© 1990 S. Karger AG, Basel 0378-7346/90/0293-0169S2.75/0
Gynecol Obstet Invest 1990;29:169-172
Pregnancy-Associated Plasma Protein-A-Induced Inhibition of Human Leukocyte Elastase: An Artifact Paul Bischofa, Alain Gervaixb, Arielle Meisser*, Suzanne Suterb “Département de Gynécologie et d’Obstétrique, et bDépartement de Pédiatrie et Génétique, Université de Genève, Suisse
Key Words. Pregnancy-associated plasma protein A • Leukocyte elastase
Introduction Pregnancy-associated plasma protein A (PAPP-A) is a macromolecular protein first described by Lin et al. [1] in 1974 and purified to homogeneity by others [2-4], It is a homotetrameric glycoprotein (19% CHO) of 800,000 molecular weight, migrating in the ci2-position in an elec tric field. It is thus biochemically but not immunologically similar to cu-macroglobulin (a2M). In contrast to a2M, PAPP-A has the unique property of binding hepa rin [5] and this property was used to develop a purifica tion technique [6] based on heparin-Sepharose affinity chromatography. During pregnancy, PAPP-A, is produced by the trophoblast almost exclusively towards the maternal circu lation [7, 8], Peripheral levels of PAPP-A increase expo nentially during pregnancy with no signs of plateauing near term [9, 10]. Although the biological function of PAPP-A is still unknown, several in vitro effects have been reported. PAPP-A has been described as an inhibi
tor in many in vitro assays: complement activity [11], mitogen-induced lymphoblastogenesis [12], coagulation [13] and granulocyte elastase [14], We have recently reported that the inhibition of complement and coagula tion attributed to PAPP-A was in fact due to heparin contaminating the purified PAPP-A preparations [15]. This contaminant was released during the purification process by the heparin-Sepharose column. One must assume that the inhibition by the same PAPP-A prepara tions of lymphocyte transformation was also due to hep arin contamination since heparin has been described as an inhibitor of phytohemagglutinin-induced lymphocyte transformation [16]. In their report about the inhibitory effects of PAPP-A on granulocyte elastase, Sinosich et al. [14] purified PAPP-A by a technique similar to ours, thus using an affinity step on heparin-Sepharose. Since we have observed that this column releases heparin [15] and since heparin is known to be an inhibitor of elastase [17], we wondered if this inhibitory effect was really due to PAPP-A or if it could be attributed to heparin.
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
Abstract. Pregnancy-associated plasma protein A (PAPP-A), was reported to be an inhibitor in many in vitro systems. Since it was shown that the inhibition of coagulation and complement activity attributed to PAPP-A was in fact due to a contamination by heparin occurring during the purification process, we undertook the present study to see whether the reported PAPP-A-induced inhibition of human leukocyte elastase (HLE) could also be attributed to heparin contamination. PAPP-A was purified from maternal pregnancy EDTA plasma by a method which was previously shown to eliminate contaminating heparin: this preparation was inactive in the HLE assay. But PAPP-A isolated by heparin-Sepharose chromatography, or a PAPP-A-free washing of the heparin-Sepharose column were both inhibitors of HLE. Furthermore the inactive PAPP-A preparation, when incubated with the PAPP-A-free washing of the heparin-Sepharose column, yielded a high molecular weight preparation which inhibited HLE. It is concluded that PAPP-A is not an inhibitor of HLE and that the inhibition of HLE previously attributed to PAPP-A was due to contaminating heparin.
