- Email: [email protected]
Assaying proteinases with azocoll
ANALYTICAL
BIOCHEMISTRY
l&446-450
Assaying
(I 984)
Proteinases
with Azocoll
RICARD~CHAVIRA,JR.,THOMASJ.BURNETT,*
ANDJAMESH.HAGEMAN'
Department of Chemistry, New Mexico State University, Las Cruces, New Mexico 88003 and *Stauffer Chemical Company, 1200 S. 47th Street, Richmond, Call$omia 94804 Received July 22, 1983 Azocoll, an insoluble, ground collagen to which a bright-red azodye is attached has been widely used for the assay of proteolytic enzymes. Earlier studies showed that hydrolysis of azocoll progressed linearly as a function of proteinase concentration but in an exponentially increasing manner as a function of time. No explanation for the latter behavior has been offered. We have found that assays of both crude extracts of Bacillus subtilis and commercial preparations of subtilisin BPN’ gave linear rates of hydrolysis of azocoll as a function of protease concentration; however, both gave increasing rates of hydrolysis of azocoll as a function of time. In attempting to improve and standardize proteolytic assaysusing azocoll we have found: (a) the absorption maximum of solubilized azocoll at pH 7.8 is 5 16 nm and is not signiticantly altered at acid pH; (b) assayswhich are perfectly linear as a function of time can be obtained by using azocoll that has been vigorously prewashed with buffer; (c) the soluble hltrate removed by prewashing can regenerate the nonlinear time courses previously observed; and (d) the rate of hydrolysis of azocoll can be varied by a factor of 3 by varying the rates of agitation of the assay tubes. In summary, to obtain reproducible, linear assaysit was essential to prewash commercial azocoll and agitate reaction tubes vigorously. KEY WORDS:azocoll; proteinase; Bacillus subtilis: protease substrates; subtilisin; enzyme assays.
Azocoll, an insoluble protein-dye conjugate, has been widely used for the assay of proteolytic enzymes (l-7). Azocoll is hydrolyzed readily by a variety of proteinases and yields soluble, colored peptides in proportion to enzyme concentrations at fixed incubation times (for example, Biologics Technical Information Dot. No. 3805-482, Calbiochem-Behring Corp., Refs. (4,8)). However, Cabib and Ulane (9) reported that azocoll was hydrolyzed at an increasing rate as a function of incubation time when either yeast chitin synthetase activating factor or trypsin was assayed. We have observed a similar nonlinear time course for the hydrolysis of azocoll by extracts of Bacillus subtilis. We show here that the nonlinearity is due to inhibitory material present in commercial preparations, examine the effect of agitation on rates of reaction, and describe a means of obtaining linear assays as a function of time. ’ To whom correspondence should be addressed. 0003-2697184 $3.00 Copyright Q 1984 by Academic Ress. Inc. All rights of reproduction in any form reserved.
MATERIALS
AND METHODS
Azocoll was purchased from CalbiochemBehring Corporation and from Sigma Chemical Company. Trizma base, subtilisin BPN’ (Type VII) and CaClz were purchased from Sigma. All other chemicals used were of reagent grade. B. subtilis 168 (trp-) was grown in supplemented nutrient broth and harvested 6 h after the end of logarithmic growth (10). Cells were washed, extracted, and centrifuged as previously described (10) except that 1 M KC1 was included in the wash buffers to minimize contamination of extracts with extracellular proteases. The clear, cell-free supernatant fraction was used in assays without dialysis or further treatment. Standard assays for azocoll hydrolysis were carried out as follows. Azocoll (0.250 g) was suspended in 50 ml of buffer (0.05 M TrisHCl, 1 mM CaClz, pH 7.8) and stirred rapidly with a magnetic stirring bar apparatus in a
446
AZOCOLL
HYDROLYSIS
447
