Protein Translocation Assays to Probe Protease Function and Screen for Inhibitors

Abstract

In this chapter, you will learn how to use translocation biosensors to investigate protease functions in living cells. We here present modular protein translocation biosensors tailored to investigate protease activity and protein-protein interactions. Besides the mapping of protease function, the biosensors are also applicable to identify chemicals and/or (nano)materials modulating the respective protein activities and can also be exploited for RNAi-mediated genetic screens.

1 Introduction

Fluorescent protein biosensors are powerful tools for dissecting the complexity of cellular processes. As cellular biosensors have the advantage of acting in a physiological and/or pathophysiological environment, such as cancer cells, these are used to define the dynamics of cellular regulation, especially when combined with automated multi-parameter imaging technologies [1]. The increase in the use of cell-based assays during all major steps of drug discovery and development has increased the demand for cellular biosensors. Such biosensors are expected to allow the detection of a wide variety of signaling molecules and bear the potential for novel assay applications. Intensifying the use of kinetic, in contrast to snapshot drug screening assays , is expected to reveal subtle, but discrete effects of compounds, aiding the interpretation of their mode of action and leading to an improved understanding of key regulatory cellular pathways, based on “functional cellular responses” [2].

Redistribution approaches, a cell-based assay technology that uses protein translocation as the primary readout, have great potential to study the activity of cellular signaling pathways and other intracellular events. Protein targets are labeled with (different) autofluorescent proteins (e.g., the green fluorescent protein—GFP), and the assays are read using high content or high-throughput microscope-based instruments [2]. Protein translocation assays have the potential for the profiling of lead series, primary screening of compound libraries, or even as readouts for gene-silencing studies. However, any realistic applications of high-content and high-throughput cell-based assays critically depend on robust and reliable biological readout systems with a high signal-to-noise ratio. Hence, the spatial and functional division into the nucleus and the cytoplasm marks not only two dynamic intracellular compartments vital for the cell but that can also be easily distinguished by microscopy and thus, be exploited for translocation biosensor assays.

We here present two cell-based translocation-based biosensor-systems to investigate protease activity and protein-protein interaction networks (Fig. 1, Table 1) in living cells.

Fig. 1

Modular composition of the biosensors. The biosensor proteins consist of functional modules, tailored to meet the biosensors’ specific applications. The biosensor backbone is composed of NLS, GFP, GST, and NES. For the CB, a protease cleavage site (PCS) and a Myc-epitope is integrated between GST and the NES. If desired, the protease of interest (POI, shown in red) can be ectopically coexpressed as a fusion with mCherry. The PB system consists of two components. Molecule I is based on the backbone, containing in addition binding partner (BP) 1. BP2 is expressed as a nucleolar fusion protein containing the RevM10 protein and BFP (shown in blue)

Table 1 Overview of the different biosensors

The biosensor “prototype” is composed of functional modules, tailored to meet the specific biosensors’ applications (Fig. 1, Table 1). These modules are autofluorescent proteins, transport signals, and gluthatione S-transferase (GST). As such, all biosensors can be visualized by fluorescence microscopy in living or fixed cells. Their intracellular localization is controlled by a rational combination of opposing transport signals, namely nuclear localization signals (NLS) and nuclear export signals (NES), mediating nuclear import and export, respectively. Integration of GST increases the molecular weight up to >55 kDa, preventing passive diffusion. In the following paragraphs, we will describe the composition and applications of the different biosensors.

1.1 Protease Cleavage Biosensor (CB)

In any physiological or patho-physiological state, numerous proteins are being processed or degraded in a highly controlled fashion [3]. Intrinsic hydrolytic cleavage is performed by proteases, playing critical roles in innumerable biological processes. As protease signaling is mostly irreversible, all proteases are strictly regulated. Consequently, protease deregulation often leads to patho-physiological states that in principle could be medicated by specific protease inhibitors or activators [3, 4]. Proteases are therefore important drug targets in the pharmaceutical industry as well as potential disease markers [3, 4].

