A rapid and accurate method for evaluating the degradation of pan-Akt in cells by PROTACs using NanoLuc luciferase

Abstract

Proteolysis targeting chimera (PROTAC) is a protein degradation technique that has been increasingly used in the development of new drugs in recent years. Akt is a classical serine/threonine kinase, and its role outside of the kinase has gradually gained attention in recent years, making it one of the proteins targeted by PROTACs. Currently, there are many methods used for the evaluation of intracellular protein degradation, but each has its own advantages or disadvantages. This study aimed to investigate the feasibility of evaluating the degradation of pan-Akt proteins in cells by PROTACs (MS21 and MS170) using the NanoLuc luciferase method. After conducting a thorough comparison between this method and the classical western blot assay in various cells, as well as testing the stability of the experiments between multiple batches, we found that NanoLuc luciferase is a highly accurate, stable, low-cost and easy-to-operate method for the evaluation of intracellular pan-Akt degradation by PROTACs with a short cycle time and high cellular expandability. Given the numerous advantages of this method, it is hypothesized that it could be extended to evaluate the degradation of more target proteins of PROTACs. In summary, the NanoLuc luciferase is a suitable method for early protein degradation screening of PROTAC compounds.

Introduction

Proteolysis targeting chimera (PROTAC) is a technology that utilizes the intracellular ubiquitin proteasome system to specifically degrade target proteins. The classic PROTAC molecule consists of three parts including a protein-of-interest binding moiety, an E3 ubiquitin ligase binding moiety, and a linker between them. Unlike most small molecule inhibitors traditionally used in the clinic, PROTACs have unique advantages. They do not depend on the active sites of the target proteins for protein degradation, can reengage in the degradation of new proteins after inducing degradation of the target proteins, and are able to overcome resistance caused by target overexpression or mutations in the active sites [1]. Since the concept of PROTAC was first proposed by Professors Crews and Deshaies in 2001 [2], PROTAC technology has been vigorously developed, and dozens of molecules (including ARV-110 targeting AR, ARV-471 targeting αER-α, NX-2127, and BGB-16673 targeting BTK) have entered into clinical trials [3–6]. Meanwhile, a variety of new protein degradation technologies have also been derived based on PROTAC technology, such as transcription factor-based PROTACs, Light-Controllable PROTACs, phosphoPROTACs, lysosomal hydrolysis-targeted chimeras, and autophagy targeting chimeras [7].

The protein kinase B, also known as Akt, is a serine/threonine kinase that is encoded by three closely related genes in humans: Akt1, Akt2, and Akt3. It is a central node in the PI3K/Akt/mTOR signaling pathway, regulating fundamental cellular and physiological processes such as cell growth, survival, gene transcription, and protein synthesis. Excessive Akt activation is closely associated with the development of various malignant tumors, including breast cancer, prostate cancer, lung cancer, colorectal cancer, ovarian cancer, and melanoma [8]. Currently, several ATP-competitive pan-Akt (Akt1, Akt2, and Akt3) inhibitors are active in the clinical stage, including capivasertib (AZD5363) [9], ipatasertib (GDC-0068) [10], afuresertib [11], and NTQ1062 [12]. Capivasertib has been approved by the FDA for the treatment of adult patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative, locally advanced or metastatic breast cancer with one or more PIK3CA/AKT1/PTEN-alterations in combination with fulvestrant. In addition to its typical enzymatic activity, recent studies have found the non-catalytic function of Akt in cell survival and drug resistance. In 2004, Remy et al. demonstrated that binding of Akt to Smad3 inhibits Smad3-mediated transcription and protects against TGF-β-induced apoptosis, a process that is not dependent on the kinase activity of Akt [13]. Furthermore, it has been shown that enzyme-inactivated Akt mutants promote growth factor-independent survival of primary human melanocytes [14]. Therefore, the development of PROTAC molecules targeting Akt is necessary. Indeed, some scientists have already initiated efforts in this direction. For example, INY-03-041, a PROTAC molecule based on GDC-0068 and lenalidomide, is a pan-Akt degrader that promotes sustained Akt degradation and prolonged inhibition of downstream signaling, with longer pharmacological action than GDC-0068 [15]. MS21, a PROTAC molecule developed based on AZD5363, exhibited stronger ex vivo antitumor effects than AZD5363 in various tumor cells and CDX models [16]. Xu et al. demonstrated that the Akt3-selective protein degrader 12I, but not a kinase inhibitor, was able to significantly reverse lung cancer resistance to osimertinib, and selective Akt3 degradation may serve as a novel strategy to overcome drug resistance in various human tumors [17]. Therefore, developing PROTAC small molecules targeting Akt is a promising direction for drug development.

Given the unpredictability of PROTAC degraders in inducing degradation of target proteins, it is important to introduce reliable degradation assays to screen for PROTAC molecules with different degradation capacities. The most common methods used for intracellular protein level detection include western blot, reporter gene assays, and mass spectrometry (MS)-based proteomics [18]. Among them, western blot is a classical method for the detection of protein degradation. However, it is not suitable for high-throughput screening of protein degradation of PROTAC molecules due to its cumbersome operation, long cycle time, and semi-quantification. MS, on the other hand, is mainly used to detect the specificity of PROTAC degradation. The reporter gene method is a useful tool for the detection of intracellular target proteins. It offers advantages such as ease of operation, short cycle time, and high throughput, which makes it increasingly used in the screening of PROTACs for intracellular protein degradation activity. The more mature methods are the HiBiT and the NanoLuc luciferase technologies developed by Promega.