loo-ml beaker. (A suspension of 5 mg/ml of termined spectrophotometrically at 278 nm azocoll is similar to levels used by others pre- assuming EtSb,, = 11.7 (12). viously.) Stirring was rapid enough to give what appeared to be a uniform suspension. RESULTS Portions of 1 to 3.0 ml were rapidly removed using an open-end plastic syringe (1 ml) and Azocoll which has been completely digested transferred to glass assay tubes (13 mm di- by subtilisin has absorption maxima at 5 16 ameter). Smaller tubes were unsatisfactory as and 415 nm (relative absorptivities of 1.2: 1) at pH 7.8 in 50 mM Tris-HCl. In this study, the azocoll clung to the sides of the tubes upon however, we have measured absorbances at rapid agitation. Tubes were placed vertically 520 nm because all previous work by others in a rack held in a rotary water bath shaker (New Brunswick Scientific Model G-86) at has been at this wavelength. We found that 1 37°C. The tubes were allowed to prewarm for mg/ml of azocoll yielded an ASu, of 0.593 when 15 min before initiating the reaction by ad- completely digested, it was previously reported dition of enzyme (20-40 ~1). Reactions were that a 1-mg/ml suspension of azocoll yielded stopped by rapidly (5-10 s) drawing the so- an ASzo of 0.464 (14). Nearly the same speclution into a Pasteur pipet (2 ml) fitted with trum was observed in 0.3 M trichloroacetic a small pack of glass wool at the neck. The acid (X,, , 520 nm; L&,, 0.491). clear filtrate, freed from the insoluble azocoll, In studying levels of intracellular proteinase in B. subtilis 168 during the course of growth was then removed to a clean tube by means of a second Pasteur pipet. Control tubes, con- and sporulation, it was necessary to vary the taining azocoll but no enzyme, were incubated times of incubation of extracts with azocoll for the same length of time as the test samples. because of the wide variation in levels of proAbsorbances of filtrates were read against teinase (10). Although rates of hydrolysis of azocoll were linear with the concentrations of buffer on a Beckman Model 24 spectrophotometer fitted with a rapid sampler. Average cell-free extracts used, rates of hydrolysis were absorbances (A& from at least 3 control fil- not linear as a function of time. A very distinct trates were subtracted from the absorbances lag phase was apparent (Fig. 1). We thought Q&,) of test filtrates. Absorption spectra were that this lag might be a result of the presence taken on a Perkin-Elmer 320 spectrophotomof more than one proteinase in the cell-free eter. Assays with cell-free extracts only, were extracts of B. subtilis (13). To eliminate this carried out in Beckman microfuge tubes (2 possibility the experiment described in Fig. 1 ml). Assays were stopped by immersing tubes was repeated using a commercial preparation in an ice-water bath. Chilled tubes were then of subtilisin BPN’; again, the lag was observed centrifuged in a Microfuge B (Beckman In(T. J. Burnett, Ph.D. dissertation, NMSU, strument Co.) centrifuge, and absorbance at 1983). This suggested that some inherent 520 nm of the supematant fraction was meaproperty of the azocoll preparation was resured as above. sponsible for the aberrant time course. To test Protein concentrations of cell-free extracts the idea that the lag might be caused by inof B. subtilis were measured by the method hibitory, soluble peptides present in the azoof Bradford (11) using the Bio-Rad Protein collagen, a portion of azocoll was washed by Assay Kit I (Bio-Rad Corp.) with an immupreincubating it in the assay buffer for 90 min. noglobulin standard. Subtilisin was prepared It was filtered and resuspended in fresh buffer for use by weighing out 1 mg/ml of subtilisin immediately before using in an assay. When BPN’ into buffer (0.05 M Na-acetate, pH 6.0, the assay tubes were not shaken at any time 1 mM CaCl*) and dialyzing for 10 h against during the assay, prewashing the azocoll rethe same buffer with 3 changes of buffer. The sulted in a linear release of colored soluble concentration of subtilisin in assays was de- material as a function of time as shown in
CHAVIRA,
BURNETT,
Ttme(hr)
FIG. I. Hydrolysis of azocoll by cell-free extracts of B. subtilis as a function of time. Each assaytube contained 2.5 mg of azocoll in 0.5 ml of 50 mM Trir+HCl, 1.0 mM CaClr, pH 7.8. The reaction was started by adding cellfree extract from B. subtilis (Materials and Methods) and placing in a water bath at 37°C. Tubes were agitated in a vortex mixer every 15 min. Assays were stopped by placing tubes on ice for 3 min and centrifuging in a Beckman microfuge B for 2 min. Absorbances at 520 nm were read against control tubes containing no cell-free extract. Assays contained 0.078 mg (0) or 0.195 mg (A) of cellfree extract of B. subtilis. Values reported are averages of triplicate measurements which have been corrected for corresponding control tubes containing no extract; standard deviations did not exceed 10% of the mean values.
