- Email: [email protected]
A mutant streptokinase lacking the C-terminal 42 amino acids is less immunogenic
Immunology Letters 70 (1999) 213 – 218
www.elsevier.com/locate/
A mutant streptokinase lacking the C-terminal 42 amino acids is less immunogenic Isis Torre`ns a,*, Ariana G. Ojalvo a, Alina Seralena a, Orlando Hayes b, Jose` de la Fuente b b
a Di6ision of Pharmaceutical, Centro de Ingenierı`a Gene`tica y Biotecnologı`a, P.O. Box 6162, Ha6ana, Cuba Di6ision of Mammalian Cell Genetics, Centro de Ingenierı`a Gene`tica y Biotecnologı`a, P.O. Box 6162, Ha6ana, Cuba
Received 10 August 1999; accepted 6 September 1999
Abstract Streptokinase (SK) is the most widely used compound for the treatment of myocardial infarction and the least expensive thrombolytic agent, but a drawback to its use is the widespread presence of anti-SK antibodies (Abs). Clinical failure of the activation of the fibrinolytic system by SK has been reported due to the presence of a high titer of anti-SK neutralizing Abs. Patients receiving SK therapy develop high anti-SK antibody titers, which might provoke severe allergic reactions. These Abs are sufficient to neutralize a standard dose of SK up to four years after initial SK administration. This is a clinical problem because of the increasing number of patients who have been treated once with SK for acute myocardial infarction (AMI) and are likely to require plasminogen activator treatment in the future. In previous in vitro studies, we have shown that a deletion mutant (mut-C42), lacking the 42 C-terminal residues, was significantly less antigenic when compared with the native molecule (SKC-2). In this study, 14 monkeys were subjected to treatment with SKC-2 and mut-C42 in order to compare their humoral response by determining SK neutralizing activity in monkey’s sera. All monkeys developed anti-SKC-2 Ab titers, but in the case where treatment induced Abs directed against the C-terminus of SKC-2, neutralizing activity against the native protein was significantly higher than that developed against mutant SK mut-C42. © 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Streptokinase; Immunology; Antibodies; Neutralizing antibodies
1. Introduction Thrombotic complications of cardiovascular diseases are a main cause of death and consequently, thrombolytic therapy has become an accepted form of treatment for AMI [1]. Thrombolytic agents are plasminogen (Plg) activators that convert Plg, the inactive proenzyme of the fibrinolytic system in blood, to proteolytic enzyme plasmin (Pl), which dissolves the fibrin of a blood clot [2]. Streptokinase (SK) is a streptococcal protein (MW 47 000) specially used for the treatment of venous thromboembolic diseases and AMI wherein it has been demonstrated to be virtually as efficacious as its more expensive alternatives, namely urokinase (UK) and tissue-type Plg activator (TPA) [1]. * Corresponding author. Fax: +53-7-336008/214764. E-mail address: [email protected] (I. Torre`ns)
SK is a bacterial protein and therefore antigenic in humans. Since streptococcal infections are common, adults are immunized with SK, and antibodies (Abs) to SK can be detected in most of them [3]. SK therapy generates significant T-cell responses to SK [4–7] and the neutralizing capacity of the Abs rise significantly [8–13]. Clinical failure of the activation of the fibrinolytic system by SK has been reported due to the presence of a high titer of neutralizing anti-SK Abs [14]. After SK administration, IgG Abs rise to a peak at two weeks and slowly fall over the next 6–12 months [15], but 50% of patients still have Ab levels sufficient to neutralize a standard dose of SK up to four years after initial SK administration [10,15,16]. High circulating anti-SK Ab titers are likely to influence the efficacy of the thrombolytic therapy [17] and cause allergic reactions [18–21]. Therefore, SK should not be re-administered, except perhaps in the first 2–3 days after initial treatment [16]. Twenty percent of patients with
0165-2478/99/$ - see front matter © 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 7 8 ( 9 9 ) 0 0 1 5 1 - 0
214
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
AMI have had a previous infarction, and reinfarction within 1 year after infarction occurs in 9% of cases [16,22]. Therefore, the engineering of a SK molecule with reduced immunoreactivity would allow its effective readministration in patients with reinfarction. SK is a heterologous protein, and it is not obvious that its immunoreactivity can be reduced by protein engineering. Furthermore, a deletion mutant lacking 46 C-terminal amino acids has about one-quarter of the specific activity of SK [23], suggesting that it is very sensitive to inactivation by site directed mutagenesis. In a previous study we have shown that native SKC2 (SK encoded by skc-2 gene from Streptococcus equisimilis group C, ATCC 9542 [24]) contains an immunodominant epitope in the C-terminal region [25]. The SKC-2 mutant protein (mut-C42), lacking 42 Cterminal residues was obtained after its expression in E. coli, without loss of Plg-specific activity (submitted for publication). We have compared the new protein mutC42 with the native SKC-2 regarding their antigenic properties, by means of direct binding and competition assays using human sera from patients treated with Heberkinase (Heber Biotec S.A., Havana, Cuba). Heberkinase contains a recombinant SKC-2 obtained after the expression of the skc-2 gene in E. coli [24]. Our results showed that recognition of mut-C42 by antiSKC-2 Abs from patients sera was 51% and 68% of reactivity to native SKC-2, as assessed by direct binding and competition assays, respectively (submitted for publication). Both, mutant and native proteins were also tested in a neutralizing activity assay using patients’ sera. For most of the individuals, mut-C42-neutralizing activity titer (NAT) significantly decreased with respect to SKC-2-NAT, ranging from 30 to 91% of the native protein value (submitted for publication). This previous study suggested that it is possible to produce an engineered variant of SK being functional and less antigenic than the native molecule in vitro. If this variant is also less antigenic in vivo it could constitute a preferred alternative to native SK for thrombolytic therapy in patients with thromboembolic diseases. In the present study, we report that mut-C42 biological activity was less affected by neutralizing Abs when compared with native SKC-2, after administration in non-human primates.
2. Materials and methods
2.1. Animal Fourteen monkeys (Cercopithecus aethiops) of either sex, between 2 and 3 years old, weighing 1.8 – 2.5 kg, were selected for the study. The experimental methods were approved by the Animal Care and Use Committee of the Center for Genetic Engineering and Biotechnol-
ogy and were conducted in accordance with the Health Guide for the Care and Use of Laboratory Animals.
