Advances in luminescence-based technologies for drug discovery

Abstract

Introduction:

Luminescence-based technologies, specifically bioluminescence and chemiluminescence, are powerful tools with extensive use in drug discovery. The production of light during chemiluminescence and bioluminescence, in contrast to fluorescence, does not require an excitation light source, resulting in high signal-to-noise ratio, less interference from the background, and no issues from phototoxicity and photobleaching. These characteristics of luminescence technologies offer unique advantages for experimental designs, allowing for greater flexibility to target a wide range of proteins and biological processes for drug discovery at different stages.

Areas covered:

This review provides a basic overview of luciferase-based technologies and details recent advances and uses cases of luciferase and luciferin variations and their applicability in the drug discovery toolset. The authors further expand upon specific applications of luciferase technologies, including chemiluminescent and bioluminescent-based microscopy. Finally, the authors end the paper with forward-looking statements on the field of luminescence and how it may shape the translational scientists’ work moving forward.

Expert opinion:

The demand for improved luciferase and luciferin pairs correlates strongly with efforts to improve the sensitivity and robustness of high throughput assays. As luminescent reporter systems improve, so will the expansion of use cases for luminescence-based technologies for early-stage drug discovery. With the synthesis of novel, non-enzymatic chemiluminescence-based probes, which previously were restrained to only basic research applications, they may now be readily implemented in drug discovery campaigns.

1. Introduction

Robust and reproducible assays are fundamental to an efficient and successful early-stage drug discovery campaign. Once a target is identified and validated, the efficiency of a hit discovery and lead optimization is now entirely dependent on the deployed screening assays, which must be biologically relevant, reproducible, high-quality, and be as high throughput as possible to maximize efficiency of data collection. Assays based on photon emission (luminescence) have higher sensitivity and broader adaptability to biological targets compared to photon absorption-based methodologies, and as a result, luminescence technologies have become essential tools in biomedical research and automated high-throughput screening (HTS). It’s worth noting that while several types of luminescence exist in nature, such as photoluminescence (induced by ultraviolet or visible light), triboluminescence (resulted from mechanical impact), and radioluminescence (excited by ionizing radiation), the most commonly encountered processes in the biomedical sciences remain fluorescence, bioluminescence and chemiluminescence.

Chemiluminescence is a phenomenon of light emission derived from a chemical reaction when chemically excited electrons return to the ground state. Bioluminescence, a type of chemiluminescence, refers to the photon-emitting processes occurring in living organisms often used to ward off predators, attract prey, or as a means of communication. The machinery responsible for bioluminescence is evolutionarily conserved among marine creatures such as bacteria, planktons, algae, crustaceans, squids, and fish, but is also found in terrestrial organisms like bacteria, fungi, worms, and insects [1,2]. These bioluminescent reactions rely on enzymes, termed luciferases, that catalyze the oxidation of a family of substrates called luciferins, resulting in the emission of a photon. The luciferase family of proteins, as well as their luciferin substrates, are as structurally diverse as the organisms from which they are derived, each with their own fingerprint wavelength and quantum yield.

One of the main advantages that bioluminescence and chemiluminescence methods have over related fluorescence technologies is the lack of external light excitation, which causes high background signal leading to a reduced signal-to-noise ratio (SNR). As a result, bioluminescent and chemiluminescent assays have higher sensitivity and selectivity. Importantly, these processes also lack any photobleaching and phototoxicity effects, all common drawbacks of fluorophore-based methods.

Users looking to harness bioluminescence in their research are afforded many platforms for introducing genetic material for a variety of luciferases into cells, enabling a wide range of target and method compatibility with the luciferase reporter system. Recently, advances in protein engineering and synthetic chemistry have resulted in incremental improvements and modifications of luciferases and luciferins. These augmented luminescence systems, in combination with improved imaging and capture technologies, have led to a marked increase over the last decade in the usage of bioluminescence-based methodologies, exemplified by the number of publications on bioluminescence and studies employing it (Figure 1).

Figure 1.

PubMed timeline for publications containing the keyword “bioluminescence” by year. This figure was created using GraphPad Prism version 8.0.2 for Windows, GraphPad Software, San Diego, California USA, www.graphpad.com

This review will provide a summary of the commonly employed bioluminescence reporters and substrates with a primary focus on recent advances in the reporters and manners in which they are employed. To that end, we curated publications using PubMed indexes that report on new developments in chemiluminescence and bioluminescence technologies and their use in early drug discovery with a focus on January 2012 through August 2022. We first discuss the developments that have resulted in luminescent technologies with improvements yielding higher signal brightness, stability, spectral shift and protein thermal stability with respect to the wildtype luciferase species. Next, we introduce new or improved high-throughput screening assays developed because of the advances in luciferase reporter systems, as well as detail other important non-enzymatic chemiluminescent systems (Figure 2). We further summarize the newest developments in the field of bioluminescence- and chemiluminescence-based microscopy at single cell level. While bioluminescence and chemiluminescence are widely used in in vivo imaging, these in vivo methodologies are not included in this review. Finally, we leave the reader with an outlook of the advancing luminescence-based technologies and their application to early-stage drug discovery efforts.