Fig. 2. Apparently, material causing the lag period in the time-course curve was eliminated by prewashing the azocoll. In reviewing the literature, we found that workers have used different rates of agitation during the incubation of azocoll with proteinase; no shaking (1,9), gentle shaking (15), and vigorous shaking (8) of incubation tubes have been used. None of the workers specified the precise conditions or rates of agitation. To determine the effect of rates of agitation on the initial rates of hydrolysis of prewashed azocoll, we carried out the experiment described in Fig. 3. The rate of agitation clearly had a dramatic effect on the rate of the reaction. Rates of shaking greater than about 330 rpm caused little increase in the rates of reaction. We found also that it was essential to use a tube of sufficient diameter to prevent azocoll from clinging to the sides of the tube. We next tested the idea that soluble pep tides, or other materials in the azocoll, were inducing the lag in the time-course curve. Concentrated portions of filtrate obtained
AND HAGEMAN
from prewashing of azocoll as described above were added back to prewashed azocoll, and the time course of hydrolysis of these samples by subtilisin was compared to that of washed azocoll suspended in fresh buffer (Fig. 4). These assays were carried out with vigorous shaking, but a similar effect was observed when no shaking was used (data not shown). In initial experiments, when assay tubes were not shaken, the extent of the lag period varied from one batch of azocoll to the next. The data of Table 1 show that the background absorbances of washes from different commercial preparations of azocoll do vary considerably. Such variations in materials released upon incubation in buffer might account for the variations observed in the extent of the lag period.
OO
60
120
1
Time(mln)
FIG. 2. Effect of prewashing azocoll on its rate of hydrolysis by B. subtilisin. Azocoll(O.525 g) was suspended in 105 ml of buffer (0.05 M Tris-HCl, 1.O mM CaClr , pH 7.8) and allowed to stand without shaking at 37’C for 90 min. The suspension was Iiltered through Whatman No. 1 filter paper and the filtrate discarded. The precipitate was resuspended in 105 ml fresh buffer and stirred with a magnetic stirrer to produce a uniform suspension. Portions (3.0 ml) were transferred into test tubes (13 mm diam) and allowed to prewarm for 15 min at 37°C. The assay was initiated by addition of 0.054 mg (120 ~1) of subtilisn BPN’. Reactions were stopped by filtration as described under Materials and Methods. Control tubes containing no enzyme were run in parallel. Absorbances of tiltrates from sample and control tubes were read against buffer. Values for unwashed (0) and prewashed (0) axocoll are averages of duplicates which were corrected by subtracting values from controls.
AZGCOLL
449
HYDROLYSIS TABLE RELEASE OF MATERIAL BATCHES OF AXEOLL
1
ABSORBING AT 520 nm FROM IN ABSENCE OF PROTEINASE
Source of Azocoll Calbiochem Sigma No. Sigma No. Sigma No. Sigma No. 0
20 Tlme(min)
40
60
No. 1 1 2 3 4
A520“ll
h
0.037 0.095 0.055 0.025 0.05 1
LIAzocoll was suspended (5 mg/ml) in buffer (0.05 M Tris-HCl, pH 7.8), with mM CaC12. After 1.0 h at 37°C sample was rapidly filtered through glasswool; absorbance of filtrate was read against buffer.
FIG. 3. Effect of rate of agitation on hydrolysis of prewashed azocoll by subtilisin. Washing of azocoll and assays of azocoll hydrolysis were as described for Fig. 2 except 1.Oml assayvolumes were used. Test tubes ( 13 mm diam) scribed above was used as the substrate. We were placed vertically in a rack held in a rotary bath thought this might be due to variations in reshaker (New Brunswick Scientific Model G-86) at 37°C. sidual colored soluble material associated with The rates of agitation were (0) 100 rpm, (A) 250rpm, the variations in batches of azocoll (Table 1). (0) 330 t-pm. Dotted lines are data from an independent experiment; the upper and lower lines were for tubes To attempt to totally eliminate the lag period shaken at 400 and 0 rpm, respectively. When the lag we prewashed the azocoll for 2 h using vigobserved in this experiment was eliminated (see Fig. 5) orous shaking. When this procedure was folrates of 330 and 400 rpm were the same.
Finally, in some experiments, a slight lag was seen (Fig. 3) even when azocoll that had been prewashed by the soaking procedure de-
lowed linear rates of hydrolysis were obtained (Fig. 5). The background absorbances of control tubes did not change the time and had an average value of only 0.004 units at 520 nm (compare with values in Table 1). DISCUSSION
T~mdmln)
FIG. 4. Effect of readdition of concentrated, soluble wash from azocoll on rate of hydrolysis of prewashed azocoll. Azocoll was washed and assayed as described in Fig. 2 except tubes (13 mm diam) were shaken at 400 rpm. In one case (0) the washed azocoll is resuspended in fresh assay buffer and in the other (A) washed azocoll was suspended in the wash obtained from incubating 0.375 g azocoll in 25 ml of buffer (3 times concentration used in standard assay) for 90 min without shaking.
