Quantification of particle-conjugated or -encapsulated peptides on interfering reagent backgrounds

Abstract

Particle-based technologies are increasingly used in diagnostics and therapeutics. The particles employed in these applications are usually composed of polymers such as poly(lactide-co-glycolide) and functionalized with peptides or proteins. Peptide or protein conjugation to particles is frequently achieved using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide while dimethyl sulfoxide is used to retrieve surface-attached or encapsulated peptides or proteins by solubilizing the particles. We examined strategies based on the bicinchoninic acid, Coomassie Plus and 3-(4-carboxybenzoyl)quinoline-2-carboxaldehyde assays for the quantification of surface-attached or encapsulated peptides or proteins. We determined that the 3-(4-carboxybenzoyl)quinoline-2-carboxaldehyde assay was a highly sensitive and accurate substitute for radioactivity, suitable for measuring multiple particle-bound or -encapsulated peptides or proteins in the presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide or poly(lactide-co-glycolide) in dimethyl sulfoxide that interfere with the more commonly-used bicinchoninic acid and Coomassie Plus assays. Our strategy enables the accurate quantification of peptides or proteins loaded onto or into particles – an essential component of designing particle-based platforms for diagnostics and therapeutics.

Introduction

Particle-based technologies underlie numerous diagnostic and therapeutic applications for the improvement of human health (1). Particles functionalized with deoxyribonucleic acid (2), polysaccharides (3) or glycoproteins (4,5) are used in a wide variety of applications that include gene therapy and pathogen-specific antibiotic and tumor-selective chemotherapeutic delivery. Peptides or proteins are frequently immobilized on or encapsulated within particles for use in drug delivery, vaccines (6) and recently, immunological tolerance induction (7,8). Particles are commonly fabricated from the biodegradable polymer poly(lactide-co-glycolide) (PLG) (9) for the delivery of peptides or proteins, including growth factors and targeting antibodies (10).

We examined strategies for the quantification of peptides or proteins conjugated to the surfaces of (7,8) or encapsulated within PLG particles. Accurate quantification of peptides or proteins loaded onto or into particles is essential for particle characterization and refinement of particle design parameters. Peptides or proteins are frequently quantified using the bicinchoninic acid (BCA), Lowry and Coomassie Plus assays (4,5,10). However, common reagents for conjugating or retrieving peptides or proteins to or from particles interfere with these assays, leading to inaccurate measurements (11,12). 1-ethyl-3-(-3-dimethylaminopropyl)carbodiimide (EDC) (12–14) is often utilized for the bioconjugation of peptides or proteins to particles (15). We measured the interference of EDC with the BCA, Coomassie Plus and 3-(4-carboxybenzoyl)quinoline-2-carboxaldehyde (CBQCA) assays. Peptides or proteins conjugated to or encapsulated within PLG particles were retrieved by dissolving the particles in dimethylsulfoxide (DMSO) and we measured the interference of PLG dissolved in DMSO with the Coomassie Plus and CBQCA assays. We determined that the CBQCA assay is a versatile, robust and sensitive strategy for the accurate quantification of peptides or proteins loaded onto or into particles, in the presence of common conjugation, encapsulation and retrieval reagents that interfere with more traditional assays.

Materials and methods

Materials

The BCA and Coomassie Plus assays and cell culture-grade water were purchased from Thermo Fisher Scientific, Inc. (Rockford, IL, USA). The CBQCA assay was purchased from Life Technologies Corporation (Grand Island, NY, USA). Sodium borate buffer was purchased from Polysciences, Inc. (Warrington, PA, USA). EDC was purchased from EMD Millipore (Billerica, MA, USA). Recombinant human insulin (INS), recombinant human lysozyme (LYS), myelin basic protein (MBP), bovine serum albumin (BSA), ovalbumin (OVA), dichloromethane (DCM) and DMSO were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). PLP_139–151 (H₂N-HSLGKWLGHPDKF-CONH₂) and OVA_323–339 (H₂N-ISQAVHAAHAEINEAGR-CONH₂) were synthesized in the Peptide Synthesis Core Facility of the Institute for BioNanotechnology in Medicine at Northwestern University. Radiolabeled PLP_139–151 (H₂N-HS[4,5-3H]LGKWLGHPDKF-CONH₂, 1 mCi/mL, 60 Ci/mmol) and OVA_323–339 ([G-³H]H₂N-ISQAVHAAHAEINEAGR-CONH₂, 1 mCi/mL, 20 Ci/mmol) were purchased from American Radiolabeled Chemicals, Inc. (St. Louis, MO, USA). Bio-Safe II^™ biodegradable scintillation cocktail was purchased from Research Products International Corporation (Mount Prospect, IL, USA).

