2Protein Purification, Promega Corporation, Madison, WI, USA
3Promega Biosciences, Inc., San Luis Obispo, CA, USA
A bioluminescent general protease assay was developed using a combination of five luminogenic peptide substrates. The peptide-conjugated luciferin substrates were combined with luciferase to form a homogeneous, coupled-enzyme assay. This single-reagent format minimized backgrounds, gave stable signals, and reached peak sensitivity within 30 min. The bioluminescent assay was used to detect multiple proteases representing serine, cysteine, and metalloproteinase classes. The range of proteases detected was broader and the sensitivity greater, when compared with a standard fluorescent assay based on cleavage of the whole protein substrate casein. Fifteen of twenty proteases tested had signal-to-background ratios >10 with the bioluminescent method, compared with only seven proteases with the fluorescent approach. The bioluminescent assay also achieved lower detection limits (≤100 pg) than fluorescent methods. During protein purification processes, especially for therapeutic proteins, even trace levels of contamination can impact the protein's stability and activity. This sensitive, bioluminescent, protease assay should be useful for applications in which contaminating proteases are detrimental and protein purity is essential.
Proteases are enzymes that catalyze the hydrolysis of protein peptide bonds. Ubiquitous in nature, proteases are present in all living organisms and have multiple and essential functions in cellular processes (1). They form a large, diverse group that is classified into six classes based on catalytic mechanism: serine, cysteine, aspartic, metallo, threonine, and glutamic (1). They are referred to as endoproteinases, responsible for cleaving internal peptide bonds, or as exopeptidases, which cleave terminal amino acids from either the N or C terminus (aminopeptidases and carboxypeptidases, respectively). A comprehensive listing and classification of proteases can be found in the MEROPS database (http://merops.sanger.ac.uk) (2). Due to their widespread and abundant nature, proteases are unavoidable and occur as contaminants during protein purification processes, negatively impacting the purified protein's yield, purity, activity, and stability. It is necessary to remove these deleterious activities either through the protein purification process itself or by inhibition. Broad-spectrum inhibitors or mixtures of inhibitors are typically used to inactivate proteases, but there are many downstream applications in which the presence of inhibitors is detrimental. In particular, the presence of neither proteases nor inhibitors is acceptable when producing biologics and biosimilars for the therapeutic market. A sensitive method for detecting trace amounts of proteolytic activity is important for validating the effectiveness of protein purification methods during process development stages, as well as confirming the absence of protease in final purified proteins.
Activity assays for a specific protease typically use a single synthetic substrate consisting of a peptide of known sequence conjugated to a fluorophore (3). However, when assaying multiple proteases, proteases of unknown identities, or proteases with unknown or different sequence requirements, a single peptide substrate is not suitable or sufficient. These applications, including protein purification, have used whole proteins, such as gelatin and casein, as universal protease substrates (4-6). Several colorimetric and fluorometric general protease assays are based on the cleavage of casein in solution (7,8). Using f luorophore-labeled casein, caseinolytic activity can be measured, by increased fluorescence due to either the release of small fragments that remain in the supernatant after trichloroacetic acid (TCA) precipitation (7) or the release of fluorescent small fragments from heavily conjugated, and therefore quenched, undigested protein (8). Drawbacks of the casein-based assays include multiple manipulations such as TCA precipitation steps, long incubations (e.g., overnight), and poor sensitivity.
Bioluminescent protease assays employing peptide-conjugated aminoluciferins were developed as an alternative to peptide-conjugated fluorophore assays to capitalize on the sensitivity of bioluminescence and to avoid fluorescence interference issues (9-11). The aminoluciferin derivatives, or ‘pro-luciferins,’ are not utilized by firefly luciferase until first processed by a protease that can release the peptide and free aminoluciferin. The first reported bioluminescent protease assays based on peptide-conjugated aminoluciferins were two-step assays, whereby the protease cleavage was allowed to occur prior to the addition of luciferase (9,10). Subsequently, homogeneous bioluminescent protease assays were developed that coupled the protease and luciferase activities into a single step (12). The coupled system generated a continuous, stable signal that was proportional to the amount of protease. Using this technology, assays have been developed for several proteases including caspases, calpain, dipeptidyl peptidase IV, and the three catalytic activities of the proteasome (12-15). In all cases, the bioluminescent protease assays have been demonstrated to be substantially more sensitive than fluorescent assays using comparable peptide-conjugated substrates (15).
The improved sensitivity of bioluminescent protease assays led to the idea of creating a general protease assay by combining luminogenic peptide substrates. The mixture needed to contain substrates that could detect a broad range of proteases and retain good sensitivity. This meant optimizing the concentrations of, and limiting the number of, substrates to keep the additive endogenous background to a minimum. The substrates for the detection of the three catalytic activities of the proteasome were considered to be an excellent starting point, because the proteasome hydrolyzes hundreds to thousands of proteins using three distinct catalytic activities that cleave peptide bonds following basic, hydrophobic, or acidic residues (16,17). Substrates for these three catalytic activities would likewise be expected to be cleaved by many endoproteinases that recognize basic, hydrophobic, or acidic residues. We combined the three proteasome substrates with two other single amino acid-conjugated aminoluciferins designed for the detection of exopeptidases. In this report, we demonstrate the detection of a broad range of proteases from multiple classes using this mixture of five luminogenic substrates. Comparison of this bioluminescent system to fluorescent, casein-based methods shows that the range of detection is broader, the sensitivity is greater, and the assay is simpler and faster to use. Also, an additional advantage of the substrate mixture strategy is the ability to use the substrates individually to confirm and further characterize any proteolytic activity.