Bischof/Gervaix/Meisser/Suter
170
Materials DEAE-Sepharose, heparin-Sepharosc, polyacrylamide gradient gels (2-16%) and high molecular weight standards were all from Pharmacia (Uppsala, Sweden). Ultrogel AcA 34 was from IBF Biotechnics (Villeneuve-La-Garennc, France), Amicon (XM 100 A) membranes were from Amicon corporation (Oosterhout, Holland), dialysis membranes (Spectrapore) were from Spectrum (Los An geles, Calif.). Antihuman PAPP-A antiserum was prepared as de scribed previously [11]. Purified human leukocyte elastase (HLE) was a gift from Dr. H.P. Schnebli (Ciba-Geigy Ltd., Basel, Switzerland) and the pNAsubstrate (methoxysuccinyl-L-alanyl-L-alanyl-L-prolyl-L-valinc-pnitroanilidc) was from Calbiochem (Lucerne, Switzerland). Purification o f PAPP-A Late maternal pregnancy EDTA plasma (ethylene-diaminotetraacetate 2 mg/ml) was dialyzed against buffer A (0.05 M phosphate containing 0.1 M NaCl and 0.02% azide pH 7.4) and subjected to ion-exchange chromatography on a DEAE-Sepharose column (5 X 15 cm). Elution was carried out with a NaCl gradient (0.1 -1.0 M) in buffer A. PAPP-A-containing fractions, as measured by radioimmu noassay [ 18], were pooled and dialyzed against buffer A. The dialyzed pool was then applied to a heparin-Sepharose col umn ( 5 X 1 5 cm). Washing and elution of this column was carried out in the same way as for the ion exchange. The pooled fractions containing immunoreactive PAPP-A were dialyzed against buffer A and concentrated on an Amicon membrane. This pool was named heparin-Sepharosc pool hereafter. An aliquot of this pool was sub jected to a calibrated [ 15] AcA 34 column (0.9 X 60 cm) and elution carried out with 0.8 M NaCl in buffer A. The PAPP-A-containing fractions were pooled (named AcA 34 pool hereafter), dialyzed against buffer A, concentrated and stored at -2 0 °C. Preparation o f PAPP-A-Heparin Complexes The heparin-Sepharose column was washed with buffer A (in the absence of PAPP-A) and 800 pi of this washing was incubated (2 h at room temperature) with 200 pi of the AcA 34 pool. The mixture was then applied to the AcA 34 column and elution carried out in low salt (0.1 M NaCl in buffer A). Fractions 11-22 (1 ml each) con taining the immunoreactive PAPP-A (fig. 1, pool A) were concen trated and stored at -2 0 °C. Fractions 24-35 (fig. 1, pool B) which did not contain significant amounts of PAPP-A were concentrated in the same way. Polyacrylamide Gel Electrophoresis Polyacrylamide gel electrophoresis was performed on gradient gels (2-16%) in the presence or absence of 0.5% SDS and 4 M urea with or without 5% P-mercaptoethanol as already reported [15], Gels were stained with silver nitrate. HLE Assay HLE activity was measured with a specific substrate (pNA-substrate) as described previously [19]. Briefly, purified HLE (10 nM) in Bulbecco’s PBS pH 7.4 was incubated with 1 0 4 M pNA-substrate and the reaction followed in a spectrophotometer at 387 nm during 2.5 min at room temperature. For inhibition studies, the PAPP-A
samples were preincubated (3 min at room temperature) with HLE in a molar ratio (1:1 to 10:1 PAPP-A:HLE) and the mixture added to the substrate. Results were expressed as percent of controls (HLE without inhibitors).
Results Purification and Polyacrylamide Gel Electrophoresis Elution of the AcA 34 column in the presence of 0.8 M NaCl yielded a symmetrical PAPP-A peak at a MW of about 800,000 [result not shown]. These frac tions were pooled, concentrated and incubated with the PAPP-A-free washing of the heparin-Sepharose column. Submitted to the AcA 34 column but eluted in low salt (0.1 M NaCl in buffer A), the incubate gave a superimposable elution profile (fig. 1). The AcA 34 pool, when run on a nondenaturing poly acrylamide gel (insert fig. 1), gave one unique band of about 800,000 MW. Under denaturing conditions, the PAPP-A preparation was resolved in one major band of about 400,000 MW (fig. 2, lanes 2, 3), one faint band (second from the top) which is a contamination with a2M and a third faint band which is also recognized by anti-PAPP-A antibodies when immunoblotted [MW 200,000, results not shown]. Under reducing conditions (fig. 2, lanes 4, 5), the PAPP-A preparation was resolved in two very close bands of about 180,000 MW. Inhibition o f Human Granulocyte Elastase The PAPP-A from the heparin-Sepharose pool inhib ited the activity of HLE in a dose-dependent manner with an inhibition of 35% at a concentration of 50 nM of PAPP-A (table 1). The elution of this heparin-Sepharose pool on an AcA 34 column in the presence of high salt (0.8 M NaCl) yielded a PAPP-A-containing pool (AcA 34 pool) which did not inhibit HLE up to a concentra tion of 100 nM of PAPP-A. This inactive pool was incu bated with the PAPP-A-free washing of the heparinSepharose column and rerun in low salt on the AcA 34 column. The PAPP-A-containing pool (AcA pool A) inhibited HLE in a dose-dependent manner (table 1). The AcA 34 pool B, which was obtained by pooling low molecular weight fractions from the same elution profile (fig. 1), gave a pool which did not contain PAPP-A (< 2 0 ng/ml)but which inhibited HLE at 100% (table 1). Finally the PAPP-A-free washing of the heparin-Sepha rose column did also inhibit the activity of HLE (ta ble 1).