In this study, we established an intracellular pan-Akt protein level assay using NanoLuc luciferase technology to evaluate the efficacy of Akt-targeted PROTAC molecules (Akt-PROTACs) in degrading Akt1, Akt2, and Akt3 proteins. The results indicate that the NanoLuc luciferase method effectively assessed the degradation activity of Akt-PROTACs on Akt1, Akt2, and Akt3 in HEK-293, PC-3, and MDA-MB-468 cells, and the degradation curves of the method were in good agreement with those of the western blot, with good reproducibility between different batches. It is a method worth popularizing to evaluate the degradation efficacy of Akt-PROTACs on Akt1, Akt2, and Akt3 proteins. The schematic diagram of the principle for detecting the degradation of target proteins by PROTACs using NanoLuc luciferase technology and western blot is shown in Fig. 1.

Figure 1.

The schematic diagram of the principle for detecting the degradation of target proteins by PROTACs using NanoLuc luciferase technology and western blot

Materials and methods

Materials and reagents

The pNLF1-N [CMV/Hygro] Vector (pNLF1-N, N1351), pNLF1-C [CMV/Hygro] Vector (pNLF1-C, N1361), FuGENE HD Transfection Reagent (E2312), and Nano-Glo Luciferase Assay System (N1130) were all obtained from Promega. Primary antibodies against Akt1 (75692S), Akt2 (3063S), Akt3 (14982S), and beta-actin (4970S) were purchased from Cell Signaling, and the Goat Anti-Rabbit second antibody (ab205718) was purchased from Abcam. Highly sensitive ECL chemiluminescence detection kit (E412-01) was obtained from Vazyme. Cell culture-related reagents including DMEM medium (C11960500BT), F-12K nutrient mixture (21127-022), Opti-MEM I (31985-070), and fetal bovine serum (10099-141C) were all provided by Gibco. The 96-well white tissue-culture plates (6005680) were from PerkinElmer, and high-resolution precast gels (4%–12%, 36254ES10) were from Yeasen. Compounds including MG-132 (474790) and MLN4924 (S7109) were purchased from Sigma and Selleck, respectively, while MS21, MS21N1, MS170, TAC-027, and TAC-030 were all synthesized by Nanjing Chia-Tai Tianqing Pharmaceutical Co. Ltd.

Cell culture and treatment

Cell lines of HEK-293, PC-3, and MDA-MB-468 were all obtained from Nanjing Cobioer Biosciences (all originated from the American Type Culture Collection). The cells were cultured in DMEM-HG (HEK-293 and MDA-MB-468) and F-12K nutrient mixture (PC-3), supplemented with 10% fetal bovine serum. The cells were cultured in a humidified incubator (5% CO₂, 37°C) and were passaged when the confluence reached 80%–90%. For the western blot assay, cells in logarithmic growth phase were inoculated into 6-well plates with 5 × 10⁵ cells per well and cultured at 5% CO₂ and 37°C for 24 h. On the next day, the cells in five wells were given different concentrations of the compound (1000 nM, 333.33 nM, 111.11 nM, 37.37 nM, and 12.35 nM), and the remaining well received an equal volume of DMSO (final concentration of DMSO in each well was 0.5%). The cells were collected 24 h later to extract total protein for western blot detection.

Plasmid construction

The CDS regions of the full-length human Akt1 (NM_005163.2), Akt2 (NM_001626.6), and Akt3 (NM_005465.7) genes were cloned into the pNLF1-N and pNLF1-C vectors, respectively. The enzyme cleavage sites for the cloning were all 5′-EcoRI-XbaI-3′. The constructed plasmids were named pNLF1-N-Akt1, pNLF1-N-Akt2, pNLF1-N-Akt3, pNLF1-C-Akt1, pNLF1-C-Akt2, and pNLF1-C-Akt3. All the plasmids were constructed by General Biologicals (Anhui) Co. and verified by sequencing.

Plasmid transfection

Assuming that 2 μg of plasmid was transfected into a total number of 1 × 10⁶ cells, they were digested and centrifuged (1000 rpm, 5 min) when the cells were cultured to 80%–90% fusion. After removal of supernatant, cells were counted using a fluorescent cell counter (CountStar-Rigel2) after addition of fresh medium, and cells were adjusted to a density of 1 × 10⁵/mL for a total of 10 mL. Opti-MEM I medium (200 μL) was added to the eppendorf tubes, followed by the desired amount of pNLF1-N-Akt1/2/3 or pNLF1- C-Akt1/2/3 plasmid into Opti-MEM I medium (If the total amount of plasmid is less than 2 μg, pcDNA3.1 (+) plasmid was used to make up the total amount of plasmid to 2 μg. If pcDNA3.1 (+) plasmid is unavailable, other plasmids without the target genes and luciferase gene can be used instead.), and mix thoroughly. Add 6 μL of FuGENE HD transfection reagent (the ratio of volume of transfection reagent to the total amount of plasmid is 3:1), mix gently by inversion for 5–10 times, and incubate for 20 min at room temperature (RT) to form a liposome/DNA complex. At the end of incubation, the liposome/DNA complex was added directly into 10 mL of cell suspension, mixed thoroughly, and inoculated into 96-well plates. The plates were incubated at 37°C in 5% CO₂ incubator for 24–72 h before starting the experiment.

