Strains, media, and chemicals
E. coli NEB 10-beta (C3019I, New England BioLabs, MA) was used for all routine cloning. E. coli Marionette-Clo53 was used for all selection experiments. E. coli Marionette-X, a Marionette-compatible derivative of NEB Express (C2523I, New England BioLabs, MA) was used for large-scale peptide expression experiments. TB media (T0311, Teknova, CA) supplemented with 0.4% glycerol (BDH1172-4LP, VWR, OH, USA) was used for peptide expression and modification. 2xYT liquid media (B244020, BD, NJ) and 2xYT + 2% agar (B214010, BD, NJ) plates were used for routine cloning and strain maintenance. SOB liquid media (S0210, Teknova, CA) was used for making competent cells. SOC liquid media (B9020S, New England BioLabs, MA) was used for outgrowth. Unless noted otherwise, cells were induced with the following chemicals: cuminic acid (268402, Millipore Sigma, MO) added as 1000× stock (200 mM) in EtOH or DMSO; 3-oxohexanoyl-homoserine lactone (3OC6-AHL) (K3007, Millipore Sigma, MO) added as a 1000× stock (1 mM) in DMSO; anhydrotetracycline (aTc) (37919, Millipore Sigma, MO) added as a 1000× stock (100 µM) in DMSO; isopropyl β-D-1-thiogalactopyranoside (IPTG) (I2481C, Gold Biotechnology, MO) added as 1000× stock (1 M) in water. Cells were selected with the following antibiotics: carbenicillin (Carb, C-103-5, Gold Biotechnology, MO) added as 1000× stock (100 mg/mL in H2O); kanamycin (Kan, K-120-10, Gold Biotechnology, MO) as 1000× stock (50 mg/mL in H2O); spectinomycin (Spec, S-140-5, Gold Biotechnology, MO); and chloramphenicol (Cm, C-105-5, Gold Biotechnology, MO). Liquid chromatography was performed with Optima Acetonitrile (A996-4, Thermo Fisher Scientific, MA) and water (Milli-Q Advantage A10, Millipore Sigma, MO) supplemented with LC-MS Grade Formic Acid (85178, Thermo Fisher Scientific). The following solvents/chemicals were also used: Ethanol (V1001, Decon Labs, PA), Methanol (3016-16, Avantor, PA), Ammonium bicarbonate (A6141 Millipore Sigma, MO), dimethyl sulfoxide (DMSO) (32434, Alfa Aesar, MA), Imidazole (IX0005, Millipore Sigma, MO), sodium chloride (X190, VWR, OH), sodium phosphate monobasic monohydrate (20233, USB Corporation, OH), sodium phosphate dibasic anhydrous (204855000, NJ), guanidine hydrochloride (50950, Millipore Sigma, MO), tris (75825, Affymetrix, OH), TCEP (51805-45-9, Gold Biotechnology, MO), and EDTA (0.5 M stock, 15694, USB Corporation, OH). DNA oligos and gBlocks were ordered from Integrated DNA Technologies (IDT) (CA).
