A Continuous Colorimetric Assay for Rhinovirus-14 3C Protease Using Peptidep-Nitroanilides as Substrates

ANALYTICAL BIOCHEMISTRY ARTICLE NO.

252, 238–245 (1997)

AB972315

A Continuous Colorimetric Assay for Rhinovirus-14 3C Protease Using Peptide p-Nitroanilides as Substrates Q. May Wang,1 Robert B. Johnson, Gregory A. Cox, Elcira C. Villarreal, and Richard J. Loncharich Infectious Diseases Research, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46285

Received December 23, 1996

Human rhinovirus encoded 3C protease is an attractive target for antiviral drug development. However, lack of a convenient and selective assay for 3C protease has been a hindrance in characterization of this enzyme and evaluation of a large number of potential inhibitors. In the present study we describe development of a simple, continuous colorimetric assay for this enzyme using peptide p-nitroanilides (pNA) as substrates. Several peptides mimicking the native 3C cleavage site of HRV-14 polyprotein have been synthesized with an N-acylated p-nitroaniline at position P1* and examined as substrates for the purified 3C protease. In these peptides, amino acids downstream from the original cleavage site have all been replaced with a chromophoric p-nitroaniline moiety which is directly linked to the bond undergoing enzymatic cleavage, thereby generating a new cleavage site Gln-pNA for the enzyme. Hydrolysis of these pNA peptides by 3C at the newly formed scissile bond releases free p-nitroaniline which is yellow-colored and can be continuously monitored at a visible wavelength. Kinetic parameters of 3C protease toward these peptides have been measured and analyzed. In addition, the pNA peptides have been modeled within the active site of the 3C protease to investigate the ability of the pNA group to act as a replacement for Gly-Pro in the prime side. The selectivity and applicability of this assay and its advantages over the previously described methods have been demonstrated and discussed. Since multiple tests can be performed simultaneously in one microtiter plate, the assay is ideal for evaluation of a large number of samples. q 1997 Academic Press

Belonging to the picornavirus family, human rhinoviruses (HRVs)2 contain more than a hundred distinct 1 To whom correspondence should be addressed. Fax: 317-2761743. E-mail: [email protected]. 2 Abbreviations used: HRVs, human rhinoviruses; pNA, p-nitroaniline; TFA, trifluoroacetic acid.

serotypes and are the major etiological agents of the common cold in humans (1). These small, plus-strand RNA viruses express their genetic information into a single large viral polyprotein which is subsequently cleaved by virally encoded proteases to generate the mature viral enzymes and structural proteins (2, 3). The first cleavage of the viral polyprotein is processed cotranslationally by the 2A protease, while the majority of the maturation cleavages are performed by the 3C protease (2, 3). As a cysteine protease, viral 3C protein contains a cysteine as a nucleophile at its active site. However, it is structurally more related to the chymotrypsin family of serine proteases rather than typical cysteine proteases (4, 5). Evidence has accumulated that the viral 3C protease is not only an important protease, but also an RNA-binding protein implicated in both viral protein maturation and viral replication (3). Therefore, HRV-14 3C protease has been viewed as an ideal target for antiviral drug development due to its important roles in viral infection as well as its unique protein structure which is distinct from any other known cellular proteases. Various methods for measurement of 3C protease activity in vitro have been described. Detection of 3C protease activity using intact viral polyprotein or engineered viral polyprotein fragments is specific but less quantitative (6). HPLC assays utilizing various synthetic peptides as substrates have been widely used for the enzyme because they are quantitative and feasible for enzyme kinetic studies (7–9). However, analysis of the cleavage products by HPLC is usually time-consuming and inappropriate for large numbers of samples. A discontinuous colorimetric assay based on the reaction between 2,4,6-trinitrobenzenesulfonic acid and primary amines has been developed for 3C proteases of hepatitis A virus and HRV-14 (10, 11). The significant overlap of absorbance spectra of the unreacted reagent and the trinitrophenyl products makes the product quantitation very tedious. Recently, a magnetic bead assay for HRV-14 3C protease has been de-

