Monoclonal antibodies as probes to detect conformational changes in the rat cysteine proteinase inhibitor cystatin A

JOURNALOF IMMUNOLOGICAL METHODS ELSEVIER

Journal of ImmunologicalMethods 168 (1994) 69-78

Monoclonal antibodies as probes to detect conformational changes in the rat cysteine proteinase inhibitor cystatin A Atsushi Takeda

*'~, A t s u o I w a s a w a b, Y o s h i k o N a k a m u r a Kazuyasu Nakaya c

a, K u m i k o O m a t a

c,

a Department of Clinical Pathology and b Tissue Culture Laboratory, Showa University, Fujigaoka Hospital, Showa University School of Medicine, 1-30 Fujigaoka, Midori-ku, Yokohama-shi, Kanagawa 227, Japan, c Department of Pharmaceutical Sciences, Showa University, Shinagawa-ku, Tokyo 146, Japan

(Received 4 May 1993, revised received 6 July 1993, accepted 10 September 1993)

Abstract Five monoclonal antibodies (MAbs), 77, 114, 138, 175 and 187, were established for rat cystatin A. MAbs 77, 114, 138 and 175 were shown to belong to the IgG1 subclass, whereas MAb 187 was an IgM. These MAbs partially suppressed inhibitory activity of rat cystatin A to papain. Their epitopes were mapped in detail on the molecule by examining the reactivities of the MAbs with NHz-terminally truncated forms and fragments of rat cystatin A by an enzyme-linked immunosorbent assay (ELISA), and by reactivity with the inhibitor on immunoblotting. In competitive binding assays the MAbs did not compete with each other, indicating that the epitopes recognized by these MAbs were substantially different. The conformational epitope recognized by the three MAbs 114, 138 and 175 belonged to one group that was highly sensitive to denaturation, but those epitopes were unchanged by NH2-terminal truncation. MAb 187 was able to recognize a linear epitope present in amino acid residues 15-50 in the NH2-terminal region. MAbs 77 and 114 reacted weakly with mouse cystatin A but not at all with human cystatain A, whereas MAb 187 reacted similarly with mouse cystatin A but at about half that level with human. The MAbs produced in this study should be useful tools for detecting conformational changes in the rat cystatin A molecule. Key words: Cysteine proteinase inhibitor; Cystatin A; Monoclonal antibody

I. Introduction Cystatins are endogenous inhibitors of cysteine proteinases and form a superfamily that is classi-

* Corresponding author. Tel.: 045 971 1151; Fax: 045 973 1019. Abbreviations: MAb, monoclonal antibody; TFA, trifluoroacetic acid; FPLC, fast protein liquid chromatography;CM, carboxylmethylated; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis.

fled into three families (families 1,2 and 3) according to the homology of the amino acid sequences (Barrett et al., 1986a, b; Sali and Turk, 1987). They are characterized by two well-conserved regions in the amino acid sequences; one is a Gly residue in the NH2-terminal segment and the other is Q V V A G or its derivatives near the middle of the cystatin sequences. Recently, X ray crystallographic analyses of chicken egg-white cystatin, and of the complex of recombinant human cystatin B with papain revealed that the

0022-1759/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0022-1759(93)E0244- C

70

A. Takeda et aL /Journal of lmmunological Methods 168 (1994) 69-78

binding site of the inhibitors consists of a hydrophobic edge-shaped structure connecting the conserved regions and the C-terminal ancillary region (Bode et al., 1988; Stubbs et al., 1990). The NH2-terminal segment is like a substrate that interacts with the active site cleft of the papain. There are several lines of evidence that one or more of these conserved regions in the cystatin molecule participate in the inhibition of target proteinases. Previous studies have shown that the NH2-terminal segments of rat and human cystatin A, as well as human cystatins C and S, and chicken egg-white cystatin, were essential for inhibition of cysteine proteinase activity (Takeda et al., 1985; Isemura et al., 1987; Abrahamson et al., 1987). The NH2-terminal segments of human cystatin B and oryzacystatin did not influence the inhibition of proteinase activity (Abe et al., 1988; Thiele et al., 1990). These results suggest that the role of the NH2-terminal segment in the cystatins could differ from one cystatin to another in the inhibition of proteinases, despite their high sequence homology. Also, the structural requirements for inhibition of target proteinases suggests the use of recombinant cystatin amino acid residues having mutations in the conserved QVVAG or related sequences (Nikawa et al., 1989; Jerela et al., 1990; Arai et al., 1991; Genenger et al., 1991), or synthetic inhibitors that include the above region (Teno et al., 1987; Marks et al., 1990; Moreau et al., 1990). Monoclonal antibodies are useful tools for studying structure-function relationships (Ishiguro et al., 1987; Morris et al., 1989; Kitajima et al., 1991) and cellular or subcellular localization (Thor et al., 1984; Lindberg and Rheinwald, 1989; Waseen and lane, 1990). Here we describe the establishment of five mouse-hybridoma cell lines that secrete monoclonal antibodies against rat cystatin A (family 1) and the characterization of their MAbs. These reagents were shown to recognize the conformational epitopes on the surface of the rat cystatin A molecule and the reactivities of two of the MAbs were sensitive to conformational changes induced by NH2-terminal truncation. We have used these MAbs as immunochemical probes to analyze structure-function relationships of rat cystatin A and for comparisons with

the structural properties of rat, mouse and human epidermal cystatin A.

