Porcine pancreas kallikrein (pKLK, 250 units) and amine-terminated magnetic particles (50 mg mL⁻¹) were purchased from Sigma-Aldrich (St. Louis, MO, United States). Buffer components and other chemicals were acquired from Sigma-Aldrich, Merck (Darmstadt, Germany), Synth (São Paulo, Brazil), or Acros (Geel, Belgium). The water used in all the preparations was obtained from a MILLI-Q^® system (Millipore^®, São Paulo, Brazil). All the chemicals and solvents used here were analytical or HPLC grade and were employed without any further purification.

The magnetic separator (model DynaMagTM-Spin Magnet) was purchased from Thermo Fisher Scientific (São Paulo, SP, Brazil). Orbital Microplate shaker (IKA, model MS3 D, Staufen, Germany). HPLC-MS analyses were carried out on a high-performance liquid chromatography system (model NexeraXR) purchased from Shimadzu (Kyoto, Japan) and equipped with three LC 20ADXR pumps, an SIL-20A automatic injector, a DGU-20A degasser, a 10-port two-position high-pressure switching valve (Valco Instruments Co. Inc. Houston, United States), and a CBM-20A system controller. The HPLC system was coupled to an ion trap mass spectrometer (model AmaZon speed) acquired from Bruker Daltonics (Bremen) and equipped with an electrospray ionization source (ESI) interface. The system was controlled with the Bruker Compass Hystar software (version 4.5), and data were acquired and analyzed by using the Compass DataAnalysis software (version 4.3).

Porcine pancreas kallikrein (pKLK) was immobilized on the surface of amine-terminated magnetic beads (MBs). Briefly, 20 mg of MBs was washed three times with 1 mL of 100 mM KH₂PO₄ buffer, pH 7.0. After each wash, the MBs were magnetically separated by using a magnetic separator, and the supernatant was removed. Then, 1 mL of glutaraldehyde (5% in 100 mM KH₂PO₄ buffer, pH 7.0) was added to the MBs and incubated under smooth rotation at room temperature for 3 h. After that, the supernatant was removed, and the MBs were washed again three times with 1 mL of 25 mM KH₂PO₄ buffer, pH 8.0. Next, 400 μL of pKLK solution was added to the MBs and incubated under smooth rotation at 4°C for 16 h. After immobilization, the supernatant was removed, and IMER-pKLK-MB was washed three times with 100 mM Tris-HCl buffer, pH 8.0, and stored in the same buffer (20 mg mL⁻¹) at 4°C until it was used.

After enzyme immobilization as described above, 100 μL of the supernatant was added to 100 μL of substrate Z-Phe-Arg-AMC (200 μM) and incubated for 15 min under agitation (800 rpm) at room temperature. Afterward, 5 μL of this solution was directly injected into the MS. Additionally, 100 μL of enzyme solution before the immobilization was incubated with 100 μL of substrate under the same conditions described above. Then, 5 μL of this solution was directly injected into MS. The ESI ionization parameters were as follows: positive ionization mode, SIM mode, capillary voltage = 4500 V, end plate voltage = 550 V, drying gas flow rate = 9 L·min⁻¹, drying temperature = 275°C, and nebulizer pressure = 40 psi. The enzymatic reaction product, Z-Phe-Arg-OH, was monitored at m/z 456 [M + H]⁺. We evaluated the immobilization efficiency on the basis of immobilization yield (IY) by applying the following equation: I Y   % = A 0 − A f A 0 × 100 where A₀ corresponds to the initial activity of the enzyme solution before immobilization, and A_f corresponds to the activity of the supernatant solution after the immobilization procedure.