2.2. Titration of monkey sera by anti-SK ELISA Total serum from monkeys was collected. Polyvinyl plates (Costar, Cambridge, Massachusetts, USA) were coated with 10 mg/ml SKC-2 in coating buffer (0.1 M Na2CO3, 0.1 M NaHCO3, pH 9.6), and incubated overnight at 4°C. Then, plates were washed three times with 0.05% Tween 20 in PBS (PBS-Tween). One hundred microlitres of serial dilutions of each monkey serum was added. After incubation for 1 h at 37°C, plates were incubated with a biotinylated protein A solution at 1:3000 dilution. After incubation for 1 h at 37°C, the binding of monkey Abs to SKC-2 was measured using a horseradish peroxidase-conjugated streptavidin (Sigma). The reaction was developed using 100 ml per well of 1 mg/ml o-phenylenediamine (Sigma), 0.03% H2O2 in substrate buffer (0.1 M citric acid, 0.2 M Na2HPO4, pH 5.0). After 30 min, the reaction was stopped with 50 ml of 4 M H2SO4. Results were measured on a Multiskan system (Titertek, Helsinki, Finland) at 492 nm. The anti-SKC-2 Ab titer was determined as the maximal dilution in which positive signal was obtained. Positive signal was considered when the value was at least 2-fold the background.
2.3. Treatment of monkeys with mut-C42 and SKC-2 Treatment was carried out by subcutaneous administration of 425 mg/kg mut-C42 and SKC-2 without Freund’s adjuvant. Animals were divided in two groups: group A (eight monkeys without previous antiSKC-2 Ab titer) and group B (six monkeys with previous anti-SKC-2 Ab titer) (Fig. 1). Half of monkeys from each group were treated with SKC-2 and the other half with mut-C42. Proteins were administered at weeks 0, 2, 4, and 6 for group A, and at week 0 for group B. The humoral response was quantified at week 8 for group A, and at week 2 for group B.
2.4. mut-C42 - and SKC-2 -neutralizing acti6ity in monkey sera The chromogenic substrate (S-2251) reaction was carried out in polyvinyl plates (Costar, Cambridge, Massachusetts, USA). Serial dilutions of SKC-2 and mut-C42 (128-2 IU, 2-fold dilutions in 20 mM Tris– HCl pH8/0.5 M NaCl) were prepared in a volume of 25 ml each one. SKC-2 and mut-C42 curves were mixed with 25 ml of each monkey diluted serum and a negative control. For monkeys without previous anti-SKC-2 Ab titer a 1:2 dilution was used, and sera from monkeys with previous anti-SKC-2 Ab titer were diluted 1:5. The negative control was a monkey serum without anti-
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
SKC-2 Ab titer. Fifty microliters of 25 mg/ml human Plg were added and allowed to mix for 10 min at room temperature. The reaction was developed by addition of 50 ml of chromogenic substrate S-2251 (Chromogenix, Antwerp, Belgium). After incubation for 30 min, the reaction was stopped with 25 ml of 20% acetic acid. Results were measured on a Multiskan system (Titertek, Helsinki, Finland) at 405 nm. The activity of SKC-2 required to obtain an absorbance of 0.7 was determined from plots of absorbance versus activity. The neutralizing activity titer (NAT) was determined as the difference between the tested serum and negative control value and was expressed as micrograms of SKC-2 and mut-C42 neutralized per milliliter of tested serum.
2.5. Statistical analysis In order to analyze SKC-2- and mut-C42-neutralizing activities in monkeys treated with SKC-2, a Student’s t-test for paired values, one-tailed distribution,
Fig. 1. Diagram of the experimental flow representing selection and treatment of animals as well as measurements carried out. n, number of animals.
215
was used. Neutralizing activities in monkeys treated with mut-C42 were analyzed using a Student’s t-test for paired values, two-tailed distribution. A Student’s t-test for two samples with unequal variance (heteroscedastic) was used for the analysis of SKC-2 and mut-C42 treatments. A value of P B 0.05 was considered to indicate statistical significance.
3. Results Sera from 14 monkeys were tested in an anti-SKC-2 ELISA and animals were divided in two groups based on the results: eight monkeys without previous antiSKC-2 Ab titers (group A) and six monkeys with previous anti-SKC-2 Ab titers (group B) (Fig. 1, Table 1). Such Abs are probably a consequence of previous contacts with streptococcus; however, no SKC-2 neutralizing capacity was detected in animals from group B (data not shown). Half of monkeys from each group were treated with SKC-2 and the other half with mutC42. The comparative antigenicity of mutant protein mut-C42 versus native SKC-2 was studied after subcutaneous administrations in groups A and B. Humoral response was quantified at week 8 after 4 administrations, for group A; and at week 2 after one administration, for group B (Fig. 1). Anti-SKC-2 Ab titers rose post-treatment, but animals from group B developed titers notably higher than those from group A (Table 1). Ab titers from group A were slightly lower for monkeys treated with mut-C42 compared with those treated with SKC-2. There are two particular monkeys (33 and 85) showing very low Ab titers. Ab titers generated by animals from group B showed no differences between treatments. Animal sera were also subjected to a neutralization assay in order to determine their neutralizing activity titer (NAT) against SKC-2 and mut-C42 (Fig. 1). Abs from most of the monkeys inhibited the formation of SKC-2-Plg and mut-C42-Plg activator complexes in vitro (Table 2). SKC-2-NAT developed by monkeys from group A were considerably lower than neutralizing capacity exhibited by group B. However, mut-C42NAT values were similar for both groups. Monkeys from group A treated with SKC-2 showed NAT values ranging between 35.43 and 54.17 mg (45.39 8.33) of SKC-2 and between 0 and 19.3 mg (9.139 8.47) of mut-C42 moiety neutralized/ml of tested serum (Table 2). Sera from monkeys treated with mut-C42 elicited NAT values ranging between 6.79 and 44 mg (24.3920) of SKC-2 and between 0 and 14.12 mg (7.59 8.69) of mut-C42 moiety neutralized per ml of tested serum. Interestingly, animals 33 and 85, showing low anti-SKC-2 Ab titers, exhibited insignificant or no NAT against both proteins.