238

AID

0003-2697/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AB 2315

/

6m45$$$301

09-12-97 07:58:25

aba

239

RHINOVIRUS 3C PROTEASE ASSAY

scribed in which an N-terminal radiolabeled peptide substrate is linked to magnetic beads through its Cterminus (12). This assay displays a discontinuous pattern and preparation of the radiolabled substrate is required. The only continuous assay developed in recent years for the HRV-14 3C enzyme is a fluorescencebased method (13). This assay was found to have relatively high background due to the inefficient quenching of the fluorescence donor and quencher. To better characterize viral 3C enzyme and explore novel inhibitors of 3C protease, an accurate, sensitive, and convenient assay for the protease suitable for screening a large number of compounds is desired. In this report, we document the development of a continuous colorimetric assay for HRV-14 3C protease using synthetic peptidyl p-nitroanilines as substrates. Several potential substrates with an N-acylated pNA moiety at P1* were synthesized and evaluated. The ability of the pNA group to act as a replacement for the GlyPro residues at the prime side is examined using structure-based ligand design techniques (18). Characterization, selectivity, and applicability of this assay as well as kinetic parameters of the enzyme toward the pNA peptides are demonstrated and discussed. MATERIALS AND METHODS

Materials. Recombinant HRV-14 3C protease was purified using the procedures published previously (13). The enzyme was further dialyzed against 25 mM Hepes, pH 7.5, 150 mM NaCl, 6 mM dithiothreitol, and 25% glycerol overnight and then stored at 0807C. Peptides were designed using the native 2C/3A cleavage site of the viral polyprotein and were custom synthesized by American Peptide Co. (CA, USA). Peptide sequences were all confirmed by both amino acid sequence analysis and mass spectrometry. Protease inhibitors were purchased from Sigma. Protease assay. A typical 3C protease assay was performed at 307C for the time indicated in a 200-ml reaction mix containing 25 mM Hepes, pH 7.5, 150 mM NaCl, 1 mM EDTA, 6 mM dithiothreitol, 250 mM pNA peptide substrate, and 3C protease at 0.4 mM. The reaction was started by the addition of either substrate or 3C protease. Cleavage of pNA peptides between glutamine at P1 and pNA at P1* by the protease releases yellow-colored free pNA, whose absorbance can be measured at a visible wavelength (405 nm) against a blank where either substrate or enzyme is not included in the reaction mix. All enzyme reactions were directly performed in microtiter plate wells and monitored using a temperature-controlled microplate reader (Molecular Devices). The determination of kinetic parameters for each substrate was performed as above, except 0.1 mM 3C protease and 50–150 mM peptide were included in the reaction. Initial velocity was measured at the

AID

AB 2315

/

6m45$$$302

09-12-97 07:58:25

TABLE 1

Cleavage Efficiency of pNA Peptides by HRV-14 3C Protease Peptide code

Sequence

pNA1 pNA2 pNA3 pNA4 pNA5 pNA6

T-L-F-Q--pNA E-T-L-F-Q--pNA D-S-L-E-T-L-F-Q--pNA E-A-L-F-Q--pNA E-V-L-F-Q--pNA D-S-L-E-V-L-F-Q--pNA

kcat /Km (M01 s01) 40.0 230 465 840 920 2365

{ { { { { {

5.3 5 8 11 10 17

(kcat /Km)rel 1.0 5.8 11.6 21.0 23.0 59.0

Note. The purified 3C protease (0.1 mM) and pNA peptide substrate at 50–150 mM were incubated under the conditions described. Reactions were monitored by absorbance at 405 nm. kcat /Km values were determined by plotting the initial reaction rates versus substrate concentrations. All 3C protease present was assumed to be active. Amino acid sequences of pNA peptides are given in single-letter amino acid code; no side chains or N-terminal groups are blocked; underlined letters represent changed amino acids; cleavage efficiency is expressed as (kcat /Km)rel using pNA1 as reference peptide.

