ANALYST. SEPTEMBER 1989, VOL. 114 1017

Determination of Aprotinin by Titration With Bovine Trypsin With End-point Detection by High-performance Liquid Chromatography

Giorgio Raspi, Antonino Lo Moro and Maria Spinetti Dipartimento di Chimica e Chimica lndustriale dell'Universita 1-56100 Pisa, Italy Marica Molinari lstituto del C. N. R. di Chimica Quantistica ed Energetica Molecolare, 1-56100 Pisa, Italy

A method for the determination of aprotinin (bovine pancreatic trypsin inhibitor, BPTl) is described. The procedure involves the formation of the BPTI-trypsin complex in the presence of an excess of BPTI, quantitative separation of the residual BPTI from the mixture by affinity chromatography and identification and evaluation of the residual BPTl by reversed-phase high-performance liquid chromatography. The method is precise with a mean coefficient of variation of 4.0 and 4.3% for intra- and inter-assay runs, respectively, and has a limit of determination of 3.0 pg of aprotinin. The proposed method can be applied to commercial samples, even in very dilute solutions, for the standardisation of aprotinin. Keywords: Aprotinin determination; bovine trypsin; high-performance liquid chromatography

Bovine pancreatic trypsin inhibitor (BPTI), commonly known as aprotinin, is widely used in clinical and experimental biochemistry: it is usually measured in plasma, serum and biological fluids following therapeutic application in man or in experimental animals.1 It is a strongly basic protein (isoelec- tric point close to 10.5) with a remarkable stability to heat, acid or alkaline treatment, organic solvents and proteolytic degradation. The determination of BPTI is usually carried out by spectrophotometric2J or titrirnetric4.5 methods, based on the inhibition by BPTI of the trypsin-catalysed hydrolysis of an appropriate substrate under specified conditions. It is known that BPTI undergoes a reversible reaction with bovine trypsin at about pH 8.0 to form a very stable complex with a 1 : 1 stoicheiometry.' The BPTI concentration in the sample is determined from the decrease in the trypsin activity caused by the sample, compared with a reference containing buffer instead of BPTI.

These procedures are indirect and non-specific and can only be used with relatively pure and concentrated inhibitor samples; in some instances, however, the determination of BPTI is made more difficult by various other factors. In this paper a method is proposed based on the direct titration of BPTI with a solution of bovine trypsin of known molarity and monitoring the decreasing residual amounts of the inhibitor by reversed-phase high-performance liquid chromatography (RP-HPLC). As the presence of BPTI in the sample is unequivocal, a calibration graph is unnecessary. This approach to the determination of BPTI is the first in which the use of the substrate is avoided; moreover, the method can be extended to the study of other enzyme - inhibitor systems for analytical applications.

Experimental Apparatus

The HPLC system consisted of a Twincle liquid chromato- graph (Jasco, Easton, MD, USA), equipped with a VL-614 injector (loop capacity: 100-1000 pl), a Jasco Uvidec 100-111 variable wavelength UV detector and a Chromatopac C-R3A data elaborator.

The column was a Merck (Darmstadt, FRG) LiChrosorb RP-18 column (250 x 4.6 mm i.d., 7 pm) kept in a Clar 055 HPLC column block heater (Violet, Rome, Italy). The UV detector was operated at 200 nm.

Econo columns, 10 X 1.0 cm i.d. (Bio-Rad Laboratories, Richmond, CA, USA), were employed for affinity chromato- graphy *

Reagents

All chemicals were of analytical-reagent grade or of the highest purity available and were stored, when necessary, as recommended by the manufacturers. Acetonitrile, sodium perchlorate and orthophosphoric acid were obtained from Merck; BPTI was obtained from Behring Diagnostics (San Diego, CA, USA), Sigma (St. Louis, MO, USA) and Bayer (Leverkusen-Bayenverk, FRG). All chemicals for inhibition and active site measurements, including bovine pancreatic trypsin, type 111, were obtained from Sigma and used as received. Water was distilled once and then de-ionised using a Milli-Q purification system (Millipore, Bedford, MA, USA). The composition of buffer solution A (pH 2.2) was 10 mM H3P04 in 0.2 M NaC104 and that of buffer solution B (pH 8.3) was 0.1 M Tris - HCI in 0.5 M NaC1.

The mobile phase for HPLC was obtained by mixing buffer solution A (pH 2.2), previously filtered through a 0.45-pm membrane filter, with acetonitrile (70 + 30). The solvents were de-aerated by bubbling helium through them.

Bovine trypsin was bound to CNBr-activated Sepharose 4B according to the method given by the manufacturer (Phar- macia, Uppsala, Sweden). Prolonged washing cycles of alternate high and low pH were necessary in order to obtain an acceptable HPLC background.

Standards

Bovine trypsin was dissolved in 1 x 10-3 M HCI, containing 0.02 M CaCI2, to yield a 2.4 mg ml-1 stock solution, which was stored at 4°C. The molarity of the stock solution was determined by titration of the active site with 4-nitrophenyl- 4'-guanidinobenzoate, according to the method of Chase and Shaw.6 By diluting this solution with 1 x 10-3 M HCI, standard solutions containing 0.5-50.0 nmol ml-1 of trypsin were obtained.