BCA, Coomassie Plus and CBQCA assays

EDC was dissolved in cell culture-grade water immediately before use. All analyses were performed in quadruplicate, unless otherwise stated, on a SpectraMax M5 microplate reader (Molecular Devices, Sunnyvale, CA). The BCA assay was performed using the microplate procedure provided by the manufacturer. The Coomassie Plus assay was performed using a microplate procedure modified from that provided by the manufacturer. Briefly, 5 μL of sample was mixed with 150 μL of Coomassie Plus reagent, incubated for 10 minutes at room temperature and read for absorbance at 595 nm. The CBQCA assay was performed using a procedure modified from that provided by the manufacturer. Briefly, 10 μL of sample was mixed with 10 μL of 5 mM ATTO-TAG CBQCA reagent, 5 μL of 20 mM KCN and 125 μL of 0.1 M sodium borate buffer, pH 9.3, incubated for 1 hour at room temperature and read for fluorescence emission at 550 nm, with excitation at 465 nm.

Spectrum analyses and kinetic studies in the BCA assay

For spectrum analyses, the absorbance of the resultant purple-colored solutions were measured from 450 nm to 650 nm, at 5 nm intervals, after 30 minutes of reaction at 37°C. For kinetic studies, the absorbance of the reaction solutions at 562 nm was measured every minute after sample addition to the BCA reagent at 37°C.

Quantification of peptide or protein loaded onto or into PLG particles using the CBQCA assay or radioactivity

PLG particles with surface carboxylic acid groups were synthesized as previously described (8). PLG particles with encapsulated peptides or proteins were synthesized by emulsifying 150 μL of 50 mg/mL peptide or protein in 4 mL of 20% w/v PLG in DCM before following the protocol previously described (8). Peptides or proteins were conjugated onto PLG particles, by incubating peptides or proteins, PLG particles and EDC at concentrations of 4 mg/mL, 50 mg/mL and 20 mg/mL, respectively, at 25°C for 1 hour. After conjugation, the PLG particles were centrifuged at 3000 g for 5 minutes and the resultant supernatant was collected. The PLG particles were then washed with deionized water thrice and the resultant supernatants and PLG particles collected. Peptides or proteins loaded onto or into PLG particles were retrieved by dissolving 50 mg of PLG particles in 1 mL of DMSO. The resultant solutions and previously collected supernatants were analyzed using the CBQCA assay to determine the conjugation and encapsulation efficiencies. Quantification using radioactivity was performed by the addition 13.5 nCi of radiolabeled PLP_139–151 or OVA_323–339 during conjugation or encapsulation. The amount of radiolabeled PLP_139–151 or OVA_323–339 in the supernatants and on the PLG particles post-conjugation and in DMSO post-dissolution of the PLP_139–151- or OVA_323–339-containing PLG particles was quantified by the addition of the PLG particles or 1 mL of the supernatant or DMSO to 1 mL of Bio-Safe II^™ biodegradable scintillation cocktail. Scintillation counting was performed on a Wallac Win Spectral 1414 liquid scintillation counter (PerkinElmer, Waltham, MA, USA). All conjugations and encapsulations were performed in triplicate.

Conjugation efficiency = (Peptide amount added - Peptide amount in supernatant)/Peptide amount added × 100%

Encapsulation efficiency = Peptide amount in PLG nanoparticles dissolved in DMSO/Peptide amount used for synthesis × 100%

Results and discussion

We were interested in the bioconjugation or encapsulation of peptides or proteins to or within PLG particles for the treatment of autoimmunity (7,8) and allergy through immunological tolerance induction. Multiple peptides (PLP_139–151 and OVA_323–339) and proteins (INS, LYS, MBP, OVA and BSA) were conjugated onto PLG particles using EDC or encapsulated within PLG particles. Unconjugated peptides or proteins were collected along with EDC in post-reaction supernatants. Peptides or proteins directly associated with the particles, i.e., conjugated or encapsulated, were retrieved by dissolving the particles in DMSO. We quantified these peptides or proteins using strategies based on the BCA, Coomassie Plus and CBQCA assays.