time points when the cleavage reaction was proceeding in a linear fashion (less than 10% substrate depletion). kcat/Km values were then determined by plotting the initial reaction rates versus variable substrate concentrations. Activities shown are averages of two measurements. HPLC analysis of protease reaction. 3C protease reactions were performed under the conditions described above. Reactions (100 ml) were stopped at the time indicated by addition of 10% acetonitrile/0.1% TFA (400 ml) and 50 ml was then injected onto a Shimadzu SCL-10A HPLC. The reaction was eluted from a C18 column (4.6 1 250 mm) at 1 ml/min with a linear gradient of 10– 60% acetonitrile/0.1% TFA in 20 min. Peaks were monitored at 214 nm. Since one of the expected cleavage products would be p-nitroaniline, free p-nitroaniline (Aldrich) was used as a standard and its HPLC profile was determined under the conditions as described above. Cells. Confluent monolayers of HeLa cells were infected with HRV-14 at a multiplicity of infection of 100 pfu (plaque-forming unit) per cell. The control consisted of mock-infected cells. After an adsorption period of 30 min at room temperature, the cells were incubated for 8 h at 357C with 5% CO2 and then scraped and harvested. After centrifugation at 1000g for 5 min, the cell pellets were collected, washed twice with PBS buffer, freezethawed, and then lysed by sonication for 10 min. The pNA peptide cleavage activity present in the supernatant was measured using the standard protease assay protocol described above. To prepare bacterial cell samples, Escherichia coli cells (BL21-DE3) transformed with an HRV-14 3C protease expression vector or untransformed control cells were grown in 100 ml tryp-

aba

240

WANG ET AL.

FIG. 1. Cleavage of peptide p-nitroanilides by 3C protease. (A) Progress curve of pNA peptide cleavage by 3C protease. Reaction in duplicate for each substrate at 250 mM was performed with 0.2 mM 3C protease as described in the text. Absorbance at 405 nm was collected at the indicated time points (n, pNA1; h, pNA2; ,, pNA3; s, pNA4). Reading from the reaction containing no substrate was taken as blank. Time course of the reaction with pNA4 peptide but not the enzyme is also shown (l). (B) Enzyme concentration-dependent cleavage of pNA4 by 3C protease. Cleavage reactions with 3C enzyme at the specified concentrations toward 250 mM pNA4 (h) or substrate solvent (s) were performed for 30 min at 307C as described. A405 nm values were taken against a blank in which 3C protease was omitted.

tone yeast media supplemented with 25 mg/ml chloramphenicol and 50 mg/ml ampicillin until the OD600 reached 0.6. The cells were induced by 0.4 mM isopropyl-1-thio-b-D-galactopyranoside at 287C for 3 h and then lysed. An aliquot of the soluble fraction of the transformed or control cell lysates was then assayed for pNA peptide cleavage activity as described above. Computational. All force field calculations were performed with the CHARMM (Chemistry at HARvard Molecular Mechanics) program (19) on a Silicon Graphics workstation. Random conformations of the substrate bound to rhinovirus 3C protease were generated, and subsequently energy was minimized within the active-site environment of the protease. Subsequently, the complexes were subjected to repeated short-term molecular dynamics and energy minimization. The resulting structures were energy minimized to zero energy gradient and then analyzed. To gain a better understanding of key interactions present in the rhinovirus 3C protease substrate complexes, we performed a hydrogen bonding analysis of the low-energy structures. Hydrogen bond strengths were estimated using the potential function in the CHARMM program (19) and the QUANTA program version 4.1.1 parameter set (Molecular Simulations, Inc.). Other methods. Protein concentration was determined by the method of Bradford (14) using bovine serum albumin as standard. Percentage of the 3C protein (Ç20 kDa) present in the preparation was deter-

