We present guidelines for developing synthetic ‘chemical transducers’ that can induce communication between naturally unrelated proteins. In addition, detailed protocols are presented for synthesizing and testing a specific ‘transducer’ that enables a growth factor to activate a detoxifying enzyme and consequently, to regulate the cleavage of an anticancer prodrug.
Signal transduction pathways, which control the response of cells to various environmental signals, are mediated by the function of signaling proteins that interact with each other and activate one other with high specificity. Synthetic agents that mimic the function of these proteins might therefore be used to generate unnatural signal transduction steps and consequently, alter the cell’s function. We present guidelines for designing ‘chemical transducers’ that can induce artificial communication between native proteins. In addition, we present detailed protocols for synthesizing and testing a specific ‘transducer’, which can induce communication between two unrelated proteins: platelet-derived growth-factor (PDGF) and glutathione-S-transferase (GST). The way by which this unnatural PDGF-GST communication could be used to control the cleavage of an anticancer prodrug is also presented, indicating the potential for using such systems in ‘artificial signal transduction therapy’. This work is intended to facilitate developing additional ‘transducers’ of this class, which may be used to mediate intracellular protein-protein communication and consequently, to induce artificial cell signaling pathways.
Signal transduction pathways play a significant role in virtually every cellular process and allow the cell to rapidly respond to environmental signals.1 These pathways are often triggered by the binding of a signaling molecule to an extracellular receptor, which results in activation of intracellular enzymes. Amplification and propagation of this signal within the cell is mediated by the function of signaling proteins that form a network of protein-protein interactions in which enzymes are reversibly activated with high specificity. Because dysregulation of these networks frequently leads to cancer development, there has been much interest in establishing 'signal transduction therapy of cancer',2 whereby drugs are designed to disrupt malignant signaling pathways. We have recently proposed an alternative approach to signal transduction therapy that relies on the ability of drugs to generate unnatural signal transduction pathways.3 In particular, we believe that by designing synthetic agents that mimic the function of signaling proteins, it would be possible to modulate the cell's function indirectly. For example, these artificial networks may enable protein biomarkers to activate enzymes that cleave prodrugs. Alternatively, these signaling protein mimetics might be able to activate unnatural cell signaling pathways, resulting in therapeutic effects.
To demonstrate the feasibility of this approach, we have recently created a synthetic 'chemical transducer'4 that enables platelet-derived growth factor (PDGF) to trigger the cleavage of an anticancer prodrug by activating glutathione-s-transferase (GST), which is not its natural binding partner. The structure of this 'transducer' consists of an anti-PDGF DNA aptamer that is modified with a bivalent inhibitor for GST. Hence, this synthetic agent belongs to a family of molecules with binding sites to different proteins,5-7 such as chemical inducers of dimerization (CIDs)8-10 and also to the group of protein-binders based on oligonucleotide-synthetic molecule conjugates.11-21
The general principles underlying the design of such systems is described herein and detailed protocols for synthesizing and testing the function of this 'transducer' with conventional enzymatic assays are provided. This work is intended to facilitate developing additional 'transducers' of this class, which may be used to mediate intracellular protein-protein communication and consequently, to induce artificial cell signaling pathways.
Figure 1 schematically describes the operating principles of synthetic 'chemical transducers' that can mediate unnatural protein-protein communications. In this illustration, a 'chemical transducer', which integrates synthetic binders for proteins I and II (binders I and II), enables protein II to trigger the catalytic activity of protein I, which is not its natural binding partner. In the absence of protein II, the transducer binds the catalytic site of the enzyme (protein I) and inhibits its activity (Figure 1, state ii). The binding of the 'transducer' to protein II, however, promotes interactions between binder I and the surface of protein II (Figure 1, state iii), which reduces its affinity toward protein I. As a result, the effective concentration of the 'free' transducer in the solution is reduced, which leads to dissociation of the transducer-protein I complex and to reactivation of protein I (Figure 1, state iv). Taken together, these steps highlight three fundamental principles underlying the design of efficient 'transducers': (1) a 'transducer' should have a specific binder for each of the protein targets, (2) the interaction between binder II and protein II should be stronger than the interaction between binder I and protein I, and (3) binder I must be able to interact with the surface of protein II. This last principle does not necessarily require that binder I alone would have a high affinity and selectivity toward protein II. Instead, it is based on our recent studies which showed that bringing a synthetic molecule in proximity to a protein is likely to promote interactions between this molecule and the surface of the protein.19,22,23
Figure 1: Operating principles of 'chemical transducers'. When the 'chemical transducer' is added to an active protein I (state i), it binds to its active site through binder I and inhibits its activity (state ii). In the presence of protein II, however, the unbound 'chemical transducer' interacts with protein II through binder II, which promotes interactions between binder I and the surface of protein II. This induced binder I-protein II interaction reduces the effective concentration of binder I, which leads to dissociation of the 'transducer'-protein I complex and to protein I reactivation (state iv). Please click here to view a larger version of this figure.