2. Materials and methods

2.1. Materials Papain, Staphylococcus aureus V8 protease and arginylendopeptidase were purchased from Sigma Chemical Co. (St. Louis, MO, USA), Worthington Biochemical (Freehold, N J, USA) and Takara Syuzo (Kyoto, Japan), respectively. Carbobenzoxy-Phe-Arg-methylcoumaryl-7-amide (Z-PheArg-MCA) was from Peptide Institute (Osaka, Japan). Cyanogen bromide (CNBr) was obtained from Tokyo Kasei (Tokyo, Japan). Rat and mouse cystatin A were purified from newborn epidermal extracts according to a method described previously (Takeda et al., 1983), using buffers containing 5% methanol. Human cystatin A was purified from the homogenate of skin scraped from the heel according to a previous method (Takeda et al., 1989). All chemicals used in the present study were of reagent grade.

2.2. Purificatton of N H 2-terminally trunca ted forms and fragments of rat cystatin A A Gly 7 form (Met~-Thr 6 deficient) was isolated by chromatography on a DEAE-cellulose column (1 × 30 cm) from an extract of newborn rat epidermis incubated for 48 h at 37°C in buffer without methanol. Two fractions identified by immunoreactivity with antiserum and inhibitory activity were designated as the Gly 7 form and the native form in order of elution (Takeda et al., 1983). The final preparations were obtained from CM-papain Sepharose column chromatography. An Ala 15 form (Metl-Glu 14 deficient) was prepared by the following procedure: rat cystatin A (8.4 mg/ml) was incubated with V8 protease at an enzyme:protein ratio of 1 : 40 (mol/mol) in 0.1 M ammonium bicarbonate for 6 h at 37°C. The reaction was terminated with (p-amidinophenyl)methanesulfonyl fluoride hydrochloride at an enzyme :inhibitor ratio of 1 : 1 (mol/mol). The reaction mixture was applied to a Bio-gel P-6 gel

A. Takeda et al. /Journal of lmmunological Methods 168 (1994) 69-78

column (1 x 90 cm) and then the column was developed with 10 mM ammonium bicarbonate. Two fractions were detected with an absorbance at 230 nm. The Ala 15 form was passed through the CM-papain Sepharose column and then purified by FPLC on a Mono Q column (PharmaciaLKB, Uppsala, Sweden). The other, N-peptide ( a c M D P G T T G I V G G V S E ) , was purified by HPLC using an ODS-120T column (Tosoh, Tokyo) with a linear gradient of 0-60% acetonitrile in 0.1% TFA. CNBr cleavage of rat cystatin A was performed as follows: rat cystatin A (3 mg) was dissolved in 0.3 ml of 70% formic acid, and seven mg of CNBr were added to the solution. The reaction mixture was incubated with constant stirring overnight at room temperature, diluted with distilled water, and then lyophilized. The dried sample was dissolved in 0.1 ml of 0.1% TFA and chromatographed by HPLC using ODS-120T column (Tosoh, Tokyo) with a linear gradient of 0-60% acetonitrile in 0.1% TFA. Three fractions were detected with an absorbance at 215 nm and were designated as CNBr fragments CB3, CB2 and CB1 in order of elution. The dried CB1 fraction was applied to a CM-papain Sepharose column using distilled water. CB1 (1.2 mg) was incubated with arginylendopeptidase at an enzyme : fragment ratio of 1 : 50 (mol/mol) in 0.1 M ammonium bicarbonate for 20 h at 37°C. The reaction mixture was applied to the ODS-120T column and then the column was chromatographed with a linear gradient of 0-60% acetonitrile in 0.1% TFA. Two fractions, identified by their absorbance at 215 nm, were designated as fragments CB1-R1 and CB1-R2 in order of elution. The NHz-terminally truncated forms and fragments were verified by NH2-terminal amino acid sequencing with an automated gasphase sequencer (Applied Biosystem, Model 470A) and the determination of amino acid composition.