Hence, our translocation -based biosensor-systems (CB) (Fig. 1, Table 1) can be used to identify chemicals that modulate the proteolytic activity of a protease of interest (POI) [1], map the POI’s cleavage-recognition site, and test potential POIs’ substrates and cofactors [5]. The CB is based on the transport biosensor, but is additionally equipped with a POI-recognizable cleavage site (PCS) recognized by the protease of interest preceding the NES. Upon processing of the biosensor by the POI, the NES is cleaved off and thus, the CB accumulates in the nucleus due to its NLS activity. Hence, cytoplasmic-nuclear translocation is indicative for protease activity in living cells [5]. An example for such an application is represented by the combination of the protease Taspase1 with its cleavage site derived from its target, the Mixed Lineage Leukemia (MLL) protein [6]. Integration of an additional Myc-epitope tag further allows detecting the cleaved NES by immunofluorescence staining using a tag-specific antibody (Fig. 1). The CB can be used to study both, endogenous or ectopically expressed proteases, by co-transfection of the CB- and the protease-encoding plasmids. To identify cells expressing the POI, we recommend using a fusion construct containing the red fluorescent protein, mCherry. Hence, one can identify cell expressing both, the biosensor and the POI.

1.2 Protein-Binding Biosensor (PB)

Protein-protein interaction networks are critical for the majority if not for all cellular events, and require distinct contact areas of both binding partners. Interfering with protein-protein interactions (PPIs) via enforced expression of dominant-negative mutants and/or application of small molecules has emerged as a promising, though challenging strategy for human therapeutics [7–9]. For a long time it has even been thought that small molecules are not feasible to target PPIs. However, this view changed over the last few years, and interest in pharmaceutical applications rose [10]. A prominent example is the interaction of p53 and mdm2 [11, 12]. If their binding is inhibited, activated p53 can accumulate in the nucleus, triggering apoptosis-mediated (tumor) cell death. Albeit most efforts focused on the inhibition of protein interactions, currently stabilization of PPIs is considered an alternative but promising approach to target disease-relevant signaling pathways [8, 13, 14]. Numerous biochemical in vitro and several cell-based methods have been developed for detecting and studying PPIs [14, 15]. As most of those are laborious and time consuming, the screening of large compound collections for PPI-modulators is still an exception.

We here present a two-component translocation -based biosensor-system (PB) (Fig. 1, Table 1) to test PPI and/or to identify compounds interfering with PPI . Molecule I is based on the green fluorescent biosensor, allowing the integration of binding partner 1 (BP1) N-terminal of the NES. The “binding partner” may represent a complete protein or distinct domains thereof. The second molecule (II) consists of binding partner 2 (BP2), fused to a nucleolar, export-deficient HIV-1 Rev protein (RevM10) [16], and the blue fluorescent protein (BFP). Although molecule I is continuously shuttling between the nucleus and the cytoplasm, it predominantly localizes to the cytoplasm due to the unequal strength of the transport signals (NES > NLS). In contrast, molecule II is anchored at the nucleolus, due to the nucleolar localization signal in the HIV-1 RevM10 protein [16]. Upon specific interaction between both binding partners in living cells, the shuttling molecule I will be entrapped in the nucleus, and thus, accumulate at the nucleolus (“nucleolar trapping”).

For the construction of biosensor derivatives for your demands, modules (NLS, NES, cleavage sites , BP) can be exchanged in the respective mammalian expression vectors (Fig. 2). These plasmids can be distributed upon request.

Fig. 2

Vector map of the CB expression plasmid. Schematics depicting the composition of the CB biosensor as well as relevant restriction sites. Transcription in eukaryotic cells is under the control of the CMV-promoter. Amp^R, ampicillin resistance; Neo^R, neomycin resistance; SP6 and T7, respective promoter sequencing primer sites

For high-throughput analysis, the generation of cell lines stably expressing the respective biosensors is of great advantage. For one, this reduces the need for biosensor plasmid transfection prior to perform screens. Additionally, stably expressing cell populations often display more homogenous biosensor expression levels, facilitating microscopic image capture and subsequent analysis. If available, positive and negative controls should be included to verify translocation biosensor-based results. For the CB system, biosensors with a mutated or unrelated protease cleavage site , and/or coexpression of a nonfunctional protease mutant are helpful negative controls [5]. Employing empty PB biosensor backbones, lacking the respective protein interaction partners, will uncover unspecific protein-protein interaction .