Plasmids and genes
Plasmids pTHSS-1282, pDAA558, and pAMK-267 were constructed from the parental pTHSSe-44 backbone, which has a pSC101 origin variant (var 2) and ampicillin resistance80. Plasmids pTHSS-1282, pDAA-558, and pAMK-267 contain a flexible linker sequence (GSSRGGKGGPGGRGGVGGGGGIGG) between the peptide/GFP and NpuC regions. Plasmids pAMK-925, pTHSS-2132, pAMK-866, and pAMK-870 were constructed from the parental pTHSS-1458 backbone, which has a colE1* origin variant and a kanamycin resistance marker80. The plasmid pEG06_044 contains the PapB enzyme, a p15A origin of replication and spectinomycin resistance. The parental backbone pTHSS-2012, which has a p15a origin and spectinomycin resistance was used for additional cloning experiments80. The plasmids pAMK-926 and pTHSS-2137 that contain the PLux promoter expressing NpuC-σC and PMI-NpuC-σC, respectively, were constructed from pTHSS-2012. The plasmids pAMK-925 and pTHSS-2132 that contain the PTac promoter expressing σN-NpuN and residues 17–124 of Mdm2 (Mdm2*)-σN-NpuN, respectively, were constructed from pTHSS-1458. The plasmid pAMK-870 that contains the constitutive PJ23105 promoter expressing Mdm2*-σN-NpuN and the P20_992 promoter expressing CAT-GFP was constructed from pTHSS-1458. The plasmid pAMK-866 that contains the constitutive PJ23105 promoter expressing 328-533 of the SARS-CoV-2 Spike protein (RBD)-σN-NpuN and the P20_992 promoter expressing CAT-GFP was constructed from pTHSS-1458. The peptide cloning plasmid pAMK-267, constructed from pTHSSe-44, contains the PLux promoter upstream of an RBS-His tag-SapI-GFP-SapI-NpuC-σC where the gfp gene can be replaced by a peptide gene through Type IIs assembly methods using the enzyme SapI (NEB). The RBS from pAMK-267 was chosen from a library of RBS variants upstream of a His tag-PMI-NpuC-σC that was tuned for co-expression with constructs containing the PJ23105 promoter expressing Mdm2*-σN-NpuN. The N-terminal His tag in pAMK-267 was left in place to provide a constant 11 aa for consistent levels of expression between different peptide sequences. The plasmid pAMK-670 that contains the PLux promoter expressing PMI-NpuC-σC was constructed from pAMK-267. The plasmid pAMK-857 that contains the PLux promoter expressing N-terminal residues 19-56 of ACE2 (ACE2α1)-NpuC-σC was constructed from pAMK-267. The plasmid pAMK-917 that contains the PLux promoter expressing the hit sequence fused to NpuC-σC was constructed from pAMK-267. Note that the pTHSSe-44 and pTHSS-1458 backbones have origin variants that alter their copy numbers, making them approximately equivalent to a p15a backbone. Genes encoding Npu intein, PMI, Mdm2*, ACE2α1, and RBD were synthesized as gBlocks. The ECF20_992 gene was sourced from a previous publication48.
Cytometry analysis
Fluorescence characterization was performed on a BD LSR Fortessa flow cytometer with the HTS attachment and BD FACSDiva software version 8.0.3 (BD, NJ). Samples were prepared by diluting overnight cultures 1:400 by adding 0.5 μl of cell culture into 200 μl of PBS containing 1 mg/mL Kan. Fluorescence measurements were made using the blue (488 nm) laser and all data was derived from the FITC-A channel (PMT voltage of 400 V). At least 30,000 events were collected for each sample, without gating, and the Cytoflow Python package (v 1.1.1) was used for downstream analysis. When presented, the median value is used.
Evaluation of the split-intein σ factor circuit
Strains of E. coli Marionette Clo harboring a combination of plasmids pTHSS-1282, pTHSS-2132, and pTHSS-2137 or pTHSS-1282, pAMK-925, and pAMK-926 were used for assessing intein splicing with or without PMI-Mdm2* induced association, respectively. Strains were grown in 1 mL of LB media + antibiotics for 20 h in a deep well 96-well plate (1896–2000, USA Scientific, FL) at 30 °C, 900 rpm in an Infors HT Multitron Pro (Infors USA, MD). Cultures were then diluted 1:100 into fresh 1 mL of LB media + antibiotics and serial 1:10 dilutions of inducers (IPTG, 10–3–103 µM; 3OC6-AHL, 10–3–103 nM) for 20 h in a deep well 96-well plate at 30 °C, 900 rpm in the Multitron Pro. 0.5 μl of saturated cell culture were then diluted into 200 μl of PBS containing 1 mg/mL Kan for cytometry analysis.
PMI variant induction profiling
PMI variants without any N-terminal sequence additions were appended to the linker-NpuC-σC and inserted into the pDAA558 backbone using standard cloning methods. These plasmids were then co-transformed into a Marionette Clo strain alongside either pAMK-878 (off-target, Spike RBD) or pAMK-877 (on-target, Mdm2*) and then grown overnight in 1 mL LB media + Kan/Carb at 30 °C, 900 rpm in a deep well 96-well plate. The next day, cultures were diluted 1:100 into 1 mL TB media + Kan/Carb containing either 0 or 1 µM 3OC6-AHL. After overnight growth in the same growth conditions, the variants were assayed as described in Cytometry Analysis.