AID

AB 2315

/

6m45$$$302

09-12-97 07:58:25

mined by densitometry analysis. All 3C protein present was assumed to be active. Electrospray ionization– mass spectrometry was conducted using a PESciex API III triple-stage quadrupole mass spectrometer. The instrument was operated in the positive-ion detection mode with an IonSpray voltage of 3500 V. Samples were diluted 1:1 with 1% acetic acid in acetonitrile and were continuously infused into the interface at a rate of 10–20 ml/min using a syringe pump. Spectra were collected over a range of 300–2000 U at 0.1-U intervals with a dwell time of 1 ms per interval. A total of 5–10 scans were averaged to yield the final spectrum. RESULTS

Based on the previous observation that synthetic peptides representing the 2C/3A cleavage site of HRV14 polyprotein can be hydrolyzed by purified 3C protease in vitro (9), a pNA peptide (pNA1), TLFQ-pNA, mimicking the above peptide was designed for the development of a continuous colorimetric assay using pnitroaniline as chromophore. In this peptide, the signal molecule p-nitroaniline was directly linked to the bond undergoing enzymatic cleavage to create a new recognition site of Gln–pNA for the protease (Table 1). Amino acids downstream from the original 2C/3A recognition site were removed in order to generate the chromophoric free amine (p-nitroaniline) from the pNA peptide upon enzymatic hydrolysis. When this pNA peptide was incubated with the purified 3C protease under the

aba

RHINOVIRUS 3C PROTEASE ASSAY

FIG. 2. HPLC analysis of 3C cleavage reaction with pNA4. Cleavage of pNA4 by 3C enzyme was performed at 307C for 0 min (A) and 60 min (B) as described under Materials and Methods. The reaction was stopped with 10% acetonitrile/0.1% TFA in a ratio of 1:4. Fifty microliters of the terminated reaction mix was injected and the eluted peaks were monitored at 214 nm on HPLC as described. As a standard, free p-nitroaniline (C) was dissolved in the same buffer (0.5 mM) and detected in the same way.

conditions described, the reaction mix became absorbent at 405 nm, a visible wavelength at which p-nitroaniline shows strong absorbance. This finding indicated that pNA1 could be recognized and cleaved at the Gln–pNA bond by the enzyme. Encouraged by this initial result, we designed another five similar pNA peptides in order to optimize this method by selecting a better substrate for 3C and to define the minimal amino acid sequence required for an efficient cleavage by the enzyme. These peptides, with five or eight amino acids upstream of the cleavage bond, were all derived from the original 2C/3A recognition site and their amino acid sequences are shown in Table 1. Alanine or valine was introduced in P4 because 3C protease prefers nonpolar amino acids with small side chains at this position (4, 9). Examination of 3C protease reaction toward these

