Target Identification of a Class of Pyrazolone Protein Aggregation Inhibitor Therapeutics for Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease with no cure, and current treatment options are very limited. Previously, we performed a high-throughput screen to identify small molecules that inhibit protein aggregation caused by a mutation in the gene that encodes superoxide dismutase 1 (SOD1), which is responsible for about 25% of familial ALS. This resulted in three hit series of compounds that were optimized over several years to give three compounds that were highly active in a mutant SOD1 ALS model. Here we identify the target of two of the active compounds (6 and 7) with the use of photoaffinity labeling, chemical biology reporters, affinity purification, proteomic analysis, and fluorescent/cellular thermal shift assays. Evidence is provided to demonstrate that these two pyrazolone compounds directly interact with 14-3-3-E and 14-3-3-Q isoforms, which have chaperone activity and are known to interact with mutant SOD1G93A aggregates and become insoluble in the subcellular JUNQ compartment, leading to apoptosis. Because protein aggregation is the hallmark of all neurodegenerative diseases, knowledge of the target compounds that inhibit protein aggregation allows for the design of more effective molecules for the treatment of ALS and possibly other neurodegenerative diseases.


■ INTRODUCTION
Amyotrophic lateral sclerosis (ALS) is a progressive and fatal neurodegenerative disease that is characterized by the degeneration of motor neurons in the central nervous system. 1,2There are two forms of ALS, sporadic ALS (SALS) and familial ALS (FALS).−10 The discovery of the dominant mutations in the SOD1 gene, which encodes the abundant cytoplasmic enzyme Cu−Zn superoxide dismutase 1, and its toxicity in FALS led to the broadening of our understanding of ALS pathology.
Mutations in SOD1 are responsible for 20% of the FALS cases whose clinical and pathological features are identical to those in SALS. 11There are more than 170 known ALS-causing mutations found in SOD1, and some retain their dismutase activity partially or entirely, demonstrating that ALS neurodegeneration is caused by the toxic properties of the mutant SOD1 independent of its activity. 12−16 The importance of the glial cells in neurodegeneration has been demonstrated by silencing expression of mutant SOD1 in these cells. 17Both mutant and wild-type SOD1 interact with Ras-related C3 botulinum toxin substrate 1 (Rac1) in microglial cells, an activator of NADPH oxidase, which mediates the production of superoxide to kill pathogens.Whereas wild-type SOD1 is involved in the normal regulatory mechanism of superoxide control, mutant SOD1 strongly binds to Rac1, locking it into an active form, which elevates the level of extracellular superoxide and enhancing motor neuron damage. 15Microglial cells are known to cause neuroinflammation in ALS. 17,18Mutant SOD1, but not wildtype SOD1, inhibits Derlin-1, an endoplasmic reticulum (ER)associated protein degradation (ERAD) component respon-sible for degradation of misfolded proteins, which leads to ER stress-activated apoptosis signal-regulating kinase 1 (ASK1)dependent cell death, crucial for disease progression in FALS. 19Furthermore, the loss of excitatory amino acid transporter 2 (EAAT2) in astrocytes has been observed in mutant SOD1 expressing rodent models, which results in the failure to rapidly clear glutamate from the synapse, causing repetitive firing of action potentials in motor neurons.This ultimately leads to the ER and mitochondrial stress due to an increase in calcium influx. 20,21Mutant SOD1 has been shown to impair the expression of monocarboxylate transporter 1 (MCT1) in oligodendrocytes, which delivers the energy metabolite lactate to the axon, thereby causing a reduced supply of energy to the motor neuron. 22Overall, these findings suggest that ALS propagates through a noncell autonomous fashion rather than damage to neurons alone.In fact, most of the above pathophysiological features, including aberrant protein aggregation, glutamate excitotoxicity, impaired mitochondrial function, and oxidative stress, serve as possible targets for developing new therapeutics for ALS.
To date, there are only three approved drugs to treat ALS, and each of these only weakly delays neurodegeneration.Riluzole (Figure 1 compound 1) was the first drug to be approved in the U.S. for ALS, which is thought to decrease the neurotoxic effect of glutamate.A drug called edaravone (Figure 1, compound 2), whose mechanism of action is unknown, was approved by the FDA but might act as a free-radical scavenger.Most recently, AMX0035 was approved for the treatment of ALS and is undergoing another Phase III clinical trial to assess the primary efficacy in people living with ALS globally.Some of the drugs tested for ALS were aimed at enhancing autophagy and glutamate-mediated excitotoxicity. 23ith this information in hand, we implemented a phenotypic drug discovery approach to identify a novel therapeutic for treating ALS.The study employed a rat PC12 cell-based assay that expresses SOD1 G93A mutation for screening a chemical library of more than 50,000 compounds to identify promising hits (Figure 1 compounds 3 and 4).The hits are based on attenuating proteasome inhibitor MG-132induced SOD1 G93A cellular toxicity.Hit-to-lead optimization of the arylsulfanyl pyrazolone (4) and cyclohexane-1,3-dione (3) scaffolds resulted in highly potent compounds with better ADMET properties than those of the hits (Figure 1 compounds 5, 6, and 7).Further studies with pyrazolone compound 6 in transgenic mice were shown to increase the median survival rate by 13%, similar to the clinical standard of care for riluzole, supporting its potential as a therapeutic candidate for ALS.Initial mechanism of action studies of these pyrazolone compounds have revealed that compounds may act as proteasome activators and do not produce a heat shock response. 24he final phase of the phenotype-based drug discovery process is target deconvolution (identification).There are several target identification approaches including direct biochemical methods, genetic interactions, and computational inference.These methods can be used individually or in combination to characterize on-or off-targets associated with a drug.We decided to implement direct biochemical methods to identify the target(s) of the pyrazolone lead compounds as potential ALS therapeutics.There are several direct biochemical techniques, such as activity-based protein profiling (ABPP), affinity purification, cellular thermal shift assay (CETSA), and photoaffinity labeling (PAL) techniques, that can be utilized for target identification.
Here we report a combination of PAL studies and CETSA experiments, which include the design and synthesis of a novel pyrazolone-based PAL probe for protein pull-down and proteomic identification of the target(s) as well as competition studies with the pyrazolone and cyclohexanedione compounds for target confirmation.Furthermore, we conducted fluorescence thermal shift assays (FTSAs) to quantify and confirm direct binding between the identified targets and the pyrazolone lead compounds.Finally, to robustly confirm the identified target and validate its drug−target engagement within the cellular context, we employed cellular thermal shift assays (CETSA).