Two-hybrid assay for RBD/Mdm2* association
To assay for protein–protein mediated splicing the following plasmid combinations were transformed into E. coli Marionette Clo and fluorescence was measured via cytometry: pAMK-866/pAMK-670 (RBD/PMI); pAMK-866/pAMK-857 (RBD/ ACE2α1); pAMK-870/pAMK-670 (Mdm2*/PMI); pAMK-870/pAMK-857 (Mdm2*/ACE2α1). Strains were grown in 1 mL of LB media + antibiotics for 20 h in a deep well 96-well plate at 30 °C, 900 rpm in a Multitron Pro. Cultures were then diluted 1:100 into fresh 1 mL of LB media + antibiotics + 1 µM 3OC6-AHL (full induction of peptide plasmid) for 20 h in a deep well 96-well plate at 30 °C, 900 rpm in the Multitron Pro. 0.5 μl of saturated cell culture were then diluted into 200 μl of PBS containing 1 mg/mL Kan for cytometry analysis.
Library generation
The Pap library was designed with diversity at the ends and middle of the peptide and included either glutamate or aspartate as a cyclization partner, for a final sequence design of XCXXX[D/E]XCXXX[D/E]X. Using the degenerate nucleotide sequences NNK for any amino acid and GAW for aspartate or glutamate, we generated a library of 1012 peptides encoded by 1014 unique codon sequences. The library of plasmids lbAMK-103, which contains the PLux promoter expressing the Pap library-NpuC-σC was constructed from pAMK-267. The pap library was amplified from pEG03_283 using degenerate oligonucleotides oAMK-915/916 (IDT). Gel purification was used to isolate the 124 bp amplicon, which was then cloned into pAMK-267 using the type IIS restriction enzyme SapI (NEB).
Linear insert and plasmid were mixed at a 1:1 molar ratio (200 fmol each) along with 10 μl 10× DNA ligase buffer, 2 μl T4 DNA ligase (HC) (20 U/μl) (M1794, Promega, WI) and 4 μl SapI in 100 μl total volume. Reactions were cycled 25 times for 2 min at 37 °C and 5 min at 16 °C then incubated for 30 min at 50 °C, 30 min at 37 °C, and 10 min at 80 °C in a DNA Engine cycler (Bio-Rad, CA). An additional 2 μl SapI was then added, and the assembly was incubated for 1 h at 37 °C. Assemblies were then purified using Zymo Spin I columns (Zymo Research, CA). Library assemblies were initially transformed into electrocompetent E. coli NEB 10β (C3020K, NEB, MA). We observed 1.5 × 107 colony forming units (CFU)/mL for lbAMK-103. Total transformants were estimated by colony counting after 107-fold dilution and plating of liquid outgrowths on selective media.