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
216
Table 1 Total anti-SKC-2 Ab titers in monkeys before and after treatment with SKC-2 or mut-C42a Treatment
SKC-2
mut-C42
Group A Animal c
Week 0
Week 8
18 21 73 321 Mean SD
0 0 0 0
64 50 256 256 156.5 115.03
6 33 78 85 Mean SD
0 0 0 0
100 16 256 16 97 113.15
Group B Animal c
Week 0
Week 2
3 42 66
5 8 30
640 640 240
Mean SD
14.33 13.65
506.66 230.94
23 26 79
2 5 10
640 260 520
Mean SD
5.67 4.04
473.33 194.25
a Group A: monkeys without previous anti-SKC-2 Ab titer. Group B: monkeys with previous anti-SKC-2 Ab titer. Half of monkeys from each group received treatment with SKC-2 and the other half with mut-C42. The anti-SKC-2 Ab titer is represented as the maximal dilution in which positive signal was obtained. Positive signal was considered when the value was at least 2-fold the background.
Monkeys from group B showed no SKC-2 neutralizing capacity before treatment (data not shown). However, after only one administration of the studied proteins, Abs from most of the monkeys inhibited the formation of SKC-2-Plg and mut-C42-Plg activator complexes in vitro (Table 2). Monkeys treated with SKC-2 showed NAT values ranging between 89.25 and 241.9 mg (160.75 9 76.77) of SKC-2 and between 0 and 21.7 mg (9.42 911.13) of mut-C42 moiety neutralized per ml of tested serum. Sera from monkeys treated with mut-C42 elicited NAT values ranging between 184.29 and 374.11 mg (290.67996.96) of SKC-2 and between 15.55 and 33.9 mg (24.69 9.17) of mut-C42 moiety neutralized per ml of tested serum. Statistical analyses supported the following results (Table 2): (a) mut-C42 was significantly less affected than SKC-2 by neutralizing Abs from monkeys treated with the native protein (P =0.0042 for group A and P= 0.0369 for group B), (b) the same result was obtained for group B animals treated with mut-C42 (P= 0.0394), (c) in contrast, monkeys from group A receiving mut-C42 treatment showed no significant differences between SKC-2- and mut-C42-neutralizing activities (P=0.0621), and (d) within each group, no statistical significance was obtained from comparison between SKC-2 and mut-C42 treatments. 4. Discussion The use of SK for the treatment of thrombosis is commonly associated with allergic reactions provoked by anti-SK Abs and induction of neutralizing Abs. This fraction from total anti-SK Ab response can neutralize large doses of SK in the first treatment, and could probably limit the use of SK for a second therapy [10,15–17].
The present study was based on evidences about the presence of an immunodominant epitope in the C-terminal region of SKC-2 [25], which can be eliminated without loss of Plg specific activity (submitted for publication). Deletion mutant mut-C42 showed to be less recognized and less neutralized by Abs present in sera from patients treated with SKC-2, with respect to the native protein (submitted for publication). Fourteen monkeys were divided in two groups according to the presence or not of previous anti-SKC-2 Abs (Fig. 1, Table 1). Half of the monkeys from each group were treated with SKC-2 and the other half with mut-C42. Treatments did not include the use of adjuvants to reproduce conditions of clinical application of SK and because of the risk of exposure of neoantigens after denaturation of the molecules. Immunological response was assessed by determination of total antiSKC-2 Ab titer and NAT. All monkeys developed anti-SKC-2 Ab titers, indicating that SKC-2 and mut-C42 are antigenic in monkeys as well as SKC-2 in humans [5–13]. Monkeys from group B, receiving only one administration, showed higher anti-SKC-2 Ab titers than animals from group A after four administrations (Table 1). This is related to the previous existence of low Ab levels in group B. Such Abs probably became expanded upon treatment. This result is similar in humans, who were previously immunized against SK by common streptococcal infections and whose Ab titers significantly rose post-SK therapy [8–13]. Therefore, the species selection was correct from the point of view of its similarity to humans. Ab generation in group A was slightly lower for animals treated with mut-C42 with respect to those receiving SKC-2 (Table 1), although these differences were not statistically significant. Group B showed no
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
217
the molecule. Monkeys from group B treated with mut-C42 showed SKC-2-NAT significantly higher than mutC42-NAT (Table 2). In this case animal sera contained low levels of anti-SKC-2 Abs before treatment. Abs directed against the C-terminal region of the protein could be present among them. Mut-C42 treatment triggered the expansion of anti-SKC-2 Ab producing cells, resulting in high levels of Abs directed against the C-terminus as well as Abs directed against other regions of the molecule. In the presence of the native protein, these Abs bind and neutralize the molecule, however, in the presence of mut-C42 they can not bind their epitopes or neutralize the protein. Animals from group A treated with mut-C42 exhibited SKC-2-NAT slightly higher than mut-C42-NAT (Table 2), but these differences were not statistically significant. In this case monkey sera did not contain anti-SKC-2 Abs before treatment and therefore, Abs acquired upon treatment were directed against several regions of the molecule, but never against the C-terminus. That is why these Abs developed similar neutralizing capacity against both proteins. It is important to notice that monkeys 33 and 85 developed insignificant or no NAT against both molecules (Table 2). Interestingly, the same animals showed very low anti-SKC-2 Ab titers, indicating that SKC-2 neutralizing Abs and total anti-SKC-2 Abs are closely related. Although a linear correlation was not found, a tendency to increase neutralizing activity with growing Ab titers was observed, suggesting that neutralizing Abs are a fraction from total Abs, and that this fraction is variable among individuals.
