The C-terminal residue of phage Vp16 PDF, the smallest peptide deformylase, acts as an offset element locking the active conformation

Prokaryotic proteins must be deformylated before the removal of their first methionine. Peptide deformylase (PDF) is indispensable and guarantees this mechanism. Recent metagenomics studies revealed new idiosyncratic PDF forms as the most abundant family of viral sequences. Little is known regarding these viral PDFs, including the capacity of the corresponding encoded proteins to ensure deformylase activity. We provide here the first evidence that viral PDFs, including the shortest PDF identified to date, Vp16 PDF, display deformylase activity in vivo, despite the absence of the key ribosome-interacting C-terminal region. Moreover, characterization of phage Vp16 PDF underscores unexpected structural and molecular features with the C-terminal Isoleucine residue significantly contributing to deformylase activity both in vitro and in vivo. This residue fully compensates for the absence of the usual long C-domain. Taken together, these data elucidate an unexpected mechanism of enzyme natural evolution and adaptation within viral sequences.

Vp16 PDF(KLF)helices DNA sequence was optimized, synthesized and cloned into pBAD/Myc-HisA plasmid (Invitrogen) using NcoI and XhoI cloning sites by GeneArt Company. The other Vp16 PDF chimeras were obtained by site-directed mutagenesis of Vp16 PDF(KLF)helices DNA sequence into pBADMyc-HisA plasmid. All corresponding Vp16 PDF chimeras were sub-cloned in pET-16b plasmid (Novagen) for protein purification.
Primers used are described in Table S3. Proteins expressed from final constructs resulted in amino acids sequences described in Table S4. Purification of chimeras has been performed as the corresponding wild type protein.

Enzymology
PDF enzymatic activity was measured by a spectrophotometric assay. The formate released by the deformylation reaction is used by FDH to convert NAD in NADH, the production of which being measured over time at 340 nm and controlled temperature (37°C). The 200µL reaction mixture contained 50 mM HEPES-NaOH pH 7.5, 12 mM NAD + , 4.5 U/mL FDH, 1 mM NiCl 2 and PDF enzyme (either 100 nM Vp16 PDF or 10 nM E. coli PDF) previously diluted in 50 mM MES-KOH pH 4.0, 80 mM NiCl 2 for Vp16 PDF or 50 mM HEPES-NaOH pH 7.5, 0.1 mg/mL BSA for E. coli PDF. The reaction was started by addition of 1-6 mM Fopeptides as indicated in the legend of Figure. Kinetics parameters (k cat , K m ) were derived from iterative non-linear least square fits using the Michaelis-Menten equation based on the experimental data (Sigma-Plot, Kinetics module).
Vp16 PDF inhibition was performed as previously described 4, 5 .

Immunological Methods
Rabbit polyclonal antibodies raised against full length E. coli MetAP and E. coli PDF were used at 1:5000 dilution for Western blot analysis as previously described 6 .

N-terminal proteomic analysis
The PAL421Tr was transformed at 30°C with pBAD plasmid expressing Vp16 PDF or E. coli PDF in the presence of ampicillin. Next day 10 mL cultures were inoculated with bacterial colony from the plate and grown for 24h in the presence of ampicillin and 0.2% arabinose at 37°C. After 24h, cultures were diluted to the OD600=0.05 and cultured at 30°C in the presence of 0.1% of arabinose for cells expressing E. coli PDF and 0.2% for Vp16 PDF to equalize the level of protein expression. Cells (4.6 x 10 10 ) were harvested and collected by centrifugation (4°C, 3300g, 30 min) at time points 2, 6, 8 and 24h. Cells were disrupted by grinding (MM300 grinder, Qiagen) in 50 mM Hepes-NaOH pH 8, 25 mM Ascorbic Acid, 25 mM L-Cystein and one tablet of antiprotease inhibitor cocktail (Roche). After centrifugation at 21,000 g for 30 min at 4°C, protein concentration in the supernatant was determined by Bradford protein assay.
For N-termini enrichment samples were prepared as previously described (Bienvenut et al., 2012). 1 mg of protein was denatured and reduced followed by cysteine alkylation with iodoacetamide. After cold acetone precipitation, proteins were resuspended in 50 mM NH 4 HCO 3 and digested by 1/100 (w/w) of TPCK treated porcine trypsin (Sigma-Aldrich) for 1.5h at 37°C, twice. Peptides were desalted with Sep-Pak columns and the retained material was eluted with 80% acetonitrile (ACN), 0.1% TFA and then evaporated to dryness. The collected material was resuspended in Strong Cation eXchange (SCX) LC buffer (5mM KH 2 PO 4 , 30% ACN and 0.05% formic acid) and injected into Alliance HPLC system using a to 22 min were analyzed as previously described 7 with an Easy Nano-LC II (Thermo Scientific) coupled to a LTQ-Orbitrap™ Velos (Thermo Scientific).

Crystallography: data collection and processing
Diffraction data were collected on single crystals at 100 K on FIP-BM30A and PROXIMA1 beamlines at ESRF (Grenoble, France) and SOLEIL (Gif-sur-Yvette, France), respectively.
Data were processed and scaled with XDS package 8 . Statistics are summarized in Table S2.
The crystal structure of Vp16 PDF crystallized in form I was solved by molecular replacement with PHASER 9 using a C-terminally truncated version of the Pseudomonas aeruginosa PDF (PDB code 1LRY 10 ) as the starting model. Structures of protein crystallized in form II was solved by rigid-body refinement in REFMAC 11 . Manual model building and refinement were done with TURBO-FRODO 12 and REFMAC, 11 respectively, yielded the final models that were validated by PROCHECK 13 . Refinement statistics are detailed in Table S2. Figures were generated using PyMOL (http://www.pymol.org).
Inspection of the density maps revealed the presence of several high electron density spots in both models, which could reasonably be interpreted as metal ions. In order to identify the nature of these ions, we performed X-ray fluorescence experiments on the crystal form II, which revealed the presence of two different metal ions, zinc and nickel. Anomalous difference maps calculated for each metal allowed us to locate zinc ions within the active site and nickel ions at the surface of the protein.

Figure S3
Comassie brilliant blue gel