Herein, we report the use of a PS-tag fused to 15-amino acid peptides for an improved ELISA method and better peptide solubility. The peptides, MACS545, MMS97, MMS100, and MMS140, were selected from the eCPX bacterial display library (CytomX Therapeutics, San Francisco, CA, USA) to bind to protective antigen (PA) of Bacillus anthracis (37-39). A PS-tag peptide sequence with a (Gly)4 spacer was added to the N or C terminus of the MACS545 peptide to compare whether the PS-tag improved the apparent binding affinity when compared with the native peptide and a biotinylated peptide. The PS-tag peptide fusion is expected to improve the hydrophilicity of peptides, and thus the peptide solubility, and is not expected to impact the structural integrity, since a flexible (Gly)4 linker is used to separate the PS-tag and the PA binding sequence.
Materials and methods
Unlabeled, biotinylated, and PS-tag peptides were purchased from RS Synthesis (Louisville, KY, USA) at >85% purity. Ninety-six–well high binding capacity neutravidin (NA) plates were purchased from Pierce Protein (Thermo Scientific, Rockford, IL, USA) for ELISA analysis with biotinylated peptides. Maxisorp (Nalge Nunc, Rochester, NY, USA) plates were used for ELISA analysis of both PS-tag–fused and unfused peptides. For direct ELISA analysis of the peptide binding affinity to a recombinant PA protein, 83 kDa PA (PA83; List Biological Laboratories, Campbell, CA, USA) was labeled directly with 44 kDa horseradish peroxidase enzyme (HRP) using EZ-Link plus activated peroxidase (Pierce Protein, Thermo Scientific), according to the manufacturer's suggested procedure for amine-reactive HRP labeling of the PA protein. Following the PA-HRP conjugation, the sample was dialyzed using a 50-kDa nominal molecular weight limit (NMWL) membrane (Pierce Protein, Thermo Scientific) to remove unreacted HRP enzyme.
Biotinylated peptide ELISA on neutravidin (NA) plate
A biotinylated MACS545 peptide was bound to the surface of an neutravidin (NA)-coated plate for binding analysis according the manufacturer's recommended procedure. Briefly, a high binding capacity NA plate was washed three times with wash buffer (25 mM Tris, 0.15 M NaCl, pH 7.2, with 0.1% w/v BSA, and 0.05% v/v Tween-20) before addition of the biotinylated peptide for a 2-h incubation at room temperature. The peptides were serially diluted 2-fold across each row beginning with the 10 µg/mL stock peptide solution (typically 10, 5, 2.5, 1.25, 0.675, 0.338, 0.169, 0.084, 0.042, and 0.021 µg/mL). Following incubation, the wells were washed three times with wash buffer, and PA-HRP was incubated in each well for 30 min at 0.2 µg/mL. After PA-HRP binding, the wells were washed three times with wash buffer before detection with 1-Step Ultra TMB ELISA substrate (Pierce Protein, Thermo Scientific). The data was recorded as total absorbance at 450 nm using a Synergy HT Microplate reader (Biotek, Winooski, VT, USA). The binding dissociation constant (KD) was determined by plotting the average total absorbance of three replicate measurements versus the concentration of peptide and fit using a sigmoid function with IGOR Pro software (WaveMetrics, Lake Oswego, OR, USA).
PS-tag and non-PS-tag ELISA
ELISA analysis for the PS-tag–fused and unfused peptides was done using a Maxisorp (Nalge Nunc) 96-well plate by initially dissolving each peptide at 10 µg/mL in 0.2 M sodium bicarbonate buffer, pH 9.5. The MACS545 unlabeled peptide required using acetonitrile for initial dissolution of 1 mg of peptide and dilution with water to a final concentration of 20% v/v acetonitrile prior to ELISA analysis (1 mg/mL final peptide concentration). The peptides were serially diluted 2-fold across each row beginning with the 10 µg/mL stock peptide solution (typically 10, 5, 2.5, 1.25, 0.675, 0.338, 0.169, 0.084, 0.042, and 0.021 µg/mL). A single row of buffer and a row of PS-tag alone were used as negative controls. Following a 2-h incubation, each well was blocked for 1 h using phosphate-buffered saline (PBS), pH 7.4, with 0.1% Tween (PBST). PA-HRP was used at 0.2 µg/mL in PBST to determine the total PA binding to each peptide. After a 45-min incubation period, the wells were washed with PBS and detected using 1-Step Ultra TMB ELISA substrate. The data was recorded as total absorbance at 450 nm using a Synergy HT Microplate reader, and the KD was determined by plotting the total absorbance versus the concentration of peptide and fit using a sigmoid function with IGOR Pro software.
