Accepted Manuscript Kinetics of trypsin catalyzed hydrolysis determined by isothermal titration calorimetry Ksenia Maximova, Joanna Trylska PII: DOI: Reference:

S0003-2697(15)00320-6 http://dx.doi.org/10.1016/j.ab.2015.06.027 YABIO 12126

To appear in:

Analytical Biochemistry

Received Date: Revised Date: Accepted Date:

23 February 2015 13 June 2015 16 June 2015

Please cite this article as: K. Maximova, J. Trylska, Kinetics of trypsin catalyzed hydrolysis determined by isothermal titration calorimetry, Analytical Biochemistry (2015), doi: http://dx.doi.org/10.1016/j.ab.2015.06.027

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Kinetics of trypsin catalyzed hydrolysis determined by isothermal titration calorimetry. Ksenia Maximova and Joanna Trylska Centre of New Technologies University of Warsaw, Banacha 2C, 02-097 Warsaw, Poland Subject category: enzymatic assays and analysis Running title: Kinetics of trypsin catalyzed hydrolysis. Corresponding authors: Ksenia Maximova, Joanna Trylska Centre of New Technologies University of Warsaw, Banacha 2C, 02-097 Warsaw, Poland Tel. +48 (22) 5543-683 Fax +48 (22) 5540-801 Email: [email protected], [email protected]

1

Abstract Isothermal titration calorimetry (ITC) was applied to determine enzymatic activity and inhibition. We measured the Michaelis-Menten kinetics for trypsin-catalyzed hydrolysis of two substrates: casein (an insoluble macromolecule substrate) and Nα-benzoyl-DLarginine β-naphthylamide (a small substrate) and estimated the thermodynamic parameters in the temperature range 20-37 0C. The inhibitory activities of reversible (small molecule benzamidine) and irreversible (small molecule phenylmethanesulfonyl fluoride and macromolecule α1-antitrypsin) inhibitors of trypsin were also determined. We have shown the usefulness of ITC for fast and direct measurement of inhibition constants and half maximal inhibitory concentrations and for predictions of the mechanism of inhibition. ITC kinetic assays could be an easy and straightforward way to estimate MichaelisMenten constants, the effectiveness of inhibitors, as well as to predict the inhibition mechanism. ITC efficiency was found similar to that of classical spectrophotometric enzymatic assays.

Key words: enzyme kinetics, isothermal titration calorimetry, trypsin, casein, benzamidine, α1-antitrypsin

2

Introduction Enzyme kinetics is an informative study of enzyme catalyzed reaction rates. Such studies enable predicting the pathway of the catalyzed reaction and developing potential inhibitors of enzyme activity. The most common laboratory methods to study enzyme kinetics and inhibition are spectrophotometric and fluorometric assays. Although widely used, these techniques are frequently too costly or time-consuming because they require labelling of substrates with the chromophore or fluorophore groups. Moreover, opaque or turbid solutions are inappropriate for spectrophotometric detection, which prevents direct studies of insoluble proteins and presupposes multistep procedures. Isothermal titration calorimetry (ITC) could be used as an alternative technique in such troublesome cases because all reactions proceed with the absorption or release of heat, which ITC is able to detect with straightforward and fast measurements [1–5]. ITC is a physical technique for the estimation of the thermodynamic parameters of binding of small molecules to larger macromolecules [6]. Typically, the ligand is titrated into the cell with a receptor and the thermal power is monitored at each step. Thermal power data can be further analyzed to determine thermodynamic parameters such as enthalpy changes and equilibrium association constants. This technique is widely used to check the binding of inhibitors to their targets [6]. However, the drawback of such binding titration ITC experiments is that they require sufficient amounts of components (in the range of µM to mM), which can be either costly or difficult to obtain. This technique has not been appreciated for the kinetic studies where the amounts are

3

comparable with those required for classical spectrophotometric enzymatic assays (in the range of pM to nM). There are few studies that present the use of ITC for kinetic measurements and they show good correlation between calorimetric and spectrophotometric data for reaction kinetics [1,3,7, 8]. The ITC approach was applied to determine kinetic parameters for opaque solutions [9], for complex, multi-substrate, and multi-protein systems [10], for reactions with insoluble substrates [11], as well as to detect reaction inhibition by products and determine the thermodynamic activation parameters [12,13]. However, applications of ITC to enzyme kinetics are still rare, probably because of scarce literature data on this particular application and optimization of the conditions. We applied the ITC technique to determine the kinetics of trypsin (24 kDa) - catalyzed hydrolysis of casein (24 kDa), which is an example of insoluble macromolecule substrate, and Nα-benzoyl-DL-arginine β-naphthylamide (BANA) (440 Da), which is an example of a small molecule substrate (Fig. 1). We have also verified reaction inhibition by products and determined the thermodynamic activation parameters. The ITC-based method was also explored to investigate enzyme inhibition by benzamidine (BA) (120 Da), selected as an example of a small molecule reversible competitive inhibitor, phenylmethanesulfonyl fluoride (PMSF) (174 Da) - as a small molecule irreversible inhibitor, and α1-antitrypsin (AT) (52 kDa) - as a macromolecule irreversible inhibitor (Fig. 1).

Materials and Methods All chemicals were purchased from Sigma Aldrich and used without further purification. Enzyme

4

The stock solution (1 mM) of trypsin from bovine pancreas ( ≥10,000 BAEE units/mg protein) was prepared in 1 mM HCl and stored at -20 0 C. Substrates The stock solution (70 mM) of BANA was prepared in DMSO. The casein solution of 45.7 mg/mL (1.905 mM) was prepared in 50 mM potassium phosphate buffer at pH 7.5 by gentle heating with stirring to 80 - 85 ºC for approximately 10 minutes until a homogeneous dispersion was achieved. Inhibitors The stock solution (64 mM) of BA was freshly prepared in 50 mM potassium phosphate buffer at pH 7.5. The stock solution (100 mM) of PMSF was freshly prepared in ethanol. The stock solution (0.4 mM) of AT from human plasma was prepared in 1% sodium azide and stored at -80 0C. The composition of the buffer solution for the substrates and the enzyme, with or without inhibitors, for the assay experiments was identical. The concentration of DMSO in assays with BANA was kept at 10 % and the concentration of EtOH in assays with PMSF was kept at 1 %. All experiments were performed in triplicates in 50 mM potassium phosphate buffer at pH 7.5 and in Tables the determined averages are shown. In all the cases the mean percentage deviation from the average was within 15 %.

