The malachite green assay protocol is a simple and cost-effective method to discover heat shock protein 90 (Hsp90) suppressors, as well as other inhibitor compounds against ATP-dependent enzymes.
Heat shock protein 90 (Hsp90) is a promising anticancer target because of its chaperoning effect on multiple oncogenic proteins. The activity of Hsp90 is dependent on its ability to hydrolyze adenosine triphosphate (ATP) to adenosine diphosphate (ADP) and free phosphate. The ATPase activity of Hsp90 is linked to its chaperoning function; ATP binds to the N-terminal domain of the Hsp90, and disrupting its binding was found to be the most successful strategy in suppressing Hsp90 function. The ATPase activity can be measured by a colorimetric malachite green assay, which determines the amount of free phosphate formed by ATP hydrolysis. Here, a procedure for determining the ATPase activity of yeast Hsp90 by using the malachite green phosphate assay kit is described. Further, detailed instructions for the discovery of Hsp90 inhibitors by taking geldanamycin as an authentic inhibitor is provided. Finally, the application of this assay protocol through the high-throughput screening (HTS) of inhibitor molecules against yeast Hsp90 is discussed.
Heat shock protein 90 (Hsp90) is a molecular chaperone that maintains the stability of proteins responsible for the development and progression of cancer. In addition, proteins responsible for the development of resistance to antineoplastic agents are also clients of Hsp901. Hsp90 is overexpressed ubiquitously in all cancer cell types (>90% of cellular proteins), compared to normal cells where it may constitute less than 2% of total proteins. Moreover, the Hsp90 of cancer cells resides in a complex with co-chaperones, whereas in a normal cell it is present predominantly in a free, un-complexed state2,3. In recent years, several Hsp90 inhibitors have been demonstrated to possess senolytic effects in in vitro and in vivo studies, where they have significantly improved the life span of mice4,5,6. All the aforementioned findings substantiate the fact that Hsp90 inhibitors could be effective in multiple cancer types, with fewer adverse effects and reduced chances of developing resistance. The chaperoning function of Hsp90 is accomplished by binding ATP at the N-terminal domain of Hsp90 and hydrolyzing it into ADP and free phosphate7. Small molecules that competitively bind to the ATP binding pocket of Hsp90 were found to successfully suppress the chaperoning effect of the protein. To date, this remains the best strategy for Hsp90 inhibition, which is supported by the fact that such inhibitors have reached clinical trials8. One of them, Pimitespib, was approved in Japan for the treatment of gastrointestinal stromal tumor (GIST) in June 20229. This is the first Hsp90 inhibitor approved since the druggability of the chaperone was established in 199410.
The malachite green assay is a simple, sensitive, fast, and inexpensive procedure for the detection of inorganic phosphate, suitable for automation and high-throughput screening (HTS) of compounds against its desired target11. The assay has been successfully employed for the screening of Hsp90 inhibitors in small lab-scale setups, as well as in a HTS12,13,14,15,16,17. The assay uses a colorimetric method that determines the free inorganic phosphate formed due to the ATPase activity of Hsp90. The basis of this quantification is the formation of a phosphomolybdate complex between free phosphate and molybdenum, which subsequently reacts with malachite green to generate a green color (Figure 1). This rapid color formation is measured on a spectrophotometer, or on a plate reader, between 600-660 nm18,19.
In the present protocol, the procedure for carrying out a malachite green assay with yeast Hsp90 and subsequent identification of inhibitors against the chaperone is described. The natural product molecule, geldanamycin (GA), with which the druggability of Hsp90 was first established, was taken as an authentic inhibitor10. HTS has become an integral part of the current drug discovery program, owing to the availability of a large number of molecules for testing. This technique has gained more significance in the past 2 years because of the urgent need for repurposing drugs for treating Covid-19 infection20,21. Therefore, a detailed outline for the HTS of molecules against yeast Hsp90 protein by adopting the malachite green assay method is presented.