■ RESULTS AND DISCUSSION
Design and Synthesis of Pyrazolone-based Photoaffinity Labeling (PAL) Probes.Photoaffinity labeling (PAL) makes it possible to covalently modify binding partner proteins by using transient UV irradiation.Numerous PAL strategies have been described. 25Generally, PAL probes are designed to have the following characteristics for use in live cell culture: 1) similar structure and efficacy to a validated lead; 2) an inducible reactive functionality for labeling interacting proteins; 3) appropriate physiochemical properties to allow membrane permeation and aqueous solubility; and 4) a conjugation site for modification with chemical biology reporters.The desire to incorporate these properties, and a wealth of SAR data from our previous investigations made it possible to predict a priori which chemical probe would be suitable for our application.We chose to incorporate an aromatic photoreactive moiety in place of the terminal 3,5dichlorophenyl group of 7, an alkynyl conjugation site in place of the methyl group of 7, and an active compound with improved solubility (as the HCl salt) compared to that of 5 or 6 (Figure 1).The photolabile 4-azido-2,3,5,6-tetrafluorophenyl group was chosen because this moiety indiscriminately inserts into any type of residue, including chemically "inert" aliphatic C−H bonds, which makes this moiety applicable when little information on binding residues is available.
Synthesis of the desired photoreactive alkynyl pyrazolone probe (abbreviated as PRAPP, Figure 1) was problematic and required a synthetic route different from that used to synthesize the parent molecule (P 3 ), although employing common synthetic transformations (Figure 2A).Pentafluorobenzaldehyde was subjected to nucleophilic aromatic substitution with sodium azide followed by reductive amination with propargylamine to obtain intermediate 9, which was also used as inactive photoreactive control 2 (IPC2).Ethyl 4-chloroacetoacetate was allowed to react with p-methoxybenzyl (PMB) alcohol, followed by reaction with hydrazine, to obtain pyrazolone intermediate 11, which was subjected to diacetyl protection before PMB deprotection to yield the corresponding alcohol (13).Functional group conversion of intermediate 13 to the corresponding bromide (14), followed by coupling with IPC2 resulted in intermediate 15, which also served as inactive photoreactive control 1 (IPC1).IPC1 is highly light-sensitive and unstable (it was stable enough to use as a control for a short period of time), which was subsequently deprotected with TFA at 50 °C to yield 16 (the PRAPP).
The PRAPP photolysis was monitored by following a 20 min irradiation with 312 nm UV light, which led to complete consumption of the PRAPP based on HPLC analysis (photolysis half-life 3.1 min, Figure 2B and C).The PRAPP was twice as potent as the parent molecule (7), having an EC 50 of 190 nM to ameliorate MG132-induced protein aggregation and cellular toxicity in PC12 SOD1 G93A cells in a manner similar to that of the validated lead pyrazolones.IPC1 and IPC2 had negligible activities (EC 50 > 16 μM).
Covalent Modification of the Target in PC12 SOD1 G93A Cells by the PRAPP.With the active photoaffinity probe and appropriate controls in hand, we next performed live cell PAL experiments to detect covalent target modification by the PRAPP.Three live cell treatment scenarios are shown in Figure 3A.Scenario 1 involved a 12 h incubation of cells with the PRAPP in the absence of MG-132 followed by subsequent UV irradiation at 312 nm for 15 min and lysis with modified RIPA buffer to collect the protein lysate.Scenarios 2 and 3 were designed to mimic the original neuroprotective assay conditions, where cells were first incubated with the PRAPP for 12 h followed by the addition of PBS (scenario 2) or MG- 132 (scenario 3) and further incubated for 24 h before UV irradiation and lysis.
For the initial experiment, a 30 μM concentration of PRAPP was used, and the protein lysates were normalized to 1 mg/mL via the Bradford assay.After incubation and irradiation, the resulting protein concentrates were treated with fluorescent azido-Alexa-647, and Cu(I)-catalyzed "click chemistry" was used to conjugate the dye with the alkyne group of the PRAPP, thereby allowing visualization of covalently bound proteins after SDS-PAGE (Figure S1.Alexa-647 was used for the fluoroconjugation dye to avoid cross-contamination by the endogenous YFP fluorescence signal (from a YFP-SOD1 G93A conjugate)).Nonspecific labeling was a concern during methodology conceptualization, and fortunately, fluorimaging revealed a prominent modified protein band at ∼25 kDa with relatively low background labeling (Figure 3B).The full gel figure corresponding to Figure 3B (Scenarios 1, 2, and 3) is provided in the Supporting Information in Figure S2.This ∼25 kDa band was detected in all three experimental scenarios, indicating that the target was there regardless of the incubation time or the presence or absence of MG-132.Interestingly, the 25 kDa band was detected in one of the controls where cells were not subjected to UV irradiation, Figure 3B (lane 2), suggesting either 1) a strong noncovalent interaction between the PRAPP and the target protein or 2) covalent modification by the PRAPP based on its inherent reactivity in the absence of UV light.The latter interpretation was disproved based on band disappearance upon dialysis overnight, demonstrating the noncovalent nature of the initial interaction and the high affinity of PRAPP toward the target.
Further experimentation revealed a concentration-dependent increase in fluorescence intensity of the 25 kDa band under all three treatment scenarios (Figure 4A and B).The correlation between the 25 kDa band intensity and the PRAPP concentration can be described according to the Hill equation, indicating saturation binding kinetics around 30 μM of the PRAPP.The above observation suggests that the 25 kDa protein target in PC12 cells is indeed the limiting factor of the binding event, and the amount does not change as a response to an increasing concentration of the PRAPP.In addition to the PRAPP concentration, the abundance of the 25 kDa protein target in PC12 cells might depend on other factors, including incubation time and MG-132 (100 nM).As shown in Supporting Figure S3A, there is a significant intensity difference in the 25 kDa protein band between scenarios 2 and 3 (Figure 3A and B), indicating an increase in the abundance of the targeted protein as a response to MG-132 induced cellular stress.
The intensity difference between MG-132 treated and untreated samples is prominent at a 10 μM concentration of PRAPP, which gradually decreases with increasing PRAPP concentration.This observation further indicates that the PRAPP concentration range used in this experiment was close to saturation kinetics (a plateau in the Hill plot indicates that the target is saturating at higher concentrations).In a separate experiment to assess the competition of MG-132 with the PRAPP, it was revealed that the intensity of the 25 kDa band increased with MG-132 concentration in SOD1 G93A PC12 cells (Supporting Figure S3B).Inhibition of the proteasome by MG-132 inhibits either the degradation of the 25 kDa protein(s) or, as a response to increased SOD1 aggregation, the cell might express more of the 25 kDa target protein.
