An In Vitro Assay for LpxC Degradation by the Membrane-Bound Protease FtsH and Adaptor LapB from Escherichia coli

Abstract

This protocol presents a sensitive, quantitative in vitro degradation assay of LpxC using reconstituted AAA+ (ATPase associated with diverse cellular activities) protease FtsH and adaptor LapB in proteoliposomes. AAA+ proteases are ATP-dependent molecular machines that maintain protein homeostasis and regulate diverse cellular processes in the cytoplasm or at cellular membranes. Compared with their cytoplasmic counterparts, membrane-bound AAA+ proteases remain poorly characterized due to technical challenges in reconstituting and measuring their activity in vitro. Here, we use the FtsH–LapB–LpxC system as a model to develop a fluorescence-based degradation assay in proteoliposomes. FtsH and LapB are reconstituted into proteoliposomes to mimic the native membrane environment, and LpxC is covalently labeled with the fluorescent dye Atto488. Degradation is initiated by combining proteoliposomes with labeled LpxC and ATP and terminated by trichloroacetic acid (TCA). A successful assay is defined by a time-dependent increase in fluorescence in the soluble supernatant fraction after TCA precipitation, corresponding to the release of fluorescent peptide fragments generated by LpxC proteolysis. This signal enables direct quantification of initial degradation rates under defined conditions. Compared with traditional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)-based assays, this method provides substantially improved sensitivity and enables quantitative kinetic analysis. Moreover, the proteoliposome system allows systematic investigation of how protein components, such as adaptors and anti-adaptors, as well as lipid composition, influence LpxC degradation. Although developed for the FtsH–LapB–LpxC system, this assay is readily adaptable to other membrane-bound AAA+ proteases and substrates, providing a general platform for studying membrane-associated proteolysis in vitro.

SUMMARY:

Escherichia coli regulates lipopolysaccharide (LPS) biosynthesis primarily by controlling the degradation of LpxC, the deacetylase that catalyzes the first committed step of LPS synthesis. The goal of this protocol is to establish a sensitive, quantitative in vitro degradation assay for LpxC, using reconstituted AAA+ protease FtsH and adaptor LapB in proteoliposomes.

INTRODUCTION:

AAA+ proteases are conserved, ATP-dependent molecular machines that play central roles in protein quality control and regulatory proteolysis^1,2. These machines contain a hexameric ring of AAA+ ATPases domain, which hydrolyze ATP, recognize, unfold, and translocate the substrates to the compartmental protease domain for degradation¹. While cytoplasmic AAA+ proteases such as ClpXP³ and Lon⁴ have been extensively characterized using well-established in vitro assays, membrane-bound AAA+ proteases remain comparatively poorly understood. This gap largely reflects the technical challenges associated with reconstituting membrane proteins into defined lipid environments and quantitatively measuring their proteolytic activity in vitro. The overall goal of this method is to establish sensitive, quantitative assays for investigating the proteolytic activity of membrane-bound proteases.

FtsH is the only essential AAA+ protease in Escherichia coli and serves as a prototypical membrane-bound AAA+ protease^5–7. Anchored to the inner membrane by N-terminal transmembrane helices, FtsH degrades a diverse set of cytoplasmic and membrane-associated substrates, thereby regulating membrane protein quality control, stress responses, and lipid homeostasis^8–11. One of its best-characterized substrates is LpxC, a soluble deacetylase that catalyzes the first committed step in lipopolysaccharide (LPS) biosynthesis^12,13. Efficient degradation of LpxC by FtsH requires the membrane protein LapB, which functions as an adaptor to specifically recognize LpxC and deliver it to the protease^14–18.

Despite its physiological importance, quantitative analysis of FtsH-mediated proteolysis in vitro has been limited, particularly for cytoplasmic substrates such as LpxC. Traditional in vitro degradation assays use FtsH purified in detergent micelles^13,19, which do not provide the membrane environment for the membrane-anchored protease. In addition, the degradation assays often rely on SDS–PAGE–based detection of substrate disappearance¹³, which can be labor-intensive, low-throughput, and insufficiently sensitive for kinetic analyses. More recently, FtsH reconstituted into bicelles has been used to study degradation of membrane substrates, as pioneered by Heedeok Hong’s group²⁰; however, a comparable system for analyzing cytoplasmic substrates remains lacking. Here, we adapt Heedeok Hong’s bicelle reconstitution to proteoliposome reconstitution to mimic the membrane environment. Besides that, the water-soluble substrate LpxC is fluorescence-labeled, which enables sensitive and quantitative measurement of the kinetics of FtsH/LapB-dependent degradation of LpxC in vitro.