Figure 2.

An overview of luminescence-based assays available to the early-stage drug discovery campaign. Created with BioRender.com

2. Luminescence systems: luciferases and luciferins

There are hundreds of luciferase-family enzymes that exist in the animal kingdom, and surprisingly, only a small fraction of these proteins and their related substrates have been characterized or studied in detail [3]. An excellent review article about different luciferase/luciferin systems and their mechanisms of luminescence was recently published by Syed and Anderson [4]. With the variety of luciferases offerings of wavelength, luminescent yield, half-life, and other characteristics, it is impossible to identify a single variant or enzyme that is applicable to every experimental setup. The following is a shortlist of the most common luciferase enzymes used in drug-discovery assays grouped by their family of luciferin substrate (Table 1):

Table 1.

List of commonly used luciferase reporters along with basic information on the enzyme and required substrates.

Luciferase	Species	Molecular Weight	Peak Emission Wavelength	ATP-Dependence	Substrates
FLuc (Firefly)	Photinus pyralis	62 kDa	550–570 nm	Yes	D-luciferin
Ultra-Glo (Firefly)	Photuris pennsylvanica	61 kDa	550–570 nm	Yes	D-luciferin
CBLuc (Click beetle)	Pyrophorus plagiophthalamus	60 kDa	540/610 nm	Yes	D-luciferin
RLuc (Sea pansy)	Renilla reniformas	36 kDa	480 nm	No	Coelenterazine
GLuc (Copepod)	Gaussia princeps	20 kDa	460 nm	Yes	Coelenterazine
CLuc (Ostracod)	Cypridina noctiluca	62 kDa	465 nm	Yes	Vargulin
OLuc (Deap-sea shrimp)	Oplophorus gracilirostris	35 kDa/19 kDa	460 nm	No	Coelenterazine
NLuc (synthetic variant of OLuc)	Oplophorus gracilirostris	19kDa	460 nm	No	Furimazine

D-luciferin reporters

• Firefly luciferase (FLuc) is perhaps the most used luciferase reporter in the HTS field derived from Photinus pyralis, the common firefly in North America. FLuc is an ATP-hydrolyzing enzyme that uses ATP, metal cation, and luciferin substrates to yield oxyluciferin in an excited state and photon emission (~560 nm) when the excited oxyluciferin relaxes back to the ground state. Firefly luciferase exhibits high luminescent activity across a broad linear range of about seven orders of magnitude [5]. Recent efforts to engineer luciferase mutants and synthetic luciferin analogs have resulted in red-shifted emission to avoid small molecule autofluorescence interference, as well as luciferins that have pH-dependent wavelength maxima [6–10].

• Click beetle luciferase (CBLuc) is another group of luciferases derived from the Pyrophorus plagiophthalamus beetle that are dependent on a luciferin substrate to catalyze the light reaction. Importantly, CBLuc variants exist that red-shift the emission maxima from green (~550 nm) to near infrared emission with naphthyl-luciferin substrates [11,12]. The red-shifted luminescence from CBLuc offers advantages to early-stage drug discovery as it overcomes common assay inner-filter effects and in vivo skin tissue penetration issues [11–14].

Coelenterazine reporters

• Renilla luciferase (RLuc) is another commonly deployed luciferase reporter derived from the octocoral, Renilla reniformis (Table 1). Interestingly, RLuc shares only minimal primary sequence homology with FLuc and light production is independent of ATP cofactor [15,16]. RLuc produces luminescence through the oxidative decarboxylation of coelenterazine to coelenteramide, resulting in the concomitant emission of blue light [17].

• Gaussia luciferase (GLuc) is a smaller (20 kDa) secreted luminescent protein found in small crustaceans called copepods (Gaussia princeps). Unlike other coelenterazine-based reporters, GLuc requires the formation of disulfide bonds integral to the activity of the enzyme, precluding its use in systems that require reducing conditions [18,19]. Interestingly, GLuc, once secreted, has a considerably longer half-life (6 days) than either FLuc or RLuc [19].