Target Validation by Competition Experiments.Competition experiments with optimized compounds P 2 and P 3 (Figure 1) were performed to validate the authenticity of the 25 kDa protein band as a pyrazolone target protein.The disappearance of the 25 kDa band would be expected to occur with increasing concentrations of the competitor if this protein was the target.
According to the results in Figure 5, pyrazolone-containing compounds P 2 and P 3 are capable of reducing the intensity of the PRAPP modified band in an exceptionally reliable concentration-dependent manner, confirming that the 25 kDa band is a true target of pyrazolone compounds.The calculated EC 50 values for the competition with P 2 and P 3 are 11.06 and 26.67 μM, respectively.In both cases, complete disappearance of the band was observed at the P 3 to PRAPP ratio of 32 and P 2 to PRAPP ratio of 6, respectively.−28 This is consistent with our observations using P 2 and P 3 , where PRAPP (EC 50 = 190 nM) is more potent than P 3 (EC 50 = 400 nM) but is less potent than P 2 (EC 50 = 67 nM).Therefore, with the more potent competitor (P 2 ), only six times excess was needed to see complete disappearance of the band, but with P 3 , 32 times excess was needed.These results also provide confirmation of a 25 kDa protein being specifically modified by PRAPP in a dynamic cellular environment and endorse this protein as having a significant role in pyrazolone-mediated neuroprotection.Importantly, competition studies with cyclohexanedione lead compound P 1 and with edaravone did not show a quantifiable concentration-dependent disappearance of the 25 kDa protein band, indicating that, despite efficacy in our model system, neither P 1 nor edaravone (Figure 1) binds to the same target as the pyrazolone compounds (Supporting Information Figure S4).(Full gel figures for competition experiments are given in supporting information, Figures S5−S6.) Protein Pull-down and Proteomics Analysis of the 25 kDa Protein Band.Protein pull-down with proteomics analysis was employed to identify the 25 kDa band covalently modified by the PRAPP.Concentrated protein lysates were conjugated with TAMRA-Dde-biotin azide and pulled-down using streptavidin magnetic beads.The TAMRA fluorophore allowed for visualization, and the fluorescent enrichment of the 25 kDa band from the pull-down and served as a guide in band excision for proteomics analysis (Figure S7).The dichlorodiphenyldichloroethylene (Dde) moiety was initially intended for use in the chemical digestion of the beads to minimize the release of nonspecifically bound proteins from the beads.However, chemical digestion failed, and the standard SDS boiling protocol was used.The results of the SDS-PAGE run of the pull-down protein are shown in Figure 6.As evidenced by the fluorescence image, it is clear that the pull-down experiment contains the desired 25 kDa target, which is absent in the control experiment.
−31 The fluorescent-guided gel band excision aids in precisely selecting the target protein from a mixture of bands, distinguishing it from nonspecific proteins, including streptavidin.On-bead digestion is another target identification technique that is typically coupled with Stable Isotope Labeling by Amino Acids in Cell Culture (SILAC) and quantitative proteomics.This technique addresses challenges such as the identification of low-abundance targets that are difficult to detect with in-gel digestion.However, we chose in-gel digestion over on-bead digestion, because our fluorescence labeling experiments indicated that the target of interest is highly abundant in the cell matrix.Proteomics analysis of the excised bands identified six potential candidates as the target protein with known chaperone function, which include five of the seven 14-3-3 protein family members (14-3-3-β/B, 14-3-3ε/E, 14-3-3-η/H, 14-3-3-γ/G, and 14−4−3-θ/Q) and heat shock protein 27 (Hsp27) (Table S1).According to the normalized total spectral value, all candidates were enriched in the pull-down samples compared with the controls; only 14-3-3-E was absent in the control.The top three protein candidates include 14-3-3-θ, with a normalized total spectral count of 25 in the pull-down sample, followed by 14-3-3-η and Hsp27, with total spectral counts of 17 and 16, respectively.Four other proteins were identified within the 23−28 kDa range, namely, RAB5C (member of the RAS oncogene family), glutathione Stransferase, Ras-related protein Rab, and triosephosphate isomerase, but they do not have a known function in the literature related to protein aggregation.Consequently, the follow-up investigation focused on 14-3-3 proteins and Hsp27.
The 14-3-3 proteins form homo-or heterodimers that are in equilibrium with the corresponding monomers.Each monomer consists of a conserved phosphoprotein binding site that interacts with its non 14-3-3 binding partners. 32The isoform structures are conserved except at the areas of the interface that form intersubunit contacts.The dimeric structure is mainly stabilized by the specific salt-bridge interactions that exist between the subunits.Theoretically, homodimers of B, Z, and Q isoforms of human 14-3-3 proteins can form six intersubunit salt-bridge interactions, while G and H isoforms can only form four. 33 The E isoform can create only two intersubunit saltbridges in its homodimeric form, whereas additional salt-bridge interactions exist in the heterodimeric conformation.Therefore, the 14-3-3-E isoform preferentially forms heterodimers rather than homodimers in the cellular environment.The dimeric structure is essential for the scaffolding function of 14-3-3 proteins, where it brings two client proteins together to form a specific structure.Furthermore, the binding partners with two 14-3-3 binding sites rely on the dimeric structure of 14-3-3.Even though potential candidates have been identified, the protein's smaller size and the presence of homo-and heterodimers make the target protein's definitive identification somewhat ambivalent.Genetic manipulation is a technique that can be used to confirm the identity of the target candidates by systematically deleting the genes corresponding to each candidate and conducting photoaffinity labeling experiments.The implementation of this technique here is challenging due to the multiple potential candidates identified by proteomics, which require the generation of a series of knockout PC12 cell lines for each candidate.However, because of the commercial availability of the HAP1 Hsp27 knockout cell line (Horizon Discovery, HZGHC004733c006), we decided first to conduct photoaffinity labeling studies in HAP1 cells to determine if Hsp27 is a definitive target of the pyrazolone compounds.
Photoaffinity Labeling Studies with HAP1 Hsp27 Knockout (KO) Cell Line.First, we conducted photoaffinity labeling studies with wild-type (WT) HAP1 cells to verify the existence of the 25 kDa covalent modification that we observed in PC12 cells.Significant nonspecific background modifications were present in HAP1 cells compared to PC12 cells, as shown in Figure 7.There were two covalently modified bands around 25 kDa (bands 1 and 2), where only band 1 showed a concentration-dependent disappearance with P 3 competition (Figure 7A), which confirmed the existence of the 25 kDa modified band in HAP1 cells.However, in Hsp27 KO HAP1 cells, the intensity of the 25 kDa covalently modified band remains unchanged compared to that in WT cells, thereby eliminating Hsp27 as the target of the pyrazolone compounds (Figure 7B).