This method should be used when researchers need to quantitatively measure the FtsH-mediated proteolysis of water-soluble substrates. It provides several advantages over existing approaches. First, fluorescence-based detection of proteolytic products, rather than substrate disappearance, substantially improves sensitivity and dynamic range, facilitating quantitative comparisons of proteolytic activity. Second, the use of proteoliposomes preserves the membrane context required for FtsH function while allowing precise control over protein and lipid composition¹⁸. Finally, although developed using the FtsH–LapB–LpxC system, this protocol is readily adaptable to other membrane-bound AAA+ proteases and their cytoplasmic substrates that require membrane-associated degradation machinery.

PROTOCOL:

1. Labeling LpxC with Atto488

1.1. LpxC preparation

1.1.1. Express and purify LpxC protein as reported previously¹⁸. Exchange the buffer of purified LpxC from Tris to 50 mM phosphate buffer, pH 7.6 ( 6 mM NaH₂PO₄, 44 mM Na₂HPO₄, 150 mM NaCl) by gel filtration using a high-resolution gel filtration column (e.g., Superdex 200 Increase 10/300 GL column).

1.1.2. Concentrate LpxC using an ultrafiltration column to a final concentration of at least 5 mg/mL at 4 °C.

1.2. Labeling LpxC with Atto488

1.2.1. Dilute 0.5 mg of LpxC in 50 mM phosphate buffer, pH 7.6, to a final volume of 225 μL. Add 25 μL of 1 M sodium bicarbonate to adjust the pH to approximately 8.3.

1.2.2. Dissolve 0.24 mg of Atto488 reactive dye in 20 μL of 0.1 M sodium bicarbonate buffer to prepare a stock solution (0.012 mg/μL). Prepare a working solution by diluting 2 μL of the stock solution with 18 μL of 0.1 M sodium bicarbonate buffer, yielding a final concentration of 0.0012 mg/μL.

1.2.3. Immediately add 4 μL of the Atto488 working solution to the protein solution (corresponding to a molar ratio of LpxC: Atto488 of 3:1). Wrap the tube with aluminum foil and incubate at 4 °C for 2 h with gentle agitation (18 rpm).

NOTE: Both the Atto488 stock and working solutions can be flash-frozen in liquid nitrogen and stored at −80 °C.

1.3. Separation of LpxC–Atto488 from free dye

1.3.1. Equilibrate a PD-10 desalting column with 10 mL of ultrapure water, followed by 15 mL of phosphate buffer, pH 7.6.

1.3.2. Apply the labeling reaction mixture to the column and allow it to fully enter the resin.

1.3.3. Elute with 5 mL of 50 mM phosphate buffer, pH 7.6. Two distinct bands corresponding to LpxC–Atto488 (first band) and free Atto488 dye (second band) should become visible. Collect fractions from the first band in ~0.5 mL aliquots.

1.3.4. Analyze eluted fractions by SDS-PAGE. Visualize fluorescence under UV illumination before Coomassie Brilliant Blue (CBB) staining.

1.3.5. Pool fractions containing LpxC–Atto488 and concentrate the labeled protein to approximately 2 mg/mL.

1.3.6. Estimate labeling efficiency by measuring the protein concentration and Atto488 concentration using a microvolume spectrophotometer.

NOTE: A ratio of Atto488 concentration (in pmol/μL) to LpxC concentration (in mg/mL) in the range of 2–5 indicates an appropriate labeling level. This corresponds to a molar ratio of Atto488 to LpxC ranging from 1:15 to 1:6 after labeling.

1.3.7. Aliquot LpxC–Atto488 into 1.5 mL microcentrifuge tubes, flash-freeze in liquid nitrogen, and store at −80 °C. At this point, the protocol can be paused for up to 3 months.

CAUTION: Liquid nitrogen is extremely cold; handle it with proper protective equipment (cryo-gloves, eye protection).

