Journal of Protein Chemistry, VoL 10, No. 2, 1991
Epitope Mapping of Snake Venom Phospholipases A2 with Pseudexin Monoclonal Antibodies Bradley G. Stiles I and J o h n L. M i d d l e b r o o k ~
Received November 13, 1990
Fifteen different monoclonal antibodies, developed against a pseudexin A, B, and C mixture, were screened for linear epitope recognition. Peptides (9-mers) spanning pseudexin B were synthesized on alanine-derivatized polyethylene pins and subsequently probed with antibody. Four antibodies recognized linear epitopes of pseudexin A, pseudexin B, and also nonidentical sequences found in other phospholipases A2 (PLA2s) as determined by enzyme-linked immunosorbent assays. Three antibodies recognized a highly conserved site important in calcium binding and the interlocking of dimeric forms of PLA2. Antibodies neutralizing lethal or enzymatic effects of PLA2 did not recognize linear epitopes. KEY WORDS: Phospholipase A2; epitope mapping; cross-reactivity; monoclonal antibodies.
!. INTRODUCTION
conserved peptide regions important in calcium binding and the interlocking of multicomponent PLA:s (Dufton et aL, 1983; Keith et aL, 1981). Investigators in a recent immunological study found two groups of PLA2s by using an enzymelinked immunosorbent assay (ELISA) and polyclonal serum against 11 different PLA2s (Middlebrook and Kaiser, 1989). Elapid venom and mammalian tissue PLA2s represent an immunological group distinctly different than the crotalid and viperid group. Extensive cross-reactivity and even protection against mouse lethality were found by using heterologous antisera. The immunological groupings corroborate an earlier proposal with polypeptide chain length and amino acid similarity studies (Dufton and Hider, 1983). Both reports suggest that epitope mapping might yield useful information in determining crossreacting sequences. In this study, monoclonal antibodies produced against a mixture of pseudexins A, B, and C from Pseudeehis porphyriacus (Australian red-bellied black snake) venom were screened for reactivity against synthesized peptides of pseudexin A, pseudexin B, and other PLA2s.
Snake venom phospholipase A2s (PLA2s) are a heterogenous group of calcium-dependent enzymes consisting of single chain molecules or complexes of two, three, or five similar or dissimilar components (Fohlman et al., 1976; Aird et al., 1985; Chang, 1985; Kini and Iwanga, 1986; Tyler et al., 1987). Most multicomponent PLA2s are held together by noncovalent bonds, except for /3-bungarotoxin, which contains a disulfide linkage between two dissimilar subunits (Kondo et al., 1978). All PLA2s hydrolyse the C2 ester bond in 3-sn-phosphoglycerides, but many snake PLA2s cause numerous pharmacological effects, like presynaptic neurotoxicity, cardiotoxicity, myotoxicity, and have indirect hemolytic and anticoagulant activity (Kini and Evans, 1989). Separate regions of the molecule may be responsible for different activities (Soons et al., 1986; Kini and Evans, 1989; Kasturi and Gowda, 1990). PLA2s share highly Department of Toxinology, Pathophysiology Division, U.S. Army Medical Research Institute of Infectious Diseases, Frederick, Maryland 21702-5011.
193 027%8033/91/0400-0193506.50/0© 1991PlenumPublishingCorporation
194 2. MATERIALS AND METHODS 2.1. Monoclonal Antibody Production BALB/c mice were immunized with peak V, a pseudexin A, B, and C mixture (Schmidt and Middlebrook, 1989), by the same procedure described for crotoxin monoclonal antibody production (Kaiser and Middlebrook, 1988). Seropositive mice were given to Dr. Brian Butman of Bionetics Inc. (Gaithersburg, Maryland) for hybridoma production using standard techniques. Positive wells were doubly cloned and ascites fluid produced from each clone. All antibodies were of the IgG1 subclass, except no. 11, which was an IgM. Antibodies were purified from ascites fluid with Protein A-agarose (Biorad, Richmond, California).