NanoLuc luciferase method to detect protein degradation

Assuming that 10 pg of pNLF1-N-Akt1 plasmid was transfected into 1 × 10⁴ HEK-293 cells. The pNLF1-N-Akt1 plasmid was diluted to a concentration of 1 ng/μL using TE Buffer, and the pcDNA3.1 (+) plasmid was adjusted to a concentration of 1000 ng/μL. Two tubes of liposome/DNA complexes were formulated according to the plasmid transfection section in Materials & Methods (T1: 2 μg pcDNA3.1 (+), T2: 1 ng pNLF1-N-Akt1 + 1999 ng pcDNA 3.1 (+)). After the incubation completed, the complexes were added into two tubes of HEK-293 cells (10 mL), mixed thoroughly and inoculated into white-bottomed impermeable 96-well assay plates at the volume of 100 μL per well (1 × 10⁴ cells). Cells transfected with T1 (T1 cells) were inoculated into two wells (as a negative control), and cells transfected with T2 (T2 cells) were inoculated into 22 wells (This design is suitable for the detection of 1 compound with 10 concentrations and double duplicate wells. If more compounds are planned to be detected, the multiplicity is increased according to this design). The 96-well plates were incubated at 37°C in a 5% CO₂ incubator for 24 h.

On the second day, the compounds to be tested were formulated into a 3.0 mM reservoir solution using DMSO, and 10 concentrations of the compounds and DMSO were added to each of the 20 wells which inoculated T2 cells using a D300e ultra-micro automated dispenser (Tecan). The initial concentration was 10 μM, with a 3-fold gradient dilution and two replicate wells for each concentration, and a final concentration of DMSO in each well was 0.5%. A certain amount of DMSO was added to each of the two wells of the inoculated T1 cells and the remaining two wells of T2 cells to make the final concentration of 0.5% (this step can also be done manually by preparing the compounds and then replacing the cells with drug-containing medium). After dosing, place the 96-well plate into an incubator and continue incubation for the indicated time (12–72 h).

After incubation with the compounds, the cells were removed and allowed to stand for 10 min at RT. The pre-thawed Nano-Glo substrate and Nano-Glo buffer from the Nano-Glo luciferase assay kit were also allowed to stand for 10 min at RT. Then, 50 μL of substrate was added to 2.5 mL of buffer to form working solution (1:50 volume ratio of substrate to buffer) under light protection. Next, add 100 μL of the prepared working solution to each well of a 96-well plate and shake the plate on a microplate shaker for 3 min at RT. Finally, chemiluminescence signal of the wells was detected using a multifunctional microplate reader (Tecan SPARK).

Enzymatic assays

The inhibitory activities of the compounds on Akt1, Akt2, and Akt3 kinases were assayed using Cisbio’s HTRF KinEASE-STK S3 Kit. The Akt1 (0.02 ng/μL), Akt2 (0.4 ng/μL), and Akt3 (0.05 ng/μL) kinases were mixed with gradient concentrations of the compounds in the assay plate, and the reaction was incubated at RT for 10 min. The reaction plate was incubated with kinase substrate and ATP for 50 min at RT. After that, Sa-XL 665 and STK3-antibody-Cryptate were added to the reaction system and incubated for 1 h at RT. The detection plate was analyzed using Envision (PerkinElmer) at 615 nm (Cryptate) and 665 nm (XL3-antibody-Cryptate) to read the fluorescence signal. Ratio_665/615 nm was used to calculate the inhibition rate of the compounds on kinase activity at different concentrations, and the IC₅₀ values of the compounds on kinases were calculated using GraphPad Prism software (Version 8.0).

Western blotting

Total proteins from HEK-293, PC-3, and MDA-MB-468 cells were extracted with RIPA lysis buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and 1× protease/phosphatase inhibitor cocktail. For immunoblotting, an equal amount of proteins was separated on 4%–12% high-resolution precast gel (Hepes-Tris) and transferred onto PVDF membranes. Then, the membranes were blocked with 5% BSA and incubated with primary antibodies (Akt1, Akt2, Akt3, and beta-actin) overnight at 4°C. The primary antibodies were diluted with 5% BSA (1:1000). On the next day, the membranes were rinsed with PBST for three times and incubated with HRP-labeled secondary antibody for 1 h at RT. After incubation, the membranes were rinsed with PBST for three times and exposed using a highly sensitive ECL chemiluminescence detection kit under a chemiluminescence imager (Tanon, 5200 Multi).

Statistical analysis

For the calculation of the half degradation rate (DC50) of a compound to a target protein using the NanoLuc luciferase assay, the mean of the minimum (Mean_min) is calculated using the chemiluminescent signal read from the T1+ DMSO well, and the mean of the maximum (Mean_max) is calculated using the chemiluminescent signal read from the T2+ DMSO well. Chemiluminescent signal read from the T2+ compound well is S_cmpd, then the equation for the degradation rate of this well is [(1−(S_cmpd−Mean_min)/(Mean_max−Mean_min))*100%]. Dose-response curves were generated using the nonlinear regression four-parameter model in GraphPad Prism, and the values of relative DC50, absolute DC50, top degradation ratio, and bottom degradation ratio were calculated. Meanwhile, CV value, which is the ratio between the standard deviation (SD) and the mean values of different assay batches, was used in this study to evaluate the degree of dispersion between different assay batches. The Z-factor of the test was calculated as [1 − (3 * (SD_max − SD_min)/(Mean_max − Mean_min))].