• NanoLuc luciferase (NLuc) is an engineered version of shrimp Oplophorus gracilirostris luciferase. It reacts with furimazine, a coelenterazine derivative, with a specific activity over 150 times that of RLuc or FLuc [20]. In addition to NLuc’s enhanced brightness, it also has improved thermal stability, pH stability, and an unbiased distribution in cells as compared to other luciferases, leading to its rapid adoption into many HTS assays.

3. Luminescence-based assays for high-throughput screening

3.1. Cell proliferation and viability assays

Perhaps the most common application of bioluminescence in HTS is in the determination of cell viability and proliferation, where researchers take advantage of the both the requirement for ATP in the Fluc-mediated oxidation of D-luciferin and the fact that only viable cells produce ATP (Figure 2). Detection of ATP allows for the rapid and robust analysis of cellular metabolism as well as other ATP-dependent processes. In most ATP-sensing assays, the amount of luciferin substrate is added in excess of available ATP so that when luciferase concentration is low, the light emissions generated are proportional to the amount of ATP in solution.

Nowadays, different vendors offer easy-to-use “Add and Read”-type kits that eliminate the washing and long-incubation steps. The kits consist of high-quality reagents at optimized concentrations that ensure the rapid and efficient ATP extraction from cells, inhibition of the endogenous ATPases, and generation of a stable luminescent signal. As mammalian and microbial cells have different levels of ATP, different kits are attuned to each type of assaying cells. With the increased application of 3D cell culture in early-stage drug discovery, the ATP-sensing cytotoxicity assays optimized for spheroids and organoids are now commercially available, as well [21].

The usual Fluc-based ATP-sensing cell viability assays are lytic end-point assays without the possibility of live-cell monitoring. To enable the continuous monitoring of cell metabolic activity, an ATP-independent assay was developed where a cell-permeable pro-substrate must be reduced by viable cells to become available for the chemical reaction. Recently, the assay was successfully applied in drug library screening in 2D cell culture and toxicity testing of chemicals in multi-cellular cultures among others [22,23].

3.2. Enzyme activity assays

Enzymes are an important target class for therapeutic intervention, and the literature is rich with examples of small molecule enzyme inhibitors as effective therapeutics against hypertension, bacterial and viral infection, cancer, and many other diseases. Therefore, much effort has been devoted to the development of rapid, sensitive, and robust enzyme activity assays. The characteristics and advantages of bioluminescence assays has led to the wide adoption of luminescence readouts in biochemical and cell-based enzymatic assays in HTS.

Traditionally, ATP-sensing assays were applied to measure the activity of ATP-consuming enzymes, where the luminescence signal is correlated to the ATP level, or inversely proportional to the enzyme activity (Figure 2). Another approach is to couple the conversion of enzyme-generated ADP into ATP and then use the same ATP-sensing method to measure the enzyme activity, generating a luminescent signal directly proportional to the enzyme activity [24]. The conversion of a substrate or product molecule into ATP, which can be quantified by bioluminescence, has enabled the development of homogenous assays for a wide range of enzymes such as methyltransferases, DNA ligases, cAMP-specific phosphodiesterases, and aminoacyl-tRNA synthetases. For instance, GTPase activities can be measured by converting any GTP molecules not hydrolyzed by the enzyme into ATP and coupling this to a luciferase reporter reaction [25]. In a newly developed glycosyltransferase assay, the nucleotides generated from a sugar-nucleotide substrate because of the glycosyltransferase activity (UMP, UDP, CMP, or GDP) are converted into ATP and measured by luminescence [26]. These assays are often only applicable to biochemical reactions with purified enzymes because the substrate or product molecule is common to all the proteins of the same family and these assays cannot discern the identity of enzymes contributing to the activity readout. On other hand, these assays can be readily used to screen an entire class of enzymes without further development.

Other bioluminescence-based enzyme assays exploit the unique structure of luciferin and how it cannot be metabolized by many mammalian and prokaryotic enzymes. For such assays, a pro-luciferin molecule which cannot enable luminescence directly is used. It must first be converted into luciferin by another enzyme before it can be used as a substrate for a luciferase. For example, DEVD-6′-aminoluciferin is a specific substrate for protease cleavage by caspase-3 or caspase-7. The cleavage by caspase-3 and -7 liberates aminoluciferin which becomes available for the luciferase, generating a luminescence signal that is directly proportional to the caspase enzyme activity. Since the substrate is specific for caspase-3 and -7, this assay can be used in both biochemical and cell-based assays and has seen widespread adoption as a method with which to measure the population of apoptotic cells [27]. Based on the same principle, a new ultra-sensitive assay to monitor monoacylglycerol lipase activity, a promising therapeutic target involved in endocannabinoid system modulation, was recently reported [28].