2.2. Peptide Synthesis Peptides were produced by a pin procedure (Geysen et aL, 1984) and PLA2 sequences were found in variousreferences (Aird et al., 1985; Bieber et aL, 1990; Dufton et aL, 1983; Mebs, 1985; Schmidt and Middlebrook, 1989). Briefly, 9-mer peptides were synthesized in duplicate on polyethylene pins (96 pins per block) by Fmo~ chemistry (Cambridge Research Biochemicals, Valley Stream, New York). Software from the Commonwealth Serum Laboratories Commission (Victoria, Australia) was used to generate the synthesis schedule. Negative control peptide sequences were randomly generated by the software. Each amino acid was coupled for 24 hr, followed by a deprotection step of dimethylformamide (DMF) : piperdine 4: 1, methanol, DMF washes, and finally the addition of another amino acid. This cycle was repeated until the completion of the desired sequence. After the final deprotection step, peptides were acetylated using DMF:acetic anhydride : diisopropylethylamine 50: 5 : 1. Side-chain protecting groups were removed by trifluoroacetic acid : phenol : ethanedithiol 95 : 2.5 : 2.5 followed by dichloromethane and methanol washes. Synthesis success was monitored by using an ELISA described below and an antibody specific for a 4-mer peptide made during each synthesis. Pins were stored under vacuum over silica. The peptides used for competition studies were synthesized and purified by Peptide Technologies Inc., Washington D.C. After reverse phase (C18) column chromatography with a 0.1% phosphoric acid/acetonitrile linear gradient, peptides were
Stiles and Middlebrook greater than 90% homogeneous. Peptides with amino acid analysis after acid hydrolysis and the percentage of acetonitrile (Acn) needed for elution from a reverse phase (C18) column are as follows: VDELD (V = 0.97, D=2.28, E = 1.01, L=0.92, eluted with 17% Acn) IDALD (I = 0.99, D = 2.04, A = 1.00, L = 1.03, eluted with 20% Acn) NTKWDIYGY (N/D = 2.01, T = 1.10, K = 0.94, I = 1.05, Y= 1.79, G = 1.14, eluted with 31% Acn) FPKLTLYSW (F = 1.07, K = 1.00, L = 2.04, T = 1.05, Y=0.76, S=0.92, eluted with 40% Acn) TPYTSLYTW (T=3.11, Y = 1.97, S= 1.00, L = 1.09, eluted with 41% Acn) FPKEKICIP ( F = 1.06, K = 1.94, E=l.17, 1=1.89, eluted with 30% Acn) YGCYCGKGG (Y = 2.00, G = 3.91, K = 0.74, eluted with 25% Acn) and YGCYCGPGG (Y=2.11, G = 4.01, eluted with 37% Acn).
2.3. Enzyme-Linked Immunosorbent Assay (ELISA) Procedures Before use, peptide-bearing pins were sonicated for 30 min at 60°C in disruption solution (1% sodium dodecyl sulfate, 0.1% 2-mercaptoethanol, 0.1M NaH2PO4 buffer, pH 7.2) and subsequently washed in distilled water and methanol. Pins were then precoated in Immulon II microtiter plates (Dynatech, Chantilly, Virginia) by a blocking solution consisting of phosphate-buffered saline (PBS), pH 7.4, 1% ovalbumin, 1% bovine serum albumin, and 0.1% Tween 20 for 1 hr on a rotary shaker at room temperature. After overnight incubation at 4°C with purified monoclonal antibody (750 ng/ml) diluted in blocking solution, pins were washed in PBS containing 0.1% Tween 20 (PBST). Sheep antimouse alkaline phosphatase conjugate (Sigma, St. Louis, Missouri) was diluted in blocking solution and incubated with pins for 1 hr at room temperature. After PBST washes, pins were added to a new ELISA plate containing p-nitrophenyl phosphate substrate in diethanolamine buffer (Kirkegaard and Perry Laboratories, Gaithersburg, Maryland). Plates were read 30min later at 405 nm. Absorbance readings represent the average of two pins per peptide. Monoclonal antibodies not giving a positive response at 750 ng/ml were tested again at 7.5/~g/ml. The ELISA for antibody titers against intact PLA2 has been previously described (Middlebrook, 1991). Briefly, Immulon II ELISA wells were coated overnight at room temperature using 0.1 ml of a 1/xg PLAz/ml carbonate buffer, pH 9.6 solution. After two PBST washes, dilutions of monoclonal antibody
PLA2 Epitope Mapping (0.6 mg/ml stock) were added and incubated for 4 hr at room temperature. Plates were washed twice and goat antimouse IgG conjugated to peroxidase (Sigma) was added and incubated for 2 hr at 4°C. Hydrogen peroxide and chromogen (2,2'-azino-di-{3ethylbenzthiazoline sulfonate} from Kirkegaard and Perry Laboratories) were added and the absorbance read at 405 nm after 1 hr. The titer given represents the inverse log of antibody dilution required to give a 1.0 absorbance reading. The competitive ELISA consisted of coating Immulon II plates overnight at 4°C with pseudexin peak V using 100 ng/ml or 250 ng/ml in carbonate buffer for antibody 2 and 14 studies, respectively. Antibodies 2 (0.31 k~g/ml) or 14 (1.89/~g/ml) were preincubated 1 hr at 37°C with varying concentrations of peptide diluted in PBS containing 0.1% Tween 20 and 0.1% gelatin. The antibody-peptide mixture was added to pseudexin coated plates for 1 hr at 37°C. After washing with PBST, alkaline phosphatase conjugate was added and incubated 1 hr at 37°C, followed by PBST washes and substrate. Absorbance was read 30 rain later at 405 nm. All samples were done in triplicate and the mean and standard deviation of absorbance readings recorded. Percent competition was calculated by comparing the absorbance of wells with and without peptide competitor.