Equipment The ITC assays were carried out on the power-compensated instruments Microcal ITC200 (MicroCal, Northampton, MA, USA) and Nano ITC (TA Instruments, New Castle, DE, USA). The Microcal ITC200 has a hastelloy cell of 200 µL volume and a syringe of 40 µL volume, and the Nano ITC has a gold cell of 190 µL volume and a

5

syringe of 50 µ L volume. The stir ring speed in the calorimeter cells was 350 rpm, and thermal power was recorded every 2 s. In each experiment the control injections of the substrate to the buffer and of the buffer to the enzyme were carried out, and the heat of the dilution was corrected. The Microcal ITC200 and Nano ITC were used with the control software provided by the manufacturers, Microcal Origin 7 and NanoAnalyze, respectively.

Enzyme assays for the determination of apparent molar enthalpy ∆Happ (Nano ITC and Microcal ITC200) and kinetic constants (Microcal ITC200)

Trypsin catalyzed hydrolysis of casein - single injection assay The trypsin solution (1.95 µM) in 50 mM potassium phosphate buffer (pH 7.5) was equilibrated at 37 0 C for 200 s. Then one aliquot of 10 µL of the casein solution (1.9 mM) was injected. The change in the instrumental thermal power was monitored until the substrate hydrolysis was complete, i.e., the signal returned to the original base line (3000 s). Data analysis gave ∆Happ = 108 kJ/mol (Nano ITC) and 103.8 kJ/mol (Microcal ITC200).

Trypsin catalyzed hydrolysis of BANA (Nano ITC) The BANA solution (0.5 mM) in 50 mM potassium phosphate buffer (10% DMSO, pH 7.5) was equilibrated at 37 0 C for 200 s. Then two aliquots of 15 µ L of the trypsin solution (143 µM) were injected. The change in the thermal power was monitored until the signal retu rned to the original base line (2500 s). Data analysis gave ∆Happ = 2 kJ/mol.

6

Trypsin catalyzed hydrolysis of BANA (Microcal ITC200) The trypsin solution (25 µM) in 50 mM potassium phosphate buffer (10% DMSO, pH 7.5) was equilibrated at 37 0 C for 200 s. Then one aliquot of 15 µL of the BANA solution (3.75 mM) was injected. The change in the thermal power was monitored until the signal returned to the original base line (2500 s). Data analysis gave ∆Happ = 5.2 kJ/mol.

Enzyme assays for the determination of apparent molar enthalpy ∆Happ and product inhibition (nano ITC) Trypsin catalyzed hydrolysis of casein – verifying product inhibition The trypsin solution (3.33 µM) in 50 mM potassium phosphate buffer (pH 7.5) was equilibrated at 37 0 C for 200 s. Then six aliquots of 0.95 µL of the casein solution (1.9 mM) were injected. The change in the thermal power was monitored until the substrate hydrolysis was complete, i.e. the signal retu rned to the original base line (1000 s). Data analysis gave ∆Happ= 141 kJ/mol.

Trypsin catalyzed hydrolysis of BANA - verifying product inhibition The trypsin solution (25 µM) in 50 mM potassium phosphate buffer (10% DMSO, pH 7.5) was equilibrated at 37 0 C for 200 s. Then seven aliquots of 4.99 µL of the BANA solution (3.75 mM) were injected. The change in the thermal power was monitored until the substrate hyd rolysis was complete and the signal returned to the original base line (500 s). Data analysis gave ∆Happ = 2.3 kJ/mol.

Multiple injection assays to determine the kinetic constants The experiments were carried out on Nano ITC if not mentioned otherwise.

7

Trypsin catalyzed hydrolysis of casein - determination of working enzyme concentration Trypsin solutions (0.20, 1.17 and 1.95 µM) in 50 mM potassium phosphate buffer (pH 7.5) were equilibrated at 37 0 C for 200 s. Then twenty aliquots of 0.95 µL of the casein solution (1.9 mM) were injected every 150 s. Data analysis gave Km = 0.013 mM; kcat = 0.023 s-1 for 1.95 µM trypsin (Microcal ITC200); Km = 0.010 mM; kcat = 0.011 s-1 for 1.95 µ M trypsin (Nano ITC).

Inhibited trypsin catalyzed hydrolysis of casein The trypsin solution (1.95 µM) with BA (0.05, 0.1, 0.2, 2.0 and 3.0 mM); or PMSF (0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1 and 1.2 mM); or AT (0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0 µM) was pre-incubated in 50 mM potassium phosphate buffer (1 % EtOH in the case of PMSF, at pH 7.5), at 37 0 C for 60 min. Then twenty aliquots of 0.95 µL of the casein solution (1.9 mM) were injected every 150 s.

Trypsin catalyzed hydrolysis of BANA - determination of working enzyme concentration The trypsin solution (0.8 or 7.5 µM) in 50 mM potassium phosphate buffer (10 % DMSO, pH 7.5) was equilibrated at 37 0 C for 200 s. Then seventeen aliquots of 2.97 µ L of the BANA solution (3.75 mM) were injected every 100 s. Data analysis gave Km = 139 µ M and kcat = 0.08 s-1 (for 7.5 µM trypsin).