2. Reconstitution of FtsH/LapB into proteoliposomes

2.1. Liposomes preparation

2.1.1 Transfer 100 mg of 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) in chloroform into a glass test tube. Evaporate the solvent under a gentle stream of N₂, followed by overnight drying under vacuum.

NOTE: POPC has a phase transition temperature of ~−2 °C and is widely used in membrane biophysics, liposome formulations, and cell model systems.

CAUTION: Chloroform can harm the eyes, skin, liver, kidneys, and nervous system. Proper protective equipment (gloves, eye protection, mask) is required during handling. Handle it in a hood using a glass syringe.

2.1.2. Add 1,315 μL of buffer A (20 mM Tris-HCl, pH 7.5, 100 mM NaCl) to the dried POPC film. Resuspend the lipids by alternating 1 min of water-bath sonication with 1 min of vortexing at maximum speed for a total of ~15 min, or until fully dispersed. (final lipid concentration: 100 mM).

NOTE: To improve lipid resuspension, adjust the immersion depth of the glass test tube in the water-bath sonicator to maximize agitation. Resuspended lipids can be stored at −20 °C. At this point, the protocol can be paused for up to 1 month.

2.1.3. Aliquot the liposome suspension into 200 μL portions and store at −20 °C.

2.2. Liposomes extrusion

2.2.1. Thaw a 200 μL liposome aliquot in a water bath at room temperature and dilute with 800 μL of buffer A to a final volume of 1 mL.

2.2.2. Subject the liposomes to three freeze-thaw cycles by alternating between liquid nitrogen and a water bath at room temperature.

2.2.3. Rinse syringes, two filter supports, and a 200 nm polycarbonate membrane with buffer A. Assemble the mini-extruder and extrude the liposomes 20 times.

NOTE: The liposome suspension should appear transparent after extrusion.

2.2.4. Divide the 1 mL extruded liposome suspension into three tubes (330 μL per tube) and dilute each with 600 μL of buffer A.

2.3. Triton X-100 swelling titration

NOTE: Triton X-100 is a low Critical micelle concentration (CMC) detergent, and it has proven to be most successful in many cases²¹.

CAUTION: Triton X-100 is a hazardous chemical, primarily causing severe eye damage, skin irritation, and harmful effects if swallowed. Proper protective equipment (gloves, eye protection) is required during handling.

2.3.1. Prepare 10% (v/v) Triton X-100 by mixing 100 μL Triton X-100 with 900 μL ultrapure water, and incubating it in a shaker to fully dissolve Triton X-100.

2.3.2. Measure the absorbance of liposomes at 540 nm (A₅₄₀) using a spectrophotometer.

2.3.3. Add 5 μL aliquots of 10% Triton X-100 sequentially, gently mixing after each addition, and measure A₅₄₀. Continue adding Triton X-100 until A₅₄₀ reaches a maximum, then add Triton X-100 an additional 4–5 times until the absorbance decreases to approximately 60% of the peak value.

2.3.4. Based on the titration curve, swell liposomes by adding 35 μL of 10% Triton X-100 and incubate at room temperature for 1 h with gentle agitation.

2.4. Reconstitution of FtsH/LapB into liposomes

2.4.1. Express and purify FtsH and LapB proteins as described previously¹⁸.

2.4.2. Wash ~2 g of a nonpolar polystyrene adsorbent (e.g., Bio-Beads SM-2) once with 10 volumes of methanol, once with 10 volumes of ethanol, and ten times with 10 volumes of ultrapure water. Store the washed adsorbent in buffer A at 4 °C.

2.4.3. For each tube of Triton X-100–swollen liposomes, add 3 μL of 500 mM TCEP, 22.5 μg of FtsH, and 22.5 μg of LapB. Adjust the total volume to 1 mL with buffer A and incubate at RT for 1 h with gentle agitation.

NOTE: Protein-to-lipid ratios of 1:20 to 1:200 (wt/wt) were routinely used during proteoliposomes reconstitution²¹. Here, a ratio of 1:100 is used.

2.4.4. Remove detergent by sequential incubation with nonpolar polystyrene adsorbent: 40 mg at RT for 0.5 h, followed by 40 mg at RT for 1 h, then 40 mg at 4 °C overnight, and finally 40 mg at 4 °C for 1 h.