differences in Ab titers between monkeys subjected to different treatments. SKC-2 and mut-C42 molecules differ in an immunodominant 42-residue C-terminal fragment. The fact that both proteins provoke a similar response despite their structural differences can be explained by taking into account their conformational features. mut-C42 molecule in vivo is likely to adopt spatial conformations such that certain regions, hidden in the native molecule, become exposed and function as highly immunogenic epitopes. Although this protein lacks the C-terminal region, it probably interacts with the immune system through other regions, inducing an Ab level very similar to that induced by the native protein. Most of the monkeys, despite the treatment received, showed neutralizing capacity against SKC-2 and mut-C42. Animals from group A showed SKC-2NAT notably lower than those from group B, regardless of the protein used as antigen (Table 2). This is closely related to the differences in anti-SKC-2 Ab titers between both groups and to the enhancement of primary response, as discussed above. In both groups, monkeys treated with SKC-2 developed neutralizing capacity against SKC-2 significantly higher than their ability to neutralize mut-C42 (Table 2), indicating that the 42-residue C-terminal region of SKC-2 contains one or more important epitopes for induction of neutralizing Abs. In this case animal sera contained Abs directed against the C-terminal region of the protein. When assayed in the presence of mutC42, these particular Abs were not able to bind to their epitopes and therefore, incapable to neutralize
Table 2 Neutralizing activity titer (NAT) of monkeys sera against SKC-2 and mut-C42 proteinsa Group A (week 8) Treatment
SKC-2
mut-C42
P
Group B (week 2)
Animal c
SKC-2
mut-C42
18 21 73 321 Mean SD
35.430 41.806 54.176 49.797 45.302 8.339
12.371 0.000 4.866 19.304 9.135 8.477
6 33 78 85 Mean SD
44.032 6.792 39.084 7.382 24.322 20.005 0.1247
14.121 0.000 15.881 0.000 7.500 8.690 0.7967
P
0.0042
0.0621
Animal c
SKC-2
mut-C42
3 42 66
241.897 89.256 151.102
6.580 0.000 21.708
Mean SD
160.752 76.777
9.429 11.131
23 26 79
374.112 184.289 313.603
33.905 24.357 15.554
Mean SD
290.668 96.967 0.1467
24.605 9.178 0.1451
P
0.0369
0.0394
a NAT was expressed as micrograms of SKC-2 and mut-C42 neutralized per milliliter of tested serum. Statistical significance of the differences for neutralizing activities in monkeys treated with SKC-2 was determined by Student’s t-test for paired values, one-tailed distribution. Neutralizing activities in monkeys treated with mut-C42 were analyzed using a Student’s t test for paired values, two-tailed distribution. Statistical significance of the differences for SKC-2 and mut-C42 treatments was determined by Student’s t-test for two-samples with unequal variance (PB0.05).
218
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
Our results indicate that in every case where treatment induced Abs directed against the C-terminus of SKC-2 (due to the presence of this region in the protein used for treatment and/or to the existence of low anti-SKC-2 Ab levels before treatment), neutralizing activity against native protein was significantly higher than the one developed against mutant mut-C42. Results obtained with animals from group B together with their similarities with the human case, as mentioned above, allow us to make some comments on patient related aspects. First, patients having neutralizing Ab levels high enough to neutralize elevated SK doses in the first treatment, rendering therapy unsuccessful, could have a better chance with an alternative SK lacking the 42 C-terminal amino acids, like mut-C42. Second, 50% of patients receiving a first treatment with SK develop Ab levels sufficient to neutralize the SK dose up to four years later [10,15,16]. These Ab are likely to interfere with therapy in a second thrombotic episode. Our results suggest that such patients could also benefit from treatment with this mutant SK. Finally, patients receiving mut-C42 in the first treatment might also have better possibilities in a second thrombotic episode because mut-C42 might generate less neutralizing activity than treatment with native SK.
Acknowledgements The authors gratefully acknowledge the excellent technical assistance of Pedro Puentes. The authors are also grateful to Dr Elder Pupo for kindly providing a pure mutant protein.
References [1] ISIS-3 (Third International Study of Infarct Survival) Collaborative Group Lancet 339 (1992) 753–781.
.
[2] D. Collen, H.R. Lijnen, Blood 78 (1991) 3114 – 3124. [3] W.S. Tillet, E.D. Edwards, R.L. Garner, J. Clin. Invest. 13 (1934) 47 – 78. [4] O.L. Bruserud, L. Sollid, P. Foyn-Jorgensen, J. Clin. Lab. Immun. 20 (1986) 69 – 74. [5] O.L. Bruserud, APMIS 98 (1990) 1077 – 1084. [6] O.L. Bruserud, S. Elsayed, G. Pawelec, Mol. Immunol. 29 (1992) 1097 – 1104. [7] L.H. Youkeles, M.Y. Soliman, D.L. Rosenstreich, J. Allergy Clin. Immun. 88 (1991) 166 – 171. [8] M. Lynch, W.A. Littler, B.L. Pentecost, R.A. Sockly, Br. Heart J. 66 (1991) 139 – 142. [9] M. Lynch, B.L. Pantecost, W.A. Littler, R.A. Stockley, Clin. Exp. Immunol. 96 (1994) 427 – 431. [10] J.M. Elliott, D.B. Cross, S.A. Cederholm-Williams, H.D. White, Am. J. Cardiol. 71 (1993) 640 – 645. [11] H.S. Lee, S. Cross, R. Davidson, T. Reid, K. Jennings, Eur. Heart J. 14 (1993) 84 – 89. [12] S. Patel, S. Jalihal, D.P. Dutka, G.K. Morris, Br. Heart J. 70 (1993) 119 – 121. [13] B. Parhami-Seren, M. Lynch, H.D. White, G.L. Reed, Mol. Immunol. 32 (1995) 717 – 724. [14] A.S. Lew, T.M.T. Neer, R.N. RodrI`guez, I.L. Geft, P.K. Shah, W. Ganz, J. Am. Coll. Cardiol. 4 (1984) 183 – 185. [15] V.J. Marder, S.N. Sherry, New Engl. J. Med. 318 (1988) 1512– 1520. [16] D. Cross, Aust. N.Z. J. Med. 23 (1993) 749 – 752. [17] D.M. Massel, A.G.G. Turpie, F.A. Ofosu, M.R. Buchanan, Circulation 84 (suppl, 2) (1991) 167. [18] K.A. Davis, P. Mathieson, C.G. Winearls, A.J. Rees, M.I. Walfort, Clin. Exp. Immunol. 80 (1990) 83 – 88. [19] K.G. McGrath, R. Patterson, J. Am. Med. Assoc. 252 (1984) 1314 – 1317. [20] K.G. McGrath, B. Zeffren, J. Alexander, K. Kaplan, R. Patterson, J. Allergy Clin. Immun. 76 (1984) 453 – 457. [21] D.H. Schweitzer, E.E. Van der Wall, H.A. Bosker, E. Scheffer, J.D. Macfarlane, Cardiology 78 (1991) 68 – 71. [22] J.T. Rivers, H.D. White, H.D. Cross, B.F. Williams, R.M. Norris, J. Am. Coll. Cardiol. 16 (1990) 340 – 348. [23] I.C. Kim, J.S. Kim, S.H. Lee, S.M. Byun, Biochem. Mol. Biol. Int. 40 (1996) 939 – 945. [24] M.P. Estrada, L. Hern·ndez, A. Pe`rez, et al., Bio-Technol. 10 (1992) 1138 – 1142. [25] I. Torre`ns, O. Reyes, A.G. Ojalvo, A. Seralena, G. Chinea, L.J. Cruz, J. de la Fuente, Biochem. Bioph. Res. Co. 259 (1999) 162 – 168.