Using insoluble substrates such as azocoll to assay enzymatic activity has two major disadvantages. First, distributing equal amounts of substrate to assay tubes requires either individual weighings for each tube or rapid distribution of a liquid suspension. In this regard, the finer mesh azocoll (100-250 mesh) sold by Calbiochem-Behring Corporation can be dispensed more uniformly in suspensions than coarser mesh products. Second, because of the nonhomogeneity of the solution, thermal motion is not sufficient to insure maximal contact of the enzyme with the substrate; thus, the degree and nature of agitation and assay tube geometry are variables which should be specified in order to reproducibly perform an assay. As Fig. 3 shows, the rate of reaction in optimally shaken assay tubes increased threefold over the rate in unshaken tubes.
450
CHAVIRA,
BURNETT,
AND HAGEMAN
that during digestion of these soluble peptides a fraction of the enzyme is active in a reaction that does not contribute to an increase in soluble colored products. After the initial lag, rates of reaction became comparable (Fig. 2). In any case, sufficiently vigorous prewashing of the azocoll yielded a substrate that gave linear kinetics with respect to time whether or not the assay tubes were agitated during the assay. 0
0
50 T~me(md
100
150
FIG. 5. Effect of extensive prewashing of azocoll on linearity of azocoll hydrolysis with time. Prewashing of azocoll was performed by suspending 0.3 g of azocoll in 60 ml of standard buffer and shaking at 400 rpm in a IOO-ml round-bottom flask for 2.0 hat 37°C and filtering. The washed azocoll was resuspended in 60 ml of fresh buffer and azocoll assayswere performed as described in Fig. 2 but using I .O-ml volumes and agitation of 400 rpm (A) or no agitation (0).
The color yield we observed for the azocoll was similar to the value previously reported ( 14) and probably varies from one lot of azoco11to another. When unwashed commercial azocoll was incubated in buffer alone, colored products were released (Table 1) in a continuous but nonlinear manner as a function of time (data not shown). However, the apparent rates of hydrolysis of unwashed azocoll by several proteases (this work and Ref. (9)) increased as a function of time even after corrections for the background absorbances of controls from corresponding incubation times were made. We suggest that the lag periods observed are due to soluble dye-peptide species associated with the azocoll (Table 1) which act as competitive inhibitors towards subtilisin and presumably other proteinases. Figures 2 and 4 show that the inhibitory effect is transient, which might suggest that the protease digests these soluble species to products which can no longer combine with the enzyme. The apparent inhibition might result from the fact
ACKNOWLEDGMENTS This work was supported in part by USPHS RR08 136 2nd by GM 19643 which are gratefully acknowledged.
REFERENCES 1. Oakley, C. L., Warrack, G. H., and van Heyningen, W. E. (1946) J. Pathol. Bacterial. 58, 229-235. 2. Rinderknecht, H., Geokas, M. C., Silverman, P., and Haverback, B. J. (1968) Clin. Chim. Acta 21,197203. 3. Millet, J. (1971) C. R. Acud. Sci. Paris 272, 18061809. 4. Orrego, C., Kerjan, P., Manta de Nadra, M. C., Szulmajster, J. (1973) J. Bacterial. 116, 636-647. 5. Cheng, Y-S. E., and Aronson, A. I. (1977) Proc. Nat. Acad. Sci. USA 14, 1254-1258. 6. Studebaker, J. F. (1979) J. Chromatogr. 185, 497503. 7. Parrish, R. F., Goodpasture, J. C., Zaneveld, L. J. D., and Polakoski, K. L. (1979) J. Reprod. Fertil. 57, 239-243. 8. Cheng, Y-S. E., and Aronson, A. I. (1977) Arch. Microbiol. 115, 61-66. 9. Cabib, E., and Ulane, R. (1973) Biochem. Biophys. Res. Commun. 50, 186-191. 10. Geele, G., Garrett, E., and Hageman, J. H. (1975) in Spores VI (P. Gerhardt, R. N. Costilow, and H. L. Sadoff, eds.), pp. 391-396. Amer. Sot. Microbiol., Washington, D. C. 11. Bradford, M. (1976) Anal. Biochem. 72,248-254. 12. Matsubura, H., Kasper, C. B., Brown, D. M., and Smith, E. L. (1965) J. Biol. Chem. 240, 11251130. 13. Srivistava, Om. P., and Aronson, A. I. (1981) Arch. Microbial. 129, 227-232. 14. Kerjan, P., Keryer, E., and Szulmajster, J. (1979) Eur. J. Biochem. 98,353-362. 15. Millet, J., and Gregoire, J. (1979) Biochimie61,385391.