2 Materials

2.1 Generation of Expression Constructs

• 1. Vector backbone (can be requested, see Table 2).

  Table 2 Biosensor—backbone vectors

• 2. Oligonucleotides.

• For complementary annealing and subsequent cloning (NES, NLS, or PCS).

• For PCR and subsequent cloning (for POI or BP2).

• For sequencing (T7 and SP6).

• 3. Enzymes.

• HindIII, NheI, KpnI, EcoRI, NotI, XhoI, Ligase, T4 PNK.

• 4. Reagents.

• Universal-Agarose: VWR International.

• TAE buffer: 40 mM Tris–HCl, 1 mM Ethylenediaminetetraacetic acid (EDTA), pH 8.0.

• Ethidium bromide: 0.5 μg/mL.

• DNA-loading dye: 20% (w/v) glycerine, 100 mM EDTA, 0.25% (w/v) bromphenol blue, 0.25% (w/v) xylencyanol.

• 1 kb DNA ladder: NEB.

• deoxyribonucleotides (dNTPs): 10 mM.

• distilled water (aqua dest.).

• QIAquick Gel Extraction Kit: QIAGEN.

• Takara DNA Ligation Kit Ver. 2.1: Clontech Laboratories.

• 5. Equipment.

• DNA electrophoresis system.

2.2 Transfection and Cell Sorting

• 1. Reagents.

  • Lipofectamine^® 2000 (transfection reagent): Invitrogen.

  • OptiMEM (reduced serum medium): Gibco, Invitrogen.

  • Dulbecco’s Modified Eagle’s Medium (DMEM): Gibco.

  • G418 (Geneticin) 30.000 U/mL: Biochrom.

  • DMEM G418 (selection medium): 1.6% (480 U/mL) G418 in DMEM.

  • Phosphate-buffered saline (PBS): Gibco.

  • Fetal calf serum (FCS): Gibco.

• 2. Equipment.

  • Cell Sorter (e.g., FACS-Calibur (BD Biosciences, Erembodegem, Belgium).

  • Fluorescence microscope (e.g., Axiovert 200 M; Carl Zeiss, Jena).

  • Laminar flow.

  • Sterile pipets.

  • Sterile cell culture flasks and plates.

  • Cell filter.

2.3 High Content Screening (HCS)

• 1. Reagents.

• PFA (paraformaldehyde) 4% in PBS: USB, Cleveland, USA.

• Hoechst 33342 25 μg/mL in PBS: Sigma-Aldrich.

• 2. TritonX 0.1% in PBS: Sigma-Aldrich.

• 3. Equipment.

• High content fluorescence microscope (e.g., ArrayScanVTI; Thermo Fisher Scientific, Waltham, USA).

• Plate washer and dispenser.

• Electronic multichannel pipets.

• Pipetting robot (e.g., Biomek^® NXp; Beckman Coulter GmbH).

• Sterile pipets.

• Sterile cell culture flasks.

• Cell culture plates with thin bottom allowing microscopic imaging.

3 Methods

3.1 Generation of Expression Constructs

3.1.1 Preparation of Vector Backbones

For inserting any sequence of interest, digest the vector backbones with the restriction enzymes as depicted in Table 3 (see Note 1 ) for 90 min at 37 °C.

Table 3 Restriction enzymes for vector preparation

3.1.2 Elution of DNA and GST-Fragment

• 1. Separate the restriction reaction by agarose gel electrophoresis.

• 2. Bind the vector DNA on a silica membrane in a high-salt buffer.

• 3. Wash with a washing-buffer.

• 4. Dry the membrane by a short centrifugation step.

• 5. Elute the vector DNA with a low-salt buffer or water.

• 6. After KpnI digestion of the vector, the GST-fragment will also be released. Collect and elute this fragment for re-integration later on.