1. Lab-scale malachite green assay
2. High-throughput screening of Hsp90 inhibitors by malachite green assay
NOTE: The protocol for high-throughput screening is similar to the lab-scale methodology. The final well volume in each case is 80 µL. However, there is a slight difference in the order of the addition of reagents. In the lab-scale based method, there are five stages of solution addition (34 µL of water, 32 µL of buffer, 2 µL compound in DMSO, 8 µL of Hsp90, and finally 4 µL of ATP solution). In contrast, with HTS there are three stages of addition (40 µL of buffer solution containing yeast Hsp90, 2 µL of DMSO in 18 µL of water containing the compounds, and finally 20 µL of ATP dissolved in water). The minimum amount of solution that can be pipetted accurately in the HTS setup is 20 µL. Hence, a difference in pipetting between lab and HTS scales is observed.
The results of the assay are interpreted in terms of absorbance due to free phosphate ion concentration. The absorbance by free phosphate due to ATP hydrolysis by the yeast Hsp90 at 620 nm is considered as 100% ATPase activity, or zero percentage protein inhibition. The inhibition of protein leads to the cessation of ATP hydrolysis (less free phosphate). which is reflected in terms of decreased absorbance at 620 nm.
Results of lab-scale malachite green assay
The standard graph for the phosphate standard is depicted in Figure 3. The activity of Hsp90 is measured in terms of its ability to hydrolyze ATP to ADP and inorganic phosphate (Pi). A higher free phosphate concentration leads to an increase in complex formation with malachite green, which causes an intense green color detectable at 620 nm (Figure 4).
The percentage inhibition is calculated by the following equation:
% Inhibition
The absorbance of Hsp90 was considered as zero percentage inhibition (100% ATPase activity). The percentage ATPase activity is used for measuring the dose-dependent inhibition of the protein by GA. It was observed that GA inhibited the protein in a dose dependent manner, with an IC50 value of 0.85 µM (Figure 5 and Table 7).
The sigmoidal curve approach of graphing software was utilized to determine the IC50 value (concentration at which the molecules exhibit 50% of Hsp90 ATPase activity).
Results of high-throughput screening (HTS) of Hsp90 inhibitors by malachite green assay
The HTS was carried out with test compounds (codes 3 to 96) in duplicate by taking GA (20 µM final well concentration) as the reference standard. The absorbance at 620 nm of the compounds was not known and hence the absorbance of the compounds alone was recorded (main plates 1 and 2; Figure 6). The absorbance value for each well of main plates 3 and 4 (with Hsp90; Figure 7) was deducted from the corresponding absorbance in plates 1 and 2 (compound alone). The absorbance of Hsp90 was considered as 100% ATPase activity.
The percentage inhibition was calculated using the following formula:
% Inhibition
The 20 µM GA concentration demonstrated 75.82% inhibition. A 100 µM of compound 82 showed 100% inhibition of the yeast Hsp90 protein, whereas compounds 6 and 95 exhibited 73.07% and 62.88% inhibition, respectively, at 100 µM (Table 8). The other compounds suppressed the protein by less than 50% at 100 µM and were considered to be inactive.
Figure 1: Reactions of malachite green (MG) reagent. (A) Structure of malachite green. (B) Reactions of malachite green reagent A (ammonium molybdate in 3 M HCl) with Pi, and subsequent reaction with reagent B (malachite green in polyvinyl alcohol). Please click here to view a larger version of this figure.
Figure 2: Experimental scheme for HTS protocol. The figure depicts the order of addition of reagents to the main assay plate from the addition plates. Please click here to view a larger version of this figure.
Figure 3: Standard plot for standard phosphate concentration. The X-axis represents Pi concentration in micromolar, and absorbance at 620 nm is depicted on the Y-axis. The error bars represent deviation between the two sample numbers (n = 2). Please click here to view a larger version of this figure.
Figure 4: Color development after addition of malachite green reagent. The green color becomes more intense with high phosphate concentration and more ATPase activity. Please click here to view a larger version of this figure.
Figure 5: IC50 calculation for geldanamycin against yeast Hsp90. Each point is a mean of two determinations (n = 2). Please click here to view a larger version of this figure.
Figure 6: Color development after addition of malachite green reagent. (A) Color development in main plate 1. The pink color in the A11 well is due to the compound color. (B) Color development in main plate 2. The pink color in the A11 well is due to the colored nature of the compound. Please click here to view a larger version of this figure.
Figure 7: Color development after addition of malachite green reagent. Color development after the addition of malachite green reagent. (A) Color development in main plate 3. (B) Color development in main plate 4. Please click here to view a larger version of this figure.