The LC-MS method was developed on the basis of a previous method described by Carvalho et al. (2022). LC analysis was carried out by using a Phenomenex Security Guard TM (model AJ04287) C18 micron guard cartridge (4.0 × 3.0 mm, Phenomenex, Torrance, California, EUA) connected to a 10-port/two-position valve in the LC-MS system to redirect the flow during pre-column cleaning to a waste line (Figure 1). The mobile phase consisted of water (solvent A and C) and methanol (solvent B); the injection volume was 5 µL. With the valve in position 1, the enzymatic reaction products were eluted from the trap column to the mass spectrometer with an initial mobile phase composed of 50% solvent B at a flow rate of 0.2 mL min⁻¹. After 2.0 min, the valve was switched to position 2, and solvent B in the mobile phase was increased to 100% at a flow rate of 1.0 mL min⁻¹, which was maintained until 5 min, to wash the column and to remove excess substrate. At 5.1 min, solvent B in the mobile phase was returned to 50% at a flow rate of 1.0 mL min⁻¹ until 6.0 min. At 6.1 min, the valve was switched to position one again, under the initial condition. The total run time was 7 min. Pump C (solvent water) at a flow rate of 0.2 mL min⁻¹ was used to deliver the mobile phase to the MS during the column washing step. All the analyses were performed at 21°C (controlled room temperature). The ESI ionization parameters were as follows: positive ionization mode, SIM mode, capillary voltage = 4,500 V, end plate voltage = 550 V, drying gas flow rate = 9 L min⁻¹, drying temperature = 275°C, and nebulizer pressure = 40 psi. The enzymatic reaction product, Z-Phe-Arg-OH, was monitored at m/z 456 [M + H]⁺.

The mobile phase and method configuration are shown in Table 1.

The IMER-pKLK-MB off-line assay conditions were optimized. More specifically, the optimal AcNH₄ concentration, pH, incubation temperature, incubation time, and IMER-pKLK-MB mass were determined.

AcNH₄ was tested at 5 or 10 mmol.L⁻¹, at pH 7.0, 8.0, 8.5, or 9.0, in triplicate. For this purpose, 0.5 mg of IMER-pKLK-MB (25 μL of the stock suspension) was added to a 0.5-mL conical microtube. IMER-pKLK-MB was washed three times with 200 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, to remove the storage buffer. After each wash, IMER-pKLK-MB was separated by using a magnetic separator for 1 min, and the supernatant was removed. Following the washing procedure, 75 μL of 5 or 10 mmol.L⁻¹ AcNH₄ at pH 7.0, 8.0, 8.5, or 9.0, and 25 μL of 200 μmol.L⁻¹ substrate solution (in AcNH₄ and at the tested pH) were added to IMER-pKLK-MB. The suspension was homogenized by vortexing and incubated under agitation at room temperature for 5 min. After incubation, the supernatant of the enzymatic reaction was removed by using a magnetic separator and analyzed by the LC-MS method described above. For the LC-MS analysis, 25 μL of the enzymatic reaction supernatant and 25 μL of 5 mmol.L⁻¹ AcNH4, pH 9.0, were added to a vial; the injected volume was 5 μL. To optimize the incubation temperature (10, 26, or 37°C), incubation time (5, 15, or 30 min), and IMER-pKLK-MB mass (0.1 or 0.5 mg, corresponding to 5 or 25 μL of the stock suspension, respectively), 5 mmol.L⁻¹ AcNH4, pH 9.0, was used, and the procedure described above was followed.

The method was qualified on the basis of criteria outlined in the literature (Cassiano et al., 2009; Food and Drug Administration, 2001) The linearity of the method was assessed by constructing a calibration curve. For this purpose, 0.1 mg of IMER-pKLK-MB was incubated with the following final substrate concentrations (n = 3): 6, 10, 15, 30, or 50 μmol.L⁻¹ in 5 mmol.L⁻¹ AcNH₄, pH 9.0, according to the procedure described above. Analyses were performed by using the LC-MS method described above. The curve was constructed by applying linear regression in Excel software, and the area of the product ion at m/z 456 [M + H]⁺ was plotted against substrate concentration. The selectivity of the method was evaluated by incubating IMER-pKLK-MB with buffer only, i.e., without substrate, under the same conditions described above. The lower limit of quantification (LLQ) and the limit of detection (LD) were assessed by incubating 0.1 mg of IMER-pKLK-MB with the following final substrate concentrations: 1, 6, 10, or 15 μmol.L⁻¹. LLQ was defined as the lowest concentration that can produce precision expressed as relative standard deviation (RSD) of less than 20%, while LD was defined as the lowest concentration generating a signal greater than twice the baseline noise.

