Bacterial Hsp90 ATPase Assays

Abstract

Bacterial Hsp90 is an ATP-dependent molecular chaperone involved in protein remodeling and activation. The E. coli Hsp90, Hsp90_Ec, collaborates in protein remodeling with another ATP-dependent chaperone, DnaK, the E. coli Hsp70. Both Hsp90_Ec and DnaK hydrolyze ATP and client (substrate) proteins stimulate the hydrolysis. Additionally, ATP hydrolysis by the combination of Hsp90_Ec and DnaK is synergistically stimulated in the presence of client (substrate). Here, we describe two steady-state ATPase assays used to monitor ATP hydrolysis by Hsp90_Ec and DnaK as well as the synergistic stimulation of ATP hydrolysis by the combination of Hsp90_Ec and DnaK in the presence of a client (substrate). The first assay is a spectrophotometric assay based on enzyme-coupled reactions that utilize the ADP formed during ATP hydrolysis to oxidize NADH. The second assay is a more sensitive method that directly quantifies the radioactive inorganic phosphate released following the hydrolysis of [γ−³³P] ATP or [γ−³²P] ATP.

1. Introduction

Hsp90 is a widely conserved and ubiquitous molecular chaperone that participates in ATP-dependent protein remodeling and activation in both eukaryotes and prokaryotes, an activity that requires the Hsp70 chaperone system [1–3]. Additionally, in eukaryotes the ATPase activity and the chaperone cycle of Hsp90 are regulated by a myriad of co-chaperones [1, 2]. In contrast to eukaryotic Hsp90, bacterial Hsp90 functions independently of Hsp90 cochaperones [4, 5].

Hsp90 is a functional homodimer, and each monomer is comprised of three domains: an N-terminal ATPase domain [6, 7], a middle domain that is involved in client binding [6, 8–10], and a C-terminal dimerization domain that also contains residues that interact with client [2, 8]. Hsp90 undergoes large-scale conformational changes upon ATP binding and hydrolysis, and these motions are coupled with the processes of client binding, remodeling, and release [6, 7, 10–12]. Apo Hsp90 in the solution exists as “V-shaped” dimers linked through the C-terminal domain [13]. However, ATP binding leads to closing of the ATP lid on the ATP-binding site, dimerization of the N-terminal domains, and compaction of the molecule [11–14]. ATP hydrolysis and ADP release trigger the dissociation of the N-domains and the return of apo Hsp90 to the “V-shaped” dimers [11–14]. Eukaryotic Hsp90 cochaperones including Aha1/Hch1, Hop/Sti1, Cdc37, and p23/Sba1 regulate the cycle of ATP binding and hydrolysis as well as imparting client protein specificity [1, 2, 12].

The Hsp70 chaperone system collaborates with Hsp90 in protein remodeling and activation [1, 3, 4]. Eukaryotic Hsp70 and its bacterial homolog, DnaK, are highly conserved, ATP-dependent molecular chaperones [15–17]. Hsp70 has both an N-terminal ATP-binding domain and a C-terminal substrate-binding domain, which are connected via a flexible linker [17]. Like Hsp90, Hsp70 undergoes large conformational changes upon ATP-binding, hydrolysis, and release [17]. Moreover, Hsp70 acts in conjunction with two cochaperones, Hsp40 (J-domain protein) and a nucleotide exchange factor (NEF). The Hsp40 cochaperone targets clients for recognition by Hsp70 and stimulates ATP hydrolysis by Hsp70, while the NEF triggers nucleotide exchange by Hsp70 [15, 18, 19].

In E. coli, Hsp90_Ec acts in conjunction with DnaK and an E. coli J-domain protein, DnaJ or CbpA. GrpE, the E. coli NEF, stimulates protein reactivation, but is not essential [4, 5]. Homologs of eukaryotic Hsp90 cochaperones have not been identified in bacteria [1]. Instead, Hsp90_Ec and DnaK directly interact, both in vivo and in vitro [4, 5], and this interaction is essential for collaboration in client remodeling and a synergistic stimulation of ATP hydrolysis [20]. Additionally, client binding and ATP hydrolysis by both chaperones is required for client remodeling and synergy in ATPase activity [20].

Two alternative methods for monitoring the ATPase activity of Hsp90_Ec, DnaK and the combination of the two chaperones are described here. One is an enzyme-coupled spectrophotometric assay that is performed in a microtiter plate. For each mole of ATP hydrolyzed, 1 mol of ADP is produced. The ADP produced is converted to ATP by pyruvate kinase (PK) in an enzymatic reaction requiring phosphoenolpyruvate (PEP) and producing pyruvate. Pyruvate is then reduced to lactate by L-lactate dehydrogenase (LDH) in an enzymatic reaction that is coupled to the oxidation of NADH to NAD+. Thus for each mole of ATP hydrolyzed, 1 mol of NADH is oxidized. The assay monitors the decrease in optical density at the NADH absorbance maxima, 340 nm, as a function of time. This decrease in NADH is directly proportional to the rate of ATP hydrolysis. The constant regeneration of ATP allows monitoring the ATP hydrolysis rate as long as PEP and NADH are present.