2.4. Computer Modeling of Porcine Pancreatic PLAz The porcine pancreatic PLA2 model was kindly provided by Dr. Dallas Hack of this Institute using insight II version 1.0.2 software from Biosym Technologies Inc., (San Diego, California) and hardware from Silicon Graphics (Sunnyvale, California). Surface solvent accessibility was calculated by a Connolly algorithm.
195 ELISA (Schmidt and Middlebrook, 1989; Middlebrook, 1990). The amino acid sequence of pseudexin B is shown in Fig. 1 (Schmidt and Middlebrook, 1989). Three of 15 monoclonal antibodies (2, 13, and 14) recognized test peptides compared to control peptides (Fig. 2). Nonreacting antibodies were further screened against 9-mer peptides spanning pseudexin B, moving one amino acid at a time and using 10-fold more antibody. A fourth antibody (no. 5) reacted with the same linear regions as 13 and 14. The epitopes recognized by the four reactive antibodies were defined (Fig. 3 and Table I). Antibody 2 recognized a 5-mer VDELD (amino acids 38-42), while antibodies 5, 13, and 14 interestingly recognized two different 9-mer sites, YGCYCGPGG and FPKLTLYSW (amino acids 25-33 and 61-69).
3.2. Reactivity of Antibodies with Heterologous Linear Sequences Since pseudexin monoclonal antibodies crossreact with other PLA2s (Middlebrook, 1991), the best possible sequence alignment of other PLA2s was made and peptides synthesized corresponding to suspected cross-reacting sites (Tables II-IV). Antibodies 2 and 14 were used for this ELISA study because they were more reactive in our initial epitope mapping experiments with pseudexin B peptides. Reactivity of antibodies 2 or 14 was found with other analogous PLA2 sequences. Antibody 2 was more discriminating in recognizing sequences than antibody 14 (Table II). A conservative change in the third position of the epitope for antibody 2 from glutamic to aspartic acid resulted in a 2.0 decrease in the absorbance reading. Other amino acid changes in the same or different positions resulted in near background readings. Antibody 14 reacted more strongly with some heterologous than homologous sequences (Table II).
3.3. Competition Studies Using Soluble Peptides 3. RESULTS
3.1. Epitope Mapping of Monoclonal Antibodies Pseudexin B 9-mer peptides, moving three amino acids at a time through the whole molecule, were initially screened for linear epitope recognition by 15 different monoclonal antibodies developed against pseudexins A, B, and C. Pseudexin B peptides were probed because the whole amino acid sequence of this most lethal isoform is known and also monoclonal antibodies react most strongly with it in an
These studies were done to confirm our previous epitope mapping results by using a different system. Soluble peptide (VDELD), representing the antibody 2 recognition site, competitively inhibited antibody 0 10 NLIQFSNMIK 50 L~RCCKIHDD 90 KSRCKC,FVCA
20 30 40 CAIPGSRPLF Q Y A D ~ ~HGTP.V.~ 60 70 80 CYGEAGKKGC FPKLTLY.8~K CTEKVPTCNA 100 110 CDAEAAKCFA KAPYIKENYN !NTKTRC
Fig. 1. Aminoacid sequenceof pseudexin B withthe monoclonal antibody recognitionsites in bold and underlined.