Calculation of the modified fraction of the library
The initial, unselected papA library was transformed and plated to resolve individual colonies. A set of 19 random colonies were picked and sequenced via colony PCR. Of the 19 sequenced colonies, 18 were properly assembled. These 18 library members were then assessed for post-translational modification via LC-MS. Plasmids were transformed into either E. coli NEB Express or E. coli Marionette-X using 30 μL of competent cells and 1 μL of each plasmid being transformed in a PCR strip tubes (1402-4700, USA Scientific, FL or 951020401, Eppendorf, NY). Transformations were incubated on ice (20–30 min), heat shocked (42 °C, 30 s), and incubated on ice again (5 min). Cells were then transferred to a deep well 96-well plate (1896–2000, USA Scientific, FL) with 120 μL of SOC media. After outgrowth (Multitron Pro, 1 h, 30 °C) in an Infors HT Multitron Pro (Infors USA, MD), 900 μL LB was added with appropriate antibiotics (at 1.1× for 1× final concentration) and incubated (Multitron Pro, 30 °C, 900 r.p.m.) until all wells reached saturation (12–30 h). Overnight cultures were diluted 1:100 into 1 ml TB media with Carb/Kan in deep well plates. After 3 h incubation (Multitron Pro, 30 °C, 900 r.p.m.), inducer was added (1 mM IPTG, 100 μl cumate), and cultures were incubated for 20 h (Multitron Pro, 30 °C, 900 r.p.m.). To purify the peptides, the 96-well plates were centrifuged (Legend XFR, 4,500 g, 4 °C, 20 min), pellets were resuspended in 600 μL lysis buffer (5 M guanidinium hydrochloride, 50 mM sodium phosphate, pH 8), and freeze-thawed (frozen at −80 °C; thawed in Multitron Pro at 37 °C, 900 r.p.m). Cell lysates were centrifuged (Legend XFR, 4,500 g, 4 °C, 60 min) and peptides affinity purified using His MultiTrap TALON plates (29-0005-96, GE Life Sciences (now Cytiva), MA), following manufacturer instructions, using 1 × 600 μL water and 2 × 600 μL lysis buffer for column equilibration, 4 × 600 μL wash buffer (50 mM Tris, 50 mM NaCl, 5 mM Imidazole pH 8), and 1 × 200 μL elution buffer (50 mM Tris, 50 mM NaCl, 150 mM Imidazole pH 8). TCEP (2 mM) was added to the peptides, they were digested with TEV protease (0.1 mg/mL) at 4 °C overnight (~16) and the samples were analyzed using the QTOF (see below). The 14 modified library sequences were then aligned and WebLogos generated (https://weblogo.berkeley.edu/logo.cgi) with default parameters, except without small sample correction.
Selection of Pap library lbAMK-103
Assembled library of plasmids lbAMK-103 was transformed into an electrocompetent Marionette Clo strain harboring the PapB modifying enzyme plasmid, pEG06_044, and the selection plasmid, pAMK-866 (all non-assembly transformation steps were > 1 × 108 efficiency). A 1 mL of liquid outgrowth of library transformants was diluted 1:50 in TB media + Carb/Kan/Spec + 1 µM 3OC6-AHL and 100 µM cumate to induce peptide + modifying enzyme, and grown at 30 °C, 250 r.p.m. for 20 h. For the first round of selection, cultures were then diluted 1:100 in TB Carb/Kan/Spec + 1 µM 3OC6-AHL and 100 µM cumate + 300 µM Cm and grown at 30 °C, 250 r.p.m. for at least 20 h (until cultures were saturated). A 0.5 µL aliquot was taken for cytometry analysis and 2 mL of culture was also taken to harvest plasmid. A 5 µL sample of purified plasmid was stored for NGS analysis and the rest was digested with 1 µL SapI (NEB) for 1 h at 37 °C to remove the background pEG06_044/pAMK-866 plasmid. The selected lbAMK-103 plasmid was then re-transformed into the strain of electrocompetent E. coli Marionette Clo strain harboring the PapB modifying enzyme, pEG06_046, and the selection plasmid, pAMK-866. The selection process was repeated once more with a Cm concentration of 800 µM and then once more with a Cm concentration of 1200 µM.
NGS analysis
Library construction for NGS was performed using the protocol for KAPA Hyper Prep Kits with PCR Library Amplification/Illumina series (KK8504, Roche). First, miniprepped library plasmids were amplified with Q5 polymerase (#M0492L, New England BioLabs, MA) with the primers oAMK-946/947 (Pap library) and oAMK-997/998 (Tgn/Lyn library). A 150 bp band was isolated via gel extraction. Indexed adapters were ligated and reamplified with 10 cycles of PCR. Gel extraction was then used to isolate the resultant 260 bp PCR product. Sample concentrations were calculated using a BioAnalyzer on a High Sense DNA chip (5067-4626, Agilent). Samples were diluted to 2 nM, denatured, and further diluted to 10 pM, with a 10% phiX spike in. Samples were run on a HiSeq 2500 using HiSeq v2 reagents for Paired End Clustering and a 200 cycle SBS kit (PE-402-4002 and FC-402-4021, Illumina) using the Illumina HCS software version 2.2.68. Forward and reverse reads were both 110 cycles, with an 8 cycle single index read. Base-calling and demultiplexing were performed using the bcl2fastq software (Illumina) with default settings. After basecalling and indexing, sequences were identified and aligned using the leader sequence and then binned by sequence.