Solutions of BPTI (ca. 10 pg ml-1) were prepared in buffer solution B and stored in a cool (4 "C), dark place; for testing the proposed procedure, the molarity of these solutions was determined by a spectrophotometric method.'

Procedure

The procedure involves the following steps: A , formation of the BPTI - trypsin complex in the presence of an excess of BPTI; B, quantitative separation of the residual BPTI from the mixture by affinity chromatography; and C, identification and evaluation of the residual BPTI by RP-HPLC.

Publ

ishe

d on

01

Janu

ary

1989

. Dow

nloa

ded

by U

nive

rsity

of

Mic

higa

n L

ibra

ry o

n 27

/10/

2014

15:

18:2

0.

View Article Online / Journal Homepage / Table of Contents for this issue

1018

Step A At least four identical aliquots, each containing approximately 3-300 pg of BPTI ( M , = 6512), are buffered at pH 8.3 with solution B: to three of these aliquots is added bovine trypsin standard solution in amounts such that the enzyme to inhibitor molar ratios were ca. 0.25,0.50 and 0.75, respectively. Careful control of the pH, with the possible addition of buffer solution B, allows the formation of the BPTI - trypsin complex. This complex is very stable over a narrow p H range centred around pH 8.3. The volume of the resulting solution does not need to be controlled because the residual BPTT will be isolated and collected in a known volume in the next step.

Step B Each solution resulting from step A is applied to a trypsin- Sepharose 4B column (bed volume ca. 1 ml), previously equilibrated with buffer solution B. The column is washed with at least ten bed volumes of the starting buffer to remove unbound substances prior to elution. The residual BPTI tightly bound to the immobilised trypsin is quantitatively recovered in a 5-ml calibrated flask by eluting, in this order, with 0.8 ml of 1.0 M H3P04 and buffer solution A up to a volume of 5 ml. Tn this step, therefore, an enrichment of the inhibitor is achieved when dilute solutions are used. Experi- ments carried out with solutions containing only the BPTI- trypsin complex at various dilutions showed that there was no significant release of BPTI.

Step C A 1000- or 10O-pl aliquot, depending on the amount of BPTI in each eluted solution, is injected on to the HPLC column and

4 8 12 Ti me/mi n

Fig. 1. Chromatogram of 6 pg ml-1 of BPTI solution. Mobile phase pH, 2.2; injection volume, lo00 pl. Other conditions as described under Experimental. A, Inhibitor

Y m 2 100

0 1 2 3 4 Amount of trypsinhmol

Fig. 2. Titration of BPTI (30 pg): peak area of the residual BPTI (peak A. Fig. 1) plotted against the amount of bovine trypsin added

ANALYST, SEPTEMBER 1989, VOL. 114

the chromatographic profile is obtained with isocratic elution at 25 +_ 0.1 "C and at a flow-rate of 1.0 ml min-1. The peak, A, due to the inhibitor, appears at 9.3 min (Fig. 1). The straight line graph obtained by plotting the peak areas corresponding to peak A as a function of the amount (nmol) of bovine trypsin added to each sample is extrapolated so that it intersects the x-axis (Fig. 2). At this point the amounts (nmol) of trypsin and of total BPTI present in the processed solution are identical.

Results The results obtained from a series of experiments carried out in order to test the proposed method are presented in Table 1. The data demonstrate the between-day (inter-assay) and within-day (intra-assay) variation in the method. The inter- assay variability was assessed singly from five replicate runs for amounts of BPTI in the range 3.0-300.0 pg and the intra-assay variability was determined in quintuplicate over the same range. The precision of the method (mean coefficient of variation) was 4.0 and 4.3% for intra- and inter-assays, respectively. The determination limit was calculated to be 3.0 pg, i.e., the level at which quantitative measurements of the peak areas were consistently possible with an injection volume of 1000 pl.

Discussion The principle of the method is the formation of a complex between BPTI and bovine trypsin: by monitoring at least one of these three components of the system during the titration, it should be possible to determine the amount of BPTI in the sample.

The possibility of following the titration by monitoring the excess of bovine trypsin used was rejected, because the enzyme is stable in weak acids but is rapidly destroyed at alkaline pH.