A second more sensitive method to measure ATP hydrolysis by Hsp90_Ec is also described. In this method, radioactive inorganic phosphate, either ³³P_i or ³²P_i, produced by hydrolysis of [γ-³³P] ATP or [γ-³²P] ATP is monitored following the formation of a phosphomolybdate complex and organic extraction.

2. Materials

All the solutions are prepared using ultrapure H₂O and analytical grade reagents from common sources (unless specifically indicated).

2.1. Enzyme-Coupled Spectrophotometric ATPase Assay

1. HKE buffer: 25 mM Hepes, pH 7.5, 50 mM KCl, and 0.1 mM EDTA (final concentrations in 100 μL reaction mixtures). A concentrated buffer (5× or 10×) can be prepared and used for accuracy and simplification. Store the concentrated buffer at room temperature. ATPase reactions are supplemented with DTT (2 mM final) during reaction assembly.

2. NADH (nicotinamide adenine dinucleotide), 25 mM stock solution in HKE buffer. NADH concentration can be determined by measuring A₃₄₀ in a spectrophotometer. The extinction coefficient of NADH is 6.22 × 10³/M/cm. Store frozen (−20 °C) in ~100 μL aliquots.

3. ATP, 100 mM stock solution. To 100 mg of ATP add 1 mL ultrapure H₂O. Adjust the pH to 7.5 using 1 M Tris base. Make a 1:1000 dilution (1 μL ATP + 999 μL ultrapure H₂O) and read the absorbance at 260 nm. The extinction coefficient of ATP is 15.4 × 10³/M/cm. A 1:1000 dilution of 100 mM ATP has an absorbance of 1.54. Adjust the volume with ultrapure H₂O as needed to obtain a concentration of 100 mM.

4. Phosphoenolpyruvate (PEP), 100 mM stock solution in HKE buffer. Store frozen (−20 °C) in ~100 μL aliquots.

5. Pyruvate kinase/lactate dehydrogenase (PK/LDH) mix, commercially available (Sigma: P0294).

6. MgCl₂, 1 M stock solution in ultrapure H₂O. Store at room temperature.

7. DTT, 1 M stock in ultrapure H₂O. Store frozen (−20 °C).

8. Proteins: DnaK, Hsp90_Ec, ribosomal protein L2 (client protein).

9. Geldanamycin (Sigma), 30 mM stock solution in DMSO.Hsp90_Ec is inhibited by 10–30 μM geldanamycin (see Note 1).

2.2. ATPase Assay Using Radioactive ATP

1. HKE buffer: 25 mM Hepes, pH 7.5, 50 mM KCl, and 0.1 mM EDTA (final concentrations in 50 μL reaction mixtures). A concentrated buffer (5× or 10×) can be prepared and used for accuracy and simplification. Store the concentrated buffer at room temperature. ATPase reactions are supplemented with DTT (2 mM final), Triton X-100 (0.005% final), and MgCl₂ (10 mM final) during reaction assembly.

2. ATP, 100 mM stock solution (see Subheading 2.1, item 3).

3. [γ-³³P] ATP (Perkin Elmer: NEG302H) or [γ-³²P] ATP (Perkin Elmer: NEG002A). Make an ~0.1 μCi/μL stock solution (typically a 1:100 dilution into ultrapure H₂O). Freeze in aliquots at −20 °C.

4. MgCl₂, 1 M stock solution in ultrapure H₂O. Store at room temperature.

5. DTT, 1 M stock in ultrapure H₂O. Store frozen (−20 °C).

6. Triton X-100, 10% solution (Thermo Scientific: 28314).

7. Tungstosilicic (silicotungstic) acid solution, 20 mM stock solution in 10 mM MgSO₄. For 50 mL, solubilize 0.12 g MgSO₄ in ultrapure H₂O. Use 2.88 g silicotungstic acid and solubilize in the prepared 50 mL of 10 mM MgSO₄. Store at 4 °C.

8. KH₂PO₄, 2 mM stock solution. For 50 mL, solubilize 13.6 mg KH₂PO₄ in ultrapure H₂O. Store at 4 °C.

9. Ammonium molybdate solution, 5% stock solution (w/v) in2 M H₂SO₄. For 50 mL, dilute stock H₂SO₄ to 2 M and then solubilize 2.5 g ammonium molybdate. Store at 4 °C.