3.3. Reporter gene assays

Luciferase-based gene reporter assays are an ubiquitous methodology appearing in the literature since the luciferase gene was reported and cloned in the 1980s. Reporter assays are generally used for medium to high-throughput approaches, where rapid and economical readouts are necessary to prioritize targets through a screening campaign. By cloning the regulatory region of a gene of interest upstream of a luciferase gene or expression vector, the luminescent signal can be used to quantify the activity of the regulatory element (cis) or proteins (trans) in the biological pathway affected by the target element (Figure 2).

One such application of the luciferase reporter system involved the insertion of luciferase and EGFP reporter genes under the UCHL1 promoter that is normally methylation-silenced in human colon cancers but can be readily demethylated to drive expression [29]. The Metridia longa luciferase (MetLuc) and EGFP genes were inserted downstream of the UCHL1 promoter to construct an assay that was highly sensitive to demethylating agents, producing luciferase and resulting in luminescence after demethylation of the promoter CpG island. The strategy of inserting luciferase reporters into CpG islands to detect aberrant DNA methylation patterns has been applied to HIV-1 promoter as well [30].

There were numerous applications of luciferase technology to the recent SARS-CoV-2 pandemic during early-stage efforts to identify and repurpose antiviral drug candidates. One recent application of the luciferase reporter was in a screening campaign for inhibitors of SARS-CoV-2 viral entry, accomplished by packaging luciferase reporter RNA and murine leukemia virus gag-pol complex within pseudotyped particles containing a SARS-CoV-2 spike protein [31]. By incubating these packaged particles with HEK293 cells expressing the spike protein’s mammalian target, ACE2, the entry and expression of viral genes can be detected by luciferase reporter activity. The design presents an alternative to screening for inhibitors of viral replication and instead focusing on compounds that act on viral entry, with additional power coming from the facile modification to evaluate new mutant spike proteins with the same reporter system. Additional approaches have integrated luciferase reporters in designing SARS-CoV-2 phenotypic assays that are BSL2 compatible through pseudovirus and replicon engineering [32].

A significant issue when using luciferase reporters in high-throughput screening is the prevalence of the reporter-specific interactions that complicates hit triaging and selection. In one example, a screening campaign to identify small molecule replacements for transcription factors indicated that over 80% of the hit molecules from a 500,000 compound library were reporter artifacts [33]. One such strategy to mitigate these interference artifacts was developed by Cheng et al. in which a coincidence reporter circuit produces two or more reporters, reducing the overall assay noise and allowing for better discrimination of signal [34]. FLuc and NLuc are a common coincidence reporter pair, owing to their dissimilar compound interference profiles, and are introduced upstream of the gene of interest and separated by a short P2A peptide that causes ribosome skipping and the production of independent and unfused reporters. Assay artifacts are separated from true hits by determining whether the compound of interest appears as a hit in both reporters or, in the case of an artifact, stabilizes only one luciferase reporter.

3.4. Complementation assays and their modifications

Luciferase complementation assays (LCA) involve splitting the luciferase coding sequence into N- and C- terminal fragments that are appended to proteins of interest so that when the target proteins interact, the luciferase fragments reconstitute into active enzyme that generates bioluminescent signal (Figure 2). By employing a gain-of-signal luminescence approach to monitor protein interaction, the assay has a high linear range and low background signal, making LCA suitable to high-throughput screening that requires a large signal to background to enable high degrees of separation in the protein-protein interaction (PPI). Complementation systems like NanoBit (NanoLuc split into two separate pieces:1.3 kDa-size small BiT (SmBiT) and 18 kDa-size large BiT (LgBiT)) enable quantifiable PPI measurements without cell lysis, a clear advantage over co-immunoprecipitation and other lytic PPI assays. The Umezawa group was the first to demonstrate the principle of luciferase complementation assays, monitoring dose-dependent interactions of insulin with phosphorylated insulin receptor substrate 1 and the N-terminal SH2 domain of PI 3-kinase in Chinese hamster ovary cells [35]. For more information, refer to previous review and methods articles detailing the LCA workflow [36–39]. Since then, the use cases for luciferase complementation in probing protein-protein interactions have expanded, recently being used to characterize and screen for inhibitors of BRAF dimerization, NOX4 and p22 heterodimerization, and compounds that inhibit the assembly of voltage-gated sodium channels [40–42].