Three other PA binding peptides with varying sequence composition (grand average hydropathy and charge) were appended with either an N- or C-terminal PS-tag to determine the general applicability of a PS-tag in peptide capture ELISA. The terminal location of the PS-tag was chosen opposite of the terminal containing the WXCFTC consensus binding sequence: MMS97 (N-terminal PS-tag), MMS100 (C-terminal PS-tag), and MMS140 (C-terminal PS-tag) (39).
Analysis of peptide structure with PS-tag
The circular dichroism (CD) absorbance of the peptides was measured on a J-815 spectrometer (Jasco, Easton, MD) using a continuous scan method from 300 to 200 nm at a 20 nm/min scan speed with 1 mm pathlength Spectrosil Far UV Quartz cuvette (Starna Cells, Altascadero, CA, USA). The samples were analyzed in triplicate with the average of the three runs presented. The PS-tag (MW, 1497.83 g/mol) was measured as a 0.1670 mM solution in BupH modified Dulbecco's PBS (Pierce Protein, Thermo Scientific) using Water CHROMASOLV for HPLC (Sigma-Aldrich, St. Louis, MO, USA). The MACS545 CD sample was made from a stock 2 mM peptide solution containing 20% acetonitrile and diluted 1:10 to 0.1905 mM in PBS. The PS-tagged MACS545 (MW, 3515 g/mol; ext, 2800 M-1cm-1) was dissolved in PBS at a concentration of 0.1106 mM for CD measurements.
Results and discussion
PS-tag peptide fusion offers potential improvement to direct ELISA using peptides as capture reagents through the enhancement of: (i) peptide adsorption to unmodified polystyrene surfaces; (ii) peptide reagent solubility; and (iii) overall capture ability. To investigate the feasibility of this approach, a series of experiments was executed to compare PS-tag peptides with native or biotinylated peptides in an ELISA format. Also, CD studies were performed to ensure that the PS-tag had no structural impact on the fused peptide, which could limit the extensibility of this peptide fusion if structure was critical for target recognition.
The native MACS545 and MACS545 fused with an N-terminal PS-tag were used simultaneously, under identical conditions, as capture agents in an ELISA for PA from B. anthracis. The PS-tagged MACS545 sample showed much higher binding affinity, KD=513 nM, compared with native MACS545, which had approximately 10-fold less signal at the same concentration and did not produce a typical concentration-dependent binding curve (Figure 1). Measurement of the PS-tag alone samples did not show nonspecific binding to the PA-HRP given the equal signal of the PS-tag alone samples to the buffer alone samples (Figure 1). The C-terminal PS-tag and N-terminal PS-tag MACS545 were analyzed simultaneously (in triplicate) to determine if the PS-tag interfered with the binding of PA to the peptide. The C- and N-terminal samples exhibited very similar binding affinities (Figure 2), with the C-terminal PS-tag having a KD=715 nM and the N-terminal PS-tag having a KD=516 nM. The N-terminal PS-tag MACS545 exhibited both higher affinity and smaller intersample variability, as seen in the narrower intersample deviation (Figure 2). This result is not surprising, since WXCFTC consensus binding sequence in MACS545 is furthest from the N terminus and would enable stronger binding and greater accessibility for binding. Comparing the PS-tagged MACS545 to an N-terminal biotinylated MACS545, there is a 10-fold difference in the binding affinity for the PS-tagged versus the biotinylated peptide, with a KD=6988 nM for the latter (Figure 2).