10. Isobutanol:toluene solution, 1:1 (v/v). Store at room temperature in a sealed glass container.

11. Proteins: DnaK, Hsp90_Ec, L2.

12. Scintillation fluid for organic solvents (e.g., Eccoscint A).

13. Geldanamycin (Sigma), 30 mM stock solution in DMSO.Hsp90_Ec is inhibited by 10–30 μM geldanamycin (see Note 1).

3. Methods

3.1. Enzyme-Coupled Spectrophotometric ATPase Assay

With the assay conditions described here, Hsp90_Ec hydrolyzes ATP at a rate of ~0.1 nmol/min and the rate is stimulated ~fourfold in the presence of L2 (client protein) [8, 20]. DnaK hydrolyzes ATP at a rate of ~0.1 nmol/min and is only slightly stimulated by L2 [20]. When the three proteins are combined, an ~twofold stimulation above the additive values is observed [20]. The ATPase activity can be measured readily using a steady-state ATPase assay in a 96-well plate. In an enzyme-coupled reaction, ADP produced by the chaperones is regenerated to ATP by a reaction that is coupled with the conversion of NADH to NAD⁺ [21, 22] (Fig. 1). The absorbance of NADH at 340 nm is monitored to measure the decrease in NADH concentration, which is quantitatively proportional to the rate of ATP hydrolysis.

1. Equilibrate a black, clear-bottom 96-well plate at 37 °C in a plate reader capable of monitoring A₃₄₀ over time at 37 °C (see Note 2).

2. Prepare a 10× mixture of PEP, PK/LDH, NADH, MgCl₂, and ATP. The final concentrations of the components in 100 μL reactions are 1 mM PEP, 1:100 dilution of the PK/LDH mix, 0.5 mM NADH, 10 mM MgCl₂, and 2 mM ATP (see Note 3). Aliquot 12 μL of the 10× mix into individual PCR tubes. These mixtures will be used in step 4 (see Note 4).

3. In individual 1.5 mL microcentrifuge tubes, assemble the chaperone mixtures (90 μL) containing HKE buffer, DTT, DnaK or Hsp90_Ec or both together in the presence or absence of L2 (see Note 5). The final concentrations in 100 μL mixtures are 25 mM Hepes, pH 7.5, 50 mM KCl, 0.1 mM EDTA, and 2 mM DTT. The proteins are typically used at equilmolar ratios of DnaK monomer:Hsp90_Ec dimer:L2 monomer at a concentration of 1 μM (see Note 6). Transfer the chaperone mixtures into wells of the 96-well plate.

4. Using a multichannel pipette, start the reactions by transferring 10 μL of the 10 mix of PEP, PK/LDH, NADH, MgCl₂, and ATP into the wells of the 96-well plate that contain the chaperone mixture. A multichannel pipette is used for this step to ensure that all the reactions have the same start time (see Note 4). Mix the reactions in the plate using a plate mixer or use the mixing function of the plate reader.

5. Measure the decrease in NADH absorbance at 340 nm as a function of time in a plate reader at 37 °C.

6. Fit the data to a linear equation using appropriate software (see Note 7). Calculate the amount of NADH (nmol) converted to NAD+ (nmol) in each reaction by comparing the slope to that of a previously determined NADH standard (see Note 8).

Fig. 1.

Enzyme-coupled spectrophotometric ATPase assay. ADP produced by Hsp90_Ec, DnaK, or the combination of chaperones with or without L2 is regenerated to ATP by a reaction that is coupled with the conversion of NADH to NAD⁺ [21, 22]. The decrease in NADH concentration is monitored

The amount of NADH (nmol) converted to NAD+ (nmol) is quantitatively proportional to the amount of ADP generated and can be used to determine the rate of ATP hydrolysis.

3.2. ATPase Assay Using Radioactive ATP

Alternatively, Hsp90_Ec ATPase can be monitored by measuring the production of inorganic phosphate, ³³P_i or ³²P_i, from the hydrolysis of [γ-³³P] ATP or [γ-³²P] ATP [23]. Free P_i, in the form of a phosphomolybdate complex, is quantitatively recovered by organic extraction.

1. Prepare a 10× mixture of 40 mM ATP and [γ-³³P] ATP (or [γ-³²P] ATP), such that after dilution each 50 μL reaction will contain 4 mM ATP and 0.1 μCi [γ-³³P] ATP (1 μL of 0.1 μCi/μL stock solution). The reactions will be started by the addition of 5 μL of this mixture in step 3 below.

2. In individual 1.5 mL microcentrifuge tubes, assemble the chaperone mixtures (45 μL) containing HKE buffer, DTT, Triton X-100, MgCl₂, DnaK or Hsp90_Ec or both together in the presence or absence of L2 (see Note 5). The final concentrations in 50 μL mixtures are 25 mM Hepes, pH 7.5, 50 mM KCl, 0.1 mM EDTA, 2 mM DTT, 0.005% Triton X-100, and 10 mM MgCl₂. The proteins are typically used at equilmolar ratios of DnaK monomer:Hsp90_Ec dimer:L2 monomer at a concentration of 1 μM (see Note 6).

