Biochimica et Biophysica Acta, 797 (1984) 187-193 187 Elsevier

BBA21751

a2-MACROGLOBULIN-PROTEINASE COMPLEXES IN CYSTIC FIBROSIS SERUM

ANALYSIS BY MONOCLONAL ANTIBODIES AND RECEPTOR-MEDIATED ENDOCYTOSIS BY NORMAL HUMAN FIBROBLASTS

PETER MARYNEN, FRED VAN LEUVEN, JEAN-JACQUES CASSIMAN * and HERMAN VAN DEN BERGHE

Center for Human Genetics, University of Leuven, Campus Gasthuisberg, O&N, Herestraat, B- 3000 Leuven (Belgium)

(Received December 6th, 1983)

Key words: ~2" Macroglobulin," Endocytosis," RIA; Proteinase," Cystic fibrosis," Monoclonal antibody

a2-Macroglobulin complexed to proteinases activated during clotting of cystic fibrosis and control sera was quantitated with the complex-specific monoclonal antibody F2B 2. Similar amounts of a2-macroglobulin complexes (between 40 and 90 btg/ml) were generated in cystic fibrosis and control sera. Endocytosis of the complexes by normal human fibroblasts was compared to the amount of complexes detected by the F2B2-radioimmunoassay. Normal uptake was observed with 13 out of 14 cystic fibrosis sera. One cystic fibrosis serum showed strongly reduced endocytosis of the complexes. Complexes isolated from this serum on immobilized F 2 B 2 failed to inhibit binding of purified a2-macroglobulin-trypsin to its receptor, demonstrat- ing deficient receptor-binding of these complexes. The low uptake complexes could not be distinguished from complexes isolated from control or other cystic fibrosis sera by isoelectric focusing, rate electrophoresis or SDS-polyacrylamide gel electrophoresis.

Introduction

The endocytosis of a2-macroglobulin by human fibroblasts is mediated through a receptor which recognizes exclusively a2-macroglobulin molecules inactivated by either proteinases [1] or primary amines [2]. We reported previously that when the uptake of naturally occurring a2-macroglobulin- protease complexes from cystic fibrosis sera by fibroblasts was examined, a variable but signifi- cantly lower extent of uptake was found compared to complexes from age-matched control sera [3]. The uptake data were normalized for the total amount of a2-macroglobulin present in the sera. The evidence suggested that in cystic fibrosis sera an anomaly existed at the level of the a2-macro- globulin complexes but further molecular char-

* To whom correspondence should be addressed.

acterization of the interaction of a2-macroglobulin with proteinases and with the receptor was re- quired before the exact molecular nature of this anomaly could be elucidated [3].

We therefore studied the receptor-recognition site generated upon inactivation (complexation) of c~2-macroglobulin. A monoclonal antibody F2B 2 was characterized, which selectively binds in- activated a2-macroglobulin [4]. Circumstantial evi- dence showed the determinant of F2B 2 to be part of the receptor-recognition site of a2-macroglobu- lin complexes [4-7].

The availability of the monoclonal antibody F 2 B 2 thus allowed us to precisely quantitate and to isolate az-macroglobulin complexes from cystic fibrosis and control sera. In the present investiga- tion control and cystic fibrosis sera were examined for the amount of a2-macroglobulin complexes present and for the endocytosis of these complexes

0304-4165/84/$03.00 © 1984 Elsevier Science Publishers B.V.

188

by fibroblasts. The complexes were isolated by affinity chromatography on the monoclonal anti- body and characterized by polyacrylamide gel electrophoresis and isoelectric focusing.

It will be shown that (i) equal amounts of a2-macroglobulin complexes are generated upon clotting of cystic fibrosis and control plasma; (ii) a2-macroglobulin complexes isolated from cystic fibrosis sera cannot be distinguished from control a2-macroglobulin complexes by polyacrylamide gel electrophoresis and isoelectric focusing; (iii) nor- mal amounts of complexes are taken up by human fibroblasts from 13 out of 14 cystic fibrosis sera, when the uptake data are normalized for the amount of a2-macroglobulin complexes present in the sera; (iv) the strongly decreased uptake of complexes occurring in one cystic fibrosis patient is due to a deficient receptor binding of the a 2- macroglobulin complexes.

Materials and Methods

Plasma, obtained by centrifugation of citrated whole blood at 1500 × g for 10 min, was clotted by addition of CaC12 to a final concentration of 25 mM and kept for 4 h at room temperature. The fibrin clots were removed by centrifugation at 10000 × g for 15 min. Control sera were obtained from the Red Cross Blood Centre, Leuven.