2.3. Preparation of mouse anti-rat cystatin A monoclonal antibodies B A L B / c mice were injected intraperitoneally with 210 /zg of the purified rat cystatin A in

71

complete Freund's adjuvant. A second injection of 100 /zg of the antigen in complete adjuvant was administered 3 weeks later. After 4 weeks the mice with the highest titers were given intraveneous booster injections of 30/xg of antigen in PBS. 3 days after the final immunization, the spleens were aseptically removed, and fusion was performed with mouse myeloma cells (P3-NS-I-1Ag4-1 strain) by the method of K6hler and Milstein (1975). Hybridoma cultures were screened for their ability to produce antibodies that bound to rat cystatin A in ELISA on microtiter plates. Positive hybridomas were subcloned twice by limiting dilution. A hybridoma producing MAb 189 was subcloned two more times by the same method. Clones were either grown for antibody production or frozen in growth medium (D-MEM containing ten-fold diluted fetal bovine serum) containing 10% dimethylsulfoxide and stored in liquid nitrogen. When substantial amounts of antibodies were required the cloned cells were inoculated intraperitoneally in B A L B / c mice pretreated with pristane. The ascitic fluid was collected and antibodies were precipitated using 50% ammonium sulfate. The precipitate was suspended in Dulbecco's phosphate-buffered saline (PBS), and dialyzed extensively against the same buffer. The isotype of individual MAbs was determined using a mouse monoclonal sub-isotyping kit (Bio-Rad laboratories, USA).

2.4. Purification of monoclonal antibodies MAbs typed as IgG class were purified on a protein G-Sepharose 4FF column using the MAb TrapG system (Pharmacia Fine Chemicals). An IgM-typed MAb was applied to a Sephacryl S-200 column (2.5 × 90 cm) equilibrated with 20 mM sodium phosphate buffer containing 0.25 M NaCI. The column was developed with the same buffer, and fractions with reactivity against the antigen were collected. Affinity-purified MAbs were purified on a rat cystatin A-coupled Sepharose 4B column which was prepared by coupling purified rat cystatin A (2 m g/ m l of gel) to CNBr-activated Sepharose 4B (Pharmacia Fine Chemicals). The fraction of MAbs obtained as described above was applied to the column (2 x 5 cm) equilibrated

A. Takeda et al. / Journal of Immunological Methods 168 (1994) 69-78

72

with PBS. The column was washed extensively with the same buffer and antibody was eluted with 0.5 M glycine-HC1 buffer (pH 2.2) containing

Table 1 Properties of monoclonal anti-rat cystatin A antibodies Competing

Isotype

antibody 77 114 138 175 187

(A)

IgG1 IgG1 IgG1 IgG1 IgM

C50 for biotinylated antibodies (/xg) 77

114

138

175

187

50

_ a 0.7 0.6 4.0 -

_ 1.1 0.5 3.2 -

_ 0.9 1.0 0.5 -

_ 141

-

T h e isotype of M A b was determined using a m o u s e monoclonal sub-isotyping kit. A competitive inhibition E L I S A procedure was performed using five anti-rat cystatin A antibodies and their biotinylated antibodies. Micrograms of competing antibody (C50) were estimated as amounts of unlabelled antibody gives 50% inhibition of labelled antibody. a No inhibition.

0.25 M NaC1. The preparations obtained were dialyzed overnight against PBS at 4°C as soon as possible. 2.5. E L I S A

1

2

3

4

5

6

(B) (kDa) "-30

--20 --12

2

3

The reactivity of the MAbs to antigen was measured by an ELISA procedure as follows: the wells of a microtitre plate (Dynatech Immulon, USA) were coated with purified preparations of rat cystatin A, its NH2-terminally truncated forms and fragments (10 ~ g / m l in 0.1 M carbonate buffer, pH 9.6). The plate was sealed and left overnight at 4°C. The contents of the wells were then aspirated and 3% BSA in PBS was added as a blocking agent. Purified antibodies were diluted to 1 /zg/ml in PBS, and added to the plate, followed by incubation for 1 h at 37°C. The plate was then washed three times with 0.05% Tween 20 in PBS (PBS-Tween), and 100 txl of goat anti-mouse Igs coupled with horseradish peroxi-

Fig. 1. Immunoblots of rat cystatin A and its NH2-terminally truncated form. A: rat cystatin A (arrow) and Gly 7 form (arrow head) (4 /~g of each) were separated by 7.5% P A G E . Lane 1 was stained with Coomassie Blue; 2 - 6 were immunoblotted with M A b s 77, 114, 138, 175, and 187, respectively. B: 7 / ~ g aliquots of rat cystatin A were separated by 15% SDS-PAGE. lanes 1, 2, and 3 were immunoblotted with M A b s 77, 114, and 187, respectively. Molecular masses were taken from molecular-mass markers run on the same blots; carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20 kDa) and cytochrome C (12 kDa).

A. Takeda et aL /Journal of lmmunological Methods 168 (1994) 69-78

dase (1/2000 in PBS) were added to the wells. The plate was incubated for 30 min at 37°C, and after removal of the conjugate was again washed with PBS-Tween. The amount of peroxidase per well was determined from its catalytic activity measured by employing 0.04% o-phenylenediamine and 1.82 mM hydrogen peroxide in citratephosphate buffer (pH 5.0). After about 30 min, the reaction was stopped by the addition of 25/~1 of 10 M sulphuric acid. Absorbance values were read photometrically at 490 nm in a microELISA autoreader (Dynatech, USA). All binding data were determined by subtracting the reading obtained using the control MAb Cj-F1 for Campyrobacter jejuni produced by Nakamura et al. (unpublished data). An indirect competitive ELISA was performed using a fixed amount of antibody. Each well received 100 /zl of rat cystatin A (10/xg/ml), and the plate was then incubated overnight at 4°C. Monoclonal antibody conjugated to biotin succimide ester (1 txg/ml final antibody concentration) was added to unlabelled antibodies at various concentrations. The antibody mixture was then incubated for 2 h on the plate previously coated with rat cystatin A. After washing the plate three times with PBS-Tween to remove free antibody, bound biotinylated antibody was estimated using avidin-peroxidase conjugate and ophenylenediamine as described above.