Inhibited trypsin catalyzed hydrolysis of BANA The trypsin solution (7.5 µM) with BA (0.1, 0.2, 0.3, 0.4 and 0.5 mM) or PMSF (0.05, 0.06, 0.07, 0.08, 0.09, 0.1, 0.11, 0.13, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8 and 0.9 µM) or AT (1, 2,

8

3, 3.75, 4, 5, 6, 7, 7.5, 8, 9 and 15 µM) was pre-incubated in 50 mM potassium phosphate buffer (10% DMSO, 1% EtOH in the case of PMSF, pH 7.5) at 37 0 C for 60 min. Then seventeen aliquots of 2.97 µL of the BANA solution (3.75 mM) were injected every 100 s.

Trypsin catalyzed hydrolysis of casein - determination of the thermodynamic activation parameters (∆H#, ∆S# and ∆G#) The trypsin solution (1.95 µM) in 50 mM potassium phosphate buffer (pH 7.5) was equilibrated at 37, 30, 25 and 20 0 C for 200 s. Then twenty aliquots of 0.95 µ L of the casein solution (1.9 mM) were injected every 150 s. Trypsin catalyzed hydrolysis of BANA - determination of the thermodynamic activation parameters (∆H#, ∆S# and ∆G#) The trypsin solution (7.5 µM) in 50 mM potassium phosphate buffer (10% DMSO, pH 7.5) was equilibrated at 37, 30 and 25 0 C for 200 s. Then seventeen aliquots of 2.97 µL of the BANA solution (3.75 mM) were injected every 100 s.

Results and Discussion Enzyme kinetics One of the most known mathematical models to explain the kinetics of enzymatic reactions was first proposed in 1903 [14] and, independently, in 1913 by Michaelis and Menten [15]. They postulated the mechanism of a single substrate reaction through Eq. 1. The most important assumption was large excess of the substrate (S) over the enzyme (E) leading to approximately constant amount of the enzyme-substrate complex (ES) over time and a steady state of the reaction. k1

E+S

ES

k cat

E+P

(1),

k -1

9

where k1 and k-1 are the rate constants of the formation and dissociation of ES, respectively; kcat is the rate constant of the products' (P) formation. The rate of the reaction at time t (Rt) can be further described by Eq. 2.


 =

 .[ ]

(2),

 [ ]

where Km is the Michaelis constant and vmax is the maximum observed velocity. The influence of various substrates, inhibitors, enzyme mutations or reaction conditions on either vmax or Km is a way to unravel the mechanism of enzymatic activity. ITC measures the kinetics of the reaction by detecting the heat absorbed or released during the enzymatic reaction. The principle of this technique involves titrating the substrate into the sample cell containing the enzyme (or vice versa), and maintaining equal temperatures in the sample cell and the reference cell (filled with buffer or water) during each injection [16]. The thermal power applied to maintain the same temperatures is equal to the heat released or absorbed during the processes occurring in the sample cell. It was shown that the heat (Q) linearly depends on the substrate moles (n) and can be characterized by Eq. 3 [1].

 =  ∙ ∆ = [] ∙  ∙ ∆

(3),

where ∆Happ is the apparent molar enthalpy of the reaction; [P]total is the concentration of the formed products; Vcell is the cell volume. Thus the thermal power at time t will be equal to the change of heat (dQ/dt) and thereby to the concentration of the product over time (Eq. 4).

 =

! 

=

["] 

∙  ∙ ∆

(4)

That is why the rate of the reaction as the function of the product formation can be estimated from the heat (Eq. 5).


 =

["] 

=

#

!

$%&'' ∙∆())



=

 .[ ]

(5)

[ ]

10

In general two approaches can be applied to run the ITC kinetic experiment – a single injection and multiple injection assays. In the first single injection case, only one portion of the substrate is injected in the cell containing the enzyme (or vice versa), and the heat of the conversion of all amounts of the substrate to the products is detected. Therefore, one observes full substrate conversion by the enzyme. In the second case, the experiment involves multiple injections of a highly concentrated solution of the substrate into the cell containing the enzyme. In this multiple injection assay, the concentration of the substrate in the reaction mixture increases stepwise until the saturation of the enzyme and maximum velocity are reached. For the trypsin-catalyzed hyd rolysis of casein and BANA, we have applied both single and multiple injection assays.

Single injection assay Figure 2 shows one injection of casein into the cell containing trypsin. Immediately after the injection, the baseline shifts and reflects the heat of the trypsin-catalyzed reaction. The consumption of casein is tracked till its completion, i.e. termination of the heat change, which is characterized by returning of the signal to the stable original baseline. For the single injection experiment, a relatively high concentration of the enzyme (in the range of nM to µM) and a relatively low concentration of the substrate (in the range of µ M to mM), as well as sufficient time after the injection, are required. The single injection assay is performed to determine ∆Happ by integrating the peak shown in Figu re 2, which is the amount of the heat produced by the reaction (Eq. 6).

∆ =

# +,-.

0∞ !

/01



23

(6)

About 3000 s was required for total decomposition of 45.7 mg/mL of casein by 1.95 µM of trypsin. The apparent molar enthalpy ∆Happ was equal to 108 kJ/mol (Nano ITC data are shown in Fig. 2; Microcal ITC200 data with ∆Happ = 103.8 kJ/mol are shown in Fig. 11

S1), which identifies that the hydrolysis of casein by trypsin is a complex process with possible several peptide bond breakdowns per one casein macromolecule [17,18]. Contrary, the single-injection monitored hydrolysis of BANA, a small substrate in which only one amide bond is cleaved, shown in Fig. 3 for Nano ITC, gave ∆Happ of 2 kJ/mol. The Microcal ITC200 data with ∆Happ = 5.2 kJ/mol are shown in Fig. S2. Due to the difference in the amount of heat observed in the casein and BANA hydrolyses, the enzyme concentration in the reaction with BANA had to be increased up to 10 µM in the cell to obtain a detectable signal. The experiment with BANA illustrates the possibility of placing the substrate in the cell and injecting the enzyme solution (Fig. 3). Indeed, this can be useful for substrates with poor solubility. Additionally, in the enzyme titration procedure, the second injection gives the heat of the dilution. Fig. 3 shows that the heat of dilution will significantly contribute in the case of low-enthalpy reactions and the raw data have to be corrected by taking it into account. Data from the single injection experiment can be transformed to the kinetic MichaelisMenten curve by the Microcal Origin 7 software supplied with Microcal ITC200 (Nano ITC does not provide software to determine kinetics from single injection experiments). Note, that for kinetics the substrate concentration has to be higher than Km ([S] > Km) and meet the steady-state condition ([S] >> [E]). The reaction rate is estimated from the decay heat curve of the substrate (Eq. 7 and Fig. 4) and instantaneous substrate concentrations at time t are calculated from Eq. 8.