1. Synthesis of the 'Chemical Transducer'
2. Controlling GST Activity by PDGF
The design, synthesis, and mechanism of action of a 'chemical transducer' that can induce artificial communication between PDGF and GST are presented in Figure 2. The structure of the 'transducer' integrates a PDGF DNA aptamer and a bis-ethacrynic amide (bEA), which is a known GST inhibitor (Figure 2a).19 These binders enable the 'transducer' to bind both PDGF and GST with different affinities, namely, with dissociation constants (kd's) of 26 nM and 144 nM, respectively.4 In addition, according to this design, the binding to PDGF should induce non-specific interactions between the bEA unit and the surface of PDGF, which would markedly reduce the potency of the bEA inhibitor.19,22,23 Figure 2b illustrates the operating mechanism underlying this system. Upon adding the 'transducer' to an active GST, the two EA units simultaneously bind both active sites of this dimeric enzyme and inhibit its activity. In the presence of PDGF, however, a PDGF-'transducer' complex is formed, which prevents the bEA unit from inhibiting GST. This consequently leads to the dissociation of the GST-'transducer' complex and to GST reactivation. GST can then be re-inhibited by adding an unmodified PDGF aptamer that displaces the 'chemical transducer' and enables it to inhibit GST again.
The ability of PDGF to control GST activity was first demonstrated by measuring GST (10 nM) activity with and without the 'chemical transducer' (500 nM) and measuring the activity of the GST-'chemical transducer' complex in the presence of different concentrations of PDGF (250, 500, and 1,000 nM) (Figure 3a). After establishing that the 'chemical transducer' induces artificial PDGF-GST communication, we next established that this artificial communication, similar to the signal transduction steps, is also reversible and rapidly adapts to changes in the environment. Reversible activation/inhibition of GST was performed by sequential additions of PDGF and unmodified PDGF aptamer to the GST-'chemical transducer' complex (Figure 3b). The response of the system to real-time changes in its environment was evaluated by measuring GST activity while adding the different inputs. A rapid increase in GST activity was observed upon the addition of PDGF (750 nM) to the GST-'transducer' complex (10 nM and 500 nM, respectively) 3.5 minutes after adding substrates (Figure 4a). Similarly, a decrease in GST activity was observed upon the addition of PDGF aptamer (5 µM) to a mixture of GST (10 nM), PDGF (750 nM), and the 'chemical transducer' (500 nM) (Figure 4b).
Finally, we demonstrated the ability of such a system to control prodrug activation in response to changes in the environment. JS-K is an anticancer prodrug activated by GST to release toxic NO (Figure 5a). The amount of NO released upon the addition of JS-K (45 µM) to the 'chemical transducer' (750 nM) and different combinations of GST (10 nM) and PDGF (2 µM) were measured (Figure 5b), thus confirming that only the presence of both GST and PDGF will result in prodrug activation.
Figure 2. 'Chemical transducer' — synthesis and operating mechanism. (a) The 'chemical transducer' is composed of a PDGF aptamer, a bivalent ethacrynic amide (EA) GST inhibitor, a fluorophore (FAM), and a quencher (Dabcyl). This aptamer-inhibitor conjugate is synthesized by attaching an azide-modified ethacrynic amide (EA) derivative to a dialkyne-modified and fluorescently labeled DNA aptamer (ODN-1). Re-printed with permission from reference4. (b) The enzymatic activity of GST is inhibited by the 'chemical transducer', due to the binding of the EA groups at the enzyme's active sites. Addition of PDGF leads to the formation of the PDGF-'chemical transducer' complex, which disrupts the 'transducer'-GST interaction, therefore restoring the enzymatic activity. The following addition of an unmodified PDGF aptamer releases the 'chemical transducer' and allows it to re-inhibit the enzyme. Please click here to view a larger version of this figure.
Figure 3: PDGF-controlled GST activity. (a) GST (10 nM) enzymatic activity in the presence (–) and absence (–) of the 'chemical transducer' (500 nM), and in the presence of the 'chemical transducer' (500 nM) with increasing concentrations (250, 500, or 1000 nM) of PDGF (—). (b) Inhibition-activation cycles of GST enzymatic activity manifested by changes in the initial velocity (V0) in response to sequential additions (IIV) of PDGF and unmodified PDGF aptamer to the GST-'chemical transducer' complex: (I) none, (II) PDGF (750 nM), (III) PDGF aptamer (4 µM), (IV) PDGF (5 µM), and (V) PDGF aptamer (10 µM). The graph presents the mean ± stdev of triplicates. Re-printed with permission from reference4. Please click here to view a larger version of this figure.