To determine K_Mapp, 0.1 mg of IMER-pKLK-MB (equivalent to 5 μL of the stock suspension) was added to a 0.5-mL conical microtube. IMER-pKLK-MB was washed three times with 200 μL of 5 mmol.L⁻¹ AcNH₄, pH 9, to remove the storage buffer. After each wash, IMER-pKLK-MB was magnetically separated by using a magnetic separator for 1 min, and the supernatant was discarded. Subsequently, IMER-pKLK-MB was incubated with 75 μL of 5 mmol.L⁻¹ AcNH4, pH 9.0, and 25 μL of substrate at 1, 6, 10, 15, 30, 40, 50, 70, 200, or 300 μmol.L⁻¹ in 5 mmol.L-1 AcNH₄, pH 9.0. Each substrate concentration was tested in triplicate. The suspension was homogenized by vortexing and incubated under agitation at room temperature for 5 min. After incubation, the enzymatic reaction supernatant was removed by using a magnetic separator and analyzed by the LC-MS method described above. For the LC-MS analysis, 25 μL of the enzymatic reaction supernatant and 25 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, were added to a vial; the injected volume was 5 μL. A curve was constructed by plotting the peak area of the product ion at m/z 456 [M + H]⁺ as a function of substrate concentration. Data were fitted by using nonlinear regression into a Michaelis-Menten plot, and K_Mapp was obtained with the GraphPad Prism 8.0 software.

Inhibition studies were carried out by using leupeptin as standard inhibitor. To determine the half maximum inhibitory concentration (IC₅₀), 0.1 mg of IMER-pKLK-MB (equivalent to 5 μL of the stock suspension) was added to a 0.5-mL conical microtube. IMER-pKLK-MB was then washed three times with 200 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, to remove the storage buffer. After each wash, IMER-pKLK-MB was magnetically separated using a magnetic separator for 1 min, and the supernatant was discarded. Next, IMER-pKLK-MB was incubated with 65 μL of 5 mM AcNH₄, pH 9.0, 25 μL of 200 μmol.L⁻¹ substrate (in 5 mmol.L⁻¹ AcNH₄, pH 9.0), and 10 μL of leupeptin at different concentrations. The final leupeptin concentration, tested in triplicate, was: 0.5, 1, 2, 5, 10, 20, 40, 70, 100, 200, or 400 μmol.L⁻¹. As control, the same assay was performed without adding leupeptin. The suspension was then homogenized by vortexing and incubated under agitation at room temperature for 5 min. After incubation, the enzymatic reaction supernatant was removed using a magnetic separator and analyzed by the LC-MS method described above. The enzymatic activity was monitored by quantifying the area of the product ion at m/z 456 [M + H]⁺. The enzymatic activities in the presence (Ai) and absence (A0) of leupeptin were compared, and the percentage of inhibition was calculated by employing the equation: %I = 100 - [(Ai/A0) × 100]. The inhibition curve was constructed by plotting the percentage inhibition versus the corresponding leupeptin concentration, and IC50 was determined from the curve built by plotting [leupeptin] versus % inhibition with the GraphPad Prism 5.0 software.