BIOCHEMISTRY
l&446-450
Assaying
(I 984)
Proteinases
with Azocoll
RICARD~CHAVIRA,JR.,THOMASJ.BURNETT,*
ANDJAMESH.HAGEMAN'
Department of Chemistry, New Mexico State University, Las Cruces, New Mexico 88003 and *Stauffer Chemical Company, 1200 S. 47th Street, Richmond, Call$omia 94804 Received July 22, 1983 Azocoll, an insoluble, ground collagen to which a bright-red azodye is attached has been widely used for the assay of proteolytic enzymes. Earlier studies showed that hydrolysis of azocoll progressed linearly as a function of proteinase concentration but in an exponentially increasing manner as a function of time. No explanation for the latter behavior has been offered. We have found that assays of both crude extracts of Bacillus subtilis and commercial preparations of subtilisin BPN’ gave linear rates of hydrolysis of azocoll as a function of protease concentration; however, both gave increasing rates of hydrolysis of azocoll as a function of time. In attempting to improve and standardize proteolytic assaysusing azocoll we have found: (a) the absorption maximum of solubilized azocoll at pH 7.8 is 5 16 nm and is not signiticantly altered at acid pH; (b) assayswhich are perfectly linear as a function of time can be obtained by using azocoll that has been vigorously prewashed with buffer; (c) the soluble hltrate removed by prewashing can regenerate the nonlinear time courses previously observed; and (d) the rate of hydrolysis of azocoll can be varied by a factor of 3 by varying the rates of agitation of the assay tubes. In summary, to obtain reproducible, linear assaysit was essential to prewash commercial azocoll and agitate reaction tubes vigorously. KEY WORDS:azocoll; proteinase; Bacillus subtilis: protease substrates; subtilisin; enzyme assays.
Azocoll, an insoluble protein-dye conjugate, has been widely used for the assay of proteolytic enzymes (l-7). Azocoll is hydrolyzed readily by a variety of proteinases and yields soluble, colored peptides in proportion to enzyme concentrations at fixed incubation times (for example, Biologics Technical Information Dot. No. 3805-482, Calbiochem-Behring Corp., Refs. (4,8)). However, Cabib and Ulane (9) reported that azocoll was hydrolyzed at an increasing rate as a function of incubation time when either yeast chitin synthetase activating factor or trypsin was assayed. We have observed a similar nonlinear time course for the hydrolysis of azocoll by extracts of Bacillus subtilis. We show here that the nonlinearity is due to inhibitory material present in commercial preparations, examine the effect of agitation on rates of reaction, and describe a means of obtaining linear assays as a function of time. ’ To whom correspondence should be addressed. 0003-2697184 $3.00 Copyright Q 1984 by Academic Ress. Inc. All rights of reproduction in any form reserved.
MATERIALS
AND METHODS
Azocoll was purchased from CalbiochemBehring Corporation and from Sigma Chemical Company. Trizma base, subtilisin BPN’ (Type VII) and CaClz were purchased from Sigma. All other chemicals used were of reagent grade. B. subtilis 168 (trp-) was grown in supplemented nutrient broth and harvested 6 h after the end of logarithmic growth (10). Cells were washed, extracted, and centrifuged as previously described (10) except that 1 M KC1 was included in the wash buffers to minimize contamination of extracts with extracellular proteases. The clear, cell-free supernatant fraction was used in assays without dialysis or further treatment. Standard assays for azocoll hydrolysis were carried out as follows. Azocoll (0.250 g) was suspended in 50 ml of buffer (0.05 M TrisHCl, 1 mM CaClz, pH 7.8) and stirred rapidly with a magnetic stirring bar apparatus in a
446
AZOCOLL
HYDROLYSIS
447