Complementation-based methods have been effective in studying SARS-CoV-2 targets in support of ongoing pandemic drug discovery efforts. One effective strategy using LCA to screen for SARS-CoV-2 3CL protease inhibitors involves attaching LgBiT to SmBiT with a linker that contains a 3CL protease cleavage site, creating a loss-of-signal assay that inversely correlates luminescent signal with protease activity [43,44]. Additional complementation bioreporters have also been designed to interrogate the interaction with the SARS-CoV-2 Spike protein and ACE2 ectodomain, identifying a group of lectins with antiviral therapeutic effect in vitro [45].

The split luciferase approach combined with bioluminescence resonance energy transfer (BRET, for more details about BRET see next section) has also been used to develop unique molecular ‘sensors’ with marked advantages to their fluorescent counterparts. In one such approach, a construct fusing LgBiT to two SmBiT with differing affinities to LgBiT was used to create a bioluminescent Zn²⁺ sensor that show vast improvement over existing fluorescent-based sensors [46]. Michielsen et al. attached a red Cy3 fluorophore to the higher affinity SmBiT (SB2) resulting in a high BRET efficiency in the absence of Zn²⁺. Only after binding of Zn²⁺ is the interaction of SB2 to LB interrupted, allowing the lower-affinity SB1 to complement with LgBiT and change the emission from red to blue wavelengths.

The luciferase complementation scheme has also been applied to thermal shift assays used in biophysical screening for compound inhibitors. In their original methods paper, Martinez et al. describe cellular thermal shift (CETSA) approach wherein the SmBiT component is fused to a target of interest and used as a detection method for remaining soluble protein [47]. When the LgBiT and the substrate are added after the CETSA thermal challenge, the SmBiT tag on the remaining soluble protein will complement with LgBiT and luminesce, providing a direct correlative readout of protein levels. This homogenous assay has provided a rapid, function-agnostic screening platform that is miniaturized to 1536-well format, already used by other labs utilizing a CETSA platform with a split NanoLuc approach (SplitLuc CETSA) [48–53].

3.5. Bioluminescence resonance energy transfer assays

Bioluminescence resonance energy transfer (BRET) is a process where energy transfer occurs from a bioluminescent donor to a fluorescent acceptor. Interestingly, the green fluorescent protein (GFP), isolated by Shimomura and colleagues in early 1960s from the Aequorea jellyfish, emits green light in those jellyfishes by absorbing the excited state energy of the luciferase aequorin, which itself catalyzes the oxidation of the substrate molecule coelenterazine triggering the chemiluminescence process [54].

Energy transfer events from a donor molecule to an acceptor molecule are the fundamentals of proximity assays widely implemented for measuring interactions between biomolecules (protein-protein, protein-DNA, protein-RNA), post-translational modifications, protein-ligand interactions/target engagement, and levels of analytes (e.g., hormones) or protein of interest. Many proximity assays used in HTS are homogenous assays, where the assay components are mixed, and the signal is acquired without any separation/washing steps resulting in shortened assay time, swift assay miniaturization, easy automation, and increased precision. Förster/fluorescence resonance energy transfer (FRET), time-resolved FRET and BRET are commonly used technologies in homogenous proximity assays.

BRET has certain advantages over FRET that are necessary conditions when deciding upon a reporter strategy. Perhaps the biggest advantage luminescence technologies have over fluorescence-based methods is the lack of an excitation light source requirement, which results in less interference from background fluorescence and no effect from photobleaching. However, due to the weak signal and limited dynamic range of the first BRET assays, the technique has not been widely applied.

There have been multiple attempts to address the limitations of BRET technologies. The original BRET pair, so-called BRET1, utilizes a RLuc-coelenterazine combination (emission maximum at 475–480 nm) as donor and a yellow fluorescent protein (YFP, emission maximum at ~530 nm) as the acceptor. The major drawback of BRET1 is the poor separation between the donor and acceptor emission spectra that results in a low signal-to-noise ratio and poor sensitivity. To increase the spectral separation, BRET2 was developed in which the native RLuc substrate was replaced with a synthetic coelenterazine analog, CTZ400A, which blue-shifts the luminescent maxima to around 400 nm when oxidized by RLuc. A. victoria GFP mutants, GFP2 or GFP10, are used as BRET2 acceptors, which have an emission maximum at 510 nm. While the spectral shift enables better distinction between donor and acceptor wavelengths, the blue-shifted donor has very low quantum yield and a fast signal decay. Attempts to improve this involve the utilization of an RLuc mutant, RLuc8, that provides higher light intensity, but the short signal lifetime remains [55].