2.4.5. Briefly centrifuge to pellet the nonpolar polystyrene adsorbent at 4 °C (400 × g for 3 s). Transfer the supernatant to ultracentrifuge tubes. Wash the nonpolar polystyrene adsorbent with 1 mL of buffer A and combine the wash with the supernatant. Add additional buffer A to a final volume of ~6 mL.

NOTE: For safety, ultracentrifuge tubes must contain at least 6 mL of liquid.

2.4.6. Balance the tubes and ultracentrifuge at 250,000 × g at 4 °C for 45 min in a fixed-angle rotor (e.g., MLA-80) to pellet proteoliposomes.

2.4.7.Discard the supernatant and resuspend the proteoliposome pellet in 70 μL of buffer A supplemented with 1 mM tris (2-carboxyethyl)phosphine (TCEP).

NOTE: The final proteoliposome volume is approximately 100 μL.

2.4.8. Flash-freeze the proteoliposomes in liquid nitrogen and store at −80 °C.

NOTE: At this point, the protocol can be paused for up to 1 month.

2.5. SDS-PAGE analysis of proteoliposomes

2.5.1. Mix proteoliposomes thoroughly and combine 5 μL of the sample with 5 μL of 2× SDS loading buffer (0.1 M Tris-HCl, pH 6.8, 5% SDS, 20% glycerol, 0.01% bromophenol blue, 200 mM DTT).

2.5.2. Load proteoliposome samples alongside 525 ng of FtsH (corresponding to 0.06 μM FtsH hexamer in 20 μL) on a 4–20% SDS-PAGE gel.

2.5.3. Stain the gel with CBB.

2.5.4. 2.Scan the gel and quantify the band intensity of FtsH for each sample. Calculate the volume of proteoliposome suspension equivalent to 525 ng of FtsH.

3. In vitro degradation of LpxC–Atto488

3.1. Perform in vitro degradation assays in a 20 μL reaction containing FtsH/LapB proteoliposomes (corresponding to 0.06 μM FtsH) and 0.5, 1, 2, 2.5, 5, 10, 20, 40 μM LpxC-Atto488 in degradation buffer (50 mM Tris-acetate, pH 8.0, 10 mM Mg-acetate, 5 mM ATP, 25 μM Zn-acetate, 80 mM NaCl, 1.4 mM β-mercaptoethanol).

3.2. For the measurement of each concentration of substrate, prepare a reaction of 70 μL.

3.2.1. Take a 20 μL fraction and quench it by immediately mixing it with 48 μL of 5% trichloroacetic acid (TCA) in 1.5 mL microcentrifuge tubes (0 min). Take the second and the third fractions and quench them at 10 and 20 min, respectively.

3.2.2. In parallel, prepare corresponding control reactions lacking proteoliposomes for background subtraction.

NOTE: Three replicate measurements are recommended in order to obtain reliable results. Minimize light exposure when handling LpxC–Atto488.

CAUTION: TCA is a highly corrosive chemical that causes severe skin burns, permanent eye damage, and intense respiratory irritation upon inhalation. Handle it with proper protective equipment (gloves, eye protection, mask).

3.3. Incubate samples at 37 °C for 30 min to precipitate proteins, then centrifuge at 18,800 × g at room temperature for 20 min.

3.4. Neutralize samples by transferring 50 μL of supernatant into a tube containing 60 μL of 500 mM Tris-HCl (pH 8.8). Mix well and transfer 100 μL to a black 96-well plate. Measure fluorescence using a plate reader with excitation at 498 nm and emission at 520 nm.

REPRESENTATIVE RESULTS:

To improve the sensitivity of the in vitro degradation assay, the substrate LpxC was labeled with the fluorescent dye Atto488. The succinimidyl ester group of Atto488 reacts efficiently with primary amino groups of LpxC under mildly alkaline conditions. Because Tris buffer contains primary amines that interfere with this reaction, LpxC purified in Tris buffer was exchanged into phosphate buffer by gel filtration before labeling. For the labeling reaction, a molar ratio of LpxC to Atto488 of 3:1 was used to achieve moderate labeling and avoid over-labeling (see Discussion). After removal of free dye, the LpxC–Atto488 fractions were analyzed by SDS–PAGE, and fluorescence was visualized under UV light (Figure 1A). The same gel was subsequently stained with CBB to confirm the presence of LpxC protein (Figure 1B). The coincidence of the fluorescent and protein bands indicates that free dye was effectively removed from the labeled LpxC preparation (Figure 1).