AID

AB 2315

/

6m45$$$303

09-12-97 07:58:25

241

pNA peptides revealed that all these peptides could be cleaved by the enzyme, although the initial reaction rates were quite different. For each of these substrates, kcat/Km was obtained from the slope of a plot of initial rate versus substrate concentrations. The results are shown in Table 1. The cleavage efficiency data, expressed in (kcat/Km)rel using pNA1 as reference substrate, revealed that pNA2–pNA6 were all better than the first peptide made for this assay (Table 1). Cleavage of pNA peptides by 3C was time-dependent and a linear increase of absorbance at 405 nm could be seen in the first 20–30 min under the conditions described (Figs. 1A and 1B). The purified enzyme cleaved these peptides in a rank order pNA4 ú pNA3 ú pNA2 @ pNA1, confirming the kinetic data. Peptides pNA5 and pNA6 were not included in this figure due to their insolubility at 250 mM, the substrate concentration tested. Further evaluation revealed that both pNA5 and pNA6 were not soluble at a concentration of Ç150 mM or higher in the typical reaction mix, although they were structurally similar to the others except for the presence of a valine at P4. Therefore, pNA4 was chosen for further characterization of the assay based on its reasonable solubility and efficient cleavage by 3C protease. To verify that 3C cleavage of pNA peptides occurred at the expected scissile bond Gln–pNA, analyses of reaction products by reverse-phase HPLC were performed. As seen in Fig. 2A, peptide pNA4 (purity over 95%) showed as a single peak with a retention time of 19.77 min under the conditions employed. Two cleavage products, with retention times of 13.86 and 14.65 min, respectively, were generated from the 3C cleavage reaction along with a decrease of the original substrate peak at 19.77 min (Fig. 2B). Since one of the products was expected to be p-nitroaniline, the elution profile of the cleavage reaction mix was compared with that of the standard free p-nitroaniline. As illustrated in Fig. 2C, standard p-nitroaniline eluted at 13.86 min, the same position as observed for one of the products. The presence of free pNA as the final product in the reaction mix indicated that the enzyme did hydrolyze the peptide at the expected Gln–pNA bond. To confirm that the peak eluted at 14.65 min was the P5–P1 peptide product, we performed mass spectrometry analysis of the pNA4 cleavage reaction mix since the P5–P1 peptide standard was not available. Mass spectrometry examinations indicated the presence of the expected P5–P1 peptide product in the reaction mix, with its size identical to the calculated one (data not shown), while the same product could not be detected in the control sample without enzyme. Similar HPLC and mass spectrometry data were obtained using pNA3 as substrate (not shown). To further support that 3C cleavage of pNA peptides occurs at the Gln–pNA scissile bond and that the active site is able to accommodate the pNA C-terminal group,

aba

242

WANG ET AL.

FIG. 3. Model of a bound pNA4 substrate. Side-by-side stereo view of the pNA4 peptide (white) modeled into the active site of the 3C protease (red). Amino acid residues of the enzyme are labeled in green.

we sought to evaluate the proposed substrates using structure-based ligand design techniques (18). Specifically, force field calculations of substrates complexed with 3C protease were performed to investigate the ability of the pNA group to act as a replacement for Gly-Pro and to examine likely protein–substrate interactions. The calculations were based on NOE distances from protein NMR studies (A. D. Kline, unpublished results) of the peptide aldehyde H-Phe-Gln-Gly-H complexed with the 3C protease (4). An example of a designed substrate, H-Glu-Ala-Leu-Phe-Gln-pNA (pNA4), from our calculations is shown in Fig. 3. The modeled substrate binds in an elongated channel formed by the two six-stranded b barrels (4). The P1* and P2* positions provide ample space for the pNA C-terminal group to bind. The main chain atoms of the P1 (Gln)– pNA moiety are folded so that the carbonyl group hydrogen bonds to the oxyanion hole. The hydrogen bond strengths of the P1 carbonyl group with the amide protons of Gly-144, Gln-145, and Cys-146 are 01.3, 01.7,

AID

AB 2315

/

6m45$$$303

09-12-97 07:58:25

and 01.8 kcal/mol, respectively, indicating strong interaction and tight fit into the active site. The amide proton of pNA4 is oriented in a standard anticonformation relative to the carbonyl oxygen, which directs the aromatic group into the P1* and P2* positions. There is substantial interaction of the aromatic ring with the aromatic ring of Phe-25 as demonstrated by a favorable 01.8 kcal/mol interaction energy. The favorable nature of the interaction is due to the ability of S1* and S2* sites of the protein to accommodate the large pNA Cterminal group. The pNA4 cleavage reaction of 3C could be fully stopped by addition of acids such as 0.5–1% acetic acid and phosphoric acid (final concentration, v/v) at any time during the reaction period without affecting pnitroaniline absorbance at 405 nm (not shown). After termination of the reaction with acid, no acid-catalyzed hydrolysis of the pNA substrates was observed within at least 6 h. The cleavage reaction of pNA4 displayed an enzyme concentration-dependent manner as illus-

aba

RHINOVIRUS 3C PROTEASE ASSAY TABLE 2

Effects of Protease Inhibitors on HRV-14 3C Activity

Inhibitor

Inhibitor type

Highest concentration tested

EDTA EGTA Egg white cystatin E-64 Iodoacetamide Pepstatin Aprotinin Benzamidine Leupeptin PMSF TLCK