3.1.3 Annealing of Complementary Oligonucleotides for Inserting NES/NLS/PCS

3.1.4 Preparation of Oligonucleotides

• 1. The NES and NLS and the PCS of the CB can be replaced by complementary DNA oligonucleotides containing the respective restriction sites (see Fig. 2). For the CB, subsequently the Myc-epitope can be inserted into the XhoI restriction site using complementary oligonucleotides, if desired.

• 2. Order the oligonucleotides with the restriction sites as depicted in Table 4.

  Table 4 Restriction sites for oligonucleotides to insert NES/NLS/PCS via oligonucleotide annealing

• 3. Phosphorylate the oligonucleotides at their 5′ ends. For each oligonucleotide, preparation of the ingredients listed in Table 5 has to be incubated for 15 min at 37 °C.

  Table 5 Oligonucleotide phosphorylation

3.1.5 Annealing of Oligonucleotides

• 1. Combine both phosphorylated complementary oligonucleotides in an aerosol-tight reaction tube.

• 2. Insert this tube in a container (cup) with boiling water and leave it on the bench until the water temperature decreased to room temperature (see Note 2 ).

• 3. Use 1 μL of the annealing product for ligation into the linear vector backbone.

• 4. Check proper integration by sequencing.

3.1.6 Integrating PCR-Amplified Sequences of BP1 and BP2

BP1 and BP2 can be replaced by inserting PCR-amplified sequences containing the restriction sites as mentioned in Table 6 (see Fig. 3).

Table 6 Restriction sites flanking amplification products for BP1 and BP2

Fig. 3

Vector maps of PB I and II. Schematics depicting the composition of the respective biosensors as well as relevant restriction sites. BP I—left, BP II—right. Transcription in eukaryotic cells is under the control of the CMV-promoter. Amp^R, ampicillin resistance; Neo^R, neomycin resistance; SP6 and T7, respective promoter sequencing primer sites

3.1.7 Ligation of PCR Amplicons into the Biosensor-Backbone

• 1. Ligate digested PCR amplicons into linear biosensor-backbone. For ligation mix the biosensor-backbone with DNA in a ratio of 0.5:2.

• 2. Incubate for 2 h at 16 °C.

• 3. Check integration by sequencing.

3.2 Generation of Cells Stably Expressing the Biosensors

For high content screening , the generation of cells stably expressing the biosensors is of great advantage (see Note 3 ). This will suspend the use of freshly transfected cells with high differential expression intensity. Procedure of screening will be shortened, as there is no need for transfection. Also the time required for imaging each will decrease due to the equalized expression level of stable cell lines. Cells stably expressing any biosensor will be named “biosensor cells” from here on. Steps for generating biosensor-cells are listed in Table 7.

Table 7 Steps for generating biosensor-cells

3.2.1 Examine Biosensor-Cells

Check cells in a fluorescent microscope for proper localization of the biosensor (see Note 5 ). Use a minimum of 40× magnification. GFP is detectable at 475/515 nm and BFP at 365/445 nm. For PB, both molecules have to be cotransfected (see Notes 6 and 7 ).

3.3 High Content Screening Assays Using Biosensors

3.3.1 Establish Cell System in 384-Well Plates (Prescreen)

Seed biosensor-cells into 384-well plates (see Note 8 ). If using HeLa-cells, seed approximately 9000 cells in 50 μL of medium. For other cell lines, the optimal seeding density has to be established.

3.3.2 Treatment and Preparation of Seeded Biosensor-Cells

• 1. If available, treat half of the wells with positive control, i.e., cleavage inhibitor for CB (see Note 9 ) or PPI -inhibitor for PB (see Note 10 ).

• 2. After incubation with the positive control, fix the cells with PFA, permeabilize with PBS 0.1% TritonX, and stain nuclei with PBS/Hoechst 33342 (25 μg/mL), as described in Table 8.

  Table 8 Steps for high content screening using the biosensors

3.3.3 Microscopy of Treated Biosensor-Cells in 384-Well Plates

• 1. Analyze the cellular localization of the biosensors using a high content microscope. Use the respective software for evaluating the intracellular localization of the biosensor. For ArrayScan VTI, this is “Molecular Translocation” or “Cytoplasm to Nucleus Translocation .”

