November 1, 2011 (Vol. 31, No. 19)
Nathan J. Evans
Thomas Kirkland Ph.D. Head of Advanced Technology Chemistry Research Promega
Chad Zimprich Research Scientist Promega
Andrew L. Niles Senior Research Scientist Promega
Bioluminescent Assays Offer Distinct Advantages over Fluorescent Approaches
Histone deacetylase (HDAC) enzymes play a critical role in normal gene-regulation events and the maintenance of homeostasis. However, their dysregulation has been implicated in a variety of disease states, including several forms of cancer, congestive heart failure, diabetes, inflammation, and neurological disorders. Because of the early success of first-generation HDAC inhibitors in both basic research and clinical applications, there is strong continuing interest in developing more potent and selective compounds.
Unfortunately, existing research tools for characterization of HDAC function suffer from poor sensitivity, detection interferences, or are time-consuming and expensive to apply. In this article, we describe a simple, robust, and sensitive bioluminescent assay method that measures the activity of various HDAC enzyme isoforms from recombinant or cell-based sources.
Two Enzymatic Mechanisms
The histone deacetylase family of enzymes consists of 18 members that are organized based upon homology with their yeast counterparts. This family is divided into four main groups: the zinc-dependent class I (HDACs 1, 2, 3, and 8), class IIa (HDACs 4, 5, 7, and 9) and class IIb (HDACs 6 and 10) enzymes; and the NAD+-dependent class III sirtuins (Sirts 1–7).
HDAC 11 shares features from both class I and II HDACs and is often grouped into a separate class IV. Although mechanistically distinct, both the zinc- and NAD+-dependent enzymes can remove ε-acetyl moieties on lysine residues from a diverse set of cellular proteins. Nuclear HDACs and sirtuins are capable of deacetylating N-terminal histone tails, resulting in a condensed chromatin structure and typically repressed gene transcription. However, cytosolic and mitochondrial deacetylases are also known to regulate numerous nonhistone proteins, leading to diverse biological effects.
Assay Design and Procedure
Promega’s HDAC-Glo™ I/II and SIRT-Glo™ Assays are based on optimized, acetylated peptide substrates derived from sequences found in histone 4 and p53 proteins, respectively, which are conjugated to aminoluciferin. These chemically engineered peptides do not react with the developer or luciferase enzymes prior to the specific HDAC or sirtuin deacetylation event. Upon deacetylation, the de-protected substrate undergoes specific proteolytic cleavage by the developer reagent to yield aminoluciferin and the peptide.
Liberated aminoluciferin product is then quantified using Ultra-Glo™ recombinant firefly luciferase. These enzymatic reactions occur virtually simultaneously to produce “glow-type” luminescence upon addition of a single reagent.
The “add-mix-measure” reagents are created by simply rehydrating the lyophilized HDAC-Glo I/II or SIRT-Glo Substrate with buffer, then adding the developer reagent. The SIRT-Glo Substrate also contains the necessary NAD+ co-factor. These complete reagents can be added directly to the deacetylase source in a 1:1 ratio, followed by a brief incubation (15–45 minutes) to achieve signal steady state. This steady-state signal is stable for several hours, allowing additional flexibility for automation and processing multiple assay plates.
To demonstrate broad isozyme responsiveness, recombinant deacetylases were titrated and analyzed with either the HDAC-Glo I/II or SIRT-Glo Assay (Figure 1). The HDAC-Glo I/II Assay was responsive to all members of the class I and II HDACs, including HDAC 11. Subsequent experiments using pharmacologically relevant HDAC inhibitors produced IC50 profiles consistent with values reported in the literature.
The SIRT-Glo Assay measured activity from sirtuins 1, 2, and 3 that was NAD+-dependent and nicotinamide-inhibitable (data not shown). Both assays achieved dynamic signal-to-noise ratios that remained linear over several orders of magnitude of enzyme concentration, thus reducing the need for extensive enzyme titrations or high enzyme content per well.
Assay Sensitivity and Ease of Use
To benchmark assay sensitivity, HDAC-Glo I/II and SIRT-Glo Assays were compared to commercially available fluorescent assay methods (Figure 2). HeLa cell nuclear extracts or recombinant sirtuin 2 were titrated and assayed per the manufacturer’s instructions.
The HDAC-Glo I/II and SIRT-Glo Assays delivered a 400-fold increase in sensitivity when compared to the fluorescent methods. This enhanced sensitivity allows for tangible cost savings due to a reduced requirement for recombinant enzyme and greater resolution of nonabundant deacetylase activities.
In addition, there was a notable time savings associated with use of the HDAC-Glo I/II and SIRT-Glo Assays due to the relatively short 15-minute time period between reagent addition and data acquisition. In contrast, available fluorescent-based assays typically require multiple steps and more than two hours to complete.
High-Throughput Screening and Facile Detection of Assay Interferences
All assays are subject to potential detection interferences that may generate false hits during high-throughput screening activities. To address possible assay interferences against the developer and luciferase enzymes, cognate, non-acetylated forms of the substrates were produced for counterscreening high-throughput screening hits against the coupled detection enzymes.
These nonacetylated peptides can be introduced directly into the prepared reagent and contacted with the test compounds to determine a compound’s HDAC selectivity profile or reveal assay interferences. Halley et al., has successfully implemented the HDAC-Glo I/II, SIRT-Glo and counterscreen assays in a 1,536-well format and reported acceptable hit and false hit rates against the FDA 640 and Hypha Discovery MycoDiverse libraries.
Cell-Based HDAC Assays
Cell-based HDAC experiments are especially relevant because they address the natural context of HDAC enzymes and can reflect the diversity of expression arising from different cell types. Furthermore, on-target HDAC potency can be linked to functional outcomes such as on- and off-target toxicities. Because the HDAC-Glo I/II Substrate is cell permeable, the assay can be conducted in either a lytic (with the addition of detergent) or nonlytic manner. Nonlytic formats are particularly attractive when the cell source is semi-precious and too difficult to obtain for parallel experimentation.
As illustrated in Figure 3, human embryonic stem cells (hESC) were contacted with various HDAC inhibitors for a period of one hour. The HDAC-Glo I/II Reagent was then applied in a nonlytic formulation and HDAC activity measured. Medium and reagent could then be removed for additional experimentation such as RT-PCR.
The hESC contained robust HDAC activity that can be inhibited in a dose-dependent manner and compounds rank ordered. The SIRT-Glo Substrate, while also cell permeable, is not recommended for cell-based applications due to HDAC cross-reactivity with the SIRT-Glo peptide sequence.
The HDAC-Glo I/II and SIRT-Glo Assays provide users with a simple, homogeneous, “add-mix-measure” procedure for the specific detection of lysine deacetylase activities. These assays generate a robust, proportional, and persistent signal that remains stable for several hours, making them useful for basic research through high-throughput applications. Both assays are broadly responsive to recombinant enzymes, with the HDAC-Glo I/II Assay providing additional functionality for use in cell-based applications. The bioluminescent readout also confers distinct advantages over alternative fluorescent approaches, including higher sensitivity, shorter incubation times, and less complicated assay procedures. Moreover, data acquisition is faster than with fluorescent methods. HDAC-Glo I/II and SIRT-Glo Assays also offer a counterscreening method with nonacetylated substrate for high-throughput activities that detect assay interferences.
Nathan J. Evans is senior research scientist, Thomas A. Kirkland is staff scientist, Chad Zimprich is research scientist IV, and Andrew L. Niles (email@example.com) is senior research scientist II at Promega.