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
Materials and Methods
PAPP-A and Human Leukocyte Elastase Inhibition
669 440 232 kD
Fractions
Discussion As shown by nondenaturing polyacrylamide gradient gel electrophoresis, our purification protocol yields a preparation which has only one band at 800,000 MW. However, in the presence of SDS a small contamination by a2M can be seen. As shown previously [ 15] the use of a gel filtration column in the presence of high salt (AcA 34 with 0.8 M NaCl) yields a heparin-free PAPP-A which does not inhibit complement activity or throm bin-induced polymerization of fibrin anymore. PAPP-A purified under these conditions did neither inhibit HLE in our assay. In contrast the PAPP-A-containing pool or the PAPP-A-free washing of the heparin-Sepharose col umn were both good inhibitors of HLE. Furthermore when the PAPP-A-free heparin-Sepharose washing was incubated with purified PAPP-A (AcA 34 pool) which did not inhibit HLE, and rerun on the AcA 34 column, inhibition of HLE was observed in the high molecular
Fig. 1. Elution profile on Biogel AcA 34 of purified PAPP-A incubated with the PAPP-A-free washing of the heparin-Sepharose column. Insert: polyacrylamide gradient gel of purified PAPP-A run under nondenaturing conditions (without SDS). Fig. 2. Polyacrylamide gradient gel of purified PAPP-A in the presence of 0.5% SDS with or without ß-mercaptoethanol. Lane 1: high molecular weight standards, lanes 2, 3: denatured PAPP-A, lanes 4, 5: denatured and reduced PAPP-A.
Table 1. Inhibition of granulocyte elastase Pools
Concentrations, nM,
OD, % of control
Heparin-Sepharose
10 20 50
81 72 65
5 10 50 100
94 82 99 80
AcA 34 pool A (low salt)
10 20 50
43 17 10
AcA 34 pool B (low salt)
75 pi, no PAPP-A
0
Heparin-Sepharose washing
75 pi, no PAPP-A
55
AcA 34 (high salt)
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
V0
171
Bischof/Gervaix/Meisser/Suter
172
Acknowledgements The authors wish to thank Mrs. I. Riederer for typing the manu script. This work was supported by grant No. 3.873.086 from the Swiss National Fund for Scientific Research.
References 1 Lin, T.M.; Halbert, S.P.; Kiefer, D.; Spellacy, W.N.; Gall, S.: Characterization of four human pregnancy-associated plasma proteins. Am. J. Obstet. Gynec. 118: 223-236 (1974). 2 Bischof, P.: Purification and characterization of pregnancy-asso ciated plasma protein-A (PAPP-A). Archs Gynecol. 227: 315326 (1979). 3 Folkersen, J.; Westergaard, J.G.; Hindersson, P.; Teisner, B.: Affinity chromatographic purification of a new high molecular weight pregnancy specific protein SP4; in Lehmann, Carcinoembryonic proteins, vol. 2, pp. 503-508 (Elsevier/North Hol land, Amsterdam 1979). 4 Sutcliffe, R.G.; Kukulska-Langlands, B.M.; Coggins, J.R.; Hunt er, J.B.; Gore, G.H.: Studies on human pregnancy-associated plasma protein-A. Purification by affinity chromatography and structural comparison with a 2-macroglobulin. Biochem. J. 191: 799-809 (1980). 5 Sinosich, M.J.; Teisner, B.; Davey, M.; Grudzinskas, J.G.: Preg nancy-associated plasma protein-A: interaction with heparin in crossed affinity immunoelectrophoresis. Aust. N.Z. J. Med. 11: 429-433 (1981). 6 Davey, M.W.; Teisner, B.; Sinosich, M.J.; Grudzinskas, J.G.: Interaction between heparin and human pregnancy-associated
7
8
9
10
11 12
13
14
15
16
17
18
19