Implementation of a similar experimental procedure with 14-3-3 isoforms was challenging due to the number of isoforms suggested by proteomics and the lack of readily available knockout cell lines.However, biological (functional) or physiological assays can be used to evaluate the direct binding between potential target(s) and the drug candidate to overcome this challenge.Therefore, we decided to evaluate the direct binding of P 2 and PRAPP with isolated potential 14- 3-3 protein isoform targets to confirm the proteomics findings.The lack of functional assays for 14-3-3 isoforms initially made it difficult to achieve this goal.The fluorescent thermal shift assay (FTSA) (also known as differential scanning fluorimetry (DSF)) is a widely employed technique to confirm direct protein−protein and protein−small molecule binding.Thus, we carried out FTSAs to demonstrate direct binding of the lead pyrazolone compound (P 2 ) and the PRAPP with the isolated 14-3-3 protein targets in vitro.
Fluorescent Thermal Shift Assays (FTSAs) of Isolated Protein Targets with P 2 and PRAPP.The direct binding of small molecules affects the protein thermal stability and shifts the melting temperature (T m ).FTSAs quantify the changes in T m using the SYPRO orange fluorescent probe.The quantum yield of the environment-sensitive dye increases significantly when binding to hydrophobic surfaces exposed when a protein unfolds.The melting temperatures of the 14-3-3 isoforms tested had T m values around 53 °C, which is close to the values reported in the literature. 34However, the FTSA results indicated that only 14-3-3-E and Q isoforms showed a significant T m shift upon incubation with P 2 (Figure 8); no binding of P 2 to isoforms B, H, and G was observed (Supporting Information Figure S8).The 14-3-3-E isoform demonstrated a major T m shift (3.2 °C), indicating a stronger binding of P 2 and stabilization, while 14-3-3-Q showed a minor  T m shift (1.1 °C).Based on these data, we estimated the apparent binding constants (K d app ) of P 2 with the two isoforms by fitting the ΔT m values and P 2 concentrations.The estimated K d app values for P 2 with 14-3-3-E and 14-3-3-Q are 2.85 and 4.32 μM, respectively.The observed major T m shift for 14-3-3-E with P 2 led us to continue the FTSA experiments to evaluate the binding of PRAPP and P 1 , the cyclohexanedione lead molecule, with 14-3-3-E.PRAPP showed a T m shift of 2.5 °C, which is similar to that for P 2 , confirming the direct and robust binding of pyrazolone compounds with the 14-3-3-E isoform (Figure S9).However, P 1 did not show a T m shift, which is consistent with the outcome of the competition experiment (Figure S10).
Overall, the FTSA experiments revealed direct binding of pyrazolone compounds P 2 and PRAPP with 14-3-3-E, as indicated by major T m shifts.In addition, 14-3-3-Q also displayed a minor T m shift with P 2 , indicating some binding interactions.These findings suggest that 14-3-3-E is the main target of the pyrazolone compounds and that 14-3-3-Q is a minor target.Even though 14-3-3 isomers B, H, and G did not show any binding with pyrazolone compounds, the presence of those isomers in the pull-down protein matrix can be attributed to 14-3-3-E forming heterodimers with isoforms B, H, and G in cells.The covalent modification of 14-3-3-E heterodimers in cells eventually pulls down B, H, and G isoforms attached to 14-3-3-E, making those proteins appear in the proteomics results.Based on this observation, it also can be concluded that the binding of pyrazolone and PRAPP does not disrupt the formation of the heterodimers if it were to bind at the interface, in which case the other isoforms would not have been pulled down.
Finally, given the major T m shift observed for 14-3-3-E with P 2 and PRAPP, we decided to focus on 14-3-3-E to evaluate these binding interactions in a cellular environment.As the final step in target identification, this kind of confirmation is essential for validating the identified target within the relevant cell model.The Cellular Thermal Shift Assay (CETSA) is a widely used technique for confirming cellular target engagement of drugs, employing a principle similar to that of FTSA.Therefore, we conducted CETSA experiments to evaluate the direct binding of the lead pyrazolone compound P 2 to 14-3-3-E in vivo within the PC12 G93A cellular environment.
Evaluation of P 2 Interactions with 14-3-3-E in PC12 G93A Cells using Cellular Thermal Shift Assay (CETSA).The CETSA experiment was first described by Molina et al. as a technique to monitor drug target engagement in cells and tissues, utilizing either intact cells/tissues or cell lysates. 35This process involves heating multiple aliquots of cell lysate to different temperatures and subsequent centrifugation to separate the soluble fraction from the precipitated proteins upon cooling.Thereafter, the presence of the target protein in the soluble fraction is quantified by Western blotting.
Our initial attempts to conduct the experiment in intact PC12-SOD1 G93A cells were unsuccessful, likely due to thermal lysis of the cells at high temperatures.Consequently, we decided to switch to the cell lysate protocol.We first conducted the experiment at a high concentration of P 2 to determine if there was a significant thermal shift associated with 14-3-3-E in the presence of P 2 in the cell lysate.Interestingly, a significant relative band intensity difference was observed for 14-3-3-E with P 2 compared to the DMSO control, between 60 and 64 °C, with the maximum difference noted at 63 °C (Figure 9A).Overall, the enhanced thermal stability of 14-3-3-E in the cell lysate, compared to that of the purified protein (T m = 52.3°C), can be attributed to the protein− protein interactions it forms with non-14-3-3 binding partners and the heterodimer formation in the cellular environment.Such interactions further stabilize the protein, shifting its T m to a higher value.However, in contrast to common observations in which small molecule binding increases the thermal stability of a protein, the addition of P 2 markedly reduced the presence of 14-3-3-E in the soluble fraction, indicating a destabilization of 14-3-3-E in the cell lysate.In fact, this kind of decreased thermal stability is rare but has been reported in the literature associated with small molecules that disrupt protein−protein interactions and complex formation. 36Since in vitro FTSA indicated increased thermal stability and noninterruption of 14-3-3-E dimer formation in the presence of P 2 , the observed in vivo destabilization might result from the potential disruption of protein−protein interactions between 14-3-3-E and its non-14-3-3 binding partners in the cellular environment due to P 2 , thereby lowering the thermal stability.This observation is highly significant for assigning a potential function associated with the 14-3-3-E and P 2 binding interactions in the context of ALS pathology.It can be hypothesized that the neuroprotective phenotypic outcome of P 2 is linked to the potential disruption of interactions between 14-3-3-E and its non-14-3-3 binding partners in PC12 G93A cells.