Stock solution | Final conc. | Dilution (Final/Stock) | Amount required for 100 ml stock (Dilution × total volume) |
1000 mM Tris-HCl, pH 7.4 | 200 mM | 0.2 | 20 mL |
1000 mM KCl | 40 mM | 0.04 | 4 mL |
1000 mM MgCl2 | 12 mM | 0.012 | 1.2 mL |
Ultra-pure water | NA | NA | Add to 100 mL total volume |
Table 1: Preparation of assay buffer.
# | 1 | 2 | 3 | 4 | 5 |
A | Ultra-pure water | Ultra-pure water | Blank | Blank | |
B | 4 µM PB | 4 µM PB | Negative Control | Negative control | |
C | 8 µM PB | 8 µM PB | GA(100 µM) + Hsp90 | GA(100 µM) + Hsp90 | |
D | 12 µM PB | 12 µM PB | GA(20 µM) + Hsp90 | GA(20 µM) + Hsp90 | |
E | 16 µM PB | 16 µM PB | GA(4 µM) + Hsp90 | GA(4 µM) + Hsp90 | |
F | 24 µM PB | 24 µM PB | GA(0.8 µM) + Hsp90 | GA(0.8 µM) + Hsp90 | |
G | 32 µM PB | 32 µM PB | GA(0.16 µM) + Hsp90 | GA(0.16 µM) + Hsp90 | |
H | 40 µM PB | 40 µM PB | GA(0.032 µM) + Hsp90 | GA(0.032 µM) + Hsp90 | |
Note: The total well volume is 80 μL. PB = Phosphate buffer standard. |
Table 2: The layout of the 96-well plate.
# | Blank | Negative control | Yeast Hsp90 + GA |
Assay buffer | 40 µL | 32 µL | 32 µL |
DMSO | 2 µL | 2 µL | – |
GA | – | – | 2 µL |
Hsp90 | – | 8 µL | 8 µL |
ATP (4mM) | 4 µL | 4 µL | 4 µL |
Ultra-pure water | 34 µL | 34 µL | 34 µL |
Total volume | 80 µL | 80 µL | 80 µL |
Table 3: Constituent of each well.
Plate ID | Explanation |
1 | Main plate containing compound |
2 | Main plate containing compound (duplicate of plate 1) |
3 | Main plate with Hsp90 and compound |
4 | Main plate with Hsp90 and compound (duplicate of plate 3) |
CP | Compound solution |
A | Buffer solution |
B | Buffer solution with Hsp90 |
C | ATP solution |
D | Malachite green reagent solution |
Note: All plates are 96-well transparent. Main plate: plates where the final ATPase reaction takes place. Plate CP, A, B, C, and D are addition plates containing specific reagents. |
Table 4: Types of plates used in the HTS assay.
# | 1 | 2 | 3 | 4 | CP | A | B | C | D | ||
Compound/GA solution in 1:10 DMSO:water | 20 µL | 20 µL | 20 µL | 20 µL | 90 µL | – | – | – | – | ||
Hsp90 solution: buffer solution (1:4) | – | – | 40 µL | 40 µL | – | – | 90 µL | – | – | ||
Buffer solution | 40 µL | 40 µL | – | – | – | 90 µL | – | – | – | ||
ATP (0.8 mM) | 20 µL | 20 µL | 20 µL | 20 µL | – | – | – | 90 µL | – | ||
Malachite green reagent | – | – | – | – | – | – | – | – | 90 µL | ||
Total volume | 80 µL | 80 µL | 80 µL | 80 µL | 90 µL | 90 µL | 90 µL | 90 µL | 90 µL | ||
Note: GA = Geldanamycin. For addition plates CP, A, B, C, and D, an extra 10 μL is taken for the perfect transfer of solutions to the main plates. |
Table 5: Constituent of each well in the HTS protocol.
# | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
A | DMSO+W | 9 | 17 | 25 | 33 | 41 | 49 | 57 | 65 | 73 | 81 | 89 |
B | GA | 10 | 18 | 26 | 34 | 42 | 50 | 58 | 66 | 74 | 82 | 90 |
C | 3 | 11 | 19 | 27 | 35 | 43 | 51 | 59 | 67 | 75 | 83 | 91 |
D | 4 | 12 | 20 | 28 | 36 | 44 | 52 | 60 | 68 | 76 | 84 | 92 |
E | 5 | 13 | 21 | 29 | 37 | 45 | 53 | 61 | 69 | 77 | 85 | 93 |
F | 6 | 14 | 22 | 30 | 38 | 46 | 54 | 62 | 70 | 78 | 86 | 94 |
G | 7 | 15 | 23 | 31 | 39 | 47 | 55 | 63 | 71 | 79 | 87 | 95 |
H | 8 | 16 | 24 | 32 | 40 | 48 | 56 | 64 | 72 | 80 | 88 | 96 |
Note: 3-96 are compound codes for evaluating Hsp90 inhibition potential. GA = Geldanamycin. W=Ultra-pure water |
Table 6: Well plate lay out for main plate 1 and 2 (without Hsp90), and 3 and 4 (with Hsp90).
Well Constituent | % ATPase activity* | % inhibition |
Hsp90 | 100 | 0.00 |
GA(100 µM) | 17.46 | 82.55 |
GA (20 μM) | 17.77 | 82.23 |
GA(4 μM) | 18.27 | 81.73 |
GA (0.8 μM) | 58.43 | 41.57 |
GA (0.16 μM) | 91.38 | 8.62 |
GA (0.032 μM) | 92.86 | 7.14 |
*A mean of two independent determination |
Table 7: Percentage inhibition and ATPase activity of geldanamycin.
Well Constituent | Abs (A) | Well Constituent | Abs (B) | B-A | % ATPase activity* | % inhibition |
Blank | 0.281 | Negative control | 0.663 | 0.383 | 100 | 0 |
GA (20 µM) | 0.295 | GA (20 µM)+Hsp90 | 0.388 | 0.093 | 24.18 | 75.82 |
Compound 6 (100 µM) | 0.3 | Compound 6 (100 µM) + Hsp90 | 0.403 | 0.103 | 26.93 | 73.07 |
Compound 82 (100 µM) | 0.49 | Compound 82 (100 µM) + Hsp90 | 0.462 | -0.028 | -7.32 | 107.32 |
Compound 95 (100 µM) | 0.327 | Compound 95 (100 µM) + Hsp90 | 0.469 | 0.142 | 37.12 | 62.88 |
Table 8: Percentage inhibition and ATPase activity of geldanamycin and selected compounds.
Hsp90 is a significant target for the discovery of novel anticancer drug molecules. Since its druggability was established in 199410, 18 molecules have reached clinical trials. At present, seven molecules are in various phases of clinical trials, either alone or in combination22. All such small molecules are N-terminal ATP binding inhibitors. The other means of inhibiting the chaperone (C-terminal inhibitors, middle domain inhibitors) have not proceeded as fast as N-terminal inhibitors. Hence, the N-terminal ATP binding inhibitor compounds still hold promise of progression to a clinically marketable molecule. Furthermore, the only Hsp90 inhibitor, approved in June, 2022 (Pimitespib for the treatment of GIST in Japan), is an N-terminal ATP-binding molecule9. However, a single molecule, to date, has not reached a marketable stage in other parts of the world23,24. There are several main reasons for this failure, including formulation issues, the cost of producing molecules in addition to toxicity such as ototoxicity, and cardiotoxicity associated with few compounds. To overcome these adverse effects, Hsp90 N-terminal selective inhibitors (alpha and beta) are now being explored25,26,27. All the aforementioned discussion warrants the HTS screening of compounds, that will suppress Hsp90 activity by binding to its ATP binding cleft.
Assays for the discovery of drugs need to be fast, cost-effective, sensitive, easily reproducible, and utilize less hazardous chemicals and/or conditions. Additionally, it should be conveniently carried out on a lab scale as well as in an HTS setup. A detailed analysis of available Hsp90 ATPase inhibitory assays was undertaken, and malachite green phosphate assay was found to be the most suitable among all in the given institutional setup. The other assay systems were not automation-friendly in an HTS system (pyruvate/lactate dehydrogenase coupled enzyme)28,29; expensive like the fluorescence polarization assay30, scintillation proximity assays, surface plasmon resonance assay, and time-resolved fluorescence resonance energy transfer (TR-FRET)28,29; time-consuming like (TR-FRET)28,29; or possessed radiation hazards like scintillation proximity assays28,29. Therefore, a simple malachite green assay as a lab-scale or HTS for the identification of novel Hsp90 N-terminal ATP binding inhibitors is presented in this protocol.