We initially estimated that the amounts of peptides or proteins conjugated onto or encapsulated within PLG particles were on the microgram scale. Therefore, we based our initial studies on the colorimetric BCA and Coomassie Plus assays because of their large working range on the microgram scale – 20 μg/mL to 2000 μg/mL for the BCA assay and 100 μg/mL to 1500 μg/mL for the BCA assay. Kumar et al. reported that EDC interfered with the Lowry assay (11). Interestingly, Vashist et al. reported that EDC did not interfere with the BCA assay (12) despite similar mechanisms of peptide and protein detection and color development for both assays. We observed that EDC interfered strongly with the BCA assay, with absorbance increasing linearly with EDC concentration, similar to BSA (Fig. 1A). We further investigated the mechanism of interference of EDC with the BCA assay. Chromophores formed by EDC and BSA exhibited similar absorption spectra, with maxima at 562 nm (Fig. 1B). Kinetic studies of chromophore formation by EDC demonstrated zero-order kinetics while studies with BSA demonstrated a gradual decrease with time (Fig. 1C). This result is consistent with observations that color development (reduction of Cu²⁺ to Cu⁺) is mediated initially by more accessible cysteine, tryptophan and tyrosine side-chains and later by less accessible peptide bonds (12). Thus, the BCA assay was not suitable for the quantification of particle-bound peptides or proteins in the presence of EDC.

Figure 1. Interference of EDC or PLG in DMSO with the BCA or Coomassie Plus assays.

(A) EDC interfered with the BCA assay and resulted in absorbance similar to BSA. (B) Absorption spectra of the purple-colored complexes produced by EDC (200 mg/mL) and BSA (250 μg/mL) in the BCA assay. (C) Kinetic studies of the reactions of EDC (200 mg/mL) and BSA (250 μg/mL) with the BCA reagent. (D) EDC did not interfere with the Coomassie Plus assay. (E) INS, LYS, OVA, BSA, PLP_139–151 and OVA_323–339, but not MBP, exhibited linear concentration-absorbance relationships with the Coomassie Plus assay. (F) PLG in DMSO interfered with the Coomassie Plus assay.

We then tested for the interference of EDC with the Coomassie Plus assay. Contrary to the manufacturer’s instructions, EDC did not interfere with the Coomassie Plus assay (Fig. 1D). The Coomassie Plus assay exhibited linear absorbance-concentration relationships for INS, LYS, OVA, BSA, PLP_139–151 and OVA_323–339 (Fig. 1E). For MBP, absorbance increased hyperbolically with MBP concentration (Fig. 1E), likely due to the hydrophobic nature of MBP leading to increased binding of Coomassie Brilliant Blue G-250 dye at lower concentrations and depletion of Coomassie Brilliant Blue G-250 dye at higher concentrations (16). Due to its mechanism of color development, the Coomassie Plus assay did not exhibit sufficient sensitivity to adequately resolve different concentrations of PLP_139–151 and OVA_323–339 (Fig. 1E) (17,18). Furthermore, PLG in DMSO interfered strongly with the Coomassie Plus assay (Fig. 1F), due to the precipitation of PLG in the acidic Coomassie Plus reagent (17,18). Therefore, the Coomassie Plus assay was not suitable for the quantification of peptides and particle-bound or -encapsulated peptides or proteins.