The possibility of monitoring the complex was investigated but the results were not encouraging. The presence of the complex and the residual RPTI in the sample at p H 8.3 can be demonstrated by injecting an aliquot of the sample directly on to the HPLC column and using a mobile phase of pH 8.0. Fig. 3 shows a typical chromatogram obtained under such condi- tions, where the peaks A and B are due, respectively, to the elution of the complex and of BPTI.7 The fraction correspond- ing to peak A was collected in order to test for the presence of the complex. After concentration on a Rotavapor at room

Table 1. Summary of intra- and inter-assay results

Difference between BPTI

BPTI added BPTI found added and Type k SD*/pg CV, YO found, YO

Intra-assay . . 3.0 3.2k0.3 9.4 +6.7 10.0 10.3f0.5 4.9 +3.0 30.0 29.4k0.7 2.4 -2.0 80.0 81.3k2.3 2.8 +1.6

150.0 152.2t4.0 2.6 +1.5 300.0 296.0k5.9 2.0 -1.3

Mean CV: 4.0%

Inter-assay . . 3.0 2.9k0.2 6.9 -3.3 10.0 10.4k0.4 4.0 +4.0

80.0 81.5k3.9 4.8 +1.9

300.0 304.5k6.7 2.2 +1.5

30.0 29.5 f 1.6 5.4 -1.7

150.0 146.2k3.5 2.4 -2.5

Mean CV: 4.3%

* n = 5

Publ

ishe

d on

01

Janu

ary

1989

. Dow

nloa

ded

by U

nive

rsity

of

Mic

higa

n L

ibra

ry o

n 27

/10/

2014

15:

18:2

0.

View Article Online

ANALYST, SEPTEMBER 1989, VOL. 114 1019

I 1 I

0 4 8 12 16 Time/mi n

Fig. 3. Chromatogram of A , the BPTI - trypsin complex and B, the excess of BPTI. BPTI (80 pg) and bovine trypsin (9.2 nmol) in 1.0 ml of buffer solution B; mobile phase pH, 8.0; injection volume, 100 p1. Other conditions as described under Experimental

temperature, the pH was lowered to 2.2 and an aliquot of the solution was injected by using a mobile phase of pH 2.2. The appearance of a peak with characteristics that were identical with those of the peak shown in Fig. 1 confirmed the presence of BPTI and hence the presence of the complex in the fraction Collected at pH 8.0. Owing to the severe tailing and poor resolution of peak A, reproducible peak areas could not be obtained. even by varying the column temperature and percentage of organic modifier in the mobile phase. More- over, problems arose from precipitates that form readily in the mobile phase and from the presence of inactive trypsin in the sample injected, which complicates the chromatographic trace and, overall, contaminates the HPLC column.

Instead, the chromatographic peak for BPTT, obtained under well-defined conditions of elution, was utilised success- fully for our purposes (Fig. 1). In fact, for this polypeptide, the conditions described here give good results in terms of recovery, resolution and reproducibility of peak areas.

Application

The proposed method can be applied to commercial samples containing preservatives or bacteriostatic agents, for the standardisation of BPTI solutions. Further, owing to the enrichment step in the procedure, the method is directly

applicable to the analysis of very dilute solutions. As with other literature methods, comparable amounts of othei trypsin inhibitors interfere seriously. Tn this instance, it i: necessary to isolate BPTI by a specific separation, foi example, by RP-HPLC under the conditions specified here.

The typical applications described above were carried out ir several series of experiments. Various commercial sample: and very dilute solutions collected from HPLC eluates 01 BPTI were assayed and reproducible values were obtained. with a relative standard deviation of 4.0%. To evaluate the proposed method, these values were compared with those obtained using a spectrophotometric methodl: the result5 obtained with the two methods were in good agreement, the difference being ca. k 1.8%.

Conclusion The proposed method for the determination of BPTT provides a suitable alternative to existing methods: the method involves a simple and effective sample clean-up procedure and a specific and rapid RP-HPLC determination so that the analyses can easily be performed with a minimum of sample handling. Using a series of affinity columns, and with a chromatographic run of about 10 min, a series of samples can be processed and analysed within a short period of time. Further, this method has the advantage of avoiding the limits resulting from the competition between the substrate and inhibitor for the active site of the enzyme.

This work was supported by CNR and MPI.

References 1. 2. 3.

4.

5 .

6.

7.

Fritz, H., and Wunderer, G., Drug Res., 1983, 33, 479. Kassel. B.. Methods Enzymol., 1970, 19, 844. Fritz, H., Trautschold, L., and Werle, E . , in Bergmeyer, H. U., Editor, “Methods of Enzymatic Analysis,” Volume 2, Verlag Chemie, Weinheim, and Academic Press, New York,

Federation Internationale Pharmaceutique Commission, 3rd Report, J. Mond. Pharm., 1968, I , 33. Ruyssen, R., and Lawers, A., “Pharmaceutical Enzymes,” Story-Scientia, Ghent, 1978, pp. 227-241. Chase, T., Jr., and Shaw, E., Biochem. Biophys. Res. Commun., 1967,29, 508. Raspi, G., Lo Moro, A., and Spinetti, M., Ann. Chim. (Rome), 1987,77, 525.

Paper 91008280 Received February 23rd, 1989

Accepted April 14th, 1989

1974, pp. 1064-1080.

Publ

ishe

d on

01

Janu

ary

1989

. Dow

nloa

ded

by U

nive

rsity

of

Mic

higa

n L

ibra

ry o

n 27

/10/

2014

15:

18:2

0.

View Article Online