Bacterial strains, plasmids, primers and growth conditions
Bacterial strains and plasmids utilized in terms of the investigations are summarized in TABLE S 1. Escherichia coli strains were cultivated in LB or M9 medium at 37°C. Antibiotics were added to the cultures, when required, with appropriate concentrations.
Overexpression And Purification Of Gqqa
The gene encoding GqqA, isolated from Komagataeibacter europeaus, was integrated in pET21a with a C-terminal TEV-protease cleavage site (ENLYFQS) and a His6-tag, as described before by 1. GqqA was produced by gene overexpression in E. coli strain BL21 (DE3). The culture was grown in LB medium with ampicillin (final concentration 100 µg ml− 1) inoculated with 25 ml pre-culture at 37°C. When the cells reached the exponential growth phase (OD600nm), protein expression was induced by supplementation of 1 mM IPTG (isopropyl-β-D-1-thiogalactopyranoside). After 3 h of incubation the cells were harvested by centrifugation applying 4,000g at 277 K for 30 min. The pellet was resuspended in lysis buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8.0). The cells were disrupted by 5×1 min sonification and centrifuged with 17000g at 277 K for 1 h. The supernatant containing the His6-tagged target protein was incubated on a Ni-NTA agarose column for 1 h on a rotary shaker at 277 K. The sample was washed twice with wash buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, pH 8.0) and eluted with 5 ml elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, pH 8.0). The sample was concentrated by centrifugation applying amicon filters and loaded onto a size-exclusion chromatography column (Superose6Increase, GE Healthcare) in a buffer consisting of 0.1 M K2HPO4, pH 5.0, 150 mM NaCl. Fractions containing the pure protein were pooled and concentrated to 10 mg ml− 1. The purity of the protein was verified by Coomassie stained SDS-PAGE (FIGURE S1 A), which shows a dominant band at approximately 30 kDa, consistent with monomeric GqqA (30.5 kDa).
To solve the crystallographic phase problem the 8 methionines of GqqA were substituted by selenomethionines 40. The preculture was centrifuged with 17,000g at 4°C for 20 min. and resuspended in 5 ml M9-medium. The culture was grown in M9-medium with Ampicillin (final concentration 100 µg ml1) inoculated with 1.25 ml resuspended pellet at 37°C. When the cells reached the exponential growth phase (OD600nm), they were supplemented with 20 ml selenomethionine (SeMet) (1 g l− 1) and 5 ml of an amino acid mixture (TABLE S2). After 15 minutes protein expression was induced by supplementation of 1 mM Isopropyl-β-D-thiogalactopyranoside (IPTG) and after 3 h of incubation at 37°C the cells were harvested by centrifugation with 4,000g at 4°C for 30 min. The selenomethionine-variant of GqqA (SeGqqA) was purified following the same procedures as applied for GqqA.
Esi-ms/ms Analysis
For enzyme assays 3-oxo-octanoyl homoserine lactone (oxo-C8-HSL) as substrate was used in 5 mM N-2-hydroxyethylpiperazine-N’-3-propanesulfonic acid (EPPS) buffer, pH 8.0. Prior to use a 100 mg/ml stock solution of oxo-C8-HSL in dimethyl sulfoxide (DMSO) was prepared. If otherwise not indicated, reactions were performed, in triplicates and corrected for non-enzymatic transformation, at − 70° and in 5 mM EPPS buffer, pH 8.0. The assays were conducted according to the following steps. One milliliter of 5 mM EPPS buffer, pH 8.0, was introduced into a 1.5 ml Eppendorf tube that was closed tightly. The substrate oxo-C8-HSL was then added to a final concentration of 0.5 mg/ml (from a 100 mg/ml stock solution in DMSO). Finally, the protein was immediately added to a final concentration of 10 µg/ml (from a stock solution of 5 mg/ml in 40 mM HEPES, pH 7.0). Aliquots were taken at different time intervals, and the reaction products were analyzed by Mass Spectrometry (MS) experiments, performed in a hybrid quadrupole-time of flight mass spectrometer (QTOF, QSTAR pulsar i, ABSciex) equipped with a micro electrospray ion source (in positive and negative mode). Samples were dissolved in methanol (0.1 ml sample plus 0.9 ml methanol) and introduced in the spectrometer using a syringe infusion pump with a 10 ul/min flow. N2 was employed in the collision cell. MS experiments in TOFMS mode were registered to identify the reaction products. The experiments were performed with positive ion detection. The instrumental parameters were set as follows: mass range 50-2000 Dalton; ion spray voltage (IS): 4500 V; ion source gas pressure (GS1): 10 psi; curtain gas pressure (Cur): 20 psi; declustering potential (DP): 30 V; focusing potential (FP): 210 V; declustering potential 2 (DP2): 15 V; collision gas: 3. Substrates and products used and generated are listed in TABLE S3 together with their molecular mass.