 =

#

!

$%&'' ∙∆())



(7) 67

[4] = [4]0 −

/8 6

$%&'' ∙∆())

(8)

12

The hyd rolysis of casein by trypsin resulted in a continuous kinetic curve shown in Fig. 4. The Michaelis-Menten parameters Km and kcat were determined as 0.060 mM and 0.063 s-1, respectively. For the BANA cleavage the respective constants were Km = 0.524 mM and kcat = 0.053 s-1 (see Fig. S2). Although the single injection method is fast and enables measuring the rate of the reaction with insoluble substrates that can be placed in the cell instead of the syringe, this assay has some disadvantages. Due to the high volume of the injected solution, a significant impact of the heat of dilution and time of the injection, during which some substrate amounts can be depleted, may be observed. Whereas precise determination of the substrate concentration during the reaction is necessary to accurately determine the kinetic parameters [19]. Further, the single injection assay is a long process and could be sensitive to inhibition by products of the reaction, giving inaccurate molar enthalpy and kinetic constants because of the decrease of the amounts of the active enzyme. That is why verifying the inhibition of the enzyme by its products should be always carried out.

Verifying the inhibition of the enzyme by the products During the single injection assay the hydrolysis of the substrate gives a continuous increase of the products in the reaction mixture. However, the products themselves can decrease the amount of the active enzyme and thus the amount of the depleted substrate and heat. This may occur once the enzyme is inhibited by its products, which either do not dissociate fully after hyd rolysis or bind to the active or other enzyme sites. The product inhibition can be verified by several injections of equal amounts of the substrate into the cell containing the enzyme; note that sufficient time is necessary after each injection to complete the hydrolysis of all the injected amount of the substrate. As shown in Fig. 5 for the casein-trypsin reaction, the heat signal decreases in every following

13

injection of casein, which signifies a decrease of the active form of the enzyme due to its inhibition by the products. Contrary, an equal amount of heat for each following injection for the BANA-trypsin hydrolysis signifies no or negligible product inhibition (see Fig. 5). Thus the apparent molar enthalpy should be calculated including, if applicable, the effect of product inhibition. ∆Happ for the BANA-trypsin reaction was assumed as the average molar enthalpy of the injections and equaled to 2.3 kJ/mol in agreement with the result from the singe injection assay (2 kJ/mol). For the casein-trypsin system, indeed a significant product inhibition was observed, giving different determined ∆Happ in the single injection assay (108 kJ/mol) and in the assay verifying the product inhibition (141 kJ/mol). In this casein-trypsin case only the heat of the second injection was considered, where only negligible inhibition took place (see Fig. 5). The heat of the first injection was not taken into account in the data analysis because of the impact of the diffusion of the solutions during the insertion of the syringe and the equilibration stage. Overall, measuring the kinetics of the reaction with negligible product inhibition is manageable by the ITC single injection assay, otherwise the product accumulation will influence the Michaelis-Menten functions and data analysis. Thus the single injection assay has to be carried out taking into account the correction for possible product inhibition in order to obtain reliable ∆Happ for further determination of kinetic constants (Eq. 5). However,

measurement and determination of kinetic constants has to be

performed in a fast experiment with high concentration of the substrate to avoid obvious product accumulation and possible inhibition. The multiple injection assay described further meets such conditions.

14

Multiple injection assay In the multiple injection assay a highly concentrated substrate solution is injected multiple times, allowing for the baseline equilibration after each injection (see Fig. 6). The enzyme concentration in the reaction cell is generally lower than in the single injection assay, so the amount of the product produced is negligible and the effect of product inhibition is minimized. That is why the multiple injection assay should be more accurate. Differently than in the single injection assay, in which sufficient time after the injection is allowed and the entire substrate is depleted with the baseline returning to the origin level, in the multiple injection assay the concentration of the substrate in the reaction mixture increases stepwise with every injection. The concept is to maintain only minor substrate cleavage during each injection and reach the satu ration of the enzyme and maximum velocity at the steady-state conditions. To achieve this, the experiment should be run with a relatively high concentration of the substrate (in the range of mM to M), but the initial substrate concentration in the cell in the beginning of the titration should be below Km to have some data points on the Michaelis-Menten plot below Km; the final substrate concentration must be higher than Km ([S] > Km) and meet the steady-state condition ([S] >> [E]). This assay requires a relatively low concentration of the enzyme (in the range of pM to µM) since higher amounts of the enzyme would result in a faster substrate depletion, which is a condition more appropriate for the single injection experiment. However, the low enzyme concentr ation will give low heat, especially for the reactions with low enthalpy change, so the multiple injection assay is sensitive to the baseline drifts [20,21]. For the BANA-trypsin reaction to obtain a measurable signal the enzyme concentration had to be increased to 7.5 µM (Fig. 6); lower enzyme concentration of 0.8 µM did not give detectable heat.