196
Stiles and Middlebrook 2.22.01.81.6E
1.4" 1.2-
~
1.0-
o
.D
Epitope Mapping of Snake Venom Phospholipases A2 with Pseudexin Monoclonal Antibodies Bradley G. Stiles I and J o h n L. M i d d l e b r o o k ~
Received November 13, 1990
Fifteen different monoclonal antibodies, developed against a pseudexin A, B, and C mixture, were screened for linear epitope recognition. Peptides (9-mers) spanning pseudexin B were synthesized on alanine-derivatized polyethylene pins and subsequently probed with antibody. Four antibodies recognized linear epitopes of pseudexin A, pseudexin B, and also nonidentical sequences found in other phospholipases A2 (PLA2s) as determined by enzyme-linked immunosorbent assays. Three antibodies recognized a highly conserved site important in calcium binding and the interlocking of dimeric forms of PLA2. Antibodies neutralizing lethal or enzymatic effects of PLA2 did not recognize linear epitopes. KEY WORDS: Phospholipase A2; epitope mapping; cross-reactivity; monoclonal antibodies.
!. INTRODUCTION
conserved peptide regions important in calcium binding and the interlocking of multicomponent PLA:s (Dufton et aL, 1983; Keith et aL, 1981). Investigators in a recent immunological study found two groups of PLA2s by using an enzymelinked immunosorbent assay (ELISA) and polyclonal serum against 11 different PLA2s (Middlebrook and Kaiser, 1989). Elapid venom and mammalian tissue PLA2s represent an immunological group distinctly different than the crotalid and viperid group. Extensive cross-reactivity and even protection against mouse lethality were found by using heterologous antisera. The immunological groupings corroborate an earlier proposal with polypeptide chain length and amino acid similarity studies (Dufton and Hider, 1983). Both reports suggest that epitope mapping might yield useful information in determining crossreacting sequences. In this study, monoclonal antibodies produced against a mixture of pseudexins A, B, and C from Pseudeehis porphyriacus (Australian red-bellied black snake) venom were screened for reactivity against synthesized peptides of pseudexin A, pseudexin B, and other PLA2s.
Snake venom phospholipase A2s (PLA2s) are a heterogenous group of calcium-dependent enzymes consisting of single chain molecules or complexes of two, three, or five similar or dissimilar components (Fohlman et al., 1976; Aird et al., 1985; Chang, 1985; Kini and Iwanga, 1986; Tyler et al., 1987). Most multicomponent PLA2s are held together by noncovalent bonds, except for /3-bungarotoxin, which contains a disulfide linkage between two dissimilar subunits (Kondo et al., 1978). All PLA2s hydrolyse the C2 ester bond in 3-sn-phosphoglycerides, but many snake PLA2s cause numerous pharmacological effects, like presynaptic neurotoxicity, cardiotoxicity, myotoxicity, and have indirect hemolytic and anticoagulant activity (Kini and Evans, 1989). Separate regions of the molecule may be responsible for different activities (Soons et al., 1986; Kini and Evans, 1989; Kasturi and Gowda, 1990). PLA2s share highly Department of Toxinology, Pathophysiology Division, U.S. Army Medical Research Institute of Infectious Diseases, Frederick, Maryland 21702-5011.
193 027%8033/91/0400-0193506.50/0© 1991PlenumPublishingCorporation
194 2. MATERIALS AND METHODS 2.1. Monoclonal Antibody Production BALB/c mice were immunized with peak V, a pseudexin A, B, and C mixture (Schmidt and Middlebrook, 1989), by the same procedure described for crotoxin monoclonal antibody production (Kaiser and Middlebrook, 1988). Seropositive mice were given to Dr. Brian Butman of Bionetics Inc. (Gaithersburg, Maryland) for hybridoma production using standard techniques. Positive wells were doubly cloned and ascites fluid produced from each clone. All antibodies were of the IgG1 subclass, except no. 11, which was an IgM. Antibodies were purified from ascites fluid with Protein A-agarose (Biorad, Richmond, California).