Figure 4: Real-time control of GST activity. (a) Enhancement of GST enzymatic activity detected immediately after the addition of 750 nM PDGF (-) to a solution containing GST (10 nM) and 'chemical transducer' (500 nM) (-) at t = 3.5 min. (b) Addition of a unmodified PDGF aptamer (5 µM) (-) to a solution containing GST (10 nM), PDGF (750 nM), and the 'chemical transducer'(500 nM) (-) at t = 1.5 min leads to an immediate decrease in the enzymatic reaction rate. Re-printed with permission from reference4. Please click here to view a larger version of this figure.
Figure 5: Controlled prodrug activation. (a) GST activation of JS-K prodrug to release toxic NO. (b) NO release in the presence of the 'chemical transducer' and different combinations of GST (10 nM) and PDGF (2 µM). The graph presents the mean ± stdev of triplicates. Changed with permission from reference4. Please click here to view a larger version of this figure.
We presented a method for designing and testing of a ‘chemical transducer’ that can induce artificial communication between two naturally unrelated proteins, GST and PDGF, without modifying the native proteins. The unnatural GST-PDGF communications could be detected in real time by using enzymatic assays that follow the changes in the activity of GST in the presence of the ‘chemical transducer’ and increasing the concentrations of PDGF. In addition to detecting the activation of GST by PDGF, these assays were used to follow the deactivation of GST by a competing aptamer, as well as the PDGF-mediated cleavage of an anticancer prodrug. Other switchable protein binders that can control the activity of proteins by responding to oligonucleotide inputs have been described.26 The protocols describe herein, however, demonstrates a means to follow the function of a new class of protein binders that respond to the presence of natural proteins. In this way, the regulation of an enzyme by a protein biomarkers can be detected in real time, which opens the way for using such systems to selectively activate prodrugs and other substrates of GST.
It is important to keep in mind that the kinetics of GST is very rapid and hence, to obtain reliable results from this enzymatic assay, it is essential to minimize the time between the substrate addition and the beginning of the measurement. In addition, because small changes in enzyme or substrate concentrations may have significant effects on the kinetic measurement, it is recommended to perform all experiments in triplicate and to perform positive and negative control experiments in order to avoid experimental errors and misinterpretation of the results.
One limitation of using enzymatic assays to follow unnatural protein-protein communication is the need to have an enzyme as one of the protein partners. In addition, the current assays follow changes in absorbance and hence, they are less suitable for detecting the activation of the enzyme in living cells, which would require using fluorogenic substrates. However, considering that this field is in its infancy, we believe that similar design principles and experimental protocols could also be used for developing future ‘chemical transducers’, which may be used to alter the response of cells to environmental signals or to provide them with new properties. If such changes will also result in therapeutic effects, these developments may lead to artificial signal transduction therapy, which relies on drugs that can mediate unnatural protein-protein communication.3
The authors have nothing to disclose.
This research was supported by the Minerva Foundation, the HFSP Organization, and a European Research Council Grant (Starting Grant 338265).
1-chloro-2,4-dinitrobenzene | Sigma-Aldrich | 237329 | |
Acetic acid | Bio Lab | 01070521 | |
Acetnitrile | J.T.Baker | 9017-03 | |
Ascorbic acid | Sigma-Aldrich | A4544 | |
Copper(II) Sulfate pentahydrate | Merck-Millipore | 102790 | |
Dimethyl sulfoxide | Merck-Millipore | 802912 | |
Dulbecco's Phosphate Buffered Saline | Biological Industries | 02-023-5A | |
Ethacrynic acid | Tokyo Chemical Industry Co. Ltd | E0526 | |
Glutathione-s-transferase M1-1 | Israel Structural Proteomics Center (Weizmann Institute of Science, Rehovot, Israel) | ||
JS-K | Sigma-Aldrich | J4137 | |
L-glutathione reduced | Sigma-Aldrich | G4251 | |
Magnesium Chloride | J.T.Baker | 0162 | |
nitrate/nitrite colorimetric assay kit | Cayman Chemical | 780001 | |
Oligonucleotides | W. M. Keck Foundation Biotechnology at Yale University | custom order | |
PDGF-BB | Israel Structural Proteomics Center (Weizmann Institute of Science, Rehovot, Israel) | ||
TBTA | Sigma-Aldrich | 678937 | |
Triethylamine | Sigma-Aldrich | T0886 | |
Desalting column | GE Healthcare | illustra MicroSpin G-25 Columns | |
HPLC | Waters | 2695 separation module | |
HPLC column | Waters | XBridgeTM OST C18 column (2.5 μM, 4.6 mm × 50 mm) | |
HPLC column | Waters | XBridgeTM OST C18 column (2.5μM, 10 mm × 50 mm) | |
Plate reader | BioTek | synergy H4 hybrid |