plasma protein-A (PAPP-A): a simple purification procedure. Analyt. Biochem. 131: 18-24 (1983). Smith, R.; Bischof, P.; Hughes, G.; Klopper, A.: Studies on preg nancy-associated plasma protein-A in the third trimester of preg nancy. Br. J. Obstet. Gynaec. 86: 882-887 (1979). Bischof, P.; DuBerg, S.; Sizonenko, M.T.; Schindler, A.M.; Bég uin, F.; Herrmann, W.L.; Sizonenko, P.C.: In vitro production of pregnancy-associated plasma protein-A (PAPP-A) by human decidua and trophoblast. Am. J. Obstet. Gynec. 148: 13-19 (1984). Folkersen, J.; Grudzinskas, J.G.; Hindersson, P.; Teisner, B.; Westergaard, J.G.: Pregnancy-associated plasma protein-A: cir culating levels during normal pregnancy. Am. J. Obstet. Gynec. 139: 910-914 (1981). Bischof, P.; Duberg, S.; Herrmann, W.L.; Sizonenko, P.C.: Am niotic fluid and plasma concentrations of pregnancy-associated plasma protein-A (PAPP-A) throughout pregnancy: comparison with other feto-placental products. Br. J. Obstet. Gynaec. 89: 358-363 (1982). Bischof, P.: Pregnancy-associated plasma protein-A: an inhibitor of the complement system. Placenta 2: 29-34 (1981). Bischof, P.; Lauber, K.; de Wurstemberger, B.; Girard, J.P.: Inhibition of lymphocyte transformation by pregnancy-associ ated plasma protein-A (PAPP-A). J. clin. Lab. Immunol. 7: 6165 (1982). Bischof, P.; Meisser, A.; Haenggeli, L.; Reber, G.; Bouvier, C.; Béguin, F.; Herrmann, W.L.; Sizonenko, P.C.: Pregnancy-associ ated plasma protein-A inhibits thrombin-induced coagulation of plasma. Thromb. Res. 32: 45-55 (1983). Sinosich, M.J.; Davey, M.W.; Ghosh, P.; Grudzinskas, J.G.: Specific inhibition of human granulocyte elastase by human pregnancy-associated plasma protein-A. Biochem. Int. 5: 777— 786 (1982). Meisser, A.; Geinoz, A.; Bischof, P.: In vitro effects of pregnan cy-associated plasma protein-A: artifacts due to heparin. Biol. Reprod. 39: 373-378 (1988). Novitskaya, S.A.; Petrova, I.V.; Serebryakov, N.G.; Vasina, I.G.: Effect of heparin on blast-formation capacity of human blood lymphocytes. Bull. Exp. Biol. Med. (CCT) 80: 1068-1069 (1975). Marossy, K.: Interaction of the Chymotrypsin- and elastase-like enzymes of the human granulocyte with glycosaminoglycans. Biochim. biophys. Acta. 659: 351-361 (1981). Bischof, P.; Haenggeli, L.; Sizonenko, M.T.; Herrmann, W.L.; Sizonenko, P.C.: A radioimmunoassay for the measurement of pregnancy-associated plasma protein-A (PAPP-A) in humans. Biol. Reprod. 24: 1076-1081 (1981). Nakajima, K.; Powers, J.C.; Ashe, B.M.; Zimmerman, M.: Map ping the extended substrate binding site of Cathepsin G and human leucocyte elastase. Studies with peptide substrates re lated to the 4 proteases inhibitor reactive site. J. biol. Chem. 254: 4027-4032 (1979). Received: May 9, 1989 Accepted: September 5, 1989 P. Bischof, MD Laboratoire d’Hormonologie Maternité C H -1211 Geneva 4 (Switzerland)
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
weight fractions (immunoreactive for PAPP-A). The low molecular weight pool which did not contain PAPP-A was also inhibitory. This indicates that the washing of the heparin-Sepharose column contained heparin and that this material was eluted on the AcA 34 column either bound to PAPP-A (pool A) or unbound (pool B). Consequently the observed inhibition of HLE was not due to PAPP-A but to contaminating heparin. Since Sinosich et at. [14] used also a heparin-Sepharose col umn in their purification protocol and since they did not use a high salt molecular sieve chromatography after wards, it can be assumed that their preparations were also contaminated by heparin and that the reported inhi bition of HLE by PAPP-A was in fact due to heparin. If PAPP-A is really such a specific inhibitor of HLE as claimed previously [14], one should have been able to isolate PAPP-A-HLE complexes. In the present study, when the purified PAPP-A was labelled with 125I [18], no binding could be observed to anti-HLE antibodies nor did HLE displace bound PAPP-A from its antibody [unpubl. results]. This study thus strongly suggests that PAPP-A is not an inhibitor of HLE as reported previously.