2.2. Peptide Synthesis Peptides were produced by a pin procedure (Geysen et aL, 1984) and PLA2 sequences were found in variousreferences (Aird et al., 1985; Bieber et aL, 1990; Dufton et aL, 1983; Mebs, 1985; Schmidt and Middlebrook, 1989). Briefly, 9-mer peptides were synthesized in duplicate on polyethylene pins (96 pins per block) by Fmo~ chemistry (Cambridge Research Biochemicals, Valley Stream, New York). Software from the Commonwealth Serum Laboratories Commission (Victoria, Australia) was used to generate the synthesis schedule. Negative control peptide sequences were randomly generated by the software. Each amino acid was coupled for 24 hr, followed by a deprotection step of dimethylformamide (DMF) : piperdine 4: 1, methanol, DMF washes, and finally the addition of another amino acid. This cycle was repeated until the completion of the desired sequence. After the final deprotection step, peptides were acetylated using DMF:acetic anhydride : diisopropylethylamine 50: 5 : 1. Side-chain protecting groups were removed by trifluoroacetic acid : phenol : ethanedithiol 95 : 2.5 : 2.5 followed by dichloromethane and methanol washes. Synthesis success was monitored by using an ELISA described below and an antibody specific for a 4-mer peptide made during each synthesis. Pins were stored under vacuum over silica. The peptides used for competition studies were synthesized and purified by Peptide Technologies Inc., Washington D.C. After reverse phase (C18) column chromatography with a 0.1% phosphoric acid/acetonitrile linear gradient, peptides were
Stiles and Middlebrook greater than 90% homogeneous. Peptides with amino acid analysis after acid hydrolysis and the percentage of acetonitrile (Acn) needed for elution from a reverse phase (C18) column are as follows: VDELD (V = 0.97, D=2.28, E = 1.01, L=0.92, eluted with 17% Acn) IDALD (I = 0.99, D = 2.04, A = 1.00, L = 1.03, eluted with 20% Acn) NTKWDIYGY (N/D = 2.01, T = 1.10, K = 0.94, I = 1.05, Y= 1.79, G = 1.14, eluted with 31% Acn) FPKLTLYSW (F = 1.07, K = 1.00, L = 2.04, T = 1.05, Y=0.76, S=0.92, eluted with 40% Acn) TPYTSLYTW (T=3.11, Y = 1.97, S= 1.00, L = 1.09, eluted with 41% Acn) FPKEKICIP ( F = 1.06, K = 1.94, E=l.17, 1=1.89, eluted with 30% Acn) YGCYCGKGG (Y = 2.00, G = 3.91, K = 0.74, eluted with 25% Acn) and YGCYCGPGG (Y=2.11, G = 4.01, eluted with 37% Acn).
2.3. Enzyme-Linked Immunosorbent Assay (ELISA) Procedures Before use, peptide-bearing pins were sonicated for 30 min at 60°C in disruption solution (1% sodium dodecyl sulfate, 0.1% 2-mercaptoethanol, 0.1M NaH2PO4 buffer, pH 7.2) and subsequently washed in distilled water and methanol. Pins were then precoated in Immulon II microtiter plates (Dynatech, Chantilly, Virginia) by a blocking solution consisting of phosphate-buffered saline (PBS), pH 7.4, 1% ovalbumin, 1% bovine serum albumin, and 0.1% Tween 20 for 1 hr on a rotary shaker at room temperature. After overnight incubation at 4°C with purified monoclonal antibody (750 ng/ml) diluted in blocking solution, pins were washed in PBS containing 0.1% Tween 20 (PBST). Sheep antimouse alkaline phosphatase conjugate (Sigma, St. Louis, Missouri) was diluted in blocking solution and incubated with pins for 1 hr at room temperature. After PBST washes, pins were added to a new ELISA plate containing p-nitrophenyl phosphate substrate in diethanolamine buffer (Kirkegaard and Perry Laboratories, Gaithersburg, Maryland). Plates were read 30min later at 405 nm. Absorbance readings represent the average of two pins per peptide. Monoclonal antibodies not giving a positive response at 750 ng/ml were tested again at 7.5/~g/ml. The ELISA for antibody titers against intact PLA2 has been previously described (Middlebrook, 1991). Briefly, Immulon II ELISA wells were coated overnight at room temperature using 0.1 ml of a 1/xg PLAz/ml carbonate buffer, pH 9.6 solution. After two PBST washes, dilutions of monoclonal antibody
PLA2 Epitope Mapping (0.6 mg/ml stock) were added and incubated for 4 hr at room temperature. Plates were washed twice and goat antimouse IgG conjugated to peroxidase (Sigma) was added and incubated for 2 hr at 4°C. Hydrogen peroxide and chromogen (2,2'-azino-di-{3ethylbenzthiazoline sulfonate} from Kirkegaard and Perry Laboratories) were added and the absorbance read at 405 nm after 1 hr. The titer given represents the inverse log of antibody dilution required to give a 1.0 absorbance reading. The competitive ELISA consisted of coating Immulon II plates overnight at 4°C with pseudexin peak V using 100 ng/ml or 250 ng/ml in carbonate buffer for antibody 2 and 14 studies, respectively. Antibodies 2 (0.31 k~g/ml) or 14 (1.89/~g/ml) were preincubated 1 hr at 37°C with varying concentrations of peptide diluted in PBS containing 0.1% Tween 20 and 0.1% gelatin. The antibody-peptide mixture was added to pseudexin coated plates for 1 hr at 37°C. After washing with PBST, alkaline phosphatase conjugate was added and incubated 1 hr at 37°C, followed by PBST washes and substrate. Absorbance was read 30 rain later at 405 nm. All samples were done in triplicate and the mean and standard deviation of absorbance readings recorded. Percent competition was calculated by comparing the absorbance of wells with and without peptide competitor.