Following the establishment of the thermal shift of 14-3-3-E in the presence of P 2 at high concentrations, we conducted isothermal dose−response (ITDR) experiments to evaluate the concentration dependence of the relative band intensity of 14-3-3-E in the soluble fraction at 63 °C.As shown in Figure 9B, a clear decrease in the relative band intensity of 14-3-3-E was observed with increasing concentrations of P 2 , confirming the concentration dependence of 14-3-3-E protein stability.Interestingly, this response was reached at relatively high P 2 concentrations, an observation also noted in the original CETSA protocol that used cell lysate compared with intact cells.For instance, drugs with nanomolar inhibitor potencies, such as raltitrexed and methotrexate, displayed a maximum response between 100−250 μM drug concentrations.The authors attributed this observation to factors such as the higher temperatures used in the experiment, potential protein binding, metabolic factors at higher lysate concentrations, and the nonequilibrium nature of protein precipitation upon unfolding, suggesting the use of the ITDR CETSA experiment as a fingerprint for target engagement across a range of drug concentrations.Overall, the CETSA experiments we conducted confirm the P 2 interactions with 14-3-3-E in a cellular context and validate 14-3-3-E as a definitive target of the pyrazolone compounds.
Importance of 14-3-3-E and 14-3-3-Q in ALS Pathogenicity.The role of protein aggregation as the primary causative agent in neurodegenerative diseases remains a topic of debate, with some studies suggesting that smaller soluble oligomers might be the principal toxic species rather than larger aggregates. 37Additionally, there are observations indicating that protein aggregates may represent a cellular protective response to sequester misfolded proteins, challenging the traditional perspective of their pathogenic role. 38Even though the precise mechanism of ALS pathogenesis is unclear, protein aggregation is still one of the main pathological features, which is also common in other neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD), and Huntington's disease (HD).There are mainly two classes of aggregate-prone protein phenotypes distinguished by a cellular quality control mechanism, namely, soluble normal misfolded proteins and insoluble amyloidogenic proteins. 39ammalian cells are known to localize these aggregates at distinct inclusion sites. 40,41Normal soluble misfolded proteins consist of ubiquitinated proteins with mutations or stressinduced damage, and they accumulate in the Juxta Nuclear Quality control compartment (JUNQ) adjacent to the nucleus.Amyloidogenic proteins form insoluble aggregates and are sequestered in Insoluble Protein Deposit-like (IPOD-like) inclusions. 39 recent study has shown that in the SOD1 G93A ALS mammalian cell line, aggregation of insoluble SOD1 G93A aggregates in JUNQ interferes with its quality control function.39 Furthermore, it has been demonstrated that sequestration of these SOD1 G93A aggregates in an insoluble inclusion (e.g., in IPOD) reduces the harmful effects of aggregation on the cell viability.Another recent independent study found that in the SOD1 G93A ALS transgenic mouse model, 14-3-3 proteins colocalize with mutant SOD1 aggregates, making them more insoluble in the spinal cord than in wild-type mice.42 The same study found that 14-3-3-E and 14-3-3-Q isoforms interact with mutant SOD1 aggregates in the JUNQ of N2a neuroblastoma cells and become insoluble.Consequently, the cytosolic abundance of 14-3-3-E and 14-3-3-Q isoforms decreases, promoting the translocation of Bax (Bcl-2-associated X protein) into the mitochondria and releasing cytochrome c into the cytosol, causing apoptosis (Figure 10A).A proposed function of 14-3-3 proteins is to inhibit apoptosis by interacting with Bad (Bcl-2-associated death promoter) or Bax, which prevents the localization of these proteins into the mitochondria.Furthermore, overexpression of 14-3-3-E and 14-3-3-Q in the same cell line dramatically decreased SOD1 G93A -induced cytotoxicity.Based on these literature findings, we hypothesize that the pyrazolone compounds disrupt the 14-3-3-E and 14-3-3-Q interactions with misfolded SOD1 G93A and prevent cosequestration into JUNQ, keeping the cytosolic 14-3-3 level adequate to maintain its regular function (Figure 10B).As previously discussed, the observed decrease in the thermal stability of 14-3-3-E in the PC12 G93A cell lysate upon P 2 exposure, coupled with the absence of SOD1 in the pull-down matrix, further supports our hypothesis that the pyrazolone compounds disrupt the 14-3-3-E-SOD1 G93A interactions.14-3-3 proteins are also proposed to function as molecular chaperones that interact with aggregation-prone proteins or to target misfolded proteins to aggresome structures. Thse results strongly support our experimental finding of 14-3-3-E (in vitro/in vivo) and 14-3-3-Q (in vitro) as molecular targets of pyrazolone compounds and their functional importance in ALS pathogenesis.

■ CONCLUSIONS
ALS is a progressive and fatal neurological disorder that affects neurons in the central nervous system.We launched a phenotypic drug discovery campaign to discover a novel therapeutic that inhibits protein aggregation for treating ALS.One class of compounds that we optimized was the pyrazolone, which was active in extending the life of the ALS mouse model and was shown to maintain the health of axons in upper motor neurons from a mouse with mutant SOD1 toxicity in vitro (unpublished data).As the last stage of our phenotype-based ALS drug discovery approach, we have successfully conducted direct biochemical studies to find molecular targets of the pyrazolone-based neuroprotective compounds optimized from hits in our screening campaign.First, we designed a tetrafluoroazidobenzene-containing photoaffinity probe based on neuroprotective pyrazolone compound P 3 , and from subsequent photoaffinity labeling studies, we identified a 25 kDa covalently modified protein band in the PC12 SOD1 G93A cell line.The covalent modification is concentration-dependent, and the competition experiments revealed that both pyrazolone compounds (P 2 and P 3 ) and the photoaffinity probe (PRAPP) compete for the same 25 kDa protein target.Subsequent protein pull-down experiments, followed by proteomics analysis of the 25 kDa modified protein band, revealed that this protein belongs to the 14-3-3 family of proteins.To confirm the identity and obtain direct binding evidence, we expressed and purified the 14-3-3 family proteins identified by proteomics and conducted FTSAs with the pyrazolone compounds and the PRAPP.These experiments identified 14-3-3-E as the primary target and 14-3-3-Q as the minor target of the pyrazolone lead compounds.Given the major T m shift observed for 14-3-3-E with P 2 in vitro by FTSA, we conducted CETSA experiments to robustly confirm the cellular drug−target engagement between P 2 and 14-3-3-E.The CETSA experiment revealed a decrease in the thermal stability of 14-3-3-E in the cell lysate in the presence of P 2 , indicating the potential disruption of protein−protein interactions between 14-3-3-E and its cellular non-14-3-3 binding partners.The functional importance of 14-3-3-E and 14-3-3-Q isoforms in ALS has been extensively studied in the literature using SOD1 G93A transgenic mice models and N2a neuroblastoma cells.The 14-3-3-E and 14-3-3-Q isoforms are known to interact with mutant SOD1 aggregates in the JUNQ of N2a neuroblastoma cells that express SOD1 G93A and become insoluble, which causes a decrease in cytosolic 14-3-3 abundance, leading to apoptosis.Therefore, we hypothesize that the pyrazolone compounds prevent apoptosis by disrupting 14-3-3-E and 14-3-3-Q interactions with misfolded SOD1 G93A aggregates and elevating the cytosolic levels of 14-3-3-E and 14-3-3-Q.This is the first definitive identification of the target(s) of a small molecule with activity in an ALS phenotypic cell-based model and in the ALS mouse model.