15

The minimal concentration to detect trypsin-catalyzed hydrolysis of casein was 0.2 µ M (Fig. 7). However, with such a low 0.2 µM concentration of trypsin we observed the product inhibition (see Fig. 7). Therefore, the working concentration of the enzyme was increased and set to 1.95 µM, the concentration at which the product inhibition was negligible (see Fig. S3 (bottom), Table S1). The equilibrated baseline position after each injection was used to construct the plot of the reaction rate from Eq. 5 as a function of the total substrate concentration (assuming that no appreciable substrate degradation takes place during the measurement) and Michaelis-Menten constants were determined (see Fig. 7, Table 1). The experiment was repeated on the Microcal ITC200 (see Fig. S3 (top)) and the results were compared with the single injection measurements (Table 1). The kinetic constants from the single and multiple injection assays differ because of the impact of the product inhibition in the former assay. The multiple injection data were comparable for both instruments so this assay was used for further kinetic experiments. For the depletion of BANA by trypsin the obtained Km was equal to 139 µM and kcat was 0.08 s-1 (see Fig. 6). Note, that in the multiple injection assay there is no significant substrate consumption during the measurement because of the short time between the injections. However, the instrument response time has to be taken into account for correct kinetic experiments [22]. It was shown that if the reaction time is longer than 10 times the instrument response time, the discrepancies caused by the thermal lag can be neglected [4,5,23]. Thus for microcalorimetric instruments with the response time of 10 - 15 s the minimum acceptable spacing between the injections should not be lower than 100 - 150 s.

Enzyme inhibitors

16

We used ITC to determine the kinetic constants characterizing substrate binding and converting it to products in the catalytic cycle of the enzyme. However, ITC has not been appreciated for the enzyme inhibition studies, so we have verified ITC for this purpose. Since either irreversible or reversible inhibition decreases enzymatic activity, the difference in heat of the substrate hydrolysis by the enzyme without and with the inhibitor could be used to quantitatively monitor the inhibition effect. An irreversible inhibitor covalently modifies the enzyme, so the concentration of the enzyme-substrate complex decreases and permanently reduces the enzymatic activity, which is reflected by a decrease in the maximum velocity, vmax, or the turnover number, kcat [24]. For reversible inhibitors many modes of binding (competitive, uncompetitive, noncompetitve, mixed and partial mixed) can be distinguished because they differently affect Km and kcat [24]. For example, increasing the concentration of a competitive inhibitor in the reaction mixture increases the Km parameter and does not change vmax. In order to explore the ITC technique for the inhibitory activity, we selected benzamidine (BA) as an example of the trypsin reversible and competitive inhibitor [25– 28]. We carried out multiple injections of the BANA substrate to the pre-incubated with BA trypsin solution (of 0.1, 0.2, 0.3, 0.4 and 0.5 mM) at 37 0 C for 60 minutes (see Fig. 8 and Table 2). The data showed increasing Km and a trend of the unchanged kcat , which is in agreement with the competitive inhibition mechanism. The inhibitor effectiveness can be estimated by its inhibition constant Ki, which refers to the equilibrium dissociation constant of the enzyme-inhibitor complex [29]. Generally, to measure the inhibitor binding affinity, the half maximal inhibitory concentration (IC50) is determined because the procedure is less complicated. To estimate Ki the rate of the reaction has to be measu red while independently varying both the inhibitor and substrate concentrations (comparing with IC50 where the measurements are performed

17

at one concentration of the substrate and various concentrations of the inhibitor). That is why to determine Ki about 100 experiments have to be run (including repetitions). However, the process can be simplified by running the calorimetry titration assays, in which the concentration of the substrate changes during one experiment, and Ki for the competitive inhibition can be calculated from Eq. 9.

9:)) = 9: (1 +

[> ]

?

)

(9),

where Kmapp is the Michaelis constant at the inhibitor concentration [I]. Thus based on the Michaelis-Menten curves shown in Fig. 8 for the BANA-trypsin reaction with BA as the inhibitor, the average Ki was equal to 102 µM. Benzamidine as a reversible inhibitor was also verified in the casein-trypsin system (see Fig. S4 and Table S2). However, the applied inhibitory concentration of benzamidine had to be millimolar because of a more complex mechanism of the trypsin action on such a large macromolecule as casein [17,27]. Higher benzamidine concentration was also necessary in other experiments with natural substrates that were hydrolyzed by trypsin [27]. As mentioned above, not all inhibitors bind reversibly to enzymes. In some cases the enzymes are inactivated by the formation of irreversible covalent complexes with inhibitors. Such processes are typically multistep and more complicated so the apparent inhibitory activity is estimated by IC50, in other words by detecting the changes in vmax or kcat. A frequently used serine protease irreversible inhibitor is PMSF [30]. We used PMSF in the trypsin-catalyzed experiments with casein and BANA as substrates (see Fig. 9 and Table 3). In comparison with benzamidine as the competitive inhibitor, the PMSF inhibitor showed a notable decrease in the reaction rates and similar K m parameters upon increasing the PMSF concentration. This indicates an irreversible mechanism of inhibition by PMSF. The IC50 value for PMSF, as a small irreversible 18

inhibitor of the casein-trypsin reaction, was determined as 0.6 mM, which was higher than its IC50 of 0.1 mM obtained for the BANA-trypsin system (Fig. S5). We observed the same effect on the kinetic constants for the AT inhibitor (Fig. 10, Table 4) that is also a known irreversible inhibitor of proteases [31–33]. Interestingly, IC50 for the AT 52 kDa inhibitor was lower in the case of AT inhibition of the reaction with the macromolecule casein as a substrate (IC50 of about 2.5 µM) than in the reaction with a small BANA substrate (5 µM) (Fig. S6). Again, this suggests the complexity of the casein hydrolysis. To elucidate this mechanism in detail additional studies are required.