2.4. Computer Modeling of Porcine Pancreatic PLAz The porcine pancreatic PLA2 model was kindly provided by Dr. Dallas Hack of this Institute using insight II version 1.0.2 software from Biosym Technologies Inc., (San Diego, California) and hardware from Silicon Graphics (Sunnyvale, California). Surface solvent accessibility was calculated by a Connolly algorithm.
195 ELISA (Schmidt and Middlebrook, 1989; Middlebrook, 1990). The amino acid sequence of pseudexin B is shown in Fig. 1 (Schmidt and Middlebrook, 1989). Three of 15 monoclonal antibodies (2, 13, and 14) recognized test peptides compared to control peptides (Fig. 2). Nonreacting antibodies were further screened against 9-mer peptides spanning pseudexin B, moving one amino acid at a time and using 10-fold more antibody. A fourth antibody (no. 5) reacted with the same linear regions as 13 and 14. The epitopes recognized by the four reactive antibodies were defined (Fig. 3 and Table I). Antibody 2 recognized a 5-mer VDELD (amino acids 38-42), while antibodies 5, 13, and 14 interestingly recognized two different 9-mer sites, YGCYCGPGG and FPKLTLYSW (amino acids 25-33 and 61-69).
3.2. Reactivity of Antibodies with Heterologous Linear Sequences Since pseudexin monoclonal antibodies crossreact with other PLA2s (Middlebrook, 1991), the best possible sequence alignment of other PLA2s was made and peptides synthesized corresponding to suspected cross-reacting sites (Tables II-IV). Antibodies 2 and 14 were used for this ELISA study because they were more reactive in our initial epitope mapping experiments with pseudexin B peptides. Reactivity of antibodies 2 or 14 was found with other analogous PLA2 sequences. Antibody 2 was more discriminating in recognizing sequences than antibody 14 (Table II). A conservative change in the third position of the epitope for antibody 2 from glutamic to aspartic acid resulted in a 2.0 decrease in the absorbance reading. Other amino acid changes in the same or different positions resulted in near background readings. Antibody 14 reacted more strongly with some heterologous than homologous sequences (Table II).
3.3. Competition Studies Using Soluble Peptides 3. RESULTS
3.1. Epitope Mapping of Monoclonal Antibodies Pseudexin B 9-mer peptides, moving three amino acids at a time through the whole molecule, were initially screened for linear epitope recognition by 15 different monoclonal antibodies developed against pseudexins A, B, and C. Pseudexin B peptides were probed because the whole amino acid sequence of this most lethal isoform is known and also monoclonal antibodies react most strongly with it in an
These studies were done to confirm our previous epitope mapping results by using a different system. Soluble peptide (VDELD), representing the antibody 2 recognition site, competitively inhibited antibody 0 10 NLIQFSNMIK 50 L~RCCKIHDD 90 KSRCKC,FVCA
20 30 40 CAIPGSRPLF Q Y A D ~ ~HGTP.V.~ 60 70 80 CYGEAGKKGC FPKLTLY.8~K CTEKVPTCNA 100 110 CDAEAAKCFA KAPYIKENYN !NTKTRC
Fig. 1. Aminoacid sequenceof pseudexin B withthe monoclonal antibody recognitionsites in bold and underlined.
196
Stiles and Middlebrook 2.22.01.81.6E
1.4" 1.2-
~
1.0-
o
.D