loo-ml beaker. (A suspension of 5 mg/ml of termined spectrophotometrically at 278 nm azocoll is similar to levels used by others pre- assuming EtSb,, = 11.7 (12). viously.) Stirring was rapid enough to give what appeared to be a uniform suspension. RESULTS Portions of 1 to 3.0 ml were rapidly removed using an open-end plastic syringe (1 ml) and Azocoll which has been completely digested transferred to glass assay tubes (13 mm di- by subtilisin has absorption maxima at 5 16 ameter). Smaller tubes were unsatisfactory as and 415 nm (relative absorptivities of 1.2: 1) at pH 7.8 in 50 mM Tris-HCl. In this study, the azocoll clung to the sides of the tubes upon however, we have measured absorbances at rapid agitation. Tubes were placed vertically 520 nm because all previous work by others in a rack held in a rotary water bath shaker (New Brunswick Scientific Model G-86) at has been at this wavelength. We found that 1 37°C. The tubes were allowed to prewarm for mg/ml of azocoll yielded an ASu, of 0.593 when 15 min before initiating the reaction by ad- completely digested, it was previously reported dition of enzyme (20-40 ~1). Reactions were that a 1-mg/ml suspension of azocoll yielded stopped by rapidly (5-10 s) drawing the so- an ASzo of 0.464 (14). Nearly the same speclution into a Pasteur pipet (2 ml) fitted with trum was observed in 0.3 M trichloroacetic a small pack of glass wool at the neck. The acid (X,, , 520 nm; L&,, 0.491). clear filtrate, freed from the insoluble azocoll, In studying levels of intracellular proteinase in B. subtilis 168 during the course of growth was then removed to a clean tube by means of a second Pasteur pipet. Control tubes, con- and sporulation, it was necessary to vary the taining azocoll but no enzyme, were incubated times of incubation of extracts with azocoll for the same length of time as the test samples. because of the wide variation in levels of proAbsorbances of filtrates were read against teinase (10). Although rates of hydrolysis of azocoll were linear with the concentrations of buffer on a Beckman Model 24 spectrophotometer fitted with a rapid sampler. Average cell-free extracts used, rates of hydrolysis were absorbances (A& from at least 3 control fil- not linear as a function of time. A very distinct trates were subtracted from the absorbances lag phase was apparent (Fig. 1). We thought Q&,) of test filtrates. Absorption spectra were that this lag might be a result of the presence taken on a Perkin-Elmer 320 spectrophotomof more than one proteinase in the cell-free eter. Assays with cell-free extracts only, were extracts of B. subtilis (13). To eliminate this carried out in Beckman microfuge tubes (2 possibility the experiment described in Fig. 1 ml). Assays were stopped by immersing tubes was repeated using a commercial preparation in an ice-water bath. Chilled tubes were then of subtilisin BPN’; again, the lag was observed centrifuged in a Microfuge B (Beckman In(T. J. Burnett, Ph.D. dissertation, NMSU, strument Co.) centrifuge, and absorbance at 1983). This suggested that some inherent 520 nm of the supematant fraction was meaproperty of the azocoll preparation was resured as above. sponsible for the aberrant time course. To test Protein concentrations of cell-free extracts the idea that the lag might be caused by inof B. subtilis were measured by the method hibitory, soluble peptides present in the azoof Bradford (11) using the Bio-Rad Protein collagen, a portion of azocoll was washed by Assay Kit I (Bio-Rad Corp.) with an immupreincubating it in the assay buffer for 90 min. noglobulin standard. Subtilisin was prepared It was filtered and resuspended in fresh buffer for use by weighing out 1 mg/ml of subtilisin immediately before using in an assay. When BPN’ into buffer (0.05 M Na-acetate, pH 6.0, the assay tubes were not shaken at any time 1 mM CaCl*) and dialyzing for 10 h against during the assay, prewashing the azocoll rethe same buffer with 3 changes of buffer. The sulted in a linear release of colored soluble concentration of subtilisin in assays was de- material as a function of time as shown in
CHAVIRA,
BURNETT,
Ttme(hr)
FIG. I. Hydrolysis of azocoll by cell-free extracts of B. subtilis as a function of time. Each assaytube contained 2.5 mg of azocoll in 0.5 ml of 50 mM Trir+HCl, 1.0 mM CaClr, pH 7.8. The reaction was started by adding cellfree extract from B. subtilis (Materials and Methods) and placing in a water bath at 37°C. Tubes were agitated in a vortex mixer every 15 min. Assays were stopped by placing tubes on ice for 3 min and centrifuging in a Beckman microfuge B for 2 min. Absorbances at 520 nm were read against control tubes containing no cell-free extract. Assays contained 0.078 mg (0) or 0.195 mg (A) of cellfree extract of B. subtilis. Values reported are averages of triplicate measurements which have been corrected for corresponding control tubes containing no extract; standard deviations did not exceed 10% of the mean values.