Figure 1: Atto488 labeling of LpxC.

(A) In-gel fluorescence of LpxC–Atto488 fractions eluted from a PD-10 desalting column. (B) Coomassie Brilliant Blue (CBB) staining of the same gel shown in panel A.

To enable efficient reconstitution of FtsH and LapB into liposomes, preformed liposomes were partially destabilized with Triton X-100. As the concentration of Triton X-100 increased, A₅₄₀ initially increased and reached a maximum, corresponding to detergent saturation of the liposomes. Further addition of Triton X-100 resulted in a decrease in A₅₄₀, reflecting liposome solubilization (Figure 2). Based on empirical optimization, efficient membrane protein reconstitution was achieved when the A₅₄₀ value decreased to approximately 60% of the peak value.

Figure 2: Swelling titration of liposomes induced by Triton X-100.

The absorbance at 540 nm (A₅₄₀) of the liposomes was plotted as a function of Triton X-100 volume.

To assess reconstitution efficiency and quantify FtsH incorporation into proteoliposomes, proteoliposome samples, together with 525 ng of purified FtsH as a reference, were analyzed by SDS–PAGE and visualized by CBB staining. Because equal mass amounts of FtsH and LapB were used during reconstitution, comparable band intensities of FtsH and LapB indicate similar reconstitution efficiencies for both proteins (Figure 3). A darkly stained region at the bottom of the gel likely corresponds to POPC lipids.

Figure 3: SDS–PAGE analysis of FtsH and FtsH/LapB reconstituted into proteoliposomes (FtsH/LapB-PL).

Purified FtsH (525 ng) and FtsH/LapB-PL (5 μL) were mixed with SDS loading buffer, separated on a 4–20% SDS–PAGE gel, and stained with Coomassie Brilliant Blue (CBB). This figure was assembled from two sections cropped from the same gel.

Compared with monitoring substrate disappearance, measuring the formation of degradation products provides a more sensitive and accurate method to quantify proteolytic rates. In this protocol, the proteolytic products of LpxC-Atto488, which are short peptides of less than 5 kDa²², were quantified for product formation. Following TCA precipitation, degradation products of LpxC remain in the supernatant. The fluorescence intensity of the supernatant was measured using a plate reader and serves as a readout for the in vitro proteolytic activity of the FtsH/LapB complex toward LpxC. The representative measurement results of three replicates are shown in Supplementary Table 1. The fluorescence increase observed during the first 10 min is comparable to that in the subsequent 10-minute interval, indicating this protocol works well, and the initial velocity measurements are reliable (Supplementary Table 1). After subtracting the background, the net increased fluorescence at 20 min of different LpxC-Atto488 concentrations of three replicates, as well as the standard deviation (STDEV), is shown in Table 1. These data indicate that the proteolytic activity of LpxC by FtsH/LapB increases with increasing LpxC concentration. The data can be further plotted to generate a Michaelis-Menten curve¹⁸.

Table 1:

Net fluorescence increase at 20 min for different concentrations of LpxC–Atto488.

LpxC-Atto488(μM)	repeat 1	repeat 2	repeat 3
0.5	162	154	162
1	241	238	294
2.5	448	467	511
5	836	811	723
10	848	850	931
20	1137	1039	1109
40	1234	1088	1244

The results represent three independent measurements, with standard deviations (STDEV) indicated. The data show that the proteolytic activity of LpxC by FtsH/LapB increases with increasing LpxC concentration.

Together, this protocol establishes a workflow for quantitatively measuring the degradation of LpxC by the membrane-bound AAA+ protease FtsH and its adaptor LapB. The procedure begins with labeling LpxC with Atto488, followed by reconstitution of FtsH/LapB into proteoliposomes and subsequent measurement of degradation activity. Using this system, we show that the proteolytic activity of the FtsH/LapB complex toward LpxC increases with substrate concentration, and the resulting data are suitable for Michaelis-Menten kinetic analysis. Beyond the FtsH-LapB-LpxC system, this protocol provides a generalizable platform for studying other membrane-bound AAA+ proteases and their cytoplasmic substrates, facilitating future mechanistic investigations of membrane-associated proteolysis under defined conditions.