Validation of sequences from NGS
Hit peptides from NGS were resynthesized as gBlocks (IDT). These gBlocks were used as template for PCR to introduce SapI restriction sites compatible for re-cloning into the pAMK-267 library backbone. Newly reconstructed library members were transformed into Marionette-Clo cells containing modifying enzyme and selection plasmids and were then plated on media containing Carb/Kan/Spec. Individual transformants were then cultured in TB media + Carb/Kan/Spec in a deep well 96-well plate (1896-2000, USA Scientific, FL) and incubated overnight (Multitron Pro, 30 °C, 900 rpm). These cultures were then subcultured 1:100 in TB media + Carb/Kan/Spec either fully induced (1 µM 3OC6-AHL, and 100 µM cumate) or uninduced and incubated for 20 hr (Multitron Pro, 30 °C, 900 rpm) before taking 0.5 µL for standard flow cytometry analysis.
Analysis of Pap2c_1 mutants
Truncations of Pap2c_1 were generated via insertion into pAMK-267 with standard cloning methods. These variants were then transformed into a Marionette Clo strain with the PapB modifying enzyme plasmid, pEG06_044, and the selection plasmid, pAMK-866, and then grown overnight in 1 mL LB media + Kan/Carb at 30 °C, 900 rpm in a deep well 96-well plate. The next day, cultures were diluted 1:100 into 1 mL TB media + Kan/Carb containing either 0 or 1 µM 3OC6-AHL. After overnight growth in the same growth conditions, the variants were assayed as described in Cytometry Analysis.
Site-saturation mutagenesis of Pap2c_1
A library containing all possible single site substitutions was created from the TEV-GIRAY-core parent construct (pAMK-1056). 15 PCR reactions were run to create NNK substitutions at each position except C8 and E12. These 15 PCRs were then combined in equimolar amounts and assembled into the pAMK-267 backbone using SapI (see Library Generation). Colony counting indicated a library size of 1.3 × 106. This library was then miniprepped and transformed into the Marionette Clo strain harboring the PapB modifying enzyme plasmid, pEG06_044, and the selection plasmid, pAMK-866. After overnight growth in LB media 30 °C, the library with modifying enzymes and selection plasmid was then induced by diluting 20 µL into 1 mL TB media + Carb/Kan/Spec and 3.2 nM 3OC6-AHL. Overnight growth was conducted in a deep well 96-well plate at 30 °C with 900 rpm shaking. After the culture reached saturation (20 hours), a selection was conducted by diluting 10 µL into 1 mL TB media + Carb/Kan/Spec, 3.2 nM 3OC6-AHL, and 250 µM Cm. The library was also diluted into media without any Cm to provide a “pre-selection” library. These cultures were grown until saturated (20 hours) before being miniprepped. The pre- and post-selection library plasmids were then amplified via PCR using the osDAA154 and osDAA155 primers (Supplementary Table 3) and 200 bp bands we gel isolated. These fragments were then submitted for paired-end reads on an Illumina MiSeq, utilizing V2 chemistry. Paired-end alignments were conducted with PANDAseq and only sequences without any mismatches between forward and reverse reads were kept for downstream processing, resulting in >50 K reads per sample. Read fractions for each core sequence variant were calculated by dividing the number of reads for each DNA sequence by total number of reads in the sample. Enrichment values were then calculated from these by dividing the read fraction of each sequence in the post-selection library by its read fraction in the pre-selection library. Degenerate DNA sequences corresponding to the same amino acid sequence were averaged to get a single enrichment value for each core variant. These amino acid sequence enrichments were then normalized to the enrichment for the parent core sequence to derive “Enrichment relative to parent”.