Metalloprotease Metalloprotease Cys protease Cys protease Cys protease Asp protease Ser protease Ser protease Ser/Cys protease Ser/Cys protease Ser/Cys protease

50 mM 50 mM 8 mM 100 mM 10 mM 20 mM 15 mM 50 mM 1 mM 10 mM 1 mM

IC50a (mM) NI b NI NI NI 1.0 { 0.1 NI NI NI 0.75 { 0.05 8.0 { 0.2 ú1.0

Note. Viral 3C enzyme at 0.4 mM and pNA4 peptide at 250 mM were incubated at 307C for 30 min with the indicated protease inhibitor or the solvent in which it dissolved. Cleavage reactions with or without inhibitors were monitored by A405 nm . a IC50 : these values represent the inhibitor concentration required to reduce the protease activity by 50% compared to the control group containing no inhibitor. At least six different concentrations of each inhibitor were examined to generate IC50 values. Shown is the average of two determinations. b NI: no inhibition was found at the highest concentration examined.

trated in Fig. 1B, while reactions with enzyme or substrate alone did not bring significant absorbance change at 405 nm (Fig. 1), indicating that the substrate itself was stable under the conditions employed and the hydrolysis of the Gln–pNA bond was specifically performed by the enzyme. This method could be used for evaluation of 3C inhibitors as shown in Table 2. The results obtained using this colorimetric assay were consistent with those generated using other methods (15). For example, both the pNA colorimetric assay and the previously described HPLC methods (7) showed that E-64, a potent irreversible inhibitor of certain cysteine proteases including papain, was unable to inhibit 3C. Using this assay, we could quantitatively determine IC50 values for the reagents exhibiting inhibitory activities toward 3C protease (Table 2). Moreover, this assay has been routinely used in our laboratory for 3C enzyme inhibition kinetic studies. To see if this colorimetric assay could be used for 3C protease activity measurement in crude cell homogenates, the pNA4 peptide cleavage activity in the bacterial cells transformed with a 3C expression vector was measured and compared with that in the control cells. As shown in Fig. 4A, the pNA4 hydrolysis activity in the transformed cells was more than 15-fold higher than that observed in the control cells, demonstrating that the assay worked specifically toward 3C in the crude systems. Recombinant 3C protease could be puri-

AID

AB 2315

/

6m45$$$303

09-12-97 07:58:25

243

fied from the transformed bacterial cell lysates by assaying the hydrolysis of pNA4 at the Gln–pNA bond and the formation of free p-nitroaniline (not shown). In addition, detection and measurement of 3C protease activity from HeLa cells infected with HRV-14 were successful using pNA4 as substrate (Fig. 4B). DISCUSSION