Activation energy (Ea) and thermodynamic activation parameters (∆G#, ∆H#, ∆S#) The ITC experiments can be performed in 2 - 80 0 C temperature range giving a straightforward effect of the temperature on the rates of the enzymatic reactions. To estimate the thermodynamic parameters we car ried out the multiple injection assay at 20 – 37 0 C (see Fig. 11 and Table 5). The activation energies (Ea) for the hydrolysis of casein (35.9 kJ/mol) and BANA (-25.4 kJ/mol) were calculated from the slopes (-Ea/R) of the Arrhenius plots, which were equal to -4320 and 3057 K, respectively. Next, the thermodynamic activation parameters: enthalpy (∆H#), entropy (∆S #), and Gibbs free energy (∆G#) of the activation were estimated from the equations 11-13 [34], ∆H# = Ea – RT

(11)

∆G# = RT(ln(kB/h) + lnT – lnkcatapp)

(12)

∆S# = (∆H# –∆G#)/T (13), where the energy values are in kJ/mol, kcat is in s–1, kB is the Boltzmann constant (1.3805 x 10-23 JK–1), h is the Planck constant (6.626 x 10-34 J s–1), and R is the gas constant (8.314 J mol-1 K-1). The correlation between the reaction rates and temperatures is different for the casein and BANA systems: casein prefers hydrolysis at higher temperatures and

19

BANA at lower temperatures. Similarly, the dependence on pH and ionic strength of the medium can be carried out.

Conclusions We have used ITC to determine the kinetics of hydrolysis of various substrates by trypsin and inhibitory activity of reversible and irreversible trypsin inhibitors. The comparison with other assays and techniques has shown that ITC can be a comparable and faster way to determine the reaction rates and thermodynamic parameters than standard spectrophotometric and fluorometric assays. The advantage of ITC is that one can use unlabeled and poorly soluble compounds and similar ligand and enzyme concentrations as in spectrophotometric assays. Even for the reactions characterized by low enthalpy, and thus low heat, additional background reaction can be monitored to study the desired process. Overall, the ITC kinetic experiments can be applied for accurate, fast and straightforward determination of kinetic constants for small substrates with low enthalpy reactions and for insoluble large substrates together with the estimation of product inhibition. The experiments can be performed at various conditions such as temperature, pH, ionic strength and media. Also, we have shown the efficiency of this technique for measuring the inhibitory activity and predicting the mechanism of inhibitor's action.

Acknowledgements This research received funding from the Polish-Norwegian Research Programme operated by the National Centre for Research and Development under the Norwegian Financial Mechanism 2009-2014 in the frame of Project Contract No POL-

20

NOR/198939/13/2013. The authors thank the Division of Biophysics, Faculty of Physics University of Warsaw for the use of Microcal ITC200 funded by the EU Project No. POIG.02.01.00-14-122/09, prof. Ryszard Stolarski and dr. Colette Quinn of TA Instruments-Waters LLC for helpful discussions.

References

[1]

M.J. Todd, J. Gomez, Enzyme kinetics determined using calorimetry: a general assay for enzyme activity?, Anal. Biochem. 296 (2001) 179–87.

[2]

M.L. Bianconi, Calorimetry of enzyme-catalyzed reactions, Biophys. Chem. 126 (2007) 59–64.

[3]

I. Haq, Calorimetry in the fast lane : The use of ITC for obtaining enzyme kinetic constants, 2011.

[4]

B.A. Williams, E.J. Toone, Calorimetric evaluation of enzyme kinetic parameters, J. Org. Chem. 58 (1993) 3507–3510.

[5]

P.E. Morin, E. Freire, Direct calorimetric analysis of the enzymatic activity of yeast cytochrome c oxidase, Biochemistry. 30 (1991) 8494–8500.

[6]

A. Velazquez-Campoy, S.A. Leavitt, E. Freire, Characterization of protein-p rotein interactions by isothermal titration calorimetry, Methods Mol Biol. 261 (2004) 35– 54.

[7]

J.E. Ladbury, M.L. Doyle, M.L. Bianconi, Biocalorimetry 2. Chapter 9. Titration Calorimetry as a Tool to Determine Thermodynamic and Kinetic Parameters of Enzymes, John Wiley & Sons, Ltd, Chichester, UK, 2004.

[8]

C. Aguirre, I. Condado-Morales, L.F. Olguin, M. Costas, Isothermal titration calorimetry determination of individual rate constants of trypsin catalytic activity, Anal. Biochem. 479 (2015) 18–27. inserted reference

[9]

S.N. Olsen, Applications of isothermal titration calorimetry to measure enzyme kinetics and activity in complex solutions, Thermochim. Acta. 448 (2006) 12–18.

[10] G.D. Watt, A microcalorimetric procedure for evaluating the kinetic parameters of enzyme-catalyzed reactions: Kinetic measurements of the nitrogenase system, Anal. Biochem. 187 (1990) 141–146. [11] T. Lonhienne, E. Baise, G. Feller, V. Bouriotis, C. Gerday, Enzyme activity determination on macromolecular substrates by isothermal titration calorimetry:

21

application to mesophilic and psychrophilic chitinases, Biochim. Biophys. Acta Protein Struct. Mol. Enzymol. 1545 (2001) 349–356. [12] J. Cheleski, H.J. Wiggers, A.P. Citadini, A.J. da Costa Filho, M.C. Nonato, C.A. Montanari, Kinetic mechanism and catalysis of Trypanosoma cruzi dihydroorotate dehydrogenase enzyme evaluated by isothermal titration calorimetry, Anal. Biochem. 399 (2010) 13–22. [13] L. Cai, A. Cao, L. Lai, An isothermal titration calorimetric method to determine the kinetic parameters of enzyme catalytic reaction by employing the product inhibition as probe, Anal. Biochem. 299 (2001) 19–23. [14] V. Henri, Lois Générales de l’action des diastases, 1903. [15] L. Michaelis, M.L. Menten, K.A. Johnson, R.S. Goody, The original Michaelis constant: translation of the 1913 Michaelis-Menten paper, Biochemistry. 50 (2011) [16] M.W. Freyer, E.A. Lewis, Isothermal titration calorimetry: experimental design, data analysis, and probing macromolecule/ligand binding and kinetic interactions, Methods Cell Biol. 84 (2008) 79–113. [17] J. Tauzin, L. Miclo, S. Roth, D. Mollé, J.-L. Gaillard, Tryptic hyd rolysis of bovine αS2-casein: identification and release kinetics of peptides, Int. Dairy J. 13 (2003) 15–27. [18] J.H. Northrop, The kinetics of trypsin digestion: II. Conditions under which the reaction is monomolecular, J. Gen. Physiol. 6 (1924) 417–28. [19] N.A. Demarse, M.C. Killian, L.D. Hansen, C.F. Quinn, Determining enzyme kinetics via isothermal titration calorimetry, Methods Mol. Biol. 978 (2013) 21–30. inserted reference [20] L. Murphy, M.J. Baumann, K. Borch, M. Sweeney, P. Westh, An enzymatic signal amplification system for calorimetric studies of cellobiohyd rolases, Anal. Biochem. 404 (2010) 140–8. [21] M.J. Baumann, L. Mu rphy, N. Lei, K.B.R.M. Krogh, K. Borch, P. Westh, Advantages of isothermal titration calorimetry for xylanase kinetics in comparison to chemical-reducing-end assays, Anal. Biochem. 410 (2011) 19–26. [22] A. Zogg, F. Stoessel, U. Fischer, K. Hungerbühler, Isothermal reaction calorimetry as a tool for kinetic analysis, Thermochim. Acta. 419 (2004) 1–17. [23] C.W. Hansen, L.D. Hansen, A.D. Nicholson, M.C. Chilton, N. Thomas, J. Clark, J.C. Hansen, Correction for instrument time constant and baseline in determination of reaction kinetics, Int. J. Chem. Kinet. 43 (2011) 53–61. [24] D.L. Nelson, M.M. Cox, Lehninger Principles of Biochemistry, W.H. Freeman, 2013.

22

[25] H.-K.S. Leiros, B.O. Brandsdal, O.A. Andersen, V. Os, I. Leiros, R. Helland, J. Otlewski, N.P. Willassen, A.O. Smalås, Trypsin specificity as elucidated by LIE calculations, X-ray structures, and association constant measurements, Protein Sci. 13 (2004) 1056–1070. [26] M. Mares-Guia, E. Shaw, Studies on the active center of trypsin. The binding of amidines and guanidined as models of the substrate side chain, J. Biol. Chem. 240 (1965) 1579–85. [27] F. Markwardt, H. Landmann, P. Walsmann, Comparative Studies on the Inhibition of Trypsin, Plasmin, and Thrombin by Derivatives of Benzylamine and Benzamidine, Eur. J. Biochem. 6 (1968) 502–506. [28] R. Talhout, A. Villa, A.E. Mark, J.B.F.N. Engberts, Understanding binding affinity: a combined isothermal titration calorimetry/molecular dynamics study of the binding of a series of hydrophobically modified benzamidinium chloride inhibitors to trypsin, J. Am. Chem. Soc. 125 (2003) 10570–10579. [29] A. Cornish-Bowden, A simple graphical method for determining the inhibition constants of mixed, uncompetitive and non-competitive inhibitors, Biochem J. 137 (1974) 143–144. [30] M.A.A. Dorrah, M.K.A. Elmelegi, T.T.M. Bassal, Trypsin- and chymotrypsin esterase-like activity in the midgut regions of the larval Parasarcophaga hertipes, EFFLATOUNIA. 8 (2008) 1–8. [31] N.M. Green, E. Work, Pancreatic Trypsin Inhibitor 2. Reaction with trypsin, Biochem J. 54 (1952) 347–352. [32] J.-S. Shin, M.-H. Yu, Kinetic dissection of alpha 1-antitrypsin inhibition mechanism, J. Biol. Chem. 277 (2002) 11629–35. [33] H. Loebermann, R. Tokuoka, J. Deisenhofer, R. Huber, Human a1-Proteinase Inhibitor, J. Mol. Biol. 177 (1984) 531–556. [34] T. Lonhienne, C. Gerday, G. Feller, Psychrophilic enzymes : revisiting the thermodynamic parameters of activation may explain local flexibility, Biochim. Biophys. Acta. 1543 (2000) 1–10.

23

Table 1 ITC-determined constants for trypsin-catalyzed hydrolysis of casein. Instrument Microcal ITC200 Microcal ITC200

Nano ITC

Nano ITC

Single injection

Multiple

Single

Multiple

assay

injection assay

injection

injection

assay

assay

103.8

N/A

108

N/A

k cat (s-1)

0.063

0.023

N/A

0.011

Km (mM)

0.060

0.013

N/A

0.010

∆Happ (kJ/mol)

N/A - data are not available (see text)

24

Table 2. Trypsin-catalyzed hyd rolysis of Nα-benzoyl-DL-arginine β-naphthylamide hydrochloride (BANA) with benzamidine (BA) as an inhibitor added at various concentrations. Experiments were performed on Nano ITC. Concentration of BA (mM) 0

0.1

0.2

0.3

0.4

0.5

k cat (s-1)

0.08

0.11

0.09

0.11

0.15

0.15

Kmapp (µM)

172

334

443

614

947

1390

Ki (µM)

0

106

127

117

89

71

25

Table 3. Multiple injection assay with phenylmethanesulfonyl fluoride (PMSF) as an inhibitor performed on Nano ITC. Substrate Casein

BANA

PMSF (mM)

0

0.1

0.5

1

1.1

0

0.1

0.5

0.7

kcat (s-1)

0.012

0.009

0.005

0.001

0

0.091

0.048

0.014

0

Kmapp (µM)