Dynamic Light Scattering measurements
Prior to SAXS and crystallization experiments the homogeneity of GqqA and SeGqqA in solution (3.6 mg ml− 1 in 0.1 M K2HPO4 pH 5.0, 150 mM NaCl) was verified performing dynamic light scattering (DLS) measurements at 20°C. 15 µl of purified GqqA suspension were measured in Helma cuvettes applying the SpectroSize 300 DLS system (Xtal Concepts, Germany).
Small Angle X-ray Scattering
Small angle X-ray scattering data for GqqA were collected at 20°C at beamline P12 (PETRA III, EMBL, Hamburg, Germany). Three different protein concentrations (4, 2 and 0.8 mg ml− 1) were measured applying an automated robotic sample changer and a Dectris 2D photon-counting detector (PILATUS-6M) with 3.1 m sample to detector distance. Scattering data for all three concentrations were collected, integrated, and averaged applying the SasTool software (EMBL, Hamburg, Germany) (Franke 2017. Guinier analysis and radius of gyration (RG) values were calculated using PRIMUS 41. The pair distribution functions P(r) and forward scattering intensities I(0) were processed with GNOM 42 and PRIMUS. Data collection statistics and results are summarized in TABLE S4.
Crystallization
Initial screening of GqqA and SeGqqA was performed applying the sitting-drop vapor-diffusion method and 96-well plates utilizing a Honeybee crystallization robot (Genomic Solutions, USA) and using the commercial kit JCSG-plus (Molecular Dimensions, UK). Thin GqqA crystals were identified after 3 days with 0.8 M succinic acid at pH 7.0 as reservoir solution. To obtain larger and X-ray suitable crystals streak seeding was applied to seed a droplet of 2 µl protein solution (10 mg ml− 1) and 2 µl reservoir solution using 24-well linbro plates (FIGURE S1, C and D). Seed suspensions were prepared according to the protocol of HAMPTON Research (HAMPTON Research, US). SeGqqA crystals were obtained in 10 % PEG3350. Final crystallization conditions are summarized in TABLE S5.
GqqA diffraction data collection, structure solution, and refinement
For native GqqA and SeGqqA X-ray diffraction data were collected at 100 K at beamline P11 (PETRA III, DESY, Hamburg). The diffraction data were indexed, integrated and scaled using the XDS software package 43.
Initial structure solution was possible applying diffraction data of a single SeMet-labelled GqqA (SeGqqA) crystal collected at a wavelength of 0.98 Å. This crystal diffracted up to 1.7 Å resolution and diffraction data were used for single-wavelength anomalous phasing (SAD), performed using the Autorickshaw-Server 44. Based on eight located Se-positions phasing was possible and initial electron densities allowed model building of SeGqqA applying the program Coot 45. Subsequent refinement was performed utilizing the program Refmac5 46. Interestingly this structure contained only two of the three GqqA protein domains.