Peptide purification
Peptide hit gBlocks were cloned into the peptide expression plasmid, pEG03-119 using their flanking SapI restriction sites. The peptide and modifying enzyme plasmids were co-transformed into E. coli Marionette-X, streaked onto 2xYT agar with Carb/Spec and incubated at 30 °C overnight. Individual colonies were used to inoculate 20 mL of LB media in a 125 mL shake flask and incubated overnight at 30 °C and 250 rpm in an Innova44 (Eppendorf, NY). A 5 mL aliquot of overnight starter culture was diluted in 500 mL total volume TB media with Carb/Spec in Fernbach flasks and grown at 30 °C and 250 rpm until reaching OD600 0.8–1.0, at which point 1 mM IPTG and 200 μM cumate are added. Induced cultures were grown for a further 20 h at 30 °C and 250 rpm and then centrifuged (4,000 g, 4 °C, 10 min) in a Sorvall RC 6+ centrifuge (Thermo Fisher Scientific, MA). Pellets were resuspended in 30 mL lysis buffer (5 M guanidinium hydrochloride, 300 mM NaCl, 10 mM imidazole, 50 mM sodium phosphate, pH 7.5), and freeze-thawed twice (frozen in −80 °C freezer; thawed in innova44 incubator at 30 °C, 250 rpm). Cell lysates were centrifuged (Eppendorf 5424, 20,000 g, 18 °C, 45 min) in a Sorvall RC 6+ centrifuge (Thermo Fisher Scientific, MA) and the peptides affinity purified via gravity-flow using 3 mL resin-bed volume of Ni-NTA agarose resin (88223, Thermo Fisher Scientific, MA), following manufacturer instructions, using 2 resin-bed volumes water and lysis buffer for column equilibration, 4 resin-bed volumes of wash buffer (5 M guanidinium hydrochloride, 300 mM NaCl, 25 mM imidazole, 50 mM sodium phosphate, pH 7.5), 4 resin-bed volumes of elution buffer buffer (5 M guanidinium hydrochloride, 300 mM NaCl, 250 mM imidazole, 50 mM sodium phosphate, pH 7.5). Eluate from Ni-NTA purification was then subjected to solid-phase extraction (SPE) using Strata-XL 500 mg tubes (8B-S043-HCH, Phenomenex, CA). The solid phase was first conditioned with 4 bed volumes of methanol and then water. Eluate was then loaded, washed with 8 bed volumes of 10 mM NH4CO3, and eluted with 8 bed volumes of 1:1 acetonitrile:aqueous 10 mM NH4CO3. Solvent was removed via lyophilization at −80 °C for 24–48 h. To cleave the SUMO and leader peptide from the core, the extracted peptide was resuspended in 20 mL TE buffer and 100 μl 20 mg/mL TEV protease and incubated overnight at room temperature with slow orbital shaking. The cleaved peptides were then desalted using a Strata-X PRO 500 mg SPE tubes (8B-S536-HCH, Phenomenex, CA). The solid phase was first conditioned with 4 bed volumes of methanol and then water. Eluate was then loaded, washed with 8 bed volumes of 10 mM NH4CO3, and eluted with 8 bed volumes of 1:1 acetonitrile:aqueous 10 mM NH4CO3. Solvent was removed via lyophilization at -80 °C for 24–48 h. After solvent removal, a 5 mL aliquot of the mixture resuspended in 10:90 acetonitrile:water was injected into a Agilent Technologies 1260 Infinity system HPLC (HPLC) system (Agilent Technologies, Santa Clara, CA) and separated using a 150 mm × 10 cm Aeris PEPTIDE XB-C18 column (100 Å, 5 μm) at a flow rate of 2 mL/min. Separation was carried out with a gradient program, with 0.1% formic acid as solvent A and acetonitrile with 0.1% formic acid as solvent B. The % B was held at 25% for 3 minutes, then increased to 50% over the next 17 min. The eluent was passed through a diode array detector (DAD) and absorbance at 270 nm was recorded. Detected peaks were collected using an Agilent G1364B Fraction Collector and again solvent was removed via lyophilization at −80 °C for 24–48 h. Samples were resuspended in 1 mL of 1:1 acetonitrile:aqueous 10 mM NH4CO3 in pre-weighed 2 mL microcentrifuge tubes (Eppendorf) and solvent was removed via lyophilization at −80 °C for 24–48 h. Yields were measured by comparing mass of empty tubes to tubes containing lyophilized powder.