To calculate the DC50s of target protein by compounds detected by western blot, the gray values of the bands were analyzed using ImageJ software (Version 1.52a), and the gray value ratio (GVR) of the target protein to beta-actin was calculated for the computation of the degradation rate. The gray value of the DMSO treatment group was set to the maximum value (GVR_max), and the gray value of the other compounds treatment group was set to GVR_cpmd. Then the degradation rate of the compounds was calculated as [(1 − GVR_cpmd/GVR_max) * 100%]. Dose-response curves were generated using the nonlinear regression four-parameter model in GraphPad Prism (Version 8.0), and the values of relative DC₅₀, absolute DC_50, top degradation ratio, and bottom degradation ratio were calculated.

Results

NanoLuc luciferase method well reproduced the degradation profiles of pan-Akt protein by MS21 and MS170 detected by western blot

MS21 is a pan-Akt PROTAC molecule (Fig. 2A) discovered in 2021 by Xu et al. [16]. Its Akt-binding portion is capivasertib, which is able to bind specifically to the E3 ligase Von Hippel-Lindau (VHL). MS21N1, which share the same Akt binding moiety and linker of MS21, contains a diastereoisomer of VHL-1 that is incapable of binding the VHL E3 ligase (Fig. 2A) [16]. Therefore, MS21N1 was set as a negative control of MS21 which was incapable of inducing degradation the Akt proteins. MS170, another pan-Akt PROTAC molecule, consists of ipatasertib (a small molecule that targets pan-Akt), a linker, and a ligand that recruits cereblon (CRBN) E3 ligase (Fig. 2A) [19]. Since MS21 and MS170 have different Akt-binding structures and E3 ligase ligands, they were used as tool molecules for the degradation assay of pan-Akt in this study. The basic information of the compounds of MS21 and MS170 is shown in Supplementary Table S1.

Figure 2.

Results of pan-Akt proteins degradation by MS21 and MS170 in HEK-293 cells using western blot. (A) Chemical structures of MS21, MS21N1, and MS170. The protein expression of Akt1 (B), Akt2 (C), and Akt 3 (D) treated with different concentrations of MS21 or MS170 in HEK-293 cells detected by western blot, with beta-actin as an internal reference. The degradation curves of Akt1 (E), Akt2 (F), and Akt3 (G) of MS21 in HEK-293 cells detected by western blot. The degradation curves of Akt1 (H), Akt2 (I), and Akt3 (J) of MS170 in HEK-293 cells detected by western blot

We first verified the protein degradation levels of Akt1, Akt2, and Akt3 by MS21 as well as MS170 using western blot in HEK-293 cells. The results showed that MS21 was slightly better than MS170 in degrading Akt1 (Fig. 2B, E, and H), MS21 and MS170 were comparable in degrading Akt2 (Fig. 2C, F, and I), while MS170 showed a stronger degrading potential for Akt3 (Fig. 2D, G, and J). However, the maximum degradation capacities of MS21 and MS170 for Akt1, Akt2, and Akt3 at 1 μM were basically the same, and the DC50s and Max degradation ratios for pan-Akt are shown in Table 1.

Table 1.

Pan-Akt degradation parameters of MS21 and MS170 in HEK-293 detected by western blot.

Protein	Compound	Rel DC50 (nM)	Abs DC50 (nM)	Top degradation ratio (%)	Bottom degradation ratio (%)	Max degradation ratio (%)
Akt1	MS21	262.40	345.14	75.77	8.87	75.77
	MS170	434.50	435.51	73.64	25.99	73.64
Akt2	MS21	160.10	309.74	76.43	−7.37	76.43
	MS170	181.80	220.29	74.43	19.16	74.43
Akt3	MS21	218.20	237.14	90.34	4.78	81.20
	MS170	62.69	67.92	92.52	3.51	92.52

However, due to the low throughput and complicated operation of western blot, we tried to use the NanoLuc luciferase method to test the degradation level of pan-Akt in HEK-293 cells by MS21 and MS170. First, we constructed a full-length NanoLuc N-terminal fusion protein expression plasmid of Akt1, Akt2, and Akt3 based on the pNLF1-N plasmid (named pNLF1-NanoLuc-Akt1/2/3), and its vector structure is shown in Fig. 3A. Since it is an exogenously expressed protein, the plasmid transfection time and transfection amount may have an effect on the degradation activity, so we mapped the plasmid transfection amount and transfection time before formal experiments. The results showed that transfection of HEK-293 cells with pNLF1-NanoLuc-Akt1 plasmid for 24–72 h had almost no effect on the intensity of luminescence (Fig. 3B), whereas the amount of plasmid transfection showed a positive correlation with the intensity of luminescence (Fig. 3B and C). Meanwhile, we also examined the effect of the inoculation amount of cells on the fluorescence intensity, and the results showed that there was no significant increase in the fluorescence intensity when the number of HEK-293 cells in 96-well plates was more than 10,000 cells per well (Fig. 3D). Based on the above results, we chose to evaluate the degradation of Akt1 by MS21 by transfecting four concentrations of pNLF1-NanoLuc-Akt1 plasmid in 10,000 HEK-293 cells per well. The protein degradation profile of MS21 on Akt1 detected using NanoLuc luciferase was closest to that of western blot when 10 pg of pNLF1-NanoLuc-Akt1 was transfected per 10,000 cells (Fig. 3E), and the luminescence fluorescence intensity corresponding to this plasmid concentration was located in the range of 1 × 10⁶ and 1 × 10⁷. To verify this speculation, we found the corresponding plasmid transfection amounts of pNLF1-NanoLuc-Akt2 and pNLF1-NanoLuc-Akt3 at this fluorescence intensity (10 pg per 10,000 cells and 100 pg per 10,000 cells, respectively). Thus, we evaluated the degradation activities of MS21 and MS170 on pan-Akt using the NanoLuc luciferase assay in HEK-293 cells and compared them with western blot results. The NanoLuc assay results under the above conditions were highly consistent with western blot results (Fig. 3F–K). Furthermore, the batch-to-batch stability of the method was also examined. As shown in Fig. 3F–K, the results indicated that the degradation curves of the three assays performed at different times overlapped well. The three experiments were in good agreement regarding the four indexes of relative DC50, absolute DC50, top degradation ratio, and max degradation. However, the bottom degradation ratio exhibited significant fluctuation, as presented in Supplementary Table S2. The ability to distinguish the degradation capacity of different compounds on Akt1, Akt2, and Akt3 is a crucial factor for screening Akt PROTAC molecules. To compare the NanoLuc luciferase method with the classical western blot method, we analyzed the ratio of MS21 to MS170 in DC50 and Dmax parameters. It was indicated that NanoLuc luciferase effectively distinguished the degradation ability of compounds to different Akt isoforms (Fig. 3L). The degradation of pan-Akt by MS21N1 in HEK-293 cells was also tested using both western blot and NanoLuc luciferase. The results confirmed weaker degradation of pan-Akt proteins by MS21N1, with three batches of NanoLuc luciferase experiments consistent (Supplementary Fig. S1). In summary, NanoLuc luciferase method well reproduced the degradation profiles of pan-Akt protein by MS21, MS21N1, and MS170 detected by western blot.