73

blot P membrane using a horizontal electrophoresis apparatus (Atto, Japan). The blots were soaked in a blocking solution (Dainihon Pharmaceutical, (A) NH2

1

30

i

1.Native form I *

~

2.GlyT-form

~!

I*

3. A l a 1 5 - f o r m 4.

CB1

5.

CB2

6.

CB3

I

9o ~3COOH

60

I

B

I '~

BI

l - -

7, CB1-R1 8. CB1-R2 9. N-peptide

(B)

00tl

MAb77

0

2.6. Effect of rnonoclonal antibodies on inhibitory activity of rat cystatin A to papain Rat cystatin A was mixed with serial dilutions of affinity-purified Nl_Abs. After incubation at 37°C for 30 min, inhibitory activity was measured using papain as a target proteinase and Z-PheArg-MCA as a substrate, by the method described previously (Takeda et al., 1989).

2. 7. Other analytical methods The purity of three cystatin A preparations and NH2-terminally truncated forms of rat cystatin A were checked by PAGE with and without SDS (Laemmli, 1970). For immunoblot analysis, the proteins were transferred from gel to a Clear-

o

~

_ _BI~

1 2 3 4 5 6 7 8 9

_

Fig. 2. A: structures of rat cystatin A, its NH2-terminally truncated forms and fragments used for epitope analysis. The stippled bands represent QVVAG in the amino acid sequence of rat cystatin A. Asterisks indicate the conserved Gly residue in rat cystatin A. B: reactivities of MAbs 77, 114 and 187 with rat cystatin A, its NH2-terminally truncated forms and fragments. The reactivity was expressed as a percentage of the absorbance at 490 nm of the native form. Each value was determined in triplicate by ELISA. The numbers 1-9 correspond to the moieties shown in A.

74

A. Takeda et al. /Journal of lmmunological Methods 168 (1994) 69-78

Japan) and then incubated with MAbs in PBSTween. The p r o t e i n b a n d s were detected by a colour reaction with the ProtoBlot immunoblotting system (Promega Biotec, WI, USA). The concentrations of these cystatin A preparations and the truncated forms were determined according to Lowry's method using BSA as standard. The concentrations of fragments of rat cystatin A were determined by amino acid analysis.

i

lOG

2

S

5G

~D

G-

3. Results

3.1. Production of monoclonal antibodies against rat cystatin A Five hybridoma cell lines were established as stable anti-rat cystatin A monoclonal antibodyproducing clones. The isotypes of the individual monoclonal antibodies were determined to be IgG1 for MAbs 77, 114, 138 and 175, and IgM for MAb 187, respectively (Table 1). The light chain of each MAb was • class.

3.2. Reacticities of monoclonal antibodies to rat cystatin A, its NH2-terminally truncated forms and fragments The reactivities of MAbs to rat cystatin A were examined by immunoblotting. As shown in Fig. 1A, all MAbs were positive on immunoblots of rat cystatin A and its Gly7-form after normal PAGE. MAb 187 gave very strong reactions while the others reacted weakly. After SDS-PAGE (Fig. 1B), MAb 187 reacted with rat cystatin A as a single band of 11 kDa. In contrast, MAb 114 as well as MAbs 138 and 175 showed a faint 22 kDa band. No band was observed on blots developed with MAb 77. These results indicate that antigenic epitopes recognized by MAb 187 are linear, because binding activity was not affected by the denaturing conditions. In contract the four MAbs 77, 114, 138 and 175 recognized the conformational epitopes on the surface of the rat cystatin A molecule. In order to localize the epitopes for these MAbs on the rat cystatin A molecule, we determined their reactivities with the NH~-terminally

0

I

I

i

I

-4

-3

-2

-1

I

0

UnlabeLed MAb(LOG rag/rot) Fig. 3. Competitive binding assay for MAb 138. Several dilutions of unlabelled MAbs were incubated with the purified MAb 138 (1 /zg/ml) conjugated to biotin succinimide ester. The mixture was adsorbed with solid-phase rat cystatin A. The results are expressed as a percentage of the absorbance observed with the conjugate in the absence of competing antibody. Unlabelled MAbs: 77, o; 114, r,; 138, []; 175, * ; 187, e.