The development of the NLuc luciferase and its furimazine substrate led to a novel BRET platform, termed NanoBRET, created in part to address the BRET assay’s prior limitations [56]. The application of the HaloTag (HT) as an alternative for fluorescent proteins as acceptor even further advanced the NanoBRET system. HT consists of a 33 kDa protein that can be genetically fused to a protein of interest and a chloroalkane ligand that is covalently bound to the small protein. The chloroalkane ligand can bind to a range of molecules, including fluorophores [57]. This allows the use of different fluorophores in wider spectral range. NanoBRET has immensely broadened the application of BRET assays in drug discovery campaigns. The advances and applications of NanoBRET technology were recently reviewed by Dale and colleagues [58].

NanoBRET has been successfully adopted for HTS to aid in the discovery of protein-protein interaction (Figure 2) and kinase inhibitors, and, lately, for PROTAC development [57,59,60]. In contrast to the traditional BRET technologies, NanoBRET is often utilized for target engagement assays [61–63]. In these assays, NLuc is fused to the target protein and the ligand, or a tracer molecule, is linked to the acceptor fluorophore. A luminescent signal is induced after the ligand binds to the target protein, or signal is disrupted when the ligand, which has a higher affinity to the target, replaces the tracer. The small size of NLuc and its suitability for tagging extracellular protein domains allowed to fuse it to different target proteins, which were previously hard to target. Consequently, NanoBRET has become one of the preferred methods to study GPCRs and screen for modulators of their function [61]. NanoBRET target engagement assays with a tracer are valuable not only for the validation and optimization of already identified hits but are also routinely used for hit discovery in HTS campaigns. Recently, an efficient tracer discovery workflow was developed and tested on GPCRs that utilizes machine learning-guided in silico screening for scaffolds displaying target binding with a blend of synthetic strategies to rapidly generate tracer candidates [64].

3.6. Amplified Luminescent Proximity Homogenous Assay (Alpha) technologies

Another widely used proximity assay is based on non-enzymatic chemiluminescence, termed Amplified Luminescent Proximity Homogenous Assay (Alpha). Initially developed by Ullman and colleagues, Alpha technology uses the channeling of singlet oxygen species generated from a photo-excitable donor bead to an acceptor bead when in close proximity to induce a chemiluminescent signal (Figure 2) [65]. Trademarked by PerkinElmer, Alpha technology is available as AlphaScreen and AlphaLISA, with both systems using the same donor beads but acceptor beads with different emission spectrum. An irradiation at 680 nm induces the donor beads to generate singlet oxygen which can diffuse up to 200 nm, a considerable difference to other technologies such as BRET or FRET and their proximity limits of ~10 nm [65,66]. The acceptor beads of AlphaScreen emit over a wide range from 520 nm to 620 nm, whereas AlphaLISA beads emit at 615 nm. The narrow emission spectrum of AlphaLISA beads results in improved sensitivity due the reduced interference with sample components in the assaying system.

To generate a recognition pair that can be applied to a screening campaign, the surfaces of the beads are conjugated to the assay specific molecules such as streptavidin, biotin, analyte specific antibodies, protein A, or other widely used protein affinity tags. These conjugations then allow highly specific and tight binding of the beads to the respective target molecules. The flexibility in bead conjugation, very high assay signal, low background, convenient homogenous assay format and excellent stability of the reagents make the Alpha technology well suited for high-throughput screenings during early drug discovery campaigns. AlphaScreen and AlphaLISA has been successfully applied to target GPCR family members, proteases, kinases and other enzymes involved in posttranslational and epigenetic modifications, and interactions between biomolecules (reviewed in [67]). Due to its high adaptability to different types of biomolecules, Alpha technologies has been applied for biotherapeutics development as well [68–70].

One of the newest applications of Alpha is in combination with the CETSA methodology to improve throughput [71]. CETSA, in its original format, is a low throughput technique reserved for the downstream validation of chemotypes, attributable to the multiday detection process by Western blotting. The main advantage of Alpha-HT-CETSA over the other types of HT-CETSA is the measurement of the endogenous proteins without the need for genetic modulations to introduce reporters, such as in SplitLuc CETSA. It also offers an alternative to mass spectrometry-coupled CETSA when expertise and/or access to mass spectrometry is limited. The application of Alpha-HT-CETSA on different targets is reviewed recently by Henderson et al. [72].