Due to the incompatibility of the BCA and Coomassie Plus assays with the background reagents present in our peptide or protein analytes, we investigated the use of the CBQCA assay to measure the amounts of conjugated or encapsulated peptides or proteins. The CBQCA assay (working range = 10 ng to 150 μg) measures the amount of primary amines present on peptides or proteins through a highly specific reaction between CBQCA and primary amines in the presence of cyanide (19). Therefore, we expected that the CBQCA assay would not be subject to interference from EDC or PLG in DMSO because they do not contain primary amines. Accordingly, we determined that EDC (Fig. 2A) or PLG in DMSO (Fig. 2B) did not interfere with the CBQCA assay and that the fluorescence output of the assay increased linearly with increasing amounts of our peptides and proteins of interest (Fig. 2C). Although the sensitivity of the CBQCA assay was low for OVA_323–339, the linear dynamic range and resolution of the assay could be increased by doubling the incubation time to 2 hours and increasing the amount of OVA_323–339 a hundred-fold (Fig. 2C, inset). Therefore, the CBQCA assay is a versatile and sensitive method for the quantification of multiple peptides and proteins without interference from EDC or PLG in DMSO.

Figure 2. The CBQCA assay is suitable for use with multiple peptides and proteins and is not subject to interference from EDC or PLG in DMSO.

(A) EDC did not interfere with the CBQCA assay. Fluorescence was measured in arbitrary units (a.u.). (B) PLG in DMSO did not interfere with the CBQCA assay. (C) INS, LYS, MBP, OVA, BSA, PLP_139–151 and OVA_323–339 exhibited linear concentration-fluorescence relationships with the CBQCA assay. (inset) Resolution and dynamic range of the CBQCA assay can be increased for OVA_323–339.

We successfully applied the CBQCA assay to measure the conjugation and encapsulation efficiencies of PLP_139–151, OVA_323–339, OVA and LYS onto and into PLG particles (Table 1). Our measurements were compared to those obtained using radiolabeled PLP_139–151 and OVA_323–339, the quantification approach that remains unsurpassed in terms of reproducibility and sensitivity (20). The conjugation and encapsulation efficiencies measured using the CBQCA assay were similar to those measured using radioactivity (Table 1). To establish a secondary check for the accuracy of the CBQCA assay, we quantified the amounts of PLP_139–151 or OVA_323–339 conjugated to particles and remaining in the post-reaction supernatants. Only EDC, but not PLG in DMSO, was present with unconjugated PLP_139–151 or OVA_323–339 in the post-reaction supernatants. Only PLG in DMSO, but not EDC, was present with conjugated PLP_139–151 or OVA_323–339 after the peptide-particle conjugates were dissolved in DMSO. The sum of each peptide loaded onto particles and remaining in the supernatants achieved conservation of mass, i.e., total input peptide amount = amount of unconjugated peptide in the presence of EDC + amount of conjugated peptide in the presence of PLG in DMSO (data not shown). This demonstrated the robustness to interference, sensitivity, accuracy and reproducibility of the CBQCA assay for the quantification of peptides or proteins loaded onto or into PLG particles.

Table 1.

Quantification of the conjugation and encapsulation efficiencies of PLP_139–151 and OVA_323–339, OVA and LYS loaded onto or into PLG particles. Quantification of PLP_139–151 and OVA_323–339 using the CBQCA assay was compared to radioactivity.

Peptide or protein	CBQCA	Radioactivity
PLG particles with surface conjugated peptide (Conjugation efficiency %)
PLP139–151	7.24 ± 0.17	9.37 ± 0.36 (8)
OVA323–339	9.05 ± 1.38	10.61 ± 0.16 (8)
PLG particles with encapsulated peptide or protein (Encapsulation efficiency %)
PLP139–151	4.37 ± 0.20	5.57 ± 0.21
OVA323–339	4.61 ± 0.52	4.12 ± 0.13
OVA	7.29 ± 0.70	-
LYS	26.98 ± 1.60	-

The routine use of radioactivity for the quantification of peptide and protein loading onto or into particles is impractical because of the associated costs and safety concerns (20). In addition to providing a cheaper, safer and more convenient alternative to the use of radioactivity, we conclude that the CBQCA assay can be used as a highly sensitive and accurate method for measuring a wide range of particle-bound or -encapsulated peptides or proteins in the presence of EDC or PLG in DMSO that would otherwise interfere with the more commonly-used BCA and Coomassie Plus assays. Our strategy enables the accurate quantification of peptides or proteins loaded onto or into particles – a key requirement for the design of particle-based platforms for diagnostic and therapeutic applications.