truncated forms and fragments of rat cystatin A (Fig. 2A). The reactivities of MAbs 77 and 187 were decreased to 80-85% and 40-42% in the case of the Gly 7 and Ala t5 forms, respectively, by the NH2-terminal truncations of rat cystatin A, indicating that the epitopes for both MAbs may be shared in part. The three MAbs, 114, 138 and 175, showed no change in reactivity with rat cystatin A after NH2-terminal truncation (the data for MAb 114 is shown in Fig. 2B). Surprisingly, MAb 187 showed an affinity for fragment CB1 as well as fragments CB1-R1 and CB1-R2, derived from fragment CB1, but failed to react with fragments CB2 and CB3. MAbs 77, 114, 138 and 175 did not react with any fragment of rat cystatin A. These results indicate that five MAbs recognize conformational epitopes, with the epitopes recognized by MAbs 77 and 187 being sensitive to conformational changes induced by NH 2-terminal truncation, but not the epitopes for MAbs 114, 138 and 175.

3.3. Competitive binding The antigenic epitopes recognized by these MAbs were further analyzed by a competitive binding assay. A fixed amount of biotin-labeled

A. Takeda et aL /Journal of Immunological Methods 168 (1994) 69-78 I °° 1

I

I

00

> u t~ 0

5C

75

52%, 54% and 43% for MAbs 77, 114, 138 and 187, respectively, at a MAb to rat cystatin A molar ratio of more than 6 : 1. These results were consistent with the reactivities of MAbs with the complex of rat cystatin A and CM-papain, which were decreased to 44%, 39%, 47% and 64% for MAbs 77, 114, 138 and 187, respectively. These

¢-

=

0

I

i

I

,

10C

I

2 4 6 Molar ratio (MAb/cystatin) o

Fig. 4. Effect of MAbs on inhibition of papain activity by rat cystatin A. Rat cystatin A was preincubated for 30 rain at 37°C with various amounts of MAb at a molar ratio exceeding 6: 1. Papain (0.51 nM) was then added to the mixture which was incubated for 5 min at 37°C. The Z-Phe-Arg-MCA was used as a substrate for the detection of papain activity for 10 min at 37°C. The data are expressed as residual inhibitory activity for the inhibition of papain activity by rat cystatin A in the absence of antibody. This was approximately 98% inhibition at an equivalent molar ratio. MAbs: 77, ©; 114, zx; 138, Q; 187, e.

v

: 5C

77 B

Rat

MAb was mixed with various amounts of unlabelled MAb and allowed to react with rat cystatin A bound to an ELISA plate. The results of each an analysis using labelled MAb 138 are illustrated in Fig. 3, and the complete results are summarized in Table 1. MAbs 114, 138 and 175 competed with each other for binding to rat cystatin A, indicating that they fall into one group with slightly different efficiencies. MAbs 77 and 187 clearly recognize discrete epitopes, because none of the other antibodies was able to compete with them in binding to rat cystatin A, although it remains to be determined why both MAbs competed poorly with themselves in the competitive binding assay.

3.4. Effects of monoclonal antibodies on the inhibitory activity o f rat cystatin A to papain MAbs 77, 114, 138 and 187 were tested for their ability to suppress the inhibitory activity of rat cystatin A to papain, as shown in Fig. 4. Four MAbs partially suppressed the inhibitory activity of rat cystatin A; the observed maxima were 62%,

114

1

187

10

AcMDPGTTG

20 I QEVADKVK

. . . . .

Mouse M

Human 31

P---L

40

Mouse

P

- - A - - - K

Human

P

. . . . . . .

Mouse Human

1 -M

--H

91

EA--

70

80 LHMKVL I

~9

T-A -

] ~ R A - D N K Y M

RGL

- - E N I -

90

S GD*DDLKL

L DY

1 - - F N - P T - K * - N Y E - H G L - - F K S - P

Q N E - - V - T G

100

Rat

QTNKTKNDELTDF

Mouse

- - D - - - D -

Human

--

C-

--R --

I LF

T - G - L E A - Q

MKVDVGNGRF

IV

50

R QLEEKTNEKYEKFKVVEYKSQVVAGQ

61

KI-

. . . . . . . . . .

Rat

Rat

30

I VGGVSEAKPATPE

VD

N-D

--G-

Fig. 5. A: reactivities of three MAb with cystatin As from different species. Plates were coated with 100 /zl of purified cystatin As (10/.~g/ml) from rat (open bars), mouse (hatched bars), and human (closed bars). The reactivities are expressed as percentages of the absorbance observed for the rat cystatin A at 490 nm. B: amino acid sequences of rat, mouse, and human epidermal cystatin A molecules: (-), homology in these sequences. (*), deleted positions for homologous alignment. Underlining indicates the main linear epitope region assigned to MAb 187. The shaded region indicates the sequence of peptide 8 from human cystatin A digested with Achromobacter protease I. The sequences of rat, mouse, and human proteins are taken from Takio et al. (1984), HawleyNelson et al. (1988), and Takeda et al. (1989).

76

A. Takeda et al. / Journal of lmmunological Methods 168 (1994) 69-78

results suggest that the epitopes for these MAbs are localized away from the binding site of the inhibitor with papain.