One of the limitations of Alpha platform is the light sensitivity of the donor beads, which restricts repeated signal readouts. Additionally, AlphaScreen and AlphaLISA can interfere with compounds which can quench singlet oxygen or absorb light. The standard TruHit counter screen can identify singlet oxygen quencher and light absorbing library compounds. Some compounds could disrupt the affinity capture system of the beads. Specific counter assays where the capturing entities can be used by itself are highly useful to minimize the false positive rate. For example, when a protein-protein interaction is measured by using poly-Histidine tag on one protein and GST-tag on the other protein, His-tagged GST should be used in the counter screen. The Alpha-HT-CETSA can be applied only when the binding of a ligand to the target protein induces detectable shift in the thermostability of the target. Additionally, a conformational change of the target upon ligand binding could lead to a loss of antibody recognition. Therefore, careful selection of antibodies and proper preliminary control experiments should be performed.

4. Luminescence microscopy of single cells

Bioluminescence and chemiluminescence technologies have been extensively used for in vivo imaging in past decades [73]. In cell-based assays, outlined in above passages, the luminescence signal is detected (often captured by CCD or CMOS cameras) from the entire cell population in a well. The imaging of cell structures, molecular processes, protein activities, and chemical entities at a single cell level is dominated by fluorescence-based microscopy. Despite wide success, photobleaching, autofluorescence, light scattering and phototoxicity caused by the external light source excitation are still major drawbacks of fluorescence microscopy. The usage of chemiluminescence or bioluminescence could overcome these issues.

Electrochemiluminescence (ECL), chemiluminescence triggered by electrochemical stimulus, is one of the emerging microscopy techniques[74]. It offers superior spatiotemporal control, very low background, and higher sensitivity and speed over fluorescence-based measurements. Recently, ECL microscopy was successfully used to image chemical entities such as glucose, cholesterol, reactive oxygen species (ROS) in single cells or cellular structures such as plasma membrane and focal adhesions [75–79]. For example, Liu and colleagues measured the effect of diphenyleneiodonium (DPI), an NADPH oxidase and iNOS/eNOS inhibitor, on ROS production at the single-cell level [50]. The secretion of hydrogen peroxide from the cells upon stimulation with N-formylmethionyl-leucylphenylalanine was detected through luminescence signal generated by the ECL reaction with luminol (L012) on the chitosan and nanoTiO₂ modified fluoride-doped tin oxide conductive glass. The DPI-treated cells showed significantly decreased hydrogen peroxide secretion.

Bioluminescence microscopy based on the detection of light emitted by cells expressing a luciferase gene was utilized by different groups starting in the 1990s [80,81]. However, dim photon output resulted in blurry images and very long image acquisition made tracing of molecular events in real-time difficult.

The application of short focal-length imaging lens and confining the samples in a dark box enabled Ogoh et al. to image cellular structures at micrometer scale [82]. They were also able to spectrally distinguish the luciferase-luciferin systems with different spectral properties. However, the quantum yield and spectra of luciferases are temperature and pH sensitive as proved in multiple in vitro experiments. The same changes occurred when different beetle luciferases were expressed in U2OS cells and imaged at different temperatures. Out of the tested luciferases, the emission wavelength and intensity only of NLuc remained constant.

Suzuki et al. took another approach to advance the application of luminescence microscopy by generating five new spectral variants of NLuc, termed enhanced Nano-lantern (eNL) [83]. NLuc is the brightest luciferase designed so far and as demonstrated by Ogoh et al. the spectra and quantum yield are insensitive to temperature changes, but it can be used to trace only single molecule or event at once in a cell. By harnessing BRET, they fused NLuc with five different fluorescent proteins. The eNL were used as tags to marker proteins of subcellular structures, and various cellular compartments such as cytosol, nucleus, mitochondria, inner plasma membrane, peroxisome, lysosome, nucleoli, and clathrin-coated pits were imaged at high resolution. By optical filtering and spectral unmixing five different cellular structures could be imaged simultaneously and were clearly distinguishable. Moreover, they developed an eNL-based Ca²⁺ indicator and measured the effect of astemizole, a hERG channel blocker, on human iPSC-derived cardiomyocytes over time.

Another exciting development in field of bioluminescence microscopy was reported while we were working on this article. Yao et al. utilized phasor analysis, a method used to distinguish spectrally similar fluorophores, and were able to separate signals from luciferase-luciferin pairs that cannot be resolved by color alone [84]. They further applied this method to BRET reporters mentioned above and performed longitudinal multiplexed imaging in tumor spheroids over an extended period in real-time using a single dose of luciferin.