Fig. 2. Apparently, material causing the lag period in the time-course curve was eliminated by prewashing the azocoll. In reviewing the literature, we found that workers have used different rates of agitation during the incubation of azocoll with proteinase; no shaking (1,9), gentle shaking (15), and vigorous shaking (8) of incubation tubes have been used. None of the workers specified the precise conditions or rates of agitation. To determine the effect of rates of agitation on the initial rates of hydrolysis of prewashed azocoll, we carried out the experiment described in Fig. 3. The rate of agitation clearly had a dramatic effect on the rate of the reaction. Rates of shaking greater than about 330 rpm caused little increase in the rates of reaction. We found also that it was essential to use a tube of sufficient diameter to prevent azocoll from clinging to the sides of the tube. We next tested the idea that soluble pep tides, or other materials in the azocoll, were inducing the lag in the time-course curve. Concentrated portions of filtrate obtained
AND HAGEMAN
from prewashing of azocoll as described above were added back to prewashed azocoll, and the time course of hydrolysis of these samples by subtilisin was compared to that of washed azocoll suspended in fresh buffer (Fig. 4). These assays were carried out with vigorous shaking, but a similar effect was observed when no shaking was used (data not shown). In initial experiments, when assay tubes were not shaken, the extent of the lag period varied from one batch of azocoll to the next. The data of Table 1 show that the background absorbances of washes from different commercial preparations of azocoll do vary considerably. Such variations in materials released upon incubation in buffer might account for the variations observed in the extent of the lag period.
OO
60
120
1
Time(mln)
FIG. 2. Effect of prewashing azocoll on its rate of hydrolysis by B. subtilisin. Azocoll(O.525 g) was suspended in 105 ml of buffer (0.05 M Tris-HCl, 1.O mM CaClr , pH 7.8) and allowed to stand without shaking at 37’C for 90 min. The suspension was Iiltered through Whatman No. 1 filter paper and the filtrate discarded. The precipitate was resuspended in 105 ml fresh buffer and stirred with a magnetic stirrer to produce a uniform suspension. Portions (3.0 ml) were transferred into test tubes (13 mm diam) and allowed to prewarm for 15 min at 37°C. The assay was initiated by addition of 0.054 mg (120 ~1) of subtilisn BPN’. Reactions were stopped by filtration as described under Materials and Methods. Control tubes containing no enzyme were run in parallel. Absorbances of tiltrates from sample and control tubes were read against buffer. Values for unwashed (0) and prewashed (0) axocoll are averages of duplicates which were corrected by subtracting values from controls.
AZGCOLL
449
HYDROLYSIS TABLE RELEASE OF MATERIAL BATCHES OF AXEOLL
1
ABSORBING AT 520 nm FROM IN ABSENCE OF PROTEINASE
Source of Azocoll Calbiochem Sigma No. Sigma No. Sigma No. Sigma No. 0
20 Tlme(min)
40
60
No. 1 1 2 3 4
A520“ll
h
0.037 0.095 0.055 0.025 0.05 1
LIAzocoll was suspended (5 mg/ml) in buffer (0.05 M Tris-HCl, pH 7.8), with mM CaC12. After 1.0 h at 37°C sample was rapidly filtered through glasswool; absorbance of filtrate was read against buffer.
FIG. 3. Effect of rate of agitation on hydrolysis of prewashed azocoll by subtilisin. Washing of azocoll and assays of azocoll hydrolysis were as described for Fig. 2 except 1.Oml assayvolumes were used. Test tubes ( 13 mm diam) scribed above was used as the substrate. We were placed vertically in a rack held in a rotary bath thought this might be due to variations in reshaker (New Brunswick Scientific Model G-86) at 37°C. sidual colored soluble material associated with The rates of agitation were (0) 100 rpm, (A) 250rpm, the variations in batches of azocoll (Table 1). (0) 330 t-pm. Dotted lines are data from an independent experiment; the upper and lower lines were for tubes To attempt to totally eliminate the lag period shaken at 400 and 0 rpm, respectively. When the lag we prewashed the azocoll for 2 h using vigobserved in this experiment was eliminated (see Fig. 5) orous shaking. When this procedure was folrates of 330 and 400 rpm were the same.
Finally, in some experiments, a slight lag was seen (Fig. 3) even when azocoll that had been prewashed by the soaking procedure de-
lowed linear rates of hydrolysis were obtained (Fig. 5). The background absorbances of control tubes did not change the time and had an average value of only 0.004 units at 520 nm (compare with values in Table 1). DISCUSSION
T~mdmln)
FIG. 4. Effect of readdition of concentrated, soluble wash from azocoll on rate of hydrolysis of prewashed azocoll. Azocoll was washed and assayed as described in Fig. 2 except tubes (13 mm diam) were shaken at 400 rpm. In one case (0) the washed azocoll is resuspended in fresh assay buffer and in the other (A) washed azocoll was suspended in the wash obtained from incubating 0.375 g azocoll in 25 ml of buffer (3 times concentration used in standard assay) for 90 min without shaking.