Gynecol Obstet Invest 1990;29:169-172
Pregnancy-Associated Plasma Protein-A-Induced Inhibition of Human Leukocyte Elastase: An Artifact Paul Bischofa, Alain Gervaixb, Arielle Meisser*, Suzanne Suterb “Département de Gynécologie et d’Obstétrique, et bDépartement de Pédiatrie et Génétique, Université de Genève, Suisse
Key Words. Pregnancy-associated plasma protein A • Leukocyte elastase
Introduction Pregnancy-associated plasma protein A (PAPP-A) is a macromolecular protein first described by Lin et al. [1] in 1974 and purified to homogeneity by others [2-4], It is a homotetrameric glycoprotein (19% CHO) of 800,000 molecular weight, migrating in the ci2-position in an elec tric field. It is thus biochemically but not immunologically similar to cu-macroglobulin (a2M). In contrast to a2M, PAPP-A has the unique property of binding hepa rin [5] and this property was used to develop a purifica tion technique [6] based on heparin-Sepharose affinity chromatography. During pregnancy, PAPP-A, is produced by the trophoblast almost exclusively towards the maternal circu lation [7, 8], Peripheral levels of PAPP-A increase expo nentially during pregnancy with no signs of plateauing near term [9, 10]. Although the biological function of PAPP-A is still unknown, several in vitro effects have been reported. PAPP-A has been described as an inhibi
tor in many in vitro assays: complement activity [11], mitogen-induced lymphoblastogenesis [12], coagulation [13] and granulocyte elastase [14], We have recently reported that the inhibition of complement and coagula tion attributed to PAPP-A was in fact due to heparin contaminating the purified PAPP-A preparations [15]. This contaminant was released during the purification process by the heparin-Sepharose column. One must assume that the inhibition by the same PAPP-A prepara tions of lymphocyte transformation was also due to hep arin contamination since heparin has been described as an inhibitor of phytohemagglutinin-induced lymphocyte transformation [16]. In their report about the inhibitory effects of PAPP-A on granulocyte elastase, Sinosich et al. [14] purified PAPP-A by a technique similar to ours, thus using an affinity step on heparin-Sepharose. Since we have observed that this column releases heparin [15] and since heparin is known to be an inhibitor of elastase [17], we wondered if this inhibitory effect was really due to PAPP-A or if it could be attributed to heparin.
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
Abstract. Pregnancy-associated plasma protein A (PAPP-A), was reported to be an inhibitor in many in vitro systems. Since it was shown that the inhibition of coagulation and complement activity attributed to PAPP-A was in fact due to a contamination by heparin occurring during the purification process, we undertook the present study to see whether the reported PAPP-A-induced inhibition of human leukocyte elastase (HLE) could also be attributed to heparin contamination. PAPP-A was purified from maternal pregnancy EDTA plasma by a method which was previously shown to eliminate contaminating heparin: this preparation was inactive in the HLE assay. But PAPP-A isolated by heparin-Sepharose chromatography, or a PAPP-A-free washing of the heparin-Sepharose column were both inhibitors of HLE. Furthermore the inactive PAPP-A preparation, when incubated with the PAPP-A-free washing of the heparin-Sepharose column, yielded a high molecular weight preparation which inhibited HLE. It is concluded that PAPP-A is not an inhibitor of HLE and that the inhibition of HLE previously attributed to PAPP-A was due to contaminating heparin.