■ EXPERIMENTAL SECTION
Synthesis.General Methods.All reactions were carried out under an atmosphere of dry nitrogen in oven-dried glassware unless stated otherwise.Indicated reaction temperatures refer to those of the reaction bath, while room temperature (rt) is noted as about 25 °C.All other solvents were of anhydrous quality and used as received.Commercially available starting materials and reagents were purchased from Aldrich, TCI, and Fisher Scientific and used as received unless specified otherwise.Analytical thin-layer chromatography (TLC) was performed with precoated silica gel 60 F254 plates (5 × 20 cm, 60 Å, 250 μm).Column chromatography was performed with silica gel 60 (230−400 mesh) or C18 reverse phase column with a Combi-flash instrument. 1 H and 13 C NMR spectra were recorded in deuterated solvent on a Bruker Avance III 500 MHz spectrometer (direct cryoprobe).Chemical shifts are reported in ppm with the solvent resonance as internal standard (CDCl 3 , 7.27 ppm, 77.23 ppm and DMSO-d 6 , 2.5 ppm, 39.51 ppm for 1 H and 13 C respectively).Data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, dd = doublet of doublet, t = triplet, q = quartet, br = broad, m = multiplet, abq = ab quartet), number of protons, and coupling constants.Low-resolution mass spectra (LRMS) were acquired on an Agilent 1110 MSD.High-resolution mass spectral (HRMS) data were collected inhouse using an Agilent 6210 LC-TOF (ESI).All compounds submitted for biological testing were found to be >95% pure by analytical HPLC.-2,3,5,6-tetrafluorobenzaldehyde (8).

N-(4-Azido-2,3,5,6-tetrafluorobenzyl)prop-2-yn-1-amine (9) (IPC2).
To a stirred solution of 8 (3.42 g, 15.6 mmol) in ethanol (250 mL) were added propargylamine (1.72 g, 31.2 mmol) and glacial acidic acid (0.2 mL), followed by anhydrous sodium sulfate (30 g), and the mixture was stirred for 12 h.The reaction mixture was filtered to remove the anhydrous sodium sulfate, and the resulting imine was cooled to 0 °C using an ice bath.Sodium cyanoborohydride (1.96 g, 31.2 mmol) was added to the resulting solution with stirring and allowed to warm to room temperature.After 2 h, the reaction was quenched with water and concentrated to remove ethanol.The crude product was dissolved in ethyl acetate and washed sequentially with sodium bicarbonate and brine.The organic layer was dried over anhydrous sodium sulfate and concentrated to yield a yellowish-brown viscous oil which was purified by flash chromatography on silica gel (70% ethyl acetate in hexane) to obtain 9 as a yellow solid (1.9 g, 47%).Note: This compound is highly unstable upon exposure to light.The reaction flask and the vials were covered with aluminum foil to avoid exposure to light. 1 H NMR (500 MHz, CDCl 3 ) δ 3.96 (t, J = 1.5 Hz, 2H), 3.42 (d, J = 2.4 Hz, 2H), 2.22 (t, J = 2.4 Hz, 1H) (NH proton was not visible).MS-APCI [M + H] + = 259.17.
General Procedure for Photoaffinity Labeling.PC12-SOD1 G93A cells were cultured and grown to 90% confluency (in 6-well plates) and washed once with PBS followed by the addition of serum medium containing the appropriate analyte at the indicated concentrations.Three incubation scenarios were studied (at 37 °C): Scenario 1) 12 h preincubation with PRAPP in the absence of MG-132; Scenario 2) 12 h preincubation with PRAPP in the absence of MG-132 followed by an additional 24 h incubation (control); and Scenario 3) 12 h preincubation with PRAPP followed by the addition of MG-132 (100 nM) and an additional 24 h incubation.After incubation, the medium was discarded and cells were washed with PBS (1 × 300 μL).Fresh serum-free medium was added, and the cells were irradiated with UV light (312 nm, 120 V, 60 Hz, 0.2 A [Spectroline model EB-160C], distance ∼4 cm) for 15 min.The medium was removed, cells were washed with PBS (2 × 500 μL), and 200 μL of lysis buffer containing Tris HCl (25 mM, pH 7.6), NaCl (150 mM), 1% NP40, 1% sodium deoxycholate, 0.1% SDS, and protease inhibitor cocktail (Sigma-Aldrich, 1:100 v/v) was added to each well.
Complete lysis was observed by a microscope after 10 min at 4 °C with agitation.Each lysate was centrifuged at 4 °C for 5 min at 13,000 rpm, and the resulting supernatant was concentrated with a centrifugal filter (Amicon Ultra Centrifugal Filters 3K cutoff, Millipore), normalized to 1.0 mg/mL by the Bradford assay, and analyzed immediately or stored at −80 °C for subsequent analysis.Treatment Scenario 1 was used for competition experiments with 30 μM PRAPP and varying concentrations of the competitor (P 1 , P 2 , P 3 , and edaravone) as shown in Figure 5, Figure S4, Figure S5, and Figure S6.