• 2. If positive controls were performed, calculate Z’-value using the following formula [2]:

$$ {\mathrm{Z}}^{'}=1-\left(3*\left[ S{D}^{+}+ S{D}^{-}\right]/\mathrm{R}\right) $$

For screening, the Z’-value should be 0.3 or higher.

3.3.4 High Content Screening (HCS)

• 1. Generate an appropriate amount of biosensor cells for seeding the required number of plates. Routinely check biosensor fluorescence of your cell population.

• 2. Steps for high content screening using the biosensors are listed in Table 8.

• 3. Calculate Z’ for every plate.

• 4. Identify hit compounds and retest them.

• 5. If possible, test in varying concentrations.

• 6. Evaluate validity of your novel inhibitors in further experiments (dose response curves) .

3.3.5 Optional Additional Antibody Staining

If desired, use antibody stains as described in Table 8 to detect additional targets in the cells. Thereby, it is possible to, e.g., simultaneously detect apoptotic effects.

3.4 Exemplary Result

In Fig. 4 an exemplary result for the use of CB-expressing cells to screen POI activity in HCS is shown. In this case, the CB localizes predominantly to the cytoplasm, but is continuously shuttling between the nucleus and the cytoplasm due to a nuclear export and localization signal (see Fig. 1). Co-transfection of the CB and the active POI results in its proteolytic cleavage and loss of the NES. This leads to nuclear accumulation of the fluorescent signal. In contrast the co-transfection of an inactive variant of the POI does not change the localization of the CB.

Fig. 4

Exemplary result for HCS of POI activity in CB-expressing cells. The co-expression of the CB (GFP) and the POI (mCherry) leads to a translocation of the CB in the nucleus (shown in blue after Hoechst 33342 staining) when the POI shows activity concerning cleavage. In the case of an inactive POI the CB stays predominantly in the cytoplasm

4 Notes

• 1. It is of advantage to dephosphorylate the vector ends to prevent self-ligation of the vector.

• 2. For proper annealing, insert the mix of both phosphorylated oligonucleotides directly into the tube inserted into the container (cup) containing the boiling water. Store annealed oligonucleotides at 4 °C.

• 3. Use adherent flat growing cells for high-quality microscopic images and optimal image analysis. They are easier to evaluate due to their more prominent cytoplasmic region.

• 4. Directly add G418 the day following transfection and perform first round of cell sorting as soon as possible. Do not cultivate cells too long without using them for the assay. Freeze cells as early as possible for preserving biosensor expression.

• 5. Validate the localization of the biosensor in cells with different expression levels. It may be necessary to choose the cells with a medium expression level, as too strong overexpression may cause mis-localization of the biosensor. In this case, use biosensor cells with lower expression level.

• 6. As internal transport signals of the BP may interfere with proper localization of the PB molecules, these have to be checked prior to cloning. Experience shows that it is advantageous to integrate both BPs in both PB (molecule I/II) backbones and check for their intracellular localization [8].

• 7. Cloning of the protease into a different vector backbone could be of advantage by allowing double-selection together with another antibiotic (e.g., puromycin, blasticidin).

• 8. Establish cell seeding well. Do not shake freshly seeded plates (cells will not adhere in the middle of the well or be equally distributed). Use plate lids with condensation rings. Plates should be optimized for microscopic imaging of cells. Test them before by staining cells with PBS/Hoechst 33342 (25 μg/mL).

• 9. Using cells with intrinsic POI activity might result in nuclear accumulation of the CB, dependent on the protease activity. Once processed, the cytoplasmic localization is solely generated by expression of new CB. Thus, the incubation time with the compounds has to be properly established to generate a significant CB localization difference. The CB can also be used to test cell lines for expression of POI.

• 10. Whenever possible, include positive and negative controls on every assay plate. Carefully establish the optimal incubation time for the control compound to obtain a maximum signal difference.

• 11. Do not use polystyrene plates or pipette tips because of their hydrophobic surface. Charged compounds might adhere to the polystyrene and thus, may not be available for the assay.

• 12. Take care of proper mixing the compounds with medium and not to detach the cells during this process.

• 13. Adjust washer and dispenser to slowly remove/add solutions to an outer well area to protect integrity of the cell layer.