www.elsevier.com/locate/
A mutant streptokinase lacking the C-terminal 42 amino acids is less immunogenic Isis Torre`ns a,*, Ariana G. Ojalvo a, Alina Seralena a, Orlando Hayes b, Jose` de la Fuente b b
a Di6ision of Pharmaceutical, Centro de Ingenierı`a Gene`tica y Biotecnologı`a, P.O. Box 6162, Ha6ana, Cuba Di6ision of Mammalian Cell Genetics, Centro de Ingenierı`a Gene`tica y Biotecnologı`a, P.O. Box 6162, Ha6ana, Cuba
Received 10 August 1999; accepted 6 September 1999
Abstract Streptokinase (SK) is the most widely used compound for the treatment of myocardial infarction and the least expensive thrombolytic agent, but a drawback to its use is the widespread presence of anti-SK antibodies (Abs). Clinical failure of the activation of the fibrinolytic system by SK has been reported due to the presence of a high titer of anti-SK neutralizing Abs. Patients receiving SK therapy develop high anti-SK antibody titers, which might provoke severe allergic reactions. These Abs are sufficient to neutralize a standard dose of SK up to four years after initial SK administration. This is a clinical problem because of the increasing number of patients who have been treated once with SK for acute myocardial infarction (AMI) and are likely to require plasminogen activator treatment in the future. In previous in vitro studies, we have shown that a deletion mutant (mut-C42), lacking the 42 C-terminal residues, was significantly less antigenic when compared with the native molecule (SKC-2). In this study, 14 monkeys were subjected to treatment with SKC-2 and mut-C42 in order to compare their humoral response by determining SK neutralizing activity in monkey’s sera. All monkeys developed anti-SKC-2 Ab titers, but in the case where treatment induced Abs directed against the C-terminus of SKC-2, neutralizing activity against the native protein was significantly higher than that developed against mutant SK mut-C42. © 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Streptokinase; Immunology; Antibodies; Neutralizing antibodies
1. Introduction Thrombotic complications of cardiovascular diseases are a main cause of death and consequently, thrombolytic therapy has become an accepted form of treatment for AMI [1]. Thrombolytic agents are plasminogen (Plg) activators that convert Plg, the inactive proenzyme of the fibrinolytic system in blood, to proteolytic enzyme plasmin (Pl), which dissolves the fibrin of a blood clot [2]. Streptokinase (SK) is a streptococcal protein (MW 47 000) specially used for the treatment of venous thromboembolic diseases and AMI wherein it has been demonstrated to be virtually as efficacious as its more expensive alternatives, namely urokinase (UK) and tissue-type Plg activator (TPA) [1]. * Corresponding author. Fax: +53-7-336008/214764. E-mail address: [email protected] (I. Torre`ns)
SK is a bacterial protein and therefore antigenic in humans. Since streptococcal infections are common, adults are immunized with SK, and antibodies (Abs) to SK can be detected in most of them [3]. SK therapy generates significant T-cell responses to SK [4–7] and the neutralizing capacity of the Abs rise significantly [8–13]. Clinical failure of the activation of the fibrinolytic system by SK has been reported due to the presence of a high titer of neutralizing anti-SK Abs [14]. After SK administration, IgG Abs rise to a peak at two weeks and slowly fall over the next 6–12 months [15], but 50% of patients still have Ab levels sufficient to neutralize a standard dose of SK up to four years after initial SK administration [10,15,16]. High circulating anti-SK Ab titers are likely to influence the efficacy of the thrombolytic therapy [17] and cause allergic reactions [18–21]. Therefore, SK should not be re-administered, except perhaps in the first 2–3 days after initial treatment [16]. Twenty percent of patients with
0165-2478/99/$ - see front matter © 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 7 8 ( 9 9 ) 0 0 1 5 1 - 0
214
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
AMI have had a previous infarction, and reinfarction within 1 year after infarction occurs in 9% of cases [16,22]. Therefore, the engineering of a SK molecule with reduced immunoreactivity would allow its effective readministration in patients with reinfarction. SK is a heterologous protein, and it is not obvious that its immunoreactivity can be reduced by protein engineering. Furthermore, a deletion mutant lacking 46 C-terminal amino acids has about one-quarter of the specific activity of SK [23], suggesting that it is very sensitive to inactivation by site directed mutagenesis. In a previous study we have shown that native SKC2 (SK encoded by skc-2 gene from Streptococcus equisimilis group C, ATCC 9542 [24]) contains an immunodominant epitope in the C-terminal region [25]. The SKC-2 mutant protein (mut-C42), lacking 42 Cterminal residues was obtained after its expression in E. coli, without loss of Plg-specific activity (submitted for publication). We have compared the new protein mutC42 with the native SKC-2 regarding their antigenic properties, by means of direct binding and competition assays using human sera from patients treated with Heberkinase (Heber Biotec S.A., Havana, Cuba). Heberkinase contains a recombinant SKC-2 obtained after the expression of the skc-2 gene in E. coli [24]. Our results showed that recognition of mut-C42 by antiSKC-2 Abs from patients sera was 51% and 68% of reactivity to native SKC-2, as assessed by direct binding and competition assays, respectively (submitted for publication). Both, mutant and native proteins were also tested in a neutralizing activity assay using patients’ sera. For most of the individuals, mut-C42-neutralizing activity titer (NAT) significantly decreased with respect to SKC-2-NAT, ranging from 30 to 91% of the native protein value (submitted for publication). This previous study suggested that it is possible to produce an engineered variant of SK being functional and less antigenic than the native molecule in vitro. If this variant is also less antigenic in vivo it could constitute a preferred alternative to native SK for thrombolytic therapy in patients with thromboembolic diseases. In the present study, we report that mut-C42 biological activity was less affected by neutralizing Abs when compared with native SKC-2, after administration in non-human primates.
2. Materials and methods
2.1. Animal Fourteen monkeys (Cercopithecus aethiops) of either sex, between 2 and 3 years old, weighing 1.8 – 2.5 kg, were selected for the study. The experimental methods were approved by the Animal Care and Use Committee of the Center for Genetic Engineering and Biotechnol-
ogy and were conducted in accordance with the Health Guide for the Care and Use of Laboratory Animals.