General Procedure for Visualization Using N 3 −Alexa-647.In a darkened (wrapped with aluminum foil) Eppendorf tube, each lysate (30 μL [30 μg]) was incubated with 10 μL of the freshly prepared "click" premix containing CuSO 4 (5 mM), sodium ascorbate (5 mM), TBTA (200 μM), and N 3 −Alexa-647 (50 μM) were added to PBS (50 mM, pH 7.4).The corresponding final concentrations within the lysate click incubation were: CuSO 4 (1.25 mM), sodium ascorbate (1.25 mM), TBTA (50 μM), and N 3 −Alexa-647 (12.5 μM).The mixture was vortexed briefly and incubated for 1.5 h at 37 °C in the dark.NuPAGE LDS sample loading buffer (4 × 13 μL) was added, and the mixture was denatured in boiling water for 10 min.SDS-PAGE was performed on Novex Bis-Tris gel 4− 12% with MES SDS running buffer (90 V constant for 30 min followed by 100 V for 90 min; load volume = 25 μL).Protein standards that gave appropriate visible and fluorescent bands for our application included the BioRad Precision Plus All Blue Protein Standard and Novex Pre-Stained Protein Standard.Ingel fluorescence was monitored (excitation λ = 635 nm; emission λ = 670 nm) using a Typhoon 9400 (GE Healthcare) fluorimager.Densitometric analysis and image refinement were carried out as needed using ImageJ software (NIH, open source).The contrast and brightness of images were modified using ImageJ and/or Microsoft Word to enhance visualization for publication purposes.
Proteomic Analysis.The excised protein bands of interest were digested with sequencing-grade trypsin.The samples were loaded directly onto a PepMap C18 cartridge (0.3 mm × 5 mm, 5 μm particle) trap column and then a 75 μm × 150 mm PepMap C18 analytical column (Thermo Fisher Scientific) and separated at a flow rate of 300 nL/min.Solvent A was 0.1% formic acid in water, and solvent B was 0.1% formic acid and 80% acetonitrile in water.The 75 min solvent gradient of LC was 5% B from 0 to 3 min, 5−35% B to 55 min, 35−95% B to 60 min, and washed 95% to 65 min, followed by 5% B equilibration to 75 min.The peptides were directly eluted into a Q Exactive HF mass spectrometer, and the full MS scans were acquired in the Q-Exactive mass spectrometer over a 350−1400 m/z range with a resolution of 120,000 (at 400 m/z) from 5 to 63 min.The tandem mass spectrum was acquired in the mass analyzer with a resolution of 30,000 (at 400 m/z).
All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK; version 2.6.2).Mascot was set up to search the Uniprot database (downloaded on 15/10/2019, containing 29,942 entries), assuming the digestion enzyme trypsin.Mascot was searched with a fragment ion mass tolerance of 0.30 Da, and a parent ion tolerance of 20 ppm of carbamidomethyl of cysteine was specified in Mascot as a fixed modification.Deamidation of asparagine and glutamine and oxidation of methionine were specified in Mascot as variable modifications.Scaffold (version Scaffold_4.11.1, Proteome Software Inc., Portland, OR) was used to validate MS/MSbased peptide and protein identifications.Peptide identifications were accepted if they could be established at greater than 96.0%probability by the Peptide Prophet algorithm 45,46 with Scaffold delta-mass correction.Protein identifications were accepted if they could be established at a greater than 32.0%probability of achieving a false discovery rate (FDR) less than 5.0% and contained at least two identified peptides.Protein probabilities were assigned by the Protein Prophet algorithm.Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony.Proteins sharing significant peptide evidence were grouped into clusters.
Protein Expression and Purification.His-tagged 14-3-3-E, B, and H plasmids were purchased from Addgene (plasmid ID# 31562, 39128, and 38814, respectively).E. coli (BL21) were transformed with extracted plasmids via heat shock and plated on selected media.Individual colonies were selected, grown, and sequenced to confirm gene presence.1 L cultures were grown to an OD600 of 0.6 at 37 °C and 220 rpm.Protein expression was induced with 1 mM IPTG and expressed for 12 h at 100 rpm and room temperature (approximately 25 °C).Cells were spun down at 4000 rpm for 15 min at 4 °C and resuspended in lysis buffer containing 50 mM Tris (7.4), 150 mM NaCl, 0.08% DDM, 5 mM imidazole, and 5% glycerol (with an EDTA-free complete mini protease inhibitor cocktail).Cells were lysed via sonication, and the lysate was separated from cell debris by centrifugation at 15,000 rpm, 4 °C, for 45 min.The His-tagged protein was purified using QIAGEN Ni-NTA superflow columns, and the His-tag was removed with TEV protease.His-tag-removed proteins were further purified by size exclusion chromatography.His tag removal and size exclusion chromatography purification were done at the Northwestern University Protein Production core.14-3-3-Q and 14-3-3-G proteins were purchased from Abchem (catalog numbers ab85270 and ab53869, respectively).
Fluorescent Thermal Shift Assay (FTSA) Protocol.All of the FTSA experiments were conducted in HEPES buffer (60 mM HEPES, 150 mM NaCl, pH 7.4).A DMSO balanced 8point dilution series of the compound in HEPES buffer was prepared in a 96-well assay plate (200, 100, 50, 25, 12.5, 6.25, 3.125, and 0 μM).Then, a 5 μL volume of the dilution series was transferred to a 384-well PCR plate using a multichannel pipet, and the plate was centrifuged at 1000 rpm for 2 min (the plate was covered with an adhesive aluminum seal).The protein sample was premixed at a concentration of 0.04 μg/μL with a 5X concentration of SYPRO orange in HEPES buffer, and a 5 μL aliquot was added to the compound wells.The plate was covered with an optical adhesive seal and was shaken for 2 min, followed by centrifugation at 1000 rpm for 2 min.The thermal scan was performed from 20 to 90 °C with a temperature gradient of 0.5 °C/min.The fluorescence was measured on a Bio-Rad CFX384 real-time PCR instrument.