Bischof/Gervaix/Meisser/Suter
170
Materials DEAE-Sepharose, heparin-Sepharosc, polyacrylamide gradient gels (2-16%) and high molecular weight standards were all from Pharmacia (Uppsala, Sweden). Ultrogel AcA 34 was from IBF Biotechnics (Villeneuve-La-Garennc, France), Amicon (XM 100 A) membranes were from Amicon corporation (Oosterhout, Holland), dialysis membranes (Spectrapore) were from Spectrum (Los An geles, Calif.). Antihuman PAPP-A antiserum was prepared as de scribed previously [11]. Purified human leukocyte elastase (HLE) was a gift from Dr. H.P. Schnebli (Ciba-Geigy Ltd., Basel, Switzerland) and the pNAsubstrate (methoxysuccinyl-L-alanyl-L-alanyl-L-prolyl-L-valinc-pnitroanilidc) was from Calbiochem (Lucerne, Switzerland). Purification o f PAPP-A Late maternal pregnancy EDTA plasma (ethylene-diaminotetraacetate 2 mg/ml) was dialyzed against buffer A (0.05 M phosphate containing 0.1 M NaCl and 0.02% azide pH 7.4) and subjected to ion-exchange chromatography on a DEAE-Sepharose column (5 X 15 cm). Elution was carried out with a NaCl gradient (0.1 -1.0 M) in buffer A. PAPP-A-containing fractions, as measured by radioimmu noassay [ 18], were pooled and dialyzed against buffer A. The dialyzed pool was then applied to a heparin-Sepharose col umn ( 5 X 1 5 cm). Washing and elution of this column was carried out in the same way as for the ion exchange. The pooled fractions containing immunoreactive PAPP-A were dialyzed against buffer A and concentrated on an Amicon membrane. This pool was named heparin-Sepharosc pool hereafter. An aliquot of this pool was sub jected to a calibrated [ 15] AcA 34 column (0.9 X 60 cm) and elution carried out with 0.8 M NaCl in buffer A. The PAPP-A-containing fractions were pooled (named AcA 34 pool hereafter), dialyzed against buffer A, concentrated and stored at -2 0 °C. Preparation o f PAPP-A-Heparin Complexes The heparin-Sepharose column was washed with buffer A (in the absence of PAPP-A) and 800 pi of this washing was incubated (2 h at room temperature) with 200 pi of the AcA 34 pool. The mixture was then applied to the AcA 34 column and elution carried out in low salt (0.1 M NaCl in buffer A). Fractions 11-22 (1 ml each) con taining the immunoreactive PAPP-A (fig. 1, pool A) were concen trated and stored at -2 0 °C. Fractions 24-35 (fig. 1, pool B) which did not contain significant amounts of PAPP-A were concentrated in the same way. Polyacrylamide Gel Electrophoresis Polyacrylamide gel electrophoresis was performed on gradient gels (2-16%) in the presence or absence of 0.5% SDS and 4 M urea with or without 5% P-mercaptoethanol as already reported [15], Gels were stained with silver nitrate. HLE Assay HLE activity was measured with a specific substrate (pNA-substrate) as described previously [19]. Briefly, purified HLE (10 nM) in Bulbecco’s PBS pH 7.4 was incubated with 1 0 4 M pNA-substrate and the reaction followed in a spectrophotometer at 387 nm during 2.5 min at room temperature. For inhibition studies, the PAPP-A
samples were preincubated (3 min at room temperature) with HLE in a molar ratio (1:1 to 10:1 PAPP-A:HLE) and the mixture added to the substrate. Results were expressed as percent of controls (HLE without inhibitors).
Results Purification and Polyacrylamide Gel Electrophoresis Elution of the AcA 34 column in the presence of 0.8 M NaCl yielded a symmetrical PAPP-A peak at a MW of about 800,000 [result not shown]. These frac tions were pooled, concentrated and incubated with the PAPP-A-free washing of the heparin-Sepharose column. Submitted to the AcA 34 column but eluted in low salt (0.1 M NaCl in buffer A), the incubate gave a superimposable elution profile (fig. 1). The AcA 34 pool, when run on a nondenaturing poly acrylamide gel (insert fig. 1), gave one unique band of about 800,000 MW. Under denaturing conditions, the PAPP-A preparation was resolved in one major band of about 400,000 MW (fig. 2, lanes 2, 3), one faint band (second from the top) which is a contamination with a2M and a third faint band which is also recognized by anti-PAPP-A antibodies when immunoblotted [MW 200,000, results not shown]. Under reducing conditions (fig. 2, lanes 4, 5), the PAPP-A preparation was resolved in two very close bands of about 180,000 MW. Inhibition o f Human Granulocyte Elastase The PAPP-A from the heparin-Sepharose pool inhib ited the activity of HLE in a dose-dependent manner with an inhibition of 35% at a concentration of 50 nM of PAPP-A (table 1). The elution of this heparin-Sepharose pool on an AcA 34 column in the presence of high salt (0.8 M NaCl) yielded a PAPP-A-containing pool (AcA 34 pool) which did not inhibit HLE up to a concentra tion of 100 nM of PAPP-A. This inactive pool was incu bated with the PAPP-A-free washing of the heparinSepharose column and rerun in low salt on the AcA 34 column. The PAPP-A-containing pool (AcA pool A) inhibited HLE in a dose-dependent manner (table 1). The AcA 34 pool B, which was obtained by pooling low molecular weight fractions from the same elution profile (fig. 1), gave a pool which did not contain PAPP-A (< 2 0 ng/ml)but which inhibited HLE at 100% (table 1). Finally the PAPP-A-free washing of the heparin-Sepha rose column did also inhibit the activity of HLE (ta ble 1).