Using insoluble substrates such as azocoll to assay enzymatic activity has two major disadvantages. First, distributing equal amounts of substrate to assay tubes requires either individual weighings for each tube or rapid distribution of a liquid suspension. In this regard, the finer mesh azocoll (100-250 mesh) sold by Calbiochem-Behring Corporation can be dispensed more uniformly in suspensions than coarser mesh products. Second, because of the nonhomogeneity of the solution, thermal motion is not sufficient to insure maximal contact of the enzyme with the substrate; thus, the degree and nature of agitation and assay tube geometry are variables which should be specified in order to reproducibly perform an assay. As Fig. 3 shows, the rate of reaction in optimally shaken assay tubes increased threefold over the rate in unshaken tubes.
450
CHAVIRA,
BURNETT,
AND HAGEMAN
that during digestion of these soluble peptides a fraction of the enzyme is active in a reaction that does not contribute to an increase in soluble colored products. After the initial lag, rates of reaction became comparable (Fig. 2). In any case, sufficiently vigorous prewashing of the azocoll yielded a substrate that gave linear kinetics with respect to time whether or not the assay tubes were agitated during the assay. 0
0
50 T~me(md
100
150
FIG. 5. Effect of extensive prewashing of azocoll on linearity of azocoll hydrolysis with time. Prewashing of azocoll was performed by suspending 0.3 g of azocoll in 60 ml of standard buffer and shaking at 400 rpm in a IOO-ml round-bottom flask for 2.0 hat 37°C and filtering. The washed azocoll was resuspended in 60 ml of fresh buffer and azocoll assayswere performed as described in Fig. 2 but using I .O-ml volumes and agitation of 400 rpm (A) or no agitation (0).
The color yield we observed for the azocoll was similar to the value previously reported ( 14) and probably varies from one lot of azoco11to another. When unwashed commercial azocoll was incubated in buffer alone, colored products were released (Table 1) in a continuous but nonlinear manner as a function of time (data not shown). However, the apparent rates of hydrolysis of unwashed azocoll by several proteases (this work and Ref. (9)) increased as a function of time even after corrections for the background absorbances of controls from corresponding incubation times were made. We suggest that the lag periods observed are due to soluble dye-peptide species associated with the azocoll (Table 1) which act as competitive inhibitors towards subtilisin and presumably other proteinases. Figures 2 and 4 show that the inhibitory effect is transient, which might suggest that the protease digests these soluble species to products which can no longer combine with the enzyme. The apparent inhibition might result from the fact
ACKNOWLEDGMENTS This work was supported in part by USPHS RR08 136 2nd by GM 19643 which are gratefully acknowledged.
REFERENCES 1. Oakley, C. L., Warrack, G. H., and van Heyningen, W. E. (1946) J. Pathol. Bacterial. 58, 229-235. 2. Rinderknecht, H., Geokas, M. C., Silverman, P., and Haverback, B. J. (1968) Clin. Chim. Acta 21,197203. 3. Millet, J. (1971) C. R. Acud. Sci. Paris 272, 18061809. 4. Orrego, C., Kerjan, P., Manta de Nadra, M. C., Szulmajster, J. (1973) J. Bacterial. 116, 636-647. 5. Cheng, Y-S. E., and Aronson, A. I. (1977) Proc. Nat. Acad. Sci. USA 14, 1254-1258. 6. Studebaker, J. F. (1979) J. Chromatogr. 185, 497503. 7. Parrish, R. F., Goodpasture, J. C., Zaneveld, L. J. D., and Polakoski, K. L. (1979) J. Reprod. Fertil. 57, 239-243. 8. Cheng, Y-S. E., and Aronson, A. I. (1977) Arch. Microbiol. 115, 61-66. 9. Cabib, E., and Ulane, R. (1973) Biochem. Biophys. Res. Commun. 50, 186-191. 10. Geele, G., Garrett, E., and Hageman, J. H. (1975) in Spores VI (P. Gerhardt, R. N. Costilow, and H. L. Sadoff, eds.), pp. 391-396. Amer. Sot. Microbiol., Washington, D. C. 11. Bradford, M. (1976) Anal. Biochem. 72,248-254. 12. Matsubura, H., Kasper, C. B., Brown, D. M., and Smith, E. L. (1965) J. Biol. Chem. 240, 11251130. 13. Srivistava, Om. P., and Aronson, A. I. (1981) Arch. Microbial. 129, 227-232. 14. Kerjan, P., Keryer, E., and Szulmajster, J. (1979) Eur. J. Biochem. 98,353-362. 15. Millet, J., and Gregoire, J. (1979) Biochimie61,385391.