Figure 3.

Results of pan-Akt proteins degradation by MS21 and MS170 in HEK-293 cells using NanoLuc luciferase assay. (A) Plasmid map of pNLF1-NanoLuc-Akt1. (B) Effects of different pNLF1-NanoLuc-Akt1 plasmid transfection amount and transfection time on luminescence intensity in HEK-293 cells. (C) Effects of different pNLF1-NanoLuc-Akt2/3 plasmid transfection amount on luminescence intensity in HEK-293 cells (transfection for 24 h). (D) Effects of different inoculum numbers of HEK-293 cells on luminescence intensity. (E) Comparison of the degradation curves of Akt1 protein by different pNLF1-NanoLuc-Akt1 plasmid transfection amount (transfected for 24 h) with those detected by western blot. Comparison of the degradation curves of Akt1 (F), Akt2 (G), and Akt3 (H) of MS21 in HEK-293 cells detected by NanoLuc luciferase with those detected by western blot. Comparison of the degradation curves of Akt1 (I), Akt2 (J), and Akt3 (K) of MS170 in HEK-293 cells detected by NanoLuc luciferase with those detected by western blot. (L) The ratio of MS21 to MS170 in DC50 and Dmax parameters of pan-Akt proteins in HEK-293 detected by western blot and NanoLuc luciferase

NanoLuc luciferase assay for degradation of pan-Akt proteins has potential to be expanded to multiple cell lines

To evaluate whether the NanoLuc luciferase assay could be expanded to more cells, we chose tumor cells corresponding to potential indications of Akt inhibitors (human prostate cancer cells PC-3, and human breast cancer cells MDA-MB-468) for testing. Western blot results showed that the degradation of Akt1 and Akt2 by MS21 was slightly superior to that by MS170 in PC-3 cells (Fig. 4A), while the degradation of Akt3 by MS170 was superior to that by MS21 (Supplementary Table S3). In MDA-MB-468 cells, MS170 was overall superior to MS21 for the degradation of Akt1, Akt2, and Akt3 (Fig. 4B and Supplementary Table S4). Meanwhile, we assessed the degradation of Akt1, Akt2, and Akt3 by MS21 and MS170 using NanoLuc luciferase assay simultaneously in PC-3 and MDA-MB-468 cells. Overall, the degradation profiles of Akt1, Akt2, and Akt3 by MS21 and MS170 using NanoLuc luciferase and western blot assays in PC-3 cells were generally in agreement within the set range of assayed concentrations (Fig. 4B–G). The NanoLuc luciferase method effectively discriminated the difference in degradation ability of MS21 and MS170 for all Akt isoforms, except for Akt3 in the DC50, as shown by the results of the ratios of MS21 to MS170 in DC50 and Dmax (Fig. 4H). In MDA-MB-468 cells, the curves of Akt1, Akt2, and Akt3 degradation by the two methods overlapped more closely (Fig. 4I–N) and were more consistent in the ability of MS21 and MS170 to discriminate the degradation strength of each Akt isoform (Fig. 4O). We similarly applied the NanoLuc luciferase method in PC-3 and MDA-MB-468 for protein degradation of Akt1, Akt2, and Akt3 in three replicates. The results showed that in addition to bottom degradation ratio, all three replicate experiments exhibited acceptable variability in terms of DC50, top degradation ratio, and max degradation ratio, with CV values ranging from 1.48% to 33.83% in PC-3 cells (Supplementary Table S5 and Supplementary Fig. S2) and ranging from 0.38% to 29.03% in MDA-MB-468 cells (Supplementary Table S6 and Supplementary Fig. S2). Taken together, NanoLuc luciferase has the potential to be expanded to more cells for protein degradation assessment and showed good method stability.

Figure 4.