3.5. Specificity of epitope recognition The reactivities of MAbs 77, 114 and 187 with rat, mouse and human cystatain A were examined in an ELISA (Fig. 5A). The binding specificity of MAb 187 was quite different from those of MAbs 77 and 114. MAb 187 crossreacted completely with mouse cystatin A and approx. 50% with human cystatin A. MAbs 77 and 114 reacted weakly with mouse, but not with human cystatin A. These results show that the conformational epitopes recognized by MAbs 77 and 187 are substantially different, though the reactivity of both MAbs was affected by NH2-terminal truncation of rat cystatin A as described above. CB1 almost inhibited the binding of MAb187 to native inhibitor (data not shown), indicating that the epitope for MAb 187 is located mainly within the amino acid residues, 15-31, and partially within the amino acid residues, 32-50 (Fig. 5B). We also found that these MAbs did not react with peptide 8 (sequence shadowed in Fig. 5B) from human cystatin A digested with Achromobacter protease I containing the predicted binding site sequence (Takeda et al., 1989).

4. Discussion

Several antibodies against the cystatin superfamily of proteins have previously been produced to detect and estimate the inhibitors in vertebrate tissues and body fluids (Grubb and L6fberg, 1982; Poulik et al., 1983; Kominami et al., 1984; J~irvinen et al., 1986), and to map the reactive sites of the inhibitors (Lalmanach et al., 1991,1992). The cross-reaction data obtained suggests that the conformational variability of the surface a n d / o r binding site of cystatin molecules may contribute to the slightly differing inhibitory capacities of cystatins towards target proteinases. We have produced MAbs against rat cystatin A in order to develop an immunochemical approach to investigate further the structural fea-

tures of the inhibitor and the inhibition mechanism of cysteine proteinases by different cystatin A preparations. The reactivities of MAbs to rat cystatin A, its NH2-terminally truncated forms and fragments have permitted us to map the epitopes for the five MAbs on the inhibitor molecule. All the MAbs recognized conformational epitopes on the surface of the rat cystatin A molecule; MAbs 77 and 187 recognized epitopes sensitive to conformational changes induced by NH2-terminal truncation of rat cystatin A but the three MAbs 114, 138 and 175 did not (Fig. 2). The epitopes for MAbs 77 and 187 should constitute a part of the ordered structure in rat cystatin A molecule that is susceptible to denaturation (Fig. 1B). However, neither of the antibodies competed with each other for antigen or with the other three MAbs (Table 1), which suggests that these conformational epitopes are substantially different. It is possible that the conformational epitopes recognized by these MAbs are located close to the binding site in the tertiary structure, and the MAbs may interfere with inhibitory activity by steric hindrance. There was evidence that MAb 187 also reacted with a linear sequence epitope on the rat cystatin A molecule. The major sequence recognized (amino acid residues, 15-31) is common to the NH2-terminal regions of rat, mouse and human epidermal cystatin A (Fig. 5B) and probably forms part of the five turn a helix which is enclosed by the antiparallel /3 sheet strands (Bode et al., 1988; Stubbs et al., 1990). The reactivities of MAb 187 with three different vertebrate cystatin A moieties suggest that the epitope region of rat cystatin A is very similar to that of the mouse protein although the sequence of mouse cystatin A is incomplete, and significantly different from the human protein. On the other hand, MAb 77 reacted weakly with the conformational epitope of mouse cystatin A but not at all with that of human cystatin A and the reactivities of MAb 114 were significantly species-specific with three vertebrate epidermal cystatin A preparations (Fig. 5A). These results suggest slightly different conformations around the epitope region of the three epidermal cystatin A molecules. Thus, an extended NH2-terminal segment in rat cystatin A

A. Takeda et al. /Journal of lmrnunological Methods 168 (1994) 69-78

may contribute to the essential conformation for antigenicity toward these MAbs. Rat cystatin A contains three tyrosine residues (Tyr 41, Wyr49 and Tyr 90) (Takio et al., 1984), in which Tyr 49 and Tyr 9° must feature prominently in two/3 hairpin loops as suggested by the amino acid sequence homology among cystatin A molecules (Fig. 5B), and the three dimensional structure of human cystatin B determined by X ray crystallographic analysis (Stubbs et al., 1990). The conformational changes induced by NH 2terminal truncation are compatible with the perturbation of the microenvironment of these Tyr residues in the inhibitor molecule as deduced from the spectroscopic measurements (Takeda et al., 1983). Furthermore, it has been reported that the NH2-terminal truncated Gly 7 form is less inhibitory than the native form for all target proteinases, and the more truncated mla 15 form exhibits no inhibition for them (Takeda et al., 1985). We infer that the NH2-terminal segment plays an essential role in maintaining the conformational integrity of rat cystatin A with full inhibitory activity. It is consistent with the result of crystallographic analyses in which the NH2-terminal segment of intact cystatins is flexible and serves to adopt a conformation appropriate for maximal binding with the active site cleft of proteinases (Bode et al., 1988; Machleidt et al., 1989). The MAbs characterized in this study would be useful probes with which to detect conformational changes in the rat cystatin A molecule. Furthermore, it has been confirmed, using these MAbs, that there is conformational variability among three epidermal cystatin A molecules, which may explain the slight differences in their ability to compete with the inhibitory activity of the cystatin A family of proteins.

