, we crystallized IdnL7 in the presence of the nonhydrolysable aminoacyl-AMP intermediate analog l-Ala-SA (Fig. 3 ▸ a). Nonhydrolysable aminoacyl-AMP analogs have previously been reported in the structural analyses of several adenylation enzymes (Drake et al., 2010 ▸; Niquille et al., 2018 ▸). We succeeded in determining the crystal structure of IdnL7 in complex with l-Ala-SA at 2.1 Å resolution (Table 4 ▸). IdnL7 consists of two domains: a large N-terminal domain (Met1–Gly413) and a smaller C-terminal domain (Gln420–Leu522) (Fig. 3 ▸ b). The domains are connected by a flexible hinge region (Arg414–Leu419), which allows the rotation of the C-terminal domain during the two catalytic reaction steps. The N-terminal domain forms a five-layered αβαβα-sandwich fold, whereas the C-terminal domain comprises three helices with one two-stranded β-sheet and one three-stranded antiparallel β-sheet. The C-terminal domain is arranged in the adenylation conformation in relation to the N-terminal domain. The structure contains a small number of disordered regions: Met1–Asp2, Gly123–Asp126, Gly173–Arg177 and Pro353–Asp354 in the N-terminal domain and His422–Arg425, Asp449–Gly453 and Pro516–Leu522 in the C-terminal domain.

Figure 3.

Crystal structure of IdnL7. (a) Structure of l-Ala-AMP and the intermediate analog l-Ala-SA. (b) Cartoon representation of the overall structure of IdnL7 combined with a transparent surface and the aminoacyl-SA binding pocket. The overall structure consists of an N-terminal domain (light pink) and a C-terminal domain (magenta). The l-Ala-SA molecule and the residues involved in interactions with the ligand are shown as cyan and pink sticks, respectively. Blue spheres represent the positions of water molecules. Black dashed lines indicate hydrogen bonding. (c) F _o − F _c electron-density map of l-Ala-SA contoured at 3.0σ.

According to analysis using the DALI server (Holm & Sander, 1995 ▸), the N-terminal domain of IdnL7 is similar in structure to those of other adenylation enzymes such as AlmE from the AlmEFG pathway in Vibrio cholerae (Henderson et al., 2014 ▸; r.m.s.d. of 1.9 Å for 383 C^α atoms; 27% sequence identity) and the PheA domain from gramicidin synthetase (Conti et al., 1997 ▸; r.m.s.d. of 2.1 Å for 367 C^α atoms; 26% sequence identity) (Supplementary Fig. S2). The C-terminal domain of IdnL7 is also similar to those of other adenylation enzymes, such as the 2,3-dihydroxybenzoate-AMP ligase DhbE (May et al., 2002 ▸; r.m.s.d. of 1.7 Å for 84 C^α atoms; 23% sequence identity) and PheA (Conti et al., 1997 ▸; r.m.s.d. of 2.1 Å for 80 C^α atoms; 29% sequence identity).

3.3. Ligand-binding site  

The crystal structure shows unambiguous electron density for the inhibitor l-Ala-SA (Fig. 3 ▸ c) in the active site. IdnL7 binds the adenosine moiety of l-Ala-SA in a similar manner to most of the adenylation enzymes (Conti et al., 1997 ▸; Du et al., 2008 ▸; Drake et al., 2010 ▸). The adenine moiety is buried in a hydrophobic pocket that is lined by Tyr310, Tyr411 and Ile340 on one side and a loop (residues Gly286–Ala288) on the other side (Fig. 3 ▸ b). This architecture stabilizes the purine base by hydrophobic and van der Waals interactions. Additionally, the N6 amino group of the adenine ring interacts with the main-chain carbonyl of Leu309 and the side-chain amide group of Asn308. Several water molecules participate in direct interactions with the sulfamoyladenosine moiety. N1 and N3 from the adenine ring form close contacts with water molecules. The O-2′ and O-3′ hydroxy groups of ribose are recognized by Asp399 through hydrogen bonds. The ribose O-4′ and O-5′ atoms are hydrogen-bonded to the conserved Lys500. The sulfamoyl moiety appears to be anchored to the hydroxy group of Thr313 and a water molecule via two hydrogen bonds.