Results of pan-Akt proteins degradation by MS21 and MS170 in PC-3 and MDA-MB-468 cells using NanoLuc luciferase assay. (A) The protein expression of pan-Akt treated with different concentrations of MS21 or MS170 in PC-3 and MDA-MB-468 cells detected by western blot, with beta-actin as an internal reference. Comparison of the degradation curves of Akt1 (B), Akt2 (C), and Akt3 (D) of MS21 in PC-3 cells detected by NanoLuc luciferase with those detected by western blot. Comparison of the degradation curves of Akt1 (E), Akt2 (F), and Akt3 (G) of MS170 in PC-3 cells detected by NanoLuc luciferase with those detected by western blot. (H) The ratio of MS21 to MS170 in DC50 and Dmax parameters of pan-Akt proteins in PC-3 detected by western blot and NanoLuc luciferase. Comparison of the degradation curves of Akt1 (I), Akt2 (J), and Akt3 (K) of MS21 in MDA-MB-468 cells detected by NanoLuc luciferase with those detected by western blot. Comparison of the degradation curves of Akt1 (L), Akt2 (M), and Akt3 (N) of MS170 in MDA-MB-468 cells detected by NanoLuc luciferase with those detected by western blot. (O) The ratio of MS21 to MS170 in DC50 and Dmax parameters of pan-Akt proteins in MDA-MB-468 detected by western blot and NanoLuc luciferase

NanoLuc luciferase assay can be used to screen selective Akt PROTACs

We attempted to use the NanoLuc luciferase method for the screening of pan-Akt as well as selective Akt PROTACs. The pan-Akt PROTAC molecule TAC-027 and the selective Akt3 PROTAC molecule TAC-030 were chosen (designed by our company, structures not shown). Western blot analysis confirmed the selectivity of TAC-027 and TAC-030 in degrading Akt1, Akt2, and Akt3 proteins (Fig. 5A). Consistent with western blot, the NanoLuc luciferase also well evaluated the degradation ability of TAC-027 as well as TAC-030 on Akt1, Akt2, and Akt3, and the degradation curves were in good agreement with western blot (Fig. 5B–G).

Figure 5.

Results of pan-Akt proteins degradation by TAC-027 and TAC-030 in HEK-293 cells using NanoLuc luciferase assay. (A) The protein expression of pan-Akt treated with different concentrations of TAC-027 or TAC-030 in HEK-293 cells detected by western blot, with beta-actin as an internal reference. Comparison of the degradation curves of Akt1 (B), Akt2 (C), and Akt3 (D) of TAC-027 in HEK-293 cells detected by NanoLuc luciferase with those detected by western blot. Comparison of the degradation curves of Akt1 (E), Akt2 (F), and Akt3 (G) of TAC-030 in HEK-293 cells detected by NanoLuc luciferase with those detected by western blot

Results based on N-terminal NanoLuc constructs are closer to the actual results

Since NanoLuc is a tagged protein, it can be fused to either the N-terminus or the C-terminus of the target proteins. The previous data were all derived based on N-terminal NanoLuc constructs, so does the placement of this protein in different positions have an effect on the results? Therefore, we constructed C-terminal NanoLuc-fused Akt1, Akt2, and Akt3 expression plasmids, and the typical plasmid profiles (named pNLF1-C-NanoLuc-Akt1/2/3, respectively) are shown in Fig. 6A. Subsequently, we mapped the transfection concentrations of the three expression plasmids. Compared with the N-terminal NanoLuc fusion protein, the C-terminal NanoLuc fusion protein showed lower luminescence intensity after transfection of HEK-293 cells (Figs. 3B–C and 6B). We followed the previously mapped out pattern here and chose plasmid concentrations with fluorescence intensities between 1 × 10⁶ and 1 × 10⁷ (500 pg, 1 ng, and 5 ng of pNLF1-C-NanoLuc-Akt1/2/3 plasmid per 10,000 cells, respectively) for the cell transfection (Fig. 6B). Overall, the data for pNLF1-C-NanoLuc plasmid followed the same trend as pNLF1-NanoLuc, but Abs DC50 showed a high bias, while Dmax showed a low bias (Table 2). In addition, evaluation using the pNLF1-C-NanoLuc plasmid likewise exhibited high stability between batches, further validating the stability of the method (Fig. 6C–H). The above results suggested that the evaluation of the pNLF1-NanoLuc plasmid is more recommended for the evaluation of the degradation ability of pan-Akt PROTACs.

Figure 6.

Results of pan-Akt proteins degradation by MS21 and MS170 in HEK-293 cells using NanoLuc luciferase assay (C-terminal NanoLuc fusion construct). (A) Plasmid map of pNLF1-C-NanoLuc-Akt1. (B) Effects of different pNLF1-C-NanoLuc-Akt1/2/3 plasmid transfection amount on luminescence intensity in HEK-293 cells (transfection for 24 h). Comparison of the degradation curves of Akt1 (C), Akt2 (D), and Akt3 (E) of MS21 in HEK-293 cells detected by NanoLuc luciferase assay in three independent experiments. Comparison of the degradation curves of Akt1 (F), Akt2 (G), and Akt3 (H) of MS170 in HEK-293 cells detected by NanoLuc luciferase assay in three independent experiments

Table 2.

Comparison of pan-Akt degradation parameters of MS21 and MS170 in HEK-293 detected by using western blot and NanoLuc luciferase (N-terminal or C-terminal NanoLuc fusion constructs).