Acknowledgements This work was supported in part by Research Funds for Group Work of Showa University. The authors thank Dr. A. Simpson for reading through the manuscript before its submission.

77

References Abe, K., Emori, Y., Kondo, H., Arai, S. and Suzuki, K. (1988) The NH2-terminal 21 amino acid residues are not essential for the papain-inhibitory activity of oryzacystatin, a member of the cystatin superfamily. Expression of oryzacystatin cDNA and its truncated fragments in Escherichia coli. J. Biol. Chem. 263, 7655-7659. Abrahamson, M., Ritonja, A., Brown, M.A., Grubb, A., Machleidt, W. and Barrett, A.J. (1987) Identification of the probable inhibitory reactive sites of the cysteine proteinase inhibitors human cystatin C and chicken cystatin. J. Biol. Chem. 262, 9688-9694. Arai, S., Watanabe, H., Kondo, H., Emori, Y. and Abe, K. (1991) Papain-inhibitory activity of oryzacystatin, a rice seed cysteine proteinase inhibitor, depends on the central Gln-Val-Val-Ala-Gly region conserved among cystatin superfamily members. J. Biochem. 109, 294-298. Barrett, A.J., Fritz, H., Grubb, A., Isemura, S., J~irvinen, M., Katunuma, N., Machleidt, W., Miiller-Esterl, W., Sasaki, M. and Turk, V. (1986a) Nomenclature and classification of proteins homologous with the cysteine-proteinase inhibitor chicken cystatin. Biochem. J. 236, 312. Barrett, A.J., Rawlings, N.D., Davies, M.E., Machleidt, W., Salvesen, G. and Turk, V. (1986b) Cysteine proteinase inhibitors of cystatin superfamily. In: A.J. Barrett and G. Salvesen (Eds.), Proteinase Inhibitors. Elsevier, Amsterdam, pp. 515-569. Bode, W., Engh, R., Musil, D., Thiele, U., Huber, R., Karshikov, A., Brzin, J., Kos, J. and Turk, V. (1988) The 2.0 A X-ray crystal structure of chicken egg white cystatin and its possible mode of interaction with cysteine proteinases. EMBO J. 7, 2593-2599. Genenger, G., Lenzen, S., Mentele, R., Assfalg-Machleidt, I. and Auerswald, E.A. (1991) Recombinant Q53E- and Q53N-chicken egg white cystatin variants inhibit papain, actinitine and cathepsin B. Biomed. Biochim. Acta 50, 621-625. Grubb, A. and L6fberg, H. (1982) Human y-trace, a basic microprotein: amino acid sequence and presence in the adenohypophysis. Proc. Natl. Acad. Sci. USA 79, 30243027. Hawley-Nelson, P., Roop, D.R., Cheng, C.K., Krieg, T.M. and Yuspa, S.H. (1988) Molecular cloning of mouse epidermal cystatin A and detection of regulated expression in differentiation and tumorigenesis. Mol, Carcinogen. 1,202-211. Isemura, S., Saitoh, E., Sanada, K., Isemura, M. and Itoh, S. (1987) Cystatin S and the related cysteine proteinase inhibitors in human saliva. In: V. Turk (Ed.), Cystatein Proteinases and Their Inhibitors. Walter de Gruyter, Berlin, pp. 497-505. Ishiguro, H., Higashiyama, S., Ohkubo, I. and Sasaki, M. (1987) Mapping of functional domains of human high molecular weight and low molecular weight kininogens using murine monoclonal antibodies. Biochemistry 26, 7021-7029. J~irvinen, M., Rinne, A., Alavaikko, M. and Hoppsu-Have,