The major concern in this assay is the contamination of buffers and plant extracts to be evaluated with phosphate ions, which may lead to false results. The non-enzymatic hydrolysis of ATP at the reaction pH may also lead to false results. The aforementioned problems were solved by using phosphate-free buffers (pH 7.4). Additionally, a blank was carried out with buffer + water + DMSO, and thereafter subtracted the absorbance value of the positive control/sample compounds from the blank value. A precaution that needs to be followed in this assay procedure is strict adherence to the addition of reagents, like the addition of ATP simultaneously to each well via multi-channel pipette. This is so that the reaction starts at the same time in every well, the same incubation time, and thereafter at the same measurement of absorbance after the addition of the malachite green reagent. The ATPase reaction is highly time sensitive, and hence a slight change in the reaction initiation and termination time, incubation time, and time for recording absorbance can drastically affect the assay results. Further, the addition of 10 µL of 34% sodium citrate solution (used for quenching the ATPase reaction) affects the linearity of the standard curve, and hence was not included in our assay protocol.
The only limitation of this assay is the evaluation of compounds that demonstrate absorbance higher than shown by free phosphate released from ATP by the yeast Hsp90 at 620 nm. A false positive or false negative may result in the case of such compounds. However, the chances of encountering such compounds is very rare. This can be problematic with the evaluation of plant extracts and/or their fractions in particular, where a mixture of compounds are present.
In conclusion, scale-up from lab-scale to HTS was successfully carried out with yeast Hsp90. The stepwise assay protocol on the lab-scale as well as in HTS will aid scientists worldwide to adopt this procedure for the discovery of novel Hsp90 inhibitors. Additionally, the protocol may be adapted for the discovery of ATPase inhibitors against other ATP dependent proteins like Hsp70, Grp94, Lon protease, Protease Ti, AAA proteases, etc31,32,33.
The authors have nothing to disclose.
This study was supported by the Korea Research Fellowship (KRF) program, postdoctoral fellow of the National Research Foundation of Korea (NRF), funded by the ministry of science & ICT (NRF-2019H1D3A1A01102952). The authors are thankful to KIST intramural grant and Ministry of Oceans and Fisheries grant number 2MRB130 for providing financial assistance for this project.
1M Magnesium chloride solution in water | Sigma-Aldrich | 63069-100ml | |
1M Potassium chloride solution in water | Sigma-Aldrich | 60142-100ml | |
96-well plate | SPL Life Sciences | Not applicable | |
Adenosine 5′-triphosphate disodium salt hydrate | Sigma-Aldrich | A7699-5G | |
Biomek FX laboratory automation workstation | Beckman Coulter | Not applicable | |
Compounds 3-96 | Not applicable | Not applicable | Histidine tagged yeast Hsp90 was obtained from Dr. Chrisostomos Prodromou, School of Life Sciences, University of Sussex, United Kingdom, and protein was expressed in KIST Gangneung Institute of Natural Products. Details cannot be disclosed due to patent infringement issues. |
Dimethyl sulfoxide | Sigma-Aldrich | D8418 | |
Geldanamycin, 99% (HPLC), powder | AK Scientific, Inc. | V2064 | |
Invitroge UltraPure DNase/RNase-Free Distilled Water | ThermoFisher Scientific | 10977015 | |
Malachite Green Phosphate Assay Assay kit | Sigma-Aldrich | MAK307-1KT | |
Multi-Detection Microplate Reader Synergy HT | Biotek Instruments, Inc. | Not applicable | |
Synergy HT multi-plate reader | Biotek Instruments, Inc. | Not applicable | |
Trizma hydrochloride buffer solution, pH7.4 | Sigma-Aldrich | 93313-1L | |
Yeast Hsp90 | Not applicable | Not applicable | School of Life Sciences, University of Sussex, United Kingdom and protein was expressed in KIST Gangneung Institute of Natural Products. Primary Accession number: P02829 |