Ascitic fluid from Balb /c mice injected with the F 2 B 2 hybridoma was obtained as in Marynen et al. [4]. F2B 2 monoclonal antibodies were purified by (NH4)2SO 4 precipitation of the supernatant of n-octanoic acid precipitated ascitic fluid as de- scribed [8]. F2B z antibodies were labeled with Na125I by the chloramine-T method [1].

Purified a2-macroglobulin, a2-macroglobulin. trypsin and lzSI-labeled a2-macroglobulin- trypsin were prepared according to Van Leuven et al. [1]. Na 1251 was from Amersham International. Trypsin (TPCK treated) was from Worthington.

RIA of ae-macroglobulin complexes. 200 ~1 of diluted serum was incubated overnight at 4°C with 0.1 #g of 125I-labeled-a2-macroglobulin'trypsin (6000 cpm) and 0.14 pg of F2B 2 antibodies. 20/zl of normal mouse serum (diluted 1 /40 in phos- pahte-buffered saline) and 20 /~1 of rabbit anti- mouse immunoglobulins IgG (Dakopatts, Den-

mark) were then added and the mixture was kept for 4 h at 4°C and centrifuged at 3000 × g for 10 rain. The radioactivity in the pellet and the super- natant was determined. The detection limit of the assay was 0.3/~g a2-macroglobulin, t rypsin/ml.

Rate electrophoresis. The rate electrophoresis of native proteins in 5% polyacrylamide gels and electrophoresis of proteins in 6-20% SDS-poly- acrylamide gels is described by Van Leuven et al. [2].

Isolation of ae-macroglobulin complexes from the sera. a:-Macroglobulin complexes were purified from the sera by absorption on immobilized F2B 2- antibody. F: B 2 antibodies were coupled to Affi-Gel 10 beads (Bio-Rad) according to the manufacturer (5 mg F2Bz/ml gel). 4 ml of serum was slowly passed (10 m l / h ) through a 7 × 1 cm F:B: affinity column, equilibrated with phosphate-buffered saline. The columns were washed with phosphate- buffered saline and eluted with 4 M MgC12. The eluted samples were rapidly concentrated and di- alyzed against phosphate-buffered saline. The total az-macroglobulin concentrations of the samples was measured by rocket immunoelectrophoresis.

Uptake and binding of a e-macroglobulin. Human diploid fibroblast cell cultures were obtained and cultured as described [9]. The uptake of a2-macro- globulin complexes from human sera by human fibroblast cell layers was measured exactly as de- scribed previously [3]. 25-cm: tissue culture flasks (9.105 cells) were used for the assays. Binding of 125I-labeled a2-macroglobulin, trypsin for the in- hibition experiments with purified a2-macro- globulin complexes were performed as in Van Leuven et al. [1]. Cell layers were plated at a density of 100000 cells/well in Disposo trays (Falcon). The cell layers were rinsed three times with ice-cold Dulbecco's modified Eagle's medium and incubated with 0.5 ml of this medium contain- ing 1 /xg /ml 125I-labeled az-macroglobulin. trypsin, 1 /~g /ml a2-macroglobulin complexes iso- lated with FaB 2 antibodies and 0.2% bovine serum albumin for 4 h at 4°C. The cell layers were then washed five times with ice-cold phosphate-buffered sali'ne and bound label was measured. Aspecific bmalng v~n-s determined in presence of an 100-fold excess of unlabeled a2-macroglobulin • trypsin.

Results

Quantitation of a2-macroglobulin-proteinase com- plexes in cystic fibrosis and control sera

A radioimmunoassay was developed with the complex specific monoclonal antibody F2B 2. The antigen addition curve is shown in Fig. 1. The inhibition curves obtained with control serum and cystic fibrosis serum paralleled the one obtained with purified etE-macroglobulin, trypsin, indicat- ing the absence of aspecific serum effects on the assay [10].

The amount Of a2-macroglobulin complexes in human plasma and serum was very dependent on the storage conditions. Prolonged storage of plasma and serum at - 2 0 ° C and in the presence of proteinase inhibitors, or freeze-thawing of the samples resulted in increased amounts of in- activated et2-macroglobulin detected by the FEB 2 antibodies. Therefore aE-macroglobulin complexes were quantitated and endocytosis was measured on the day of collection of the samples. The amount of aE-macroglobulin complexes in the plasma sam- ples was very low (between 0.09 and 0.3% of the total aE-macroglobulin ). In freshly clotted plasma, the amount of etE-macroglobulin complexes varied between 40 and 90/~g/ml (between 1.9 and 4% of the total a2-macroglobulin ) (Fig. 2). The relative amount of a2-macroglobulin complexes generated upon clotting of the plasma was negatively corre- lated with the total a2-macroglobulin levels. No

189

differences, either in absolute amount, or in rela- tive amount of a2-macroglobulin complexes was observed between samples from the cystic fibrosis patients and age-matched controls (Fig. 2).