The cell medium was subsequently removed, and the cells were rinsed with 1X PBS before being detached using 1 mL of 1X trypsin.To this, 5 mL of 1X PBS solution was added, and the cells were then resuspended, transferred to two separate 15 mL conical tubes, and centrifuged at 1100 rpm for 3 min.The supernatant was discarded, and the cells were resuspended in 1 mL of 1X PBS containing a proteasomal inhibitor cocktail (EDTA-free) (Sigma-Aldrich, 1:100 v/v).This solution was then transferred to two Eppendorf tubes and immediately snap-frozen by using liquid nitrogen.The cells underwent lysis through four freeze/thaw cycles, alternating between liquid nitrogen and heating at 30 °C.The cell debris was then separated from the supernatant by centrifuging at 19,000g for 20 min at 4 °C.The supernatants from both tubes were combined and split between two new Eppendorf tubes (950 μL per tube).The cell lysates were subsequently treated with either P 2 (500 μM) or DMSO (1% v/v) and incubated for 2 h at room temperature.Each sample was further divided into seven smaller aliquots (120 μL each), heated for 3 min at specified temperatures, and then allowed to cool at room temperature for another 3 min.After that, the lysate samples were centrifuged at 19,000g for 20 min at 4 °C, and the soluble fraction was collected.Subsequently, 6 μL of 5X SDS loading buffer was added to 24 μL of each sample, and the resulting mixture was denatured in boiling water for 10 min.Finally, 20 μL of this mixture was loaded onto an SDS PAGE gel.SDS PAGE was conducted using a 15-well Novex Bis-Tris 4−12% prepacked gel and 1X MES-SDS gel running buffer (comprising 50 mM MES, 50 mM Tris, 1 mM EDTA, and 0.1% SDS) at a constant voltage of 100 mV over a duration of 2 h.Postelectrophoresis, the gel was transferred to a nitrocellulose membrane at 4 °C using a constant current of 400 mA in Tris-glycine transfer buffer (made from 800 mL H 2 O, 200 mL MeOH, 25 mM Tris, and 190 mM glycine).The nitrocellulose membrane was then blocked with NFD-milk solution (1X PBS with 2.5% nonfat dry milk) and left overnight at 4 °C.After that, the membrane was rinsed with 1X PBS/0.1% Tween 20 solution (3 × 5 min), at room temperature, and incubated for 1 h at room temperature with the 14-3-3-E primary antibody (α-rabbit, Invitrogen, 1:1000 dilution in 1X PBS with 2.5 mg/mL BSA).The primary antibody solution was removed, and the membrane was washed with 1X PBS/0.1% Tween 20 (3 × 5 min).The membrane was then incubated with the secondary antibody solution (α-rabbit, horseradish peroxidase-linked, 1:5000 dilution in NFD-milk blocking solution) for 1 h.After that, the secondary antibody solution was removed, and the membrane was washed again with 1X PBS/0.1% Tween 20 (3 × 5 min).The membrane was then incubated with a solution from the PerkinElmer chemiluminescence enhancement kit for 5 min at room temperature.Immunoblot raw image data were captured using an Azure Sapphire Biomolecular Imager with a 4-second exposure.These raw immunoblot images were analyzed by using ImageJ software.Data (expressed as relative band intensity %) were normalized relative to the samples that were incubated at the lowest temperature.The ITDR CETSA experiment was conducted similarly to the aforementioned protocol at 63 °C, and the data were normalized relative to the DMSO control where [P 2 ] = 0 ■ ASSOCIATED CONTENT * sı Supporting Information The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscentsci.3c00213.
Visualization protocol for covalently modified proteins, effect of MG-132 on 25 kDa covalently modified band intensity, supplemental figures related to competition studies of PRAPP, fluoresence-guided gel excision, supplemental figures related to FTSA studies of PRAPP, P2 and P1 with 14-3-3 isoforms, 1

Figure 1 .
Figure 1.Implementation of a phenotype-based drug discovery approach to discover new ALS therapeutics and the chemical structures of riluzole, edaravone, and AMX0035.

Figure 2 .
Figure 2. Synthesis and photolysis characterization of pyrazolone PAL probes.A) Synthetic scheme for the pyrazolone PAL probes; B) HPLC traces of the photoactivation of the PRAPP after various periods of irradiation at 312 nm; C) kinetics of the PRAPP photoactivation.

Figure 3 .
Figure 3. (A) PRAPP treatment scenarios and (B) covalent modification of the target by the PRAPP with different control experiments (not added is blue; added is yellow).The compound concentrations are in μM.Alexa 647 is an azido-Alexa 647.The 25 kDa covalently modified protein band was identified from this experiment.The figure was created with BioRender.

Figure 4 .
Figure 4. Concentration-dependent covalent modification of the 25 kDa band by the PRAPP.A) Quantified relative band intensities (n = 1).The plots were fitted using the Hill equation.B) In-gel fluorescence.

Figure 5 .
Figure 5. In-gel fluorescence of the competition studies of the PRAPP with (A) P 2 and (B) P 3 ; Data are presented as means ± SD from n = 2 independent experiments; NP = neuroprotection.

Figure 6 .
Figure 6.In-gel fluorescence of the protein pull-down experiment.The 25 kDa band is visible only in pull-down lanes 4 and 5, indicating the presence of the desired target.Free TAMRA-Dde-biotin bound monomeric avidin is also visible around 15 kDa.The presence of dimeric avidin is ruled out by the absence of a band around 30 kDa in the control experiments.Lane 1: Precision plus protein all blue standard; Lane 2: Control 1 (beads only + 10 min boiling in lithium dodecyl sulfate (LDS)); Lane 3: Control 2 (lysate + TAMRA-Ddebiotin azide (−Cu 2+ ) + 10 min boiling in LDS); Lanes 4 and 5: Pulldown 1 and 2 (lysate + TAMRA-Dde-biotin azide + 10 min boiling in LDS).Left panel was created by merging green (ladder) and red (pull-down protein) channels using ImageJ.Single channel (pseduocolored) image of the same gel is depicted in the right panel.

Figure 7 .
Figure 7. Photoaffinity Labeling Experiments with HAP1 cells.(A) In-gel fluorescence of the competition of PRAPP with P 3 (band 1 showed a concentration-dependent disappearance with P 3 ) in wildtype (WT) HAP1 cells.The nonspecific background modifications remain unchanged with increasing concentration of P 3 , further confirming the 25 kDa covalently modified band as a true target; (B) comparison of the photoaffinity labeling experiments with HAP1 wild-type (WT) cells and Hsp27 knockout (KO) cells at 15 μM PRAPP concentration.

Figure 8 .
Figure 8. FTSA results of P 2 with (A) 14-3-3-E and (B) 14-3-3-Q.Melting curves are shown in the left panel, and the ΔT m vs P 2 concentration fitting curves are shown in the right panel.ΔT m values are presented as means ± SD from n = 2 independent experiments.
H NMR spectra and proteomics analysis results table (PDF) AuthorsPathum M. Weerawarna − Department of Chemistry, Chemistry of Life Processes Institute, Center for Developmental Therapeutics, Northwestern University, Evanston, Illinois 60208, United States; Present