2.2. Titration of monkey sera by anti-SK ELISA Total serum from monkeys was collected. Polyvinyl plates (Costar, Cambridge, Massachusetts, USA) were coated with 10 mg/ml SKC-2 in coating buffer (0.1 M Na2CO3, 0.1 M NaHCO3, pH 9.6), and incubated overnight at 4°C. Then, plates were washed three times with 0.05% Tween 20 in PBS (PBS-Tween). One hundred microlitres of serial dilutions of each monkey serum was added. After incubation for 1 h at 37°C, plates were incubated with a biotinylated protein A solution at 1:3000 dilution. After incubation for 1 h at 37°C, the binding of monkey Abs to SKC-2 was measured using a horseradish peroxidase-conjugated streptavidin (Sigma). The reaction was developed using 100 ml per well of 1 mg/ml o-phenylenediamine (Sigma), 0.03% H2O2 in substrate buffer (0.1 M citric acid, 0.2 M Na2HPO4, pH 5.0). After 30 min, the reaction was stopped with 50 ml of 4 M H2SO4. Results were measured on a Multiskan system (Titertek, Helsinki, Finland) at 492 nm. The anti-SKC-2 Ab titer was determined as the maximal dilution in which positive signal was obtained. Positive signal was considered when the value was at least 2-fold the background.
2.3. Treatment of monkeys with mut-C42 and SKC-2 Treatment was carried out by subcutaneous administration of 425 mg/kg mut-C42 and SKC-2 without Freund’s adjuvant. Animals were divided in two groups: group A (eight monkeys without previous antiSKC-2 Ab titer) and group B (six monkeys with previous anti-SKC-2 Ab titer) (Fig. 1). Half of monkeys from each group were treated with SKC-2 and the other half with mut-C42. Proteins were administered at weeks 0, 2, 4, and 6 for group A, and at week 0 for group B. The humoral response was quantified at week 8 for group A, and at week 2 for group B.
2.4. mut-C42 - and SKC-2 -neutralizing acti6ity in monkey sera The chromogenic substrate (S-2251) reaction was carried out in polyvinyl plates (Costar, Cambridge, Massachusetts, USA). Serial dilutions of SKC-2 and mut-C42 (128-2 IU, 2-fold dilutions in 20 mM Tris– HCl pH8/0.5 M NaCl) were prepared in a volume of 25 ml each one. SKC-2 and mut-C42 curves were mixed with 25 ml of each monkey diluted serum and a negative control. For monkeys without previous anti-SKC-2 Ab titer a 1:2 dilution was used, and sera from monkeys with previous anti-SKC-2 Ab titer were diluted 1:5. The negative control was a monkey serum without anti-
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
SKC-2 Ab titer. Fifty microliters of 25 mg/ml human Plg were added and allowed to mix for 10 min at room temperature. The reaction was developed by addition of 50 ml of chromogenic substrate S-2251 (Chromogenix, Antwerp, Belgium). After incubation for 30 min, the reaction was stopped with 25 ml of 20% acetic acid. Results were measured on a Multiskan system (Titertek, Helsinki, Finland) at 405 nm. The activity of SKC-2 required to obtain an absorbance of 0.7 was determined from plots of absorbance versus activity. The neutralizing activity titer (NAT) was determined as the difference between the tested serum and negative control value and was expressed as micrograms of SKC-2 and mut-C42 neutralized per milliliter of tested serum.
2.5. Statistical analysis In order to analyze SKC-2- and mut-C42-neutralizing activities in monkeys treated with SKC-2, a Student’s t-test for paired values, one-tailed distribution,
Fig. 1. Diagram of the experimental flow representing selection and treatment of animals as well as measurements carried out. n, number of animals.
215
was used. Neutralizing activities in monkeys treated with mut-C42 were analyzed using a Student’s t-test for paired values, two-tailed distribution. A Student’s t-test for two samples with unequal variance (heteroscedastic) was used for the analysis of SKC-2 and mut-C42 treatments. A value of P B 0.05 was considered to indicate statistical significance.
3. Results Sera from 14 monkeys were tested in an anti-SKC-2 ELISA and animals were divided in two groups based on the results: eight monkeys without previous antiSKC-2 Ab titers (group A) and six monkeys with previous anti-SKC-2 Ab titers (group B) (Fig. 1, Table 1). Such Abs are probably a consequence of previous contacts with streptococcus; however, no SKC-2 neutralizing capacity was detected in animals from group B (data not shown). Half of monkeys from each group were treated with SKC-2 and the other half with mutC42. The comparative antigenicity of mutant protein mut-C42 versus native SKC-2 was studied after subcutaneous administrations in groups A and B. Humoral response was quantified at week 8 after 4 administrations, for group A; and at week 2 after one administration, for group B (Fig. 1). Anti-SKC-2 Ab titers rose post-treatment, but animals from group B developed titers notably higher than those from group A (Table 1). Ab titers from group A were slightly lower for monkeys treated with mut-C42 compared with those treated with SKC-2. There are two particular monkeys (33 and 85) showing very low Ab titers. Ab titers generated by animals from group B showed no differences between treatments. Animal sera were also subjected to a neutralization assay in order to determine their neutralizing activity titer (NAT) against SKC-2 and mut-C42 (Fig. 1). Abs from most of the monkeys inhibited the formation of SKC-2-Plg and mut-C42-Plg activator complexes in vitro (Table 2). SKC-2-NAT developed by monkeys from group A were considerably lower than neutralizing capacity exhibited by group B. However, mut-C42NAT values were similar for both groups. Monkeys from group A treated with SKC-2 showed NAT values ranging between 35.43 and 54.17 mg (45.39 8.33) of SKC-2 and between 0 and 19.3 mg (9.139 8.47) of mut-C42 moiety neutralized/ml of tested serum (Table 2). Sera from monkeys treated with mut-C42 elicited NAT values ranging between 6.79 and 44 mg (24.3920) of SKC-2 and between 0 and 14.12 mg (7.59 8.69) of mut-C42 moiety neutralized per ml of tested serum. Interestingly, animals 33 and 85, showing low anti-SKC-2 Ab titers, exhibited insignificant or no NAT against both proteins.