5. Conclusion

The dominant detection methodology for developing high throughput assays in early-stage drug discovery has been fluorescence-based, owing in part to its robustness and ease of use. Fluorescence assays, however, are not necessarily the best performing and will typically present lower sensitivity than chemiluminescence methods. Chemiluminescent assays are not without weakness, themselves being very sensitive to experimental conditions, so their applicability in complex systems is very limited. Bioluminescent systems introduce a degree of environmental robustness emission with the signal emission occurring in an enzyme active site under biological conditions. Bioluminescence-based technologies are expected to play an essential role in future drug discovery paradigms, from hit identification to lead validation, owing to inherent advantages in the enzymatic light-producing reaction. Over the years, luciferases and their substrates have gone through multiple iterations for various applications in early-stage drug discovery. While the firefly luciferase protein has enjoyed immense popularity as a wildtype reporter, recent efforts have focused on improving specific functions or properties of nearly every luciferase described.

6. Expert Opinion

The challenge of target recognition in the complex cellular environment calls for the discovery of more specific probes, necessitating the development of robust and sensitive reporters that can easily adapt to a variety of targets. The application of bioluminescence reporters to early-stage drug discovery campaigns has seen steady increases in utilization since the first description of firefly luciferase, and that trend is likely to only accelerate as continued improvements to the luciferase enzymes and related substrates match advances to high-throughput screening architecture. There are a number of roadmaps to improving luciferase enzymes from their native form, with examples stemming from the decades of incremental improvements to P.pyralis luciferase to the engineering of NanoLuc luciferase.

There is a constant demand for more bioluminescent colors, improved enzymes, and more biocompatible substrates, creating competition with existing assay formats and reporter systems to continue driving improvements to luminescent reporter systems. Continued improvements to luciferase and luciferin pairs will follow that seek to output longer wavelength maximums red-shifted beyond compound autofluorescence wavelengths, better penetrate animal tissues, or that can be co-expressed with alternative luciferases, to produce light using synthetic luciferins adapted to specific needs, to change the signal flash kinetics, and to increase luminescent yield so that assays can continue to miniaturize as reporter sensitivity and robustness improve.

These improvements to the luciferase enzymes and their substrates are almost always followed with a new application or experimental design, much like the application of NanoLuc to nanoBRET and beyond. It is difficult to predict the future applications of luminescent technologies as it will depend on the continued development of the technology and the needs and interests of researchers in various fields. Even so, it’s perfectly reasonable to expect these technologies to afford improved imaging of cellular structures and processes at the single-cell level, enabling more detailed and accurate studies of biological phenomena. The expanded use of luminescence technologies like BRET in high-throughput screening assays is likely to identify and optimize new drug candidates more efficiently, as well as better enable the study of different diseases and disorders, including cancer and neurodegenerative conditions, to gain new insights into their underlying mechanisms and potential treatments.

The future of chemiluminescent probes, which offer significantly improved signal and sensitivity over traditional fluorescent probes, is likely to involve significant development and application of these tools to new and hard-to-assay targets. A great example of the power of chemiluminescent probes is found in assays monitoring alterations of the interior lysosomal pH, normally done with pH-sensitive fluorescence probes, such as the commercial LysoTracker kit. These pH-sensitive fluorescent probes are known to be susceptible to aggregation-caused quenching effects, photostability issues, and narrow Stokes shifts. To overcome these limitations, Wu et al. developed an efficient lysosomotropic screening platform using an iridium-based chemiluminescent probe to monitor lysosomal activity in living cells [85]. In another example of designed chemiluminescent probes, An et al. present the first tracer for monitoring reactive azanone transport in both real-time and in living systems [86].

It’s important to note the roadblocks that exist in implementing luciferase-based methodologies, especially with more specialized assays like chemiluminescent microscopy. Techniques like these come with higher barriers to entry in the form of specialized instrumentation, technical experimental design, and increased data processing demands. Of course, the balance between these elements is up to individual project goals and needs. For instance, while bioluminescence microscopy is outmatched by fluorescence microscopy methods for most targets, luminescent methods may be used when a target precludes the use of fluorescent tracers.

These roadmaps on the development of luminescence technologies are likely to catalyze new platforms and screening modalities that will continue to drive potential treatments for a wide range of diseases and disorders.

Article Highlights:

• The bioluminescent reactions conserved across the animal kingdom have seen extensive use in the early-stage drug discovery field.

• Luminescence technologies, including fluorescence, bioluminescence, and chemiluminescence, have become essential tools in biomedical research and automated high-throughput screening.

• Advances in the luminescent output, structure, and function of both the luciferase enzymes and their substrates have expanded the use cases for luminescence in assaying target activity, dynamics, and interactions.

• Recent advancements in electrochemiluminescence and bioluminescence microscopy have overcome challenges, like dim photon output and long image acquisition times, leading to the successful imaging of cellular structures at the single-cell level.

• The future of chemiluminescent probes, which offer significantly improved signal and sensitivity over traditional fluorescent probes, is likely to involve significant development and application of these tools to new and hard-to-assay targets