Diffraction data to 2.5 Å resolution were collected for native GqqA at a wavelength of 1.03 Å. The obtained SeGqqA structure was used for molecular replacement to obtain initial phases of the native GqqA structure using the program Phaser 47. The native structure was subsequently completed by building amino-acid residues 85–170 of the missing domain into the 2Fo-Fc electron density maps. The following structure refinement was performed using the program Phenix 48. In later stages of the refinement the Phenylalanine co-factor and solvent molecules were introduced and refined. Data collection statistics and refinement results are summarized in Table 1. The refined GqqA and SeGqqA structures were deposited in the Protein Data Bank with pdb codes 7AM0 and 7ALZ.
Table 1
X-ray diffraction data collection, processing and refinement statistics.
|
SeGqqA
|
GqqA
|
Diffraction source
|
P11 at PETRA III (DESY), Hamburg
|
P11 at PETRA III (DESY), Hamburg
|
Wavelength (Å)
|
0.9801
|
1.0332
|
Detector
|
Pilatus
|
Pilatus
|
Crystal-detector distance (mm)
|
270
|
479
|
Rotation range per image (°)
|
0.05
|
0.1
|
Total rotation range (°)
|
360
|
360
|
Exposure time per image (s)
|
0.1
|
0.1
|
Space group
|
P212121
|
P21
|
Unit cell parameters (Å, °)
|
a = 60.44, b = 63.73, c = 97.34,
α = β = γ = 90
|
a = 76.87, b = 65.76, c = 102.66, β = 104.27
|
Resolution range (Å)
|
50.0-1.67 (1.78–1.67)
|
49.3–2.5 (2.6–2.5)
|
Total No. of reflections
|
545065 (85204)
|
234394 (23288)
|
No. of unique reflections
|
79955 (12767)
|
34461 (3364)
|
Completeness (%)
|
99.7 (98.0)
|
99.44 (98.53)
|
Multiplicity
|
6.82 (6.67)
|
6.8 (6.9)
|
I/σ(I)
|
21.9 (3.36)
|
29.35 (3.90)
|
Rmeas.
|
6.0 (60.9)
|
4.4 (52.4)
|
CC 1/2
|
0.99 (0.91)
|
1 (0.97)
|
Wilson B-factor (Å2)
|
19.2
|
57.4
|
refinement
|
|
|
Rcryst / Rfree
|
0.16 / 0.19
|
0.22 / 0.27
|
RMS bonds (Å) / angles (°)
|
0.022/ 2.04
|
0.006/1.07
|
Ramachandran favored (%)
|
98.9
|
96.5
|
Ramachandran allowed (%)
|
1.1
|
3.2
|
Ramachandran outliers (%)
|
none
|
0.3
|
Average B-factors (Å2)
|
22.0
|
71.1
|
Values for the outer resolution shell are given in parentheses. |
In Silico Ligand Docking
Docking approaches were performed with the freely available docking software SwissDock 49 50. The software searches possible target cavities in proteins by calculating the binding energies of small molecules. N-octanoyl-L-homoserine lactone was chosen as putative ligand due to previous experiments 1 and provided in mol2 format. The GqqA dimer was used without solvent molecules but with bound L-Phe in the regulatory domain. The best hits are clustered and are visualized in FIGURE 8A.
Construction Of Gqqa Mutants
For the exchange of individual amino acids in GqqA, site-directed mutagenesis was used. For this purpose, the mutations were carried out using the phusion DNA polymerase, the construct pET21a::gqqA as template and the following mutation-specific primers:
GqqA_M1_for (5’ CGAACTTTCGTCCTTTTCGGAACAGCAGGAGATC 3’), GqqA_M1_rev (5’ TCCAGCGCACGCGCCAGT 3’), GqqA_M2_for (5’ CTCGAGCACCACCACCACC 3’), GqqA_M2_rev (5’ CCGGAAGGGCGATGCGGG 3’), GqqA_M3_for (5’ CATGCAAGGTCCGAATAGGCCCCGGCAGGCCTCGCCCGGCTGGA 3’), GqqA_M3_rev (5’ CATGCAAGGTCCGAATAGGCCC 3’), GqqA_M5_for (5’ CGGCATCATTGTCGAACTGGGACTGGACCCAG 3’), GqqA_M5_rev (5’ CGCACCTGCGCCATGGCG 3’). The GqqA_Mut4 construct was generated by using a synthesized gene fragment containing the substitution and deletion side. To remove the template DNA, the PCR product was digested with DpnI (30 min 37°C, 10 min heat inactivation at 75°C). After digestion, a phosphate group was added to the amplicon polynucleotide phosphokinase using the manufacturer's protocol (Thermo Scientific, Bremen, Germany). With the added phosphate group, the linear amplicon could be ligated with the T4 ligase by adding 1 µL 10 mM ATP. The ligation was carried out at alternating between 10°C and 30°C for 60 minutes. A heat inactivation step was followed by heat shock transformation into E. coli DH5α. Positive clones were verified by sequencing and transformed into the overexpression strain E. coli BL21(DE3).