DISCUSSION:

The first critical step of this protocol is Atto488 labeling of LpxC. A moderate labeling is critical for accurately measuring in vitro degradation activity. Based on our optimization experiments, a molar ratio of LpxC to Atto488 between 9:1 and 3:1 is optimal. Excessive Atto488 results in over-labeling, which likely perturbs the native state of LpxC. Under these conditions, the difference in degradation of LpxC–Atto488 by FtsH alone versus FtsH/LapB is markedly reduced (data not shown). Conversely, insufficient labeling produces fluorescent signals that are too weak for sensitive kinetic measurements. When adapting this protocol to measure the degradation of other water-soluble substrates, a substrate-to-Atto488 molar ratio of 3:1 provides a useful starting point.

Determining the reconstitution efficiency of FtsH/LapB into proteoliposomes is the second critical step of the protocol. To quantitatively assess LpxC degradation by FtsH/LapB, it is necessary to determine the concentration of FtsH within the proteoliposomes. In this protocol, the FtsH concentration is estimated by comparing the band intensity of a standard FtsH sample with that of FtsH in the proteoliposome samples.

Notably, LpxC–Atto488 exhibits partial self-degradation or denaturation even in the absence of FtsH/LapB. Therefore, background measurements and subtraction are other critical steps for accurate data interpretation when using this assay. In addition, Atto488 is light sensitive. To minimize photobleaching, reaction tubes should be wrapped in aluminum foil during the labeling step. During subsequent handling of LpxC–Atto488, exposure to ambient light should be minimized, ideally by working with the room lights turned off.

Monitoring the decrease of substrate bands by SDS–PAGE is a traditional approach for studying the in vitro degradation of water-soluble FtsH substrates, such as LpxC, RpoH, and CII^13,19,23. However, this method is largely qualitative rather than quantitative, as it lacks the sensitivity required to accurately determine initial velocities of substrate degradation, particularly at low substrate concentrations. In contrast, the present protocol enables sensitive and quantitative measurement of initial degradation rates. Moreover, reconstitution of FtsH/LapB into proteoliposomes provides a more physiologically relevant membrane environment. This proteoliposome-based system better preserves the functional context of membrane-bound FtsH and its interaction with the adaptor protein LapB, thereby enabling more accurate kinetic analysis²³.

This protocol provides a versatile platform for quantitatively studying membrane-associated proteolysis. It can be adapted to investigate other membrane-bound AAA+ proteases and their substrates. The ability to obtain reliable initial velocity measurements makes this method particularly valuable for mechanistic studies, including the determination of kinetic parameters and dissection of regulatory pathways. In addition, the proteoliposome-based reconstitution strategy allows controlled manipulation of lipid composition, protein stoichiometry, and accessory factors, facilitating studies of how membrane environment and cofactors influence protease function.

This protocol relies on an in vitro proteoliposome system that, while providing a membrane-like environment, does not fully recapitulate the complexity of the bacterial inner membrane, including lipid diversity and regulatory factors. In addition, FtsH concentration is estimated semi-quantitatively based on band intensity, which may introduce measurement bias. Fluorescent labeling of substrates with Atto488, although necessary for detection, may alter protein properties if not carefully controlled. Finally, the current workflow is optimized for water-soluble substrates and may not be directly applicable to membrane-embedded targets.

If fluorescence signals are weak or unstable, optimize substrate labeling by increasing the Atto488-to-LpxC ratio and/or extending the labeling time, while avoiding over-labeling that may impair protein function. High background or signal drift may indicate the presence of residual free dye; this can be mitigated by aliquoting LpxC–Atto488 samples and avoiding repeated freeze–thaw cycles. If kinetic data are noisy, adjust the protease concentration, substrate concentration range, and reaction time points to ensure measurements are collected within the initial velocity phase.

Supplementary Material

supplementary table 1

Supplementary Table 1: Fluorescence measurements from in vitro degradation assays using different concentrations of LpxC–Atto488. For each concentration, background signals and three biological replicates are shown. Net fluorescence values were calculated by subtracting the background.