10

9

10

8.5

0

195

190

191

0

78

43.5

13

0

100

53

15

0

Enzyme activity (%) 100

26

Table 4. Multiple injection assay with α1-antitrypsin (AT) as an inhibitor performed on Nano ITC. Substrate Casein

BANA

AT (µM)

0

1

2

2.5

3

0

3.75

7.5

15

kcat (s-1)

0.012

0.009

0.009

0.006

0

0.062

0.055

0.002

0

Km (µM)

10.1

9.5

7.8

7.7

0

153

212

135

0

Enzyme activity (%) 100

78

78

52

0

100

87

3.4

0

27

Table 5. Thermodynamic parameters obtained from the multiple injection assay. Substrate Casein

BANA

Temperature (0C)

37

30

25

20

37

30

25

k cat (s-1)

0.009

0.008

0.006

0.004

0.088

0.102

0.132

Km (µM)

8.05

8.67

7.70

7.15

172

180

198

Ea (kJ/mol)

35.6

∆H# (kJ/mol)

33.34

33.39

33.44

33.48

-27.99 -27.93 -27.89

∆G# (kJ/mol)

63.86

61.87

60.29

58.26

69.74

T∆S# (kJKmol-1)

-30.52

-28.48 -26.85 -24.78 -97.73 -96.42 -95.84

-25.4

28

68.49

67.95

Figure legends Figure 1. Backbone representation of a bovine pancreatic trypsin (223 amino acids, PDB code: 1S0Q), Nα-benzoyl-DL-arginine β-naphthylamide (BANA), benzamidine (BA), phenylmethanesulfonyl fluoride (PMSF), and human α1-antitrypsin (424 amino acids, PDB code: 3NE4).

Figure 2. Trypsin-catalyzed hydrolysis of casein from a single injection assay performed on Nano ITC. Red line shows the complete conversion of casein to the products by trypsin and the blue line the heat of dilution of casein in the buffer.

Figure 3. Trypsin-catalyzed hydrolysis of BANA. Red line shows the injection of trypsin into the cell with the BANA solution and the blue line the injection of trypsin into the cell containing only the buffer. Experiments were performed on Nano ITC.

Figure 4. Trypsin catalyzed hydrolysis of casein from a single injection assay performed on Microcal ITC200. Left: The complete conversion of casein to the products by trypsin (with ∆Happ = 103.8 kJ/mol). Right: The single injection assay data for the casein-trypsin reaction fitted by Microcal Origin 7 with determined Km = 0.06 mM and kcat = 0.063 s-1.

Figure 5. The ITC experiment verifying the inhibition of the enzyme by the products. Data are from Nano ITC. (A) Twenty injections of 0.95 µL of 1.9 mM casein into the cell containing 3.33 µM of trypsin. After each injection the change in the instrumental thermal power was monitored until the substrate hydrolysis was completed (1000 s). (B) Seven injections of 4.99 µ L of 3.75 mM BANA into the cell containing 25 µM of trypsin.

29

After each injection the change in the instrumental thermal power was monitored until the substrate hydrolysis was completed (500 s).

Figure 6. Trypsin-catalyzed hyd rolysis of BANA from the multiple injection assay performed on Nano ITC. (A) Raw data of the seventeen injections of 2.97 µL of 3.75 mM BANA into the cell containing 7.5 µM of trypsin. After each injection the change in the instrumental thermal power was monitored until a new baseline level equilibration (100 s). (B) Fit of the multiple injection assay raw data for the BANA-trypsin reaction by NanoAnalyze giving Km = 139 µM and kcat = 0.08 s-1.

Figure. 7. Trypsin-catalyzed hydrolysis of casein from the multiple injection assay performed on Nano ITC. (A) Twenty injections of 0.95 µL of 1.9 mM of casein into the cell containing 0.2 µM (red line); 1.17 µM (blue line); 1.95 µM (green line) of trypsin. After each injection the change in the instrumental thermal power was monitored until a new baseline level equilibration (150 s). (B) The fit of the data by NanoAnalyze.

Figure 8. Trypsin-catalyzed hyd rolysis of BANA with BA as an inhibitor performed on Nano ITC. (A) The conversion of BANA to the products by trypsin with various concentrations of added BA inhibitor (0 mM – red, 0.1 mM – blue, 0.2 mM - green, 0.3 mM – black, 0.4 mM – yellow and 0.5 mM – grey lines). (B) The data fit by NanoAnalyze.

Figure 9. Multiple injection assay with PMSF as an inhibitor performed on Nano ITC. (A) The conversion of casein to the products by trypsin dependent on the concentrations of the PMSF inhibitor (0 mM – red, 0.1 mM – blue, 0.2 mM - green, 0.5 mM – black

30

and 1 mM – yellow lines). (C) The conversion of BANA to the products by trypsin dependent on the concentration of the PMSF inhibitor (0 mM – red, 0.1 mM – blue and 0.5 mM - green lines). (B, D) The data fit by NanoAnalyze.

Figure 10. Multiple injection assay with AT as an inhibitor carried out on Nano ITC. (A) The conversion of casein to the products by trypsin dependent on the concentration of the AT inhibitor (0 µM – red, 1 µM – blue, 2 µM - green, 2.5 µM – black and 3 µM – yellow lines). (C) The conversion of BANA to the products by trypsin with AT (0 µM – red, 3.75 µM – blue, 7.5 µM - green and 15 µM – black lines). (B, D) The data fit by NanoAnalyze.

Figure 11. Raw Nano ITC data from the multiple injection assay performed to determine thermodynamic parameters for the casein-trypsin and BANA-trypsin reactions. (A) The conversion of casein to the products by trypsin at different temperatures (37 0 C – red, 30 0 C – blue, 25 0 C – green and 20 0 C – black lines). (C) The conversion of BANA to the products by trypsin at different temperatures (37 0 C – red, 30 0C – blue and 25 0 C - green lines). (B, D) The data fit by NanoAnalyze.

31

32

33

34

35

36

37

38

39

40

41

42