When the sera were stored at 4°C, the con- centration of a2-macroglobulin complexes in- creased slowly (inactivation of about 0.5% of total a2-macroglobulin per week) as indicated above. The cystic fibrosis sera however, could not be distinguished from the control sera.

Endocytosis of ae-macroglobulin complexes by hu- man fibroblasts

To investigate the apparent contradiction be- tween the lower uptake of a2-macroglobulin com- plexes from cystic fibrosis sera reported previously [3], and the equal amount of a2-macroglobulin complexes detected by the FEB 2 RIA in cystic fibrosis and control sera, uptake experiments were performed with the same samples as assayed by the F2B 2 RIA, exactly as described before [3]. The total concentration of a2-macroglobulin in the up- take medium was 200 # g /m l . The amount of a2-macroglobulin complexes varied thus from 5 /~g/ml for the sera from the children to 9 /~g /ml for the adult sera and remained constant during the 4 h assay (results not shown).

The uptake data are shown in Fig. 3. A good correlation between the amount of a2-macro- globulin endocytozed and the amount of a2-mac- roglobulin complexes present in the uptake

0.5 1

%, 5 0

100 % bound

2 ° / ,hus

0 0.2 0.4 0.8 1.6

jug O~2M-Trypsin/ml

Fig. 1. Dose-response curve of the F2B2-RIA. The inhibition curves for purified a2-macroglobulin, trypsin (O), cystic fibrosis serum (©) and control serum (13) are shown.

o°

I 2 3 mg/ml TOTAL a'2M

Fig. 2. ot2-Macroglobulin complexes in human sera. Freshly obtained human citrated plasma samples were clotted by addi- tion of 25 mM CaCI 2. The amount of a2-macroglobulin com- plexes generated was measured with the F2B 2 RIA. Total ~t2-macroglobulin was measured with rocket immuno- electrophoresis. ©, cystic fibrosis serum; O, control serum.

190

6 I

1 mcs V 0,

ONN

0 NN

MEDIUM a2M-COMPLEXES j&I/ml

Fig. 3. Endocytosis of a*-macroglobulin complexes by human

fibroblasts. The endocytosis of a,-macroglobulin complexes by

confluent human fibroblast cell layers was measured. The

amount of a,-macroglobulin complexes present in the uptake

media was measured with the F2B2 RIA. 0, cystic fibrosis

serum; 0, control serum; 0. a,-macroglobulin complexes iso-

lated from cystic fibrosis sera; n , cy2-macroglobulin complexes

isolated from control sera. N.N., serum or complexes obtained

from cystic fibrosis patient N.N.

medium was found for all control sera and for 13 out of 14 cystic fibrosis sera. Abnormal low up- take was observed with serum of one patient (N.N.)

at two different occasions. When a,-macroglobulin complexes were isolated (see further) from 3 con- trol sera and 3 cystic fibrosis sera, including N.N.,

normal uptake was observed for all samples, ex- cept N.N. (Fig. 3, squares). To investigate whether this deficient uptake was indeed due to an anomaly at the level of the cY,-macroglobulin complexes, inhibition of binding of ‘251-cY,-macroglobulin. trypsin to human fibroblasts by the isolated CQ- macroglobulin complexes was measured (Table I).

While control Lu,-macroglobulin * trypsin, a*-mac- roglobulin complexes from three control sera and two cystic fibrosis sera showed expected inhibi- tion, a decreased inhibition was observed with

cu,-macroglobulin complexes isolated from N.N. serum. This demonstrated a deficient receptor-bi-

nding of the abnormal complexes.

Isolation and characterization of the natural& oc-

curring a,-macroglobulin complexes in serum

a,-Macroglobulin complexes, generated during clotting of plasma, were isolated from 3 cystic

fibrosis sera, including the low uptake serum N.N., and 3 control sera by affinity chromatography on

TABLE I

INHIBITION OF BINDING OF ‘251-(r2-MACROGLOBULIN~TRYPSIN TO HUMAN FIBROBLAST CELL LAYERS BY

q-MACROGLOBULIN COMPLEXES ISOLATED FROM HUMAN SERA

Binding of ‘251-IabeIed a,-macroglobulin. trypsin to confluent human fibroblast cell layers was measured. The cell layers (lo5

fibroblasts) were incubated for 4 h at 4’C with medium containing 1 pg/ml ‘251-~2-macroglobulin. trypsin (spec. act. 1500 cpm/ng),

0.2% bovine serum albumin and the different a,-macroglobulin complexes stated in the table (1 pg/ml) and total bound label was

determined. Specific binding was calculated by subtracting aspecific binding, measured in presence of 100 pg/mI a,-macroglobulin.

trypsin, from total binding. The mean of duplicate assays is given.