In Vivo Qq Enzyme Assays
For a functional AHL-degrading assay, the reporter strain C. violaceum CV026 was used as previously described and with minor modifications 1 54. For this purpose, E. coli BL21 (DE3) strains containing the pET-21a::gqqA wt or the pET21a::gqqA M1-M5 mutant constructs were grown in a preculture. Fresh LB medium was inoculated with 1 % (vol/vol) of the precultures and cultured at 37°C until an OD600 of 0.6 was reached. Following this cells were induced with 0.1 mM IPTG and proteins were expressed for 16 h. After this fresh LB medium containing ampicillin and 10 µM oxo-C8-HSL was inoculated with the expression strains at a ratio of 2 % (vol/vol) and incubated at 37°C for 6 h. The culture supernatant was collected by centrifugation and 30 µL was used for bioassay. For this, a grown CV026 culture was mixed with 24-fold volume of LB agar and poured into plates. Sterile filter papers were placed on the plates. The CV026 plates with the supernatants of the cultures were incubated for 24 h at 28°C. If CV026 cells turned purple, AI molecules were present; and if CV026 cells remain colourless, the AI molecules were degraded.
β-lactamase Activity Assay
The β-lactamase activity of GqqA was determined via a disc diffusion antibiotic susceptibility test. Ampicillin 10 µg (AMP), Amoxicillin 10 µg (AML), Cefotaxime 30 µg (CTX 30) and Penicillin G 10 µg (P) sensitivity discs (Thermo Fischer Scientific, Waltham, MA, USA) were incubated for 30 min at 28°C with 30 µL of 0.1 M potassium phosphate buffer pH 8 containing 40 µg of GqqA. A control without enzyme was included. The antibiotic susceptibility test was performed on LB agar plates with S. aureus cells. After overnight incubation at 37°C the halo formation indicates a degradation of the antibiotic.
Complementation assays with an auxotrophic strain of Escherichia coli
A complementation test was performed with the phenylalanine auxotrophic E. coli strain JW2580-1. The strain was obtained from the Coli Genetic Stock Center (http:// www.cgsc.biology.yale.edu) and carries an in-frame, single-gene knockout from the Keio Collection 51. JW2580-1 was grown in LB medium overnight at 37°C. Phenylalanine auxotrophy was verified to the complementation tests by growth on minimal medium M9 supplemented with and without L-phenylalanine (2 µM) (Sigma-Aldrich, Heidelberg, Germany). gqqA was amplified using the primers gqqA_comp_f (5’ ATGAACGGGGAACGCATCATCGC 3’) and gqqA comp_rev (5’ TCAGGGTTTGCGCCGGAG 3’) and the fragment was cloned into the pDrive vector. The resulting pDrive::gqqA vector was sequenced to verify the correctness of the sequence and was then transformed into JW2580-1 by electroporation. The transformed strain as well as the auxotrophic strain were grown in LB medium overnight at 37°C and at 120 rpm; then, 5 ml of each culture was centrifuged, and cells were washed three times with M9 medium. 5 ml M9 medium with or without phenylalanine (2 µM) was inoculated with 50 µl of the washed cells and incubated up to 72h at 37°C.