Protein	Compound	Experiment	Rel DC50 (nM)	Abs DC50 (nM)	Top degradation ratio (%)	Bottom degradation ratio (%)	Max degradation ratio (%)
Akt1	MS21	WB	262.40	345.14	75.77	8.87	75.77
		N-terminal	368.97	313.37	98.30	7.81	90.96
		C-terminal	597.93	>10,000	51.09	4.77	51.09
	MS170	WB	434.50	435.51	73.64	25.99	73.64
		N-terminal	495.53	459.94	92.21	10.78	85.92
		C-terminal	1107.67	1844.36	77.56	6.40	77.75
Akt2	MS21	WB	160.10	309.74	74.43	−7.37	76.43
		N-terminal	434.47	364.77	96.34	9.23	91.70
		C-terminal	379.77	997.80	67.76	8.37	67.76
	MS170	WB	181.80	220.29	74.43	19.19	74.43
		N-terminal	751.67	618.86	100.74	4.97	88.00
		C-terminal	111.04	949.16	72.19	−3.14	72.19
Akt3	MS21	WB	218.20	237.14	90.34	4.78	81.20
		N-terminal	496.60	534.47	93.49	2.21	97.75
		C-terminal	408.10	737.09	79.39	3.03	72.86
	MS170	WB	62.69	67.92	92.52	3.51	92.52
		N-terminal	209.90	225.49	95.97	1.01	93.15
		C-terminal	79.72	89.38	95.35	−0.12	94.00

Protein degradation by PROTACs is proteasome dependent as detected by the NanoLuc luciferase assay

Protein degradation by PROTACs is dependent on the proteasome pathway. To verify whether the detection of protein degradation by this method is also affected by proteasomal activity, we used MG-132 to inhibit proteasomal activity, as well as the NAE inhibitor MLN4924 to inhibit the ubiquitin ligation process of the target proteins (NAE, a NEDD8-activating enzyme, promotes the binding of the ubiquitin-like molecule NEDD8 to the protein to regulate the activity of the protein). In this study, we found that the use of MG-132 was able to greatly inhibit the degradation of pan-Akt proteins by MS21 as well as MS170 and increased the amount of intracellular protein by more than 2-fold. Compared to the effect of MG-132, MLN4924 was also able to inhibit the degradation of pan-Akt proteins by MS21 and MS170, but the degree of inhibition was slightly weaker (Fig. 7A–F and Supplementary Table S7). This result suggested that the degradation of pan-Akt proteins detected using the NanoLuc luciferase method is indeed proteasome pathway-dependent.

Figure 7.

Pan-Akt protein degradation by MS21 and MS170 is proteasome dependent as detected by the NanoLuc luciferase assay. Effects of MG-132 and MLN4924 on the degradation of Akt1 (A), Akt2 (B), and Akt3 (C) by MS21 in HEK-293 cells using NanoLuc luciferase assay. Effects of MG-132 and MLN4924 on the degradation of Akt1 (D), Akt2 (E), and Akt3 (F) by MS170 in HEK-293 cells using NanoLuc luciferase assay. (G) Summary of batch-to-batch stability of pan-Akt protein degradation parameters of MS21 and MS170 in HEK-293, PC-3, and MDA-MB-468 cells using NanoLuc luciferase assay

Discussion

PROTAC technology has emerged as a novel therapeutic paradigm in recent years. Currently, about 20%–25% of protein targets are being investigated, with most of the work focusing on their enzymatic functions. PROTACs make up for the limitation of transcription factors, nuclear proteins, and other scaffolding proteins that are difficult to develop as small molecule drugs [1]. Currently, PROTACs have successfully degraded a variety of proteins such as BTK, BRD4, AR, ER, STAT3, IRAK4, and tau [18]. Both ARV-110 and ARV-471 showed excellent efficacy in phase II clinical trials. This technology will bring great hope for the development of clinical drugs in the future with its unique advantages.

Western blot, reporter gene assay, and MS are the main methods for evaluating the degradation ability of PROTACs on target proteins. Western blot is the most classical intracellular protein detection method, and digital western blot has been developed in recent years, which greatly simplifies and standardizes the experimental process. However, based on the detection principle of antigen-antibody binding of western blot, it still has the characteristics of being affected by antibodies and low throughput, which is not suitable for large-scale screening of PROTACs in protein degradation. In contrast, the reporter gene method is a more suitable method to detect the degradation ability of PROTACs in target proteins. HiBiT technology developed by Promega involves embedding an 11-amino acid HiBiT peptide into a specific cellular endogenous target protein via CRISPR. LgBiT is a 17 kDa protein that binds to HiBiT with high affinity to form an intact NanoBiT luciferase, which is used to detect endogenous protein levels [20]. LgBiT can be used for real-time monitoring of protein amount in living cells if expressed intracellularly, or for end-point protein amount monitoring if added to lysates. This is indeed a good method for evaluating the expression of endogenous target proteins and can be used for both real-time monitoring and endpoint detection of protein expression. However, this method requires the use of CRISPR to construct a stably transfected cell line with HiBiT-target protein fusion, which is a costly and time-consuming process, and is more limited in terms of cell line expansion. NanoLuc is a stronger luciferase than traditional luciferase such as Fluc and RLuc, which is fused to the target protein to determine the amount of the target proteins by the intensity of luminescence. Because of its high fluorescence intensity, it is capable of detecting proteins with very low expression levels, and transient transcription makes this method easy to operate and highly scalable. However, this method introduces a modified exogenous target protein into the cell, and the endogenous protein abundance may interfere with the degradation efficacy of the PROTACs, so it is controversial whether it is a suitable method for protein degradation evaluation. The aim of this study was to test the feasibility of NanoLuc luciferase method to evaluate the protein degradation capacity of PROTACs targeting pan-Akt.