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
216
Table 1 Total anti-SKC-2 Ab titers in monkeys before and after treatment with SKC-2 or mut-C42a Treatment
SKC-2
mut-C42
Group A Animal c
Week 0
Week 8
18 21 73 321 Mean SD
0 0 0 0
64 50 256 256 156.5 115.03
6 33 78 85 Mean SD
0 0 0 0
100 16 256 16 97 113.15
Group B Animal c
Week 0
Week 2
3 42 66
5 8 30
640 640 240
Mean SD
14.33 13.65
506.66 230.94
23 26 79
2 5 10
640 260 520
Mean SD
5.67 4.04
473.33 194.25
a Group A: monkeys without previous anti-SKC-2 Ab titer. Group B: monkeys with previous anti-SKC-2 Ab titer. Half of monkeys from each group received treatment with SKC-2 and the other half with mut-C42. The anti-SKC-2 Ab titer is represented as the maximal dilution in which positive signal was obtained. Positive signal was considered when the value was at least 2-fold the background.
Monkeys from group B showed no SKC-2 neutralizing capacity before treatment (data not shown). However, after only one administration of the studied proteins, Abs from most of the monkeys inhibited the formation of SKC-2-Plg and mut-C42-Plg activator complexes in vitro (Table 2). Monkeys treated with SKC-2 showed NAT values ranging between 89.25 and 241.9 mg (160.75 9 76.77) of SKC-2 and between 0 and 21.7 mg (9.42 911.13) of mut-C42 moiety neutralized per ml of tested serum. Sera from monkeys treated with mut-C42 elicited NAT values ranging between 184.29 and 374.11 mg (290.67996.96) of SKC-2 and between 15.55 and 33.9 mg (24.69 9.17) of mut-C42 moiety neutralized per ml of tested serum. Statistical analyses supported the following results (Table 2): (a) mut-C42 was significantly less affected than SKC-2 by neutralizing Abs from monkeys treated with the native protein (P =0.0042 for group A and P= 0.0369 for group B), (b) the same result was obtained for group B animals treated with mut-C42 (P= 0.0394), (c) in contrast, monkeys from group A receiving mut-C42 treatment showed no significant differences between SKC-2- and mut-C42-neutralizing activities (P=0.0621), and (d) within each group, no statistical significance was obtained from comparison between SKC-2 and mut-C42 treatments. 4. Discussion The use of SK for the treatment of thrombosis is commonly associated with allergic reactions provoked by anti-SK Abs and induction of neutralizing Abs. This fraction from total anti-SK Ab response can neutralize large doses of SK in the first treatment, and could probably limit the use of SK for a second therapy [10,15–17].
The present study was based on evidences about the presence of an immunodominant epitope in the C-terminal region of SKC-2 [25], which can be eliminated without loss of Plg specific activity (submitted for publication). Deletion mutant mut-C42 showed to be less recognized and less neutralized by Abs present in sera from patients treated with SKC-2, with respect to the native protein (submitted for publication). Fourteen monkeys were divided in two groups according to the presence or not of previous anti-SKC-2 Abs (Fig. 1, Table 1). Half of the monkeys from each group were treated with SKC-2 and the other half with mut-C42. Treatments did not include the use of adjuvants to reproduce conditions of clinical application of SK and because of the risk of exposure of neoantigens after denaturation of the molecules. Immunological response was assessed by determination of total antiSKC-2 Ab titer and NAT. All monkeys developed anti-SKC-2 Ab titers, indicating that SKC-2 and mut-C42 are antigenic in monkeys as well as SKC-2 in humans [5–13]. Monkeys from group B, receiving only one administration, showed higher anti-SKC-2 Ab titers than animals from group A after four administrations (Table 1). This is related to the previous existence of low Ab levels in group B. Such Abs probably became expanded upon treatment. This result is similar in humans, who were previously immunized against SK by common streptococcal infections and whose Ab titers significantly rose post-SK therapy [8–13]. Therefore, the species selection was correct from the point of view of its similarity to humans. Ab generation in group A was slightly lower for animals treated with mut-C42 with respect to those receiving SKC-2 (Table 1), although these differences were not statistically significant. Group B showed no
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
217
the molecule. Monkeys from group B treated with mut-C42 showed SKC-2-NAT significantly higher than mutC42-NAT (Table 2). In this case animal sera contained low levels of anti-SKC-2 Abs before treatment. Abs directed against the C-terminal region of the protein could be present among them. Mut-C42 treatment triggered the expansion of anti-SKC-2 Ab producing cells, resulting in high levels of Abs directed against the C-terminus as well as Abs directed against other regions of the molecule. In the presence of the native protein, these Abs bind and neutralize the molecule, however, in the presence of mut-C42 they can not bind their epitopes or neutralize the protein. Animals from group A treated with mut-C42 exhibited SKC-2-NAT slightly higher than mut-C42-NAT (Table 2), but these differences were not statistically significant. In this case monkey sera did not contain anti-SKC-2 Abs before treatment and therefore, Abs acquired upon treatment were directed against several regions of the molecule, but never against the C-terminus. That is why these Abs developed similar neutralizing capacity against both proteins. It is important to notice that monkeys 33 and 85 developed insignificant or no NAT against both molecules (Table 2). Interestingly, the same animals showed very low anti-SKC-2 Ab titers, indicating that SKC-2 neutralizing Abs and total anti-SKC-2 Abs are closely related. Although a linear correlation was not found, a tendency to increase neutralizing activity with growing Ab titers was observed, suggesting that neutralizing Abs are a fraction from total Abs, and that this fraction is variable among individuals.