3. Start the reactions by adding 5 μL of the ATP mixture to each reaction mixture. Incubate mixtures at 25 °C for various times (see Note 9). To determine the total input radioactivity, add 1 μL of the 0.1 μCi [γ-³³P] ATP solution to 49 μL HKE buffer and transfer 50 μL into 10 mL scintillation fluid in a scintillation vial and determine radioactivity with the experimental samples.

4. Following incubation, add 100 μL 20 mM silicotungstic acid solution to each reaction and vortex gently. Next add 200 μL 2 mM KH₂PO₄ to each reaction. Then add 100 μL 5% ammonium molybdate solution to each reaction and vortex gently. Incubate the reactions for 1 min at 37 °C.

5. Add 450 μL of the isobutanol:toluene mixture to each reaction. Vortex each tube vigorously for 15 s (see Note 10). Let the tubes sit for 1 min at room temperature to allow phase separation.

6. Remove 100 μL of the top organic layer and add to 10 mL scintillation fluid in scintillation vials. Measure radioactivity in a liquid scintillation counter (see Note 11). Use the input radioactivity control to determine the fraction of ³³P_i or ³²P_i in each sample and calculate the amount of ATP hydrolyzed (see Note 12).

4. Notes

1. To determine if Hsp90_Ec is contaminated with nonspecific ATPase activity, geldanamycin can be used. It specifically inhibits ATP hydrolysis by Hsp90_Ec [4, 24].

2. The preincubation step allows the reactions to reach 37 °C faster and reduces the lag in hydrolysis that can make monitoring the linear phase of the reaction more difficult.

3. The NADH concentration indicated, 0.5 mM, is a starting suggestion and can be adjusted to accommodate various ATP hydrolysis rates.

4. The 10× mixture (12 μL) is added into PCR tubes (preferably strips) prior to the addition to the 96-well plate. In order for all the reaction mixtures to have the same starting time point, a multichannel pipette is used to remove 10 μL of the 10× mix from the PCR tubes and make the addition into the 96-well plate.

5. For background subtraction, carry out a reaction with buffer alone and buffer with L2 only. See Note 1 for information about controls for ATPase contamination in Hsp90_Ec samples.

6. These conditions have been optimized to monitor the synergy between Hsp90_Ec and DnaK. Different ratios and concentrations of Hsp90_Ec, DnaK, and L2 may be used.

7. When fitting the data, usually only a portion of the data can be used for the linear fit. There is often a short lag at the start of the reaction, followed by a linear portion and then a plateau when NADH is exhausted. For the fit, only use data in the linear portion of the curve.

8. The NADH standard curve is determined by preparing NADH standards in duplicate as follows: (1) Transfer 3 μL of a 25 mM NADH solution into 297 μL HKE buffer to obtain 25 nmol NADH/100 μL. (2) Using150 μL of this dilution, make five twofold serial dilutions into 150 μL HKE buffer to make solutions of 12.5, 6.25, 3.13, 1.56, and 0.78 nmol NADH/100 μL. Transfer 100 μL of each concentration of NADH and HKE buffer alone into separate wells of a black, clear bottom 96-well plate. Measure absorbance at 340 nm using a plate reader. Plot A₃₄₀ as a function of NADH (nmol) in Prism or another data analysis program. Fit the data using linear regression analysis and use the resulting linear equation to calculate the rate of NADH (nmol) converted to NAD⁺ (nmol) from the experimental data.

9. This assay can be used as an end-point reaction from which the rate of ATP hydrolysis per minute can be calculated. However, it is important to determine that hydrolysis for each protein is in a linear range by measuring the amount of ATP hydrolyzed after various times of incubation and adjusting the end-point incubation time as necessary.

10. Caution should be used when opening and closing microfuge tubes during this assay. Any defect in the lid can cause leaking upon vortexing the organic solution.

11. To accurately calculate the amount of radioactive P_i in the organic layer, determine the final volume of the organic layer. Perform a mock reaction using 50 μL of HKE buffer in place of a reaction mixture (step 1, Subheading 3.2) and follow steps 2–5 as described (Subheading 3.2). Remove and determine the volume of the organic layer. This volume is necessary to calculate the amount of total P_i found in the organic layer as only 100 μL (~25%) is monitored by scintillation counting.

12. Convert cpm in each reaction mixture to nmol ATP hydrolyzed min⁻¹. Make sure to take into account the fraction of the organic layer counted, the reaction time in minutes, and the specific activity (cpm/nm) of ATP in the reaction, accounting for both the labeled and unlabeled ATP.