Fig 1.
Schematic representation of the protein complementation assay utilized for directed evolution of SrtA.
A) Murine DHFR is cloned into pET-Duet as two independent fragments consisting of the carboxyl-terminal region fused to three amino-terminal glycines in one open reading frame and amino terminus fused to an LPETG sort-tag in a second open reading frame. Endogenous methionine aminopeptidase (MAP) cleaves the initiating methionine to expose the terminal glycines. B) SrtA enzymatically ligates the two fragments to generate an active DHFR. Endogenous bacterial DHFR is inhibited by the prokaryotic specific trimethoprim. C) Initial culture conditions testing the requirement for SrtA in the DHFR complementation assay. D) Overnight growth of bacteria containing the various assay components in the presence and absence of trimethoprim was monitored using optical density at 600nm, OD600. Each trial was completed in duplicate. Cells carrying either the pET vector expressing only the split DHFR, or the pRSF vector driving expression of SrtA fail to grow in the presence of trimethoprim. Alternatively, bacteria expressing a positive control mDHFR (mDHFR-(PC)) with the internal LPETGG sequence, grow robustly in the absence of SrtA, but in the presence of trimethoprim. Similarly, bacteria expressing the split mDHFR gene along with SrtA also grow robustly following an overnight growth in trimethoprim. E) Growth of bacteria on LB-Agar plates containing Ampicillin, Kanamycin, IPTG, and trimethoprim. I) and III) BL21(DE3) cells containing pET-Duet C/N-mDHFR with empty pRSF vector, or II) and IV) BL21(DE3) cells containing pET-Duet C/N-mDHFR with pRSF-SrtA.
Fig 2.
A) Basic selection protocol for directed enzyme evolution consisting of three cycles of transformation, growth, single clone picking, and sequencing. B) The frequency and location of mutations identified during rounds 1–3. Mutations that propagated through each round are highlighted in blue. C) The location of the five mutations that were isolated are shown in space-filling format in red, the β6- β7 loop is shown in blue, and the LPET peptide is shown in ball-and-stick format in orange.
Fig 3.
A) Schematic of the continuous fluorescence assay used to measure the initial acylation kinetics as previously described. B) Schematic of the fluorescence polarization assay used to follow the complete reaction coordinate. The peptide substrate is labeled at the amino terminus with the fluorescent label tetramethylrhodomain (TMR), which is efficiently excited at 557nm and emits at 576nm. In the absence of SrtA or secondary substrate (biotin labeled GGGGGDYK peptide), the fluorescence polarization is low. Alternatively, following the reaction of substrates in the presence of SrtA, avidin binding greatly increases the increases fluorescence polarization. C) Kinetic parameters for wild-type (WT) or SrtA variants.
Fig 4.
SrtA conformational changes along the reaction coordinate as depicted in the x-ray crystal structure of SrtA bound to LPETG (shown on the left) or the NMR model bound to LPAT* (shown on the right).
The image in the middle is a model computationally morphed between the two structures. The top panels are in wireframe representation with the key catalytic residues in ball-and-stick format in blue, the three glutamic acids involved in Ca2+ chelation in red, and peptide substrate in yellow. The bottom panel is the molecular surface representation of the structure colored according to sequence conservation (among all SrtA family members). Residues that are between 76–100% conserved are shown in red, 51–75% conserved in blue, 26–50% conserved in green, and 0–25% conserved in grey. The structure on the left represents initial substrate binding. Note that the substrate sits in a highly conserved (red) pocket. Following binding, the β6- β7 loops twists and lowers/closes while the β7- β8 loop opens ~10Å in a scissor motion. Collectively, these movements guide the substrate from the top of the enzyme toward the newly formed active-site pocket in the bottom while positioning the glutamic acid residues shown in red for optimum binding of Ca2+ and properly orienting the residues in blue for catalysis.
Fig 5.
SrtA is a dynamic protein that undergoes a large conformational change upon substrate binding.
A) Residues selected for mutagenesis for spin-label incorporation are labeled and their positions are shown in the NMR (top) or x-ray crystal structure (bottom) models. B) Distance distribution plots of SrtA variants with pairs of spin labels incorporated at the indicated positions in the absence (solid line) or presence of sort-tag substrate peptide (dashed line). The colored distributions represent the predicted distances obtained from PRONOX using the x-ray (blue; PDB: 1T2W) and NMR models (gray; PDB: 2KID-1). The background-corrected dipolar evolution data (gray dots) are shown for each pair of spin labeled mutant SrtA proteins (100 μM) as recorded on a Q-band Bruker ELEXSYS 580 spectrometer (Fig B in S1 File for the raw data); the black lines represent the fits to the data in the absence of substrate and the blue lines represent the fits to the data for the SrtA protein in the presence of 10x Abz-CLEPTGG.