In this report, we have described a continuous colorimetric assay using synthetic pNA peptide as substrate for detection of rhinovirus 3C protease activity in vitro. A chromophoric molecule p-nitroaniline is placed at the P1* position within the peptides through the formation of an amide bond between the a-carboxyl group of P1 Gln and the amine group of p-nitroaniline. Hydrolysis of the amide bond between Gln and pNA by the protease releases the free p-nitroaniline which differs from its N-acylated derivative pNA substrate in its high absorbance at wavelengths around 400 nm (16). Therefore, the liberation of free p-nitroaniline from the substrates upon enzymatic cleavage can be continuously followed by its absorbance in the visible range. The molar absorptivity or extinction coefficient of free pnitroaniline has been determined as 10,360 M01 cm01 under the conditions described under Materials and Methods. The difference in molar absorptivity between the pNA4 peptide substrate and free p-nitroaniline at 405 nm has been found to be 10,290 M01 cm01. This large difference in molar absorptivity between the substrate and product confers high sensitivity on this method. Use of pNA peptides to detect cellular proteases has been widely described (16). However, despite their general popularity, this approach has not to our knowledge been applied to viral proteases including 3C. This may be partially because viral endoproteases such as 3C have been thought to be highly specific, requiring specific amino acids at the C-terminal side of the scissile bond, especially at P1* (9). It has been proposed that the strong preference for glycine at position P1* is due to main chain conformational requirements for optimal substrate binding (4). Our calculations from the enzyme–substrate model clearly reveal that the elongated channel spanning the P* side does not structurally restrict only glycine at the P1* position, as in this case where the C-terminal group occupies both the P1* and P2* positions (Fig. 3). In addition, our 3C cleavage results clearly demonstrate that HRV-14 3C protease can accept p-nitroaniline at P1* and tolerate it as the only moiety at the P* position as shown by the 3C cleavage reactions (Figs. 1 and 2). Hydrolysis of the octapeptide pNA3 by the enzyme has a kcat/Km value of 465 M01 s01 which is about 3-fold lower than that obtained for the 16-amino-acid peptide (17) containing the identical amino acids on the N-terminal side of the cleavage

aba

244

WANG ET AL.

FIG. 4. Detection of 3C protease activity in crude cell homogenates. (A) 3C activity in bacterial cells. An aliquot (30 ml) of soluble fraction of bacterial cell lysates (0.6 mg/ml) was incubated with 250 mM pNA4 peptide for a typical protease reaction. Absorbance at 405 nm after 10 min incubation was taken. (B) pNA peptide cleavage activity in HeLa cells infected with HRV-14 or control cells. Forty microliters of the soluble fraction from the control or the infected cell lysates was examined using pNA4 as substrate under the condition described. Absorbance at 405 nm after 30 min incubation was taken.

bond. When p-nitroaniline is placed at P1*, a minimum of 5 amino acids upstream from the scissile bond Gln– pNA are required for an efficient cleavage by the enzyme (Table 1). In addition, replacement of threonine with alanine or valine at P4 can result in 3.6- and 4fold increases in kcat/Km values, respectively, consistent with the previous reports using HPLC methods (9). Since the substrate concentration for pNA colorimetric assays is usually set in the millimolar range (16), the 5-amino-acid pNA4 peptide, with a reasonable solubility and kcat/Km value, was chosen for our routine 3C protease activity measurements. Viral proteases are attractive enzyme targets for antiviral chemotherapy; however, lack of convenient, sensitive, and quantitative assays for protease activity measurements has been a hindrance in characterization of these enzymes and search for potential protease inhibitors. The method documented here has advantages over the previously described procedures in its simplicity, rapidity, and high sensitivity. Also, use of radiolabeled materials is not required and the pNA peptides can be easily prepared. The simplicity of this assay and its continuous pattern makes it particularly suitable for screening a large number of samples in order to explore potential protease inhibitors. This method has also been shown to be very useful in the characterization of this enzyme, for monitoring its purification, and for enzyme kinetic studies since it can provide both qualitative and quantitative results. Moreover, our data demonstrate that this assay is selective and can be used for detecting 3C activity present in crude cell homogenates, because only the cleavage

AID

AB 2315

/

6m45$$$303

09-12-97 07:58:25

made at the Gln–pNA bond can generate a detectable signal. This is different from most of the internally quenched fluorogenic peptides with which any cleavage between the fluorescence donor/quencher pair will produce positive results (20). A similar problem also occurs in the discontinuous colorimetric assays using 2,4,6trinitrobenzenesulfonic acid to detect any primary amines formed during peptide hydrolysis (10, 11). Taken together, use of this method can alleviate the shortcomings of the previously described methods and should be helpful for both general characterization studies of 3C protease and antiviral chemotherapy using this enzyme as target. ACKNOWLEDGMENTS We thank John Richardson for performing mass spectrometry, Lou Jungheim and Ron Zimmerman for helpful discussions, and Allen Kline for sharing unpublished results.