The binding site of the aminoacyl moiety includes two charged residues, Asp216 and Lys500, which are oriented to make contact with the amino and carboxy groups, respectively. The α-amino group of the l-alanyl moiety is involved in two salt-bridge interactions (2.9 and 3.1 Å) with the side chain of Asp216 and two hydrogen bonds (2.9 and 2.7 Å) to the backbone O atoms of Gly311 and Ile317. The α-carboxy group of l-Ala-SA forms an ionic interaction with the side chain of Lys500, which is the only residue from the C-terminal domain that directly interacts with the substrate. The Asp216 and Lys500 residues are strictly invariant in amino-acid adenyl­ation enzymes and belong to the specificity-conferring code (Stachelhaus et al., 1999 ▸; Challis et al., 2000 ▸). The other residues that are involved in substrate interaction are less conserved and are dependent on the substrate properties (for example polarity and size). The substrate-binding pocket of IdnL7 is surrounded by six nonpolar residues (Phe220, Phe256, Leu283, Ala285, Leu309 and Ile317) and two polar residues (Cys217 and Thr318). Among these residues, Cys217, Ala285, Phe256, Leu309, Ile317 and Thr318 are arranged in a shallow cavity and are involved in direct interactions. Ala285 is oriented towards the methyl group of l-Ala-SA and stabilizes the position of the l-alanyl moiety through hydrophobic interactions. Cys217 also makes contact with the l-Ala-SA side chain and presumably enforces the stereochemical recognition of l-amino acids via steric repulsion of the d-enantiomers (Supplementary Fig. S3). The orientation of l-Ala-SA is also controlled by Phe215, which is a residue from outside the specificity-conferring code. Phe215 adjusts the C^α position of the alanyl moiety by van der Waals interactions.

It should be noted that IdnL7 contains the relatively small Thr318 residue at the substrate-binding pocket, providing space in front of the methyl group of the alanyl moiety. This structural observation suggests that IdnL7 can also accommodate the side chains of other small l-amino-acid substrates, such as l-serine, which is consistent with the relatively relaxed substrate specificity of IdnL7.

3.4. Structural comparison with the other small amino-acid adenylation enzymes  

IdnL7 belongs to the VinM-type family of adenylation enzymes, which includes VinM (vicenistatin; Shinohara et al., 2011 ▸), FlvM (fluvirucin B2; Miyanaga, Hayakawa et al., 2016 ▸), CmiS3 (cremimycin; Amagai et al., 2013 ▸), HitE (hitachimycin; Kudo et al., 2015 ▸), Strop_2774 (salinilactam; Udwary et al., 2007 ▸), HerL (heronamide; Zhu et al., 2015 ▸), BecL (BE-14106; Jørgensen et al., 2009 ▸), LobM (lobosamide; Schulze et al., 2015 ▸) and MlaL (ML-449; Jørgensen et al., 2010 ▸). These enzymes share high sequence identity (56–83%) and have almost identical specificity-conferring codes (Fig. 4 ▸). The Cys217, Ala285 and Thr318 residues in IdnL7 are highly conserved among VinM-type enzymes. Although VinM shows almost the same substrate specificity as IdnL7, VinM prefers l-alanine rather than l-serine and glycine (Shinohara et al., 2011 ▸). Leu283 and Ile317 of IdnL7 are replaced by Phe290 and Val324 in VinM (Fig. 4 ▸), respectively, which might cause the slightly different substrate specificity.

Figure 4.