Liquid chromatography/mass spectrometry
All chromatography was performed using the mobile phases ACN (acetonitrile supplemented with 0.1% formic acid and 0.1% water) and water (supplemented with 0.1% formic acid). The “QTOF” is an Agilent 1260 Infinity II liquid chromatograph with binary pump configured in low-dwell volume mode and column oven set to 40 °C, coupled to an Agilent 6545 QTOF mass spectrometer equipped with an Agilent electrospray ionization (ESI) source. Nitrogen gas is building-supplied and ESI source parameters are 350 °C gas temperature, 12 L/min gas flow, 30 psig nebulizer pressure, 350 °C sheath gas temperature, 8 L/min sheath gas flow, 3000 V capillary voltage, 1000 V nozzle voltage, 135 V fragmentor voltage, 15 V skimmer voltage, 600 V Oct 1 RF Vpp; the mass spectrometer was run in MS mode with reference mass enabled and tuned in positive mode with standard mass range (3200 m/z) and 2 GHz extended dynamic range. QTOF analysis was performed with a Phenomenex Aeris PEPTIDE XB-C18 2.6 μm 50 mm × 2.1 mm column. The flow rate was set at 0.5 mL/min and 5 μl sample was injected. The gradient used was 20% ACN for 0.5 min, 20–55% ACN over 5.5 min, 55–90% ACN over 0.5 minutes, 90% ACN for 1.5 min, with 0.8 min re-equilibration. LC–MS data were analyzed using MassHunter version 10.0. Accurate mass predictions of peptides were generated using the online resource, ChemCalc81.
Bio-layer interferometry
Assays were performed on an Octet Red (ForteBio) instrument at 30 °C with shaking at 1,000 rpm. Ni-NTA biosensors (18-5101, ForteBio, NY) were hydrated in 1× kinetics buffer (diluted from 10× buffer; 18-5032, ForteBio, NY) for 30 min before the measurement. Expi293F human cell-derived and purified SARS-CoV-2 RBD (RBD296-531) was loaded at 10–20 µg/mL in 1× Kinetics Buffer for 300 s prior to baseline equilibration for 180 s in 1× kinetics buffer. Association reactions of the peptide to RBD296-531 were carried out in 1× kinetics buffer at various concentrations in a two-fold dilution series from 80 mM to 1.25 mM was carried out for 900 s. Then dissociation reactions were observed for 900 s. Data were acquired using the Octet Data Acquisition software version 9.0 and response data were generated using ForteBio data analysis software version 9.0.
Competition analysis using Bio-layer interferometry
Competition of peptide AMK-1057 with hACE2, B38, and CR3022 for binding to recombinant SARS-CoV-2 RBD was evaluated using a ForteBio Octet Red instrument at 30 °C with shaking at 1,000 rpm. Ni-NTA biosensors (18-5101, ForteBio, NY) were hydrated in 1X kinetics buffer (diluted from 10X buffer; 18-5032, ForteBio NY) for 30 min before the measurement. Expi293F human cell-derived and purified SARS-CoV-2 RBD was loaded at 20 µg/mL in 1X Kinetics Buffer for 450 s prior to baseline equilibration for 300 s in 1× kinetics buffer. The RBD-loaded biosensor tips were exposed (300 s) to the hACE2 soluble receptor (300 nM), B38 (300 nM), or CR3022 IgG (300 nM, and then exposed (300 s) to either AMK-1057 peptide (10000 nM) or 1X kinetics buffer. The data was interstep corrected using the FortéBio Data Analysis Software. Additional binding by the secondary molecule indicates an unoccupied epitope (non-competitor), while no binding indicates epitope blocking (competitor).
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