78

A. Takeda et aL / Journal of Immunological Methods 168 (1994) 69-78

V.K. (1986) Cystatins A and B in normal and pathological altered human tissues. In: V. Turk (Ed.), Cysteine Proteinases and Their Inhibitors. Walter de Gruyter, Berlin, pp. 649-667. Jerala, R., Trstsnjak-Prebanda, M., Kroon-Zitko, K., Lenarcic, B. and Turk, V. (1990) Mutations in the QVVAG region of the cystatine proteinase inhibitor stefin B. Biol. Chem. Hoppe-Seyler 371 (suppl.), 157-160. Kitajima, Y., Matsuhashi, S., Nishida, H., Takasaki, Y., Takahashi, I., Hisatsugu, T. and Hori, K. (1991) Monoclonal antibodies recognizing amino-terminal and carboxyterminal regions of human aldolase A: Probes to detect conformational changes of the enzyme. J. Biochem. 109, 544-550. Kominami, E., Bando, Y., Wakamatsu, N. and Katunuma, N. (1984) Different tissue distributions of two types of thiol proteinase inhibitors from rat liver and epidermis. J. Biochem. 96, 1437-1442. K6hler, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefinded specificity. Nature 256, 495-497. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of head of bacteriophage T4. Nature 227, 680-685. Lalmanach, G., Adam, A., Moreau, T., Gutman, N. and Gauthier, K. (1991) Discrimination between rat thiostatin (T-kininogen) and one of its cystatin-like inhibitory fragments by a monoclonal antibody, and localization of the epitope. Eur. J. Biochem. 196, 73-78. Lalmanach, G., Hoebeke, J., Moreau, T., Ferrer-Di Martino, M. and Gauthier, F. (1992) An immunochemical approach to investigating the mechanism of inhibition of cystatine proteinases by members of the cystatin superfamily. J. Immunol. Methods 149, 197-205. Lindberg, K. and Rheinwald, J.G. (1989) Suprabasal 40kd keratin (K19) expression as an immunohistologic marker of premalignancy in oral epithelium. Am. J. Pathol. 134, 89-98. Machleidt, W., Thiele, U., Laber, B., Assfalg-Machleidt, I., Esterl, A., Wiegand, G., Kos, J., Turk, V. and Bode, W. (1989) Mechanism of inhibition of papain by chicken egg white cystatin. Inhibition constants of N-terminally truncated forms and cyanogen bromide fragment of the inhibitor. FEBS Lett. 243, 234-238. Marks, N., Berg. M.J., Makofske, R.C. and Danho, W. (1990) Synthetic domains of cystatins linked to enkephalins are novel inhibitors of brain cathepsins L/B. Peptides 11, 679-682. Moreau, T., Hoebeke, J., Lalmanach, G., Hattab, M. and Gauthier, F. (1990) Simulation of the inhibitory cystatin surface by a synthetic peptide. Biochem. Biophys. Res. Commun. 167, 117-122. Morris, G.E. (1989) Monoclonal antibody studies of creatine

kinase. The ART epitope: evidence for an intermediate in protein folding. Biochem. J. 257, 461-469. Nikawa, T., Towatari, T., Ike, Y. and Katunuma, N. (1989) Studies on the reactive site of the cystatin superfamily using recombinant cystatin A mutants. Evidence that QVVAG region is not essential for cysteine proteinase inhibitory activities. FEBS Lett. 255, 309-314. Poulik, M.D., Perry, D.J., Vokac, E. and Sekine, T. (1983) Post-gamma globulin. If. Radioimmunoassay determination of levels of post-gamma globulin and /32-microglobulin. Clin. Chim. Acta 128, 248-260. Sali, A. and Turk, V. (1987) Prediction of the secondary structures of stefins and cystatins, the low-molecular mass protein inhibitors of cysteine proteinases. Biol. Chem. Hoppe-Seyler 368, 493-499. Stubbs, M.T., Laber, B., Bode, W., Huber, R., Jerala, R., Lenarcic, B. and Turk, V. (1990) The refined 2.4 ,~ X-ray crystal structure of recombinant human stefin B in complex with the cysteine proteinase papain: a novel type of proteinase inhibitor interaction. EMBO J. 9, 1939-1947. Takeda, A., Kobayashi, S. and Samejma, T. (1983) Isolation and characterization of two thiol proteinase inhibitors of low molecular weight from newborn rat epidermis. J. Biochem. 94, 811-820. Takeda, A., Kaji, H., Nakaya, K., Aoki, Y., Nakamura, Y. and Samejima, T. (1985) Amino acid sequence of derivatives of newborn rat epidermal thiol proteinase inhibitor. Biochem. Int. 11, 557-564. Takeda, A., Kaji, H., Nakaya, K., Nakamura, Y. and Samejima, T. (1989) Comparative studies on the primary structure of human cystatin As from epidermis, liver, spleen, and leukocytes. J. Biochem. 105, 986-991. Takio, K., Kominami, E., Bando, Y., Katunuma, N. and Titani, K. (1984) Amino acid sequence of rat epidermal thiol proteinase inhibitor. Biochem. Biophys. Res. Commun. 121, 149-154. Teno, N., Tsuboi, S., Itoh, N., Okamoto, H. and Okada, Y. (1987) Significant effects of Z-Gln-Val-Val-OMe, common sequences of thiol proteinase inhibitors on thiol proteinases. Biochem. Biophys. Res. Commun., 143, 749-752. Thiele, U., Assfalg-Machleidt, I., Machleidt, W. and Auerswald, E.A. (1990) Biol. Chem. Hoppe-Seyler 371 (suppl), 125-136. Thor, A., Hand, P.H., Wunderlich, D., Caruso, A., Muraro, R. and Schlom, J. (1984) Monoclonal antibodies define differential ras gene expression in malignant and benign colonic diseases. Nature 311,562-565. Waseen, N.H. and Lane, D.P. (1990) Monoclonal antibody analysis of the proliferating cell nuclear antigen (PCNA). Structure conservation and the detection of a nucleolar form. J. Cell Sci. 96, 121-129.