In this study, a NanoLuc Luciferase assay was developed to detect the intensity of degradation of pan-Akt protein in cells by PROTACs. The reliability of the assay was verified using the reported pan-Akt PROTACs MS21 and MS170. These two molecules are more representative as they have different Akt ligand structures and target different E3 ligases. Since the target protein of NanoLuc fusion is an exogenous protein, the actual degradation ability of PROTAC molecules may be underestimated if too much plasmid is transferred, and may not being up to the requirements because of low baseline fluorescence intensity if too little plasmid is transferred. Therefore, it is necessary to feel the amount of plasmid transfection before using this method. It was found that when the baseline luminescence intensity was located in the range of 1 × 10⁶ and 1 × 10⁷, the degradation curves of pan-Akt calculated by this method had a better resemblance to the degradation curves of western blot. In addition, we made a detailed comparison of the main degradation parameters calculated by this assay as well as the discriminatory power of different compounds on the degradation of different Akt isoforms with western blot, and the results showed that the main parameters were in good agreement. Since HEK-293 cell is only a kind of tool cell, it is very difficult to be used for the function evaluation of the compounds. Considering that the target indications of many PROTACs are tumors, there is a strong need for testing diversity of tumor cell lines. The performance of PROTACs in terms of their ability to degrade target proteins in different cells can vary widely due to differences in target protein abundance as well as E3 ligase species. This puts a higher demand on the ability to extend the assays for PROTACs, which is one of the elements we focused on in this study. The results showed that the NanoLuc luciferase assay was well applicable to prostate cancer PC-3 cells as well as breast cancer MDA-MB-468 cells, and we also found that it could be stably applied to a wider range of tumor cell lines such as LNCaP (data not shown). Since both MS21 and MS170 are pan-Akt PROTACs, the test results of this method on TAC-030 further validated its ability to discriminate selective Akt PROTACs, which fully validates the reliability of this method.

Another risk of reporter gene method is stability, and NanoLuc is a new luciferase whose stability in experiments has not been overly reported. Therefore, we fully evaluated the batch-to-batch stability of this assay in HEK-293 cells, PC-3 cells, and MDA-MB-468 cells in this study (Fig. 7G). Of the 24 assays (72 batches), the percentages of CV values less than 25% in relative DC50, absolute DC50, top degradation ratio, and Dmax were 66.7%, 83.3%, 100%, and 100%, respectively. Although the bottom degradation ratio variance was higher, it had almost no effect on the calculation of the key results. The value of the Z-factor is a relative metric used to distinguish signal from noise. It has become a major parameter for assessing the quality of high-throughput testing methods. The Z-factors in the NanoLuc luciferase assays tested in this study were summarized in Supplementary Table S8, with a high mean value (0.86 ± 0.10, ranging from 0.57 to 1.00). Therefore, the NanoLuc luciferase is a stable and high-resolution assay for evaluating the target protein degradation by Akt PROTACs.

Since NanoLuc binds to the target protein as a fusion protein, its attachment to the N-terminus or C-terminus of the target protein is also a factor that affects the reliability of the method, and there is no one matter to predict which fusion method is superior. In this study, we compared the difference between the detection results of N-terminal fusion protein and C-terminal fusion protein at the same time. We found that the results from the N-terminal fusion protein were more comparable to the results from western blot. Since the N-terminal is closer to the promoter on the plasmid, the expression of NanoLuc in N-terminal is stronger and more stable than that of the C-terminal. It is generally preferred to use the N-terminal fusion constructs for the initial evaluation, and then consider using the C-terminal fusion constructs for the secondary evaluation if the initial evaluation results are not satisfactory.

In conclusion, NanoLuc luciferase for assessing the ability of pan-Akt PROTACs to degrade Akt proteins is an evaluative method with high accuracy, stability, simplicity, and a short cycle time (about 50 h from cell inoculation to assay). In addition, the requirement of just a NanoLuc fusion protein construct makes this method a low-cost, flexible, and cell line expandable method that is well suited for early compound screening and target cell line selection of PROTACs. It also has strong potential for expansion into early protein degradation screening of more targeted PROTACs.

Author contributions

Xiaojun Ji (Conceptualization [equal], Data curation [lead], Formal analysis [lead], Investigation [equal], Methodology [lead], Project administration [lead], Resources [lead], Software [equal], Visualization [equal], Writing—original draft [lead]), Lei Miao (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing—review & editing [equal]), Yebin Wu (Data curation [equal], Investigation [equal], Resources [equal], Validation [equal]), Tingli Zhao (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Resources [equal], Software [equal], Validation [equal]), Yaxuan Si (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal]), Xiaoyun Tan (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal]), Qiuhua Zhou (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Resources [equal], Validation [equal]), Rui Zuo (Data curation [equal], Investigation [equal], Methodology [equal], Software [equal], Validation [equal]), Junjie Pei (Data curation [equal], Investigation [equal], Methodology [equal]), Changyou Ma (Conceptualization [equal]), Jian Wu (Conceptualization [equal], Funding acquisition [equal], Project administration [equal], Resources [equal], Validation [equal], Writing—review & editing [equal]), Dan Xu (Conceptualization [equal], Funding acquisition [equal], Project administration [equal], Resources [equal], Writing—review & editing [equal])

Supplementary data

Supplementary data are available at Biology Methods and Protocols online.

Conflict of interest statement

The authors declare no conflicts of interest.

Funding

This work was supported by a grant from the Natural Science Foundation of Jiangsu (BK20210055 and BK20230172).