REFERENCES 1. Couch, R. B. (1990) in Virology (Fields, B. N., and Knipe, D. M., Eds.), pp. 607–629, Raven Press, New York. 2. Palmenberg, A. C. (1987) J. Cell. Biochem. 33, 191–198. 3. Porter, A. G. (1993) J. Virol. 67, 6917–6921. 4. Matthews, D. A., Smith, W. W., Ferre, R. A., Condon, B., Budahazi, G., Sisson, W., Villafranca, J. E., Janson, C. A., McElroy, H. E., Gribskov, C. L., and Worland, S. (1994) Cell 77, 761–771. 5. Malcolm, B. A. (1995) Protein Sci. 4, 1439–1445. 6. Heinz, B. A., Tang, J., Labus, J. M., Chadwell, F. W., Kaldor, S. W., and Hammond, M. (1996) Antimicrob. Agents Chemother. 40, 267–270.

aba

RHINOVIRUS 3C PROTEASE ASSAY 7. Orr, D. C., Long, A. C., Kay, J., Dunn, B. M., and Cameron, J. M. (1989) J. Gen. Virol. 70, 2931–2942. 8. Long, A. C., Orr, D. C., Cameron, J. M., Dunn, B. M., and Kay, J. (1989) FEBS Lett. 258, 75–78. 9. Cordingley, M. G., Callahan, P. L., Sardana, V. V., Garsky, V. M., and Colonno, R. J. (1990) J. Biol. Chem. 265, 9062–9065. 10. Malcolm, B. A., Chin, S. M., Jewell, D. A., Stratton-Thomas, J. R., Thudium, K. B., Ralston, R., and Rosenberg, S. (1992) Biochemistry 31, 3358–3363. 11. Wang, Q. M., and Johnson, R. B. (1997) Protein Peptide Lett., in press. 12. Wu, Y., and Abeles, R. H. (1995) Anal. Biochem. 229, 143–144. 13. Birch, G. M., Black, T., Malcolm, S. K., Lai, M. T., Zimmerman, R. E., and Jaskunas, S. R. (1995) Protein Expr. Purif. 6, 609– 618. 14. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. 15. Weidner, J. R., and Dunn, B. M. (1990) Arch. Biochem. Biophys. 286, 402–408.

AID

AB 2315

/

6m45$$$304

09-12-97 07:58:25

245

16. Sarath, G., de la Motte, R. S., and Wagner, F. W. (1993) in Proteolytic Enzymes (Beynon, R. J., and Bond, J. S., Eds.), pp. 25– 55, IRL Press, England. 17. Cordingley, M. G., Register, R. B., Callahan, P. L., Garsky, V. M., and Colonno, R. J. (1989) J. Virol. 63, 5037–5045. 18. Holloway, M. K., Wai, J. M., Halgren, T. A., Fitzgerald, P. M. D., Vaccs, J. P., Dorsey, B. D., Levin, R. B., Thompson, W. J., Chen, L. J., deSolms, S. J., Gaffin, N., Ghosh, A. K., Giuiliani, E. A., Graham, S. L., Guare, J. P., Hungate, R. W., Lyle, T. A., Sanders, W. M., Tucker, T. J., Wiggins, M., Wiscount, C. M., Woltersdorf, O. W., Young, S. D., Drake, P. L., and Zugay, J. A. (1995) J. Med. Chem. 38, 305–317 and references therein. 19. Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J., Swaminathan, S., and Karplus, M. (1983) J. Comp. Chem. 4, 187– 217. 20. Holskin, B. P., Bukhtiyarova, M., Dunn, B. M., Baur, P., de Chastonay, J., and Pennington, M. W. (1995) Anal. Biochem. 226, 148–155.

aba