differences in Ab titers between monkeys subjected to different treatments. SKC-2 and mut-C42 molecules differ in an immunodominant 42-residue C-terminal fragment. The fact that both proteins provoke a similar response despite their structural differences can be explained by taking into account their conformational features. mut-C42 molecule in vivo is likely to adopt spatial conformations such that certain regions, hidden in the native molecule, become exposed and function as highly immunogenic epitopes. Although this protein lacks the C-terminal region, it probably interacts with the immune system through other regions, inducing an Ab level very similar to that induced by the native protein. Most of the monkeys, despite the treatment received, showed neutralizing capacity against SKC-2 and mut-C42. Animals from group A showed SKC-2NAT notably lower than those from group B, regardless of the protein used as antigen (Table 2). This is closely related to the differences in anti-SKC-2 Ab titers between both groups and to the enhancement of primary response, as discussed above. In both groups, monkeys treated with SKC-2 developed neutralizing capacity against SKC-2 significantly higher than their ability to neutralize mut-C42 (Table 2), indicating that the 42-residue C-terminal region of SKC-2 contains one or more important epitopes for induction of neutralizing Abs. In this case animal sera contained Abs directed against the C-terminal region of the protein. When assayed in the presence of mutC42, these particular Abs were not able to bind to their epitopes and therefore, incapable to neutralize
Table 2 Neutralizing activity titer (NAT) of monkeys sera against SKC-2 and mut-C42 proteinsa Group A (week 8) Treatment
SKC-2
mut-C42
P
Group B (week 2)
Animal c
SKC-2
mut-C42
18 21 73 321 Mean SD
35.430 41.806 54.176 49.797 45.302 8.339
12.371 0.000 4.866 19.304 9.135 8.477
6 33 78 85 Mean SD
44.032 6.792 39.084 7.382 24.322 20.005 0.1247
14.121 0.000 15.881 0.000 7.500 8.690 0.7967
P
0.0042
0.0621
Animal c
SKC-2
mut-C42
3 42 66
241.897 89.256 151.102
6.580 0.000 21.708
Mean SD
160.752 76.777
9.429 11.131
23 26 79
374.112 184.289 313.603
33.905 24.357 15.554
Mean SD
290.668 96.967 0.1467
24.605 9.178 0.1451
P
0.0369
0.0394
a NAT was expressed as micrograms of SKC-2 and mut-C42 neutralized per milliliter of tested serum. Statistical significance of the differences for neutralizing activities in monkeys treated with SKC-2 was determined by Student’s t-test for paired values, one-tailed distribution. Neutralizing activities in monkeys treated with mut-C42 were analyzed using a Student’s t test for paired values, two-tailed distribution. Statistical significance of the differences for SKC-2 and mut-C42 treatments was determined by Student’s t-test for two-samples with unequal variance (PB0.05).
218
I. Torre`ns et al. / Immunology Letters 70 (1999) 213–218
Our results indicate that in every case where treatment induced Abs directed against the C-terminus of SKC-2 (due to the presence of this region in the protein used for treatment and/or to the existence of low anti-SKC-2 Ab levels before treatment), neutralizing activity against native protein was significantly higher than the one developed against mutant mut-C42. Results obtained with animals from group B together with their similarities with the human case, as mentioned above, allow us to make some comments on patient related aspects. First, patients having neutralizing Ab levels high enough to neutralize elevated SK doses in the first treatment, rendering therapy unsuccessful, could have a better chance with an alternative SK lacking the 42 C-terminal amino acids, like mut-C42. Second, 50% of patients receiving a first treatment with SK develop Ab levels sufficient to neutralize the SK dose up to four years later [10,15,16]. These Ab are likely to interfere with therapy in a second thrombotic episode. Our results suggest that such patients could also benefit from treatment with this mutant SK. Finally, patients receiving mut-C42 in the first treatment might also have better possibilities in a second thrombotic episode because mut-C42 might generate less neutralizing activity than treatment with native SK.
Acknowledgements The authors gratefully acknowledge the excellent technical assistance of Pedro Puentes. The authors are also grateful to Dr Elder Pupo for kindly providing a pure mutant protein.
References [1] ISIS-3 (Third International Study of Infarct Survival) Collaborative Group Lancet 339 (1992) 753–781.
.
[2] D. Collen, H.R. Lijnen, Blood 78 (1991) 3114 – 3124. [3] W.S. Tillet, E.D. Edwards, R.L. Garner, J. Clin. Invest. 13 (1934) 47 – 78. [4] O.L. Bruserud, L. Sollid, P. Foyn-Jorgensen, J. Clin. Lab. Immun. 20 (1986) 69 – 74. [5] O.L. Bruserud, APMIS 98 (1990) 1077 – 1084. [6] O.L. Bruserud, S. Elsayed, G. Pawelec, Mol. Immunol. 29 (1992) 1097 – 1104. [7] L.H. Youkeles, M.Y. Soliman, D.L. Rosenstreich, J. Allergy Clin. Immun. 88 (1991) 166 – 171. [8] M. Lynch, W.A. Littler, B.L. Pentecost, R.A. Sockly, Br. Heart J. 66 (1991) 139 – 142. [9] M. Lynch, B.L. Pantecost, W.A. Littler, R.A. Stockley, Clin. Exp. Immunol. 96 (1994) 427 – 431. [10] J.M. Elliott, D.B. Cross, S.A. Cederholm-Williams, H.D. White, Am. J. Cardiol. 71 (1993) 640 – 645. [11] H.S. Lee, S. Cross, R. Davidson, T. Reid, K. Jennings, Eur. Heart J. 14 (1993) 84 – 89. [12] S. Patel, S. Jalihal, D.P. Dutka, G.K. Morris, Br. Heart J. 70 (1993) 119 – 121. [13] B. Parhami-Seren, M. Lynch, H.D. White, G.L. Reed, Mol. Immunol. 32 (1995) 717 – 724. [14] A.S. Lew, T.M.T. Neer, R.N. RodrI`guez, I.L. Geft, P.K. Shah, W. Ganz, J. Am. Coll. Cardiol. 4 (1984) 183 – 185. [15] V.J. Marder, S.N. Sherry, New Engl. J. Med. 318 (1988) 1512– 1520. [16] D. Cross, Aust. N.Z. J. Med. 23 (1993) 749 – 752. [17] D.M. Massel, A.G.G. Turpie, F.A. Ofosu, M.R. Buchanan, Circulation 84 (suppl, 2) (1991) 167. [18] K.A. Davis, P. Mathieson, C.G. Winearls, A.J. Rees, M.I. Walfort, Clin. Exp. Immunol. 80 (1990) 83 – 88. [19] K.G. McGrath, R. Patterson, J. Am. Med. Assoc. 252 (1984) 1314 – 1317. [20] K.G. McGrath, B. Zeffren, J. Alexander, K. Kaplan, R. Patterson, J. Allergy Clin. Immun. 76 (1984) 453 – 457. [21] D.H. Schweitzer, E.E. Van der Wall, H.A. Bosker, E. Scheffer, J.D. Macfarlane, Cardiology 78 (1991) 68 – 71. [22] J.T. Rivers, H.D. White, H.D. Cross, B.F. Williams, R.M. Norris, J. Am. Coll. Cardiol. 16 (1990) 340 – 348. [23] I.C. Kim, J.S. Kim, S.H. Lee, S.M. Byun, Biochem. Mol. Biol. Int. 40 (1996) 939 – 945. [24] M.P. Estrada, L. Hern·ndez, A. Pe`rez, et al., Bio-Technol. 10 (1992) 1138 – 1142. [25] I. Torre`ns, O. Reyes, A.G. Ojalvo, A. Seralena, G. Chinea, L.J. Cruz, J. de la Fuente, Biochem. Bioph. Res. Co. 259 (1999) 162 – 168.