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
Materials and Methods
PAPP-A and Human Leukocyte Elastase Inhibition
669 440 232 kD
Fractions
Discussion As shown by nondenaturing polyacrylamide gradient gel electrophoresis, our purification protocol yields a preparation which has only one band at 800,000 MW. However, in the presence of SDS a small contamination by a2M can be seen. As shown previously [ 15] the use of a gel filtration column in the presence of high salt (AcA 34 with 0.8 M NaCl) yields a heparin-free PAPP-A which does not inhibit complement activity or throm bin-induced polymerization of fibrin anymore. PAPP-A purified under these conditions did neither inhibit HLE in our assay. In contrast the PAPP-A-containing pool or the PAPP-A-free washing of the heparin-Sepharose col umn were both good inhibitors of HLE. Furthermore when the PAPP-A-free heparin-Sepharose washing was incubated with purified PAPP-A (AcA 34 pool) which did not inhibit HLE, and rerun on the AcA 34 column, inhibition of HLE was observed in the high molecular
Fig. 1. Elution profile on Biogel AcA 34 of purified PAPP-A incubated with the PAPP-A-free washing of the heparin-Sepharose column. Insert: polyacrylamide gradient gel of purified PAPP-A run under nondenaturing conditions (without SDS). Fig. 2. Polyacrylamide gradient gel of purified PAPP-A in the presence of 0.5% SDS with or without ß-mercaptoethanol. Lane 1: high molecular weight standards, lanes 2, 3: denatured PAPP-A, lanes 4, 5: denatured and reduced PAPP-A.
Table 1. Inhibition of granulocyte elastase Pools
Concentrations, nM,
OD, % of control
Heparin-Sepharose
10 20 50
81 72 65
5 10 50 100
94 82 99 80
AcA 34 pool A (low salt)
10 20 50
43 17 10
AcA 34 pool B (low salt)
75 pi, no PAPP-A
0
Heparin-Sepharose washing
75 pi, no PAPP-A
55
AcA 34 (high salt)
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
V0
171
Bischof/Gervaix/Meisser/Suter
172
Acknowledgements The authors wish to thank Mrs. I. Riederer for typing the manu script. This work was supported by grant No. 3.873.086 from the Swiss National Fund for Scientific Research.
References 1 Lin, T.M.; Halbert, S.P.; Kiefer, D.; Spellacy, W.N.; Gall, S.: Characterization of four human pregnancy-associated plasma proteins. Am. J. Obstet. Gynec. 118: 223-236 (1974). 2 Bischof, P.: Purification and characterization of pregnancy-asso ciated plasma protein-A (PAPP-A). Archs Gynecol. 227: 315326 (1979). 3 Folkersen, J.; Westergaard, J.G.; Hindersson, P.; Teisner, B.: Affinity chromatographic purification of a new high molecular weight pregnancy specific protein SP4; in Lehmann, Carcinoembryonic proteins, vol. 2, pp. 503-508 (Elsevier/North Hol land, Amsterdam 1979). 4 Sutcliffe, R.G.; Kukulska-Langlands, B.M.; Coggins, J.R.; Hunt er, J.B.; Gore, G.H.: Studies on human pregnancy-associated plasma protein-A. Purification by affinity chromatography and structural comparison with a 2-macroglobulin. Biochem. J. 191: 799-809 (1980). 5 Sinosich, M.J.; Teisner, B.; Davey, M.; Grudzinskas, J.G.: Preg nancy-associated plasma protein-A: interaction with heparin in crossed affinity immunoelectrophoresis. Aust. N.Z. J. Med. 11: 429-433 (1981). 6 Davey, M.W.; Teisner, B.; Sinosich, M.J.; Grudzinskas, J.G.: Interaction between heparin and human pregnancy-associated
7
8
9
10
11 12
13
14
15
16
17
18
19
plasma protein-A (PAPP-A): a simple purification procedure. Analyt. Biochem. 131: 18-24 (1983). Smith, R.; Bischof, P.; Hughes, G.; Klopper, A.: Studies on preg nancy-associated plasma protein-A in the third trimester of preg nancy. Br. J. Obstet. Gynaec. 86: 882-887 (1979). Bischof, P.; DuBerg, S.; Sizonenko, M.T.; Schindler, A.M.; Bég uin, F.; Herrmann, W.L.; Sizonenko, P.C.: In vitro production of pregnancy-associated plasma protein-A (PAPP-A) by human decidua and trophoblast. Am. J. Obstet. Gynec. 148: 13-19 (1984). Folkersen, J.; Grudzinskas, J.G.; Hindersson, P.; Teisner, B.; Westergaard, J.G.: Pregnancy-associated plasma protein-A: cir culating levels during normal pregnancy. Am. J. Obstet. Gynec. 139: 910-914 (1981). Bischof, P.; Duberg, S.; Herrmann, W.L.; Sizonenko, P.C.: Am niotic fluid and plasma concentrations of pregnancy-associated plasma protein-A (PAPP-A) throughout pregnancy: comparison with other feto-placental products. Br. J. Obstet. Gynaec. 89: 358-363 (1982). Bischof, P.: Pregnancy-associated plasma protein-A: an inhibitor of the complement system. Placenta 2: 29-34 (1981). Bischof, P.; Lauber, K.; de Wurstemberger, B.; Girard, J.P.: Inhibition of lymphocyte transformation by pregnancy-associ ated plasma protein-A (PAPP-A). J. clin. Lab. Immunol. 7: 6165 (1982). Bischof, P.; Meisser, A.; Haenggeli, L.; Reber, G.; Bouvier, C.; Béguin, F.; Herrmann, W.L.; Sizonenko, P.C.: Pregnancy-associ ated plasma protein-A inhibits thrombin-induced coagulation of plasma. Thromb. Res. 32: 45-55 (1983). Sinosich, M.J.; Davey, M.W.; Ghosh, P.; Grudzinskas, J.G.: Specific inhibition of human granulocyte elastase by human pregnancy-associated plasma protein-A. Biochem. Int. 5: 777— 786 (1982). Meisser, A.; Geinoz, A.; Bischof, P.: In vitro effects of pregnan cy-associated plasma protein-A: artifacts due to heparin. Biol. Reprod. 39: 373-378 (1988). Novitskaya, S.A.; Petrova, I.V.; Serebryakov, N.G.; Vasina, I.G.: Effect of heparin on blast-formation capacity of human blood lymphocytes. Bull. Exp. Biol. Med. (CCT) 80: 1068-1069 (1975). Marossy, K.: Interaction of the Chymotrypsin- and elastase-like enzymes of the human granulocyte with glycosaminoglycans. Biochim. biophys. Acta. 659: 351-361 (1981). Bischof, P.; Haenggeli, L.; Sizonenko, M.T.; Herrmann, W.L.; Sizonenko, P.C.: A radioimmunoassay for the measurement of pregnancy-associated plasma protein-A (PAPP-A) in humans. Biol. Reprod. 24: 1076-1081 (1981). Nakajima, K.; Powers, J.C.; Ashe, B.M.; Zimmerman, M.: Map ping the extended substrate binding site of Cathepsin G and human leucocyte elastase. Studies with peptide substrates re lated to the 4 proteases inhibitor reactive site. J. biol. Chem. 254: 4027-4032 (1979). Received: May 9, 1989 Accepted: September 5, 1989 P. Bischof, MD Laboratoire d’Hormonologie Maternité C H -1211 Geneva 4 (Switzerland)
Downloaded by: King's College London 137.73.144.138 - 3/5/2018 5:29:59 PM
weight fractions (immunoreactive for PAPP-A). The low molecular weight pool which did not contain PAPP-A was also inhibitory. This indicates that the washing of the heparin-Sepharose column contained heparin and that this material was eluted on the AcA 34 column either bound to PAPP-A (pool A) or unbound (pool B). Consequently the observed inhibition of HLE was not due to PAPP-A but to contaminating heparin. Since Sinosich et at. [14] used also a heparin-Sepharose col umn in their purification protocol and since they did not use a high salt molecular sieve chromatography after wards, it can be assumed that their preparations were also contaminated by heparin and that the reported inhi bition of HLE by PAPP-A was in fact due to heparin. If PAPP-A is really such a specific inhibitor of HLE as claimed previously [14], one should have been able to isolate PAPP-A-HLE complexes. In the present study, when the purified PAPP-A was labelled with 125I [18], no binding could be observed to anti-HLE antibodies nor did HLE displace bound PAPP-A from its antibody [unpubl. results]. This study thus strongly suggests that PAPP-A is not an inhibitor of HLE as reported previously.