Additions to the assay medium

None 1 gg/ml cu,-macroglobulin. trypsin

cY2-Macroglobulin complexes isolated from control serum 1 2

3

a,-Macroglobulin complexes isolated from cystic fibrosis serum 1 2

3 (N.N.)

S Inhibition of

specific binding

0 (1400 cpm bound)

62

61

57

58

58

60

18

191

A B a2M.T1 2 3 4 5 6

Fig. 4. Rate electrophoresis of a2-macroglobulin complexes isolated from human sera. 8 ~g of a2-macroglobulin complexes, isolated from three control sera (2, 3, 6) and three cystic fibrosis sera (1, 4, 5) by affinity chromatography with F2B 2 antibodies were run on a 5% polyacrylamide gel in the absence (A) or in the presence of an excess F2B 2 antibodies (B) under non-denaturing circumstances, t~2M.T=purified ~2-macro- globulin- trypsin standard.

185 kDa-,-

85 kDa--,-

60 kDa-.,-

25 kDa--,-

immobilized F2B 2. Polyacrylamide gel electro- phoresis of the non-denatured isolated complexes in 5% gels showed identical patterns for all sam- ples (Fig. 4A). The patterns exhibited by the iso- lated complexes were not identical to the one observed with 'fast' a2-macroglobulin, trypsin complexes but were reminiscent of the inter- mediate mobilities observed with some proteinases as plasmin and of the multiple band patterns observed with spontaneously inactivated a2-mac- roglobulin [2]. Addition of excess F2B 2 antibodies resulted in a normal F2B 2 binding pattern [4] for all samples (Fig. 4B). Isoelectric focussing of the isolated aE-macroglobulin complexes in poly-

Fig. 51 Isoelectric focusing of a2-macroglobulin complexes gen- erated in human sera. 10/xg aE-macroglobulin complexes iso- lated from three control sera (2, 3, 6) and three cystic fibrosis sera (1, 4, 5) were focused on a commercial pH 3-10 poly- acrylamide get (LKB, Sweden) in the presence of 2 mM EDTA. a 2 M.T = purified a2-macroglobulin, trypsin standard.

Fig. 6. SDS-polyacrylamide gel electrophoresis of a2-macro- globulin complexes of human sera. 30 /.tg aE-macroglobulin complexes isolated from three control sera (2, 3, 6) and three cystic fibrosis sera (1, 4, 5) were reduced with 2-mercaptoethanol and run overnight on a 6-25% polyacrylamide gel exactly as in Van Leuven et al. [2]. a2M.T = purified aE-macroglobulin. trypsin standard.

acrylamide gels (pH 3-10) also showed identical patterns for all samples (Fig. 5). This was also the case when SDS-polyacrylamide gel electrophoresis was performed in 6-20% gradient gels under re- ducing conditions. Major bands were observed with a M r values of 185 000 and 85 000. The high molecular weight species observed are the results of inter- and intrachain molecular cross-linking upon activation of the internal thioesters of a 2- macroglobulin (Fig. 6).

Discussion

The higher levels of a2-macroglobulin com- plexes detected in the sera as compared to plasma indicate interaction of a2-macroglobulin with pro- teinases generated during clotting. Indeed, transfer of proteinases from their specific inhibitor to a 2- macroglobulin [11,12] and also direct interaction of thrombin with aE-macroglobulin in clotting plasma have been described [13]. The absence of

192

differences between the amount of c~2-macro- globulin complexes in cystic fibrosis and control sera reflects activation and complexation of a nor- mal amount of proteinases in clotting plasma of cystic fibrosis patients. The absence of a specific proteinase, as reported [14], however, would prob- ably remain undetected by the present methods.

The a2-macroglobulin complexes generated in cystic fibrosis sera cannot be distinguished by structural criteria from control a2-macroglobulin complexes. The parallel inhibition curves obtained in the RIA with purified a2-macroglobulin. trypsin, cystic fibrosis serum and control serum indicate a similar avidity of the F 2 B 2 antibodies for their determinant on a2-macroglobulin complexes in both sera. Polyacrylamide gel electrophoresis under denaturing and non-denaturing circum- stances, and isoelectric focusing of isolated com- plexes yielded identical patterns for cystic fibrosis and control samples. This structural stability (pre- servation of charge and size of the complexes) agrees with the functional stability (shielding of the bound proteinase from macromolecular sub- strates) reported for purified a2-macroglobulin. trypsin from normal and cystic fibrosis individuals [3,15,16] and contrasts with earlier reports [17-19].

The amount of a2-macroglobulin complexes present in the endocytosis assay (4-9/~g/ml) is in the linear part of the concentration-endocytosis curve for human fibroblasts [1] and remains con- stant during the assay. Therefore the ratio of the amount of internalized a2-macroglobulin to the concentration of a2-macroglobulin complexes in the supernatant is directly comparable for all