Sequence comparison of IdnL7 with other small amino-acid adenylation enzymes. The ten letters correspond to the specificity-conferring codes based on multiple sequence alignment. The sequences of the following adenylation enzymes were used for analysis: IdnL7, VinM, FlvM from Actinomadura fluva subsp. indica ATCC 53714 (Miyanaga, Hayakawa et al., 2016 ▸), CmiS3 from Streptomyces sp. MJ635-86F5 (Amagai et al., 2013 ▸), HitE from Streptomyces scabrisporus (Kudo et al., 2015 ▸), Strop_2774 from Salinispora tropica (Udwary et al., 2007 ▸), HerL from Streptomyces sp. SCSIO 03032 (Zhu et al., 2015 ▸), BecL from Streptomyces sp. DSM 21069 (Jørgensen et al., 2009 ▸), LobM from Micromonospora sp. (Schulze et al., 2015 ▸), MlaL from Streptomyces sp. MP39-85 (Jørgensen et al., 2010 ▸), PheA from Bacillus brevis (Conti et al., 1997 ▸), AlmE from Vibrio cholerae (Henderson et al., 2014 ▸), DltA from B. cereus (Osman et al., 2009 ▸), DltA from B. subtilis (Yonus et al., 2008 ▸), SfmA from S. lavendulae NRRL 11002 (Li et al., 2008 ▸), CssA from Tolypocladium inflatum (Weber et al., 1994 ▸), DptBC from S. roseosporus (Robbel & Marahiel, 2010 ▸), DhbF from B. subtilis (May et al., 2001 ▸), DptA from S. roseosporus (Robbel & Marahiel, 2010 ▸), SfmA from S. lavendulae NRRL 11002 (Li et al., 2008 ▸), EntF from E. coli (Reichert et al., 1992 ▸), CdaPS2 from S. coelicolor (Hojati et al., 2002 ▸), ArfB from Pseudomonas sp. MIS38 (Roongsawang et al., 2003 ▸), BlmIV from S. verticillus ATCC15003 (Du et al., 2000 ▸), LnmI from S. atroolivaceus S-140 (Tang et al., 2004 ▸) and BacA from B. subtilis (Eppelmann et al., 2001 ▸). Black boxes indicate key residues that are responsible for substrate accommodation in the VinM family of adenylation enzymes.

Unlike VinM-type enzymes, the adenylation enzymes DltA (Yonus et al., 2008 ▸; Du et al., 2008 ▸) and AlmE (Henderson et al., 2014 ▸), which participate in the neutralization of bacterial cell walls, are naturally selective for d-alanine and have no activity with l-alanine. DltA and AlmE possess a conserved cysteine residue at the position corresponding to Ala285 in IdnL7, which is located on the opposite side of the substrate-binding pocket to Cys217 (Supplementary Fig. S4). The position of the cysteine residue (Cys217) in the active site of IdnL7 might be important in determining the substrate conformation and specificity. Cys217 in IdnL7 is equivalent to leucine residues in DltA and AlmE. Other small amino-acid adenylation enzymes also possess larger and hydrophobic residues at this position, such as the aforementioned leucine, isoleucine and valine (Fig. 4 ▸).

NRPS adenylation domains that exhibit similar substrate profiles to IdnL7 generally possess a small residue at the position corresponding to Ala285 of IdnL7. For instance, the glycine-type adenylation enzyme DhbF_A1 (Tarry et al., 2017 ▸) from bacillibactin biosynthesis contains Gly724 at the corresponding position (Supplementary Fig. S4) and the serine-type adenylation enzyme EntF (Drake et al., 2016 ▸) from entero­bactin biosynthesis contains Ser722 (Supplementary Fig. S4).

The position corresponding to Thr318 in IdnL7 is diverse among small amino-acid adenylation enzymes (Fig. 4 ▸). Glycine-type and l-alanine-type adenylation enzymes have bulky tryptophan and tyrosine residues at this site, respectively. On the other hand, l-serine-type adenylation enzymes possess a smaller residue, such as alanine, at this position. The placement of a smaller residue at this position in IdnL7 might provide sufficient space to accommodate the side chain of l-serine.

4. Conclusions  

In this study, we conducted biochemical and structural analyses of IdnL7, a member of the VinM-type family of adenylation enzymes. The crystal structure of IdnL7 in complex with l-Ala-SA provides mechanistic insights into how IdnL7 accommodates small amino acids such as l-alanine in the substrate-binding pocket. VinM-type ligases have three conserved residues, cysteine, alanine and threonine, which seem to allow the incorporation of various small l-amino acids. Mutational analysis is under way to understand the substrate-recognition and amide-bond-forming mechanisms. At present, this study contributes to our understanding of the structure–function relationships of adenylation enzymes.

5. Related literature  

The following references are cited in the Supporting Information for this article: Anderson et al. (1964 ▸), May et al. (2005 ▸), Miller et al. (2016 ▸), Notredame et al. (2000 ▸), Robert & Gouet (2014 ▸) and Van de Vijver et al. (2008 ▸).

Supplementary Material

PDB reference: IdnL7, 6akd

Funding Statement

This work was funded by Japan Society for the Promotion of Science grant 25850050 to Akimasa Miyanaga. Ministry of Education, Culture, Sports, Science and Technology grant 16H06451 to Tadashi Eguchi.