To determine the operational stability of IMER-pKLK-MB, 0.5 mg of IMER-pKLK-MB (equivalent to 25 μL of the stock suspension) in a 0.5-mL conical microtube was subjected to 10 consecutive analysis cycles. IMER-pKLK-MB was then washed three times with 200 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, to remove the storage buffer. After each wash, IMER-pKLK-MB was separated using a magnetic separator for 1 min, and the supernatant was discarded. Following the washing procedure, 75 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, and 25 μL of 200 μmol.L-1 substrate solution (in 5 mmol.L⁻¹ AcNH₄, pH 9.0) were added to IMER-pKLK-MB. The suspension was homogenized by vortexing and incubated under agitation at room temperature for 5 min. After incubation, the enzymatic reaction supernatant was removed by using the magnetic separator and analyzed by the HPLC-MS method described above, immediately after the reaction had ended. To analyze the enzymatic activity, 25 μL of the enzymatic reaction supernatant and 25 μL of 5 mmol.L⁻¹ AcNH₄, pH 9.0, were added to a vial. The injected volume was 5 μL. To initiate a new reaction cycle, IMER-pKLK-MB was washed three times with 5 mmol.L-1 AcNH₄, pH 9, and the reaction procedure was restarted.

The storage stability of IMER-pKLK-MB was assessed by measuring its activity over time. The procedure and HPLC-MS method described above were followed. The evaluation periods included 1, 2, 3, 4, 5, 6, and 11 months after immobilization. A specific aliquot was set aside and used exclusively for this study.

We developed the HPLC-MS method to analyze reaction products while preventing ionic suppression or contamination of the ionization source by the substrate. To achieve this, we employed a C18 pre-column as a trap column, which effectively separated the products from the substrate and allowed only the products to enter the mass spectrometer, while the substrate was retained on the stationary phase. A 10-port, two-position switching valve was used for this purpose. In position 1, the reaction products are eluted from the trap column and transferred to the mass spectrometer using a methanol mobile phase (1:1, v/v), as detailed in Table 1. Changing the valve to position two allows excess substrate to be removed by washing the pre-column with mobile phase of increasing strength, containing up to 100% solvent B. We evaluated the enzymatic activity by integrating the area of the extracted ion chromatograms (EIC) for the product Z-Phe-Arg-OH, at m/z 456 [M + H]⁺, which showed a more intense signal compared to the other product (AMC, at m/z 176 with [M + H]⁺). Scheme 1 illustrates the catalytic reaction involving pKLK.

Figure 3 displays the chromatogram obtained when we analyzed the IMER-pKLK-MB activity by using the developed LC-MS method. By applying the method, we successfully separated the reaction products from the substrate and effectively removed most of the substrate from the system, thereby minimizing potential issues related to ionic suppression resulting in a fast analysis that lasted only 6 min.

We assessed the linearity of the method by constructing a calibration curve in which we plotted the product area as a function of substrate concentration. The calibration curve demonstrated a linear response for substrate concentrations ranging from 1 to 50 μmol.L⁻¹. The regression equation was y = 3606771.127x + 5817619.093; the coefficient of determination (R ²) was 0.999 (n = 3). The RSD values for the curve construction were less than 19% for all triplicates. We confirmed that the method was selective by noting the absence of a detectable signal for the product at m/z at 456 [M + H]⁺ with when we incubated only the buffer with IMER-pKLK-MB under the same assay conditions. The limit of detection (LD) and lower limit of quantification (LLQ) were 1 and 6 μmol.L-1, respectively. On the basis of literature parameters, the developed method proved adequate for monitoring the IMER-pKLK-MB activity.

Compared to traditional kallikrein activity assays that use the enzyme in solution with fluorimetric detection, the proposed method offers several advantages. It allows for enzyme reuse, reducing costs and increasing storage stability, and is effective in the presence of organic solvents. This is particularly beneficial for ligand screening applications, as it enables testing of a broader range of samples, including those that are insoluble in aqueous solutions. Additionally, samples that present fluorescence can interfere with fluorimetric assays, leading to false positives (Simeonov, 2018). This issue is avoided by using mass spectrometry, which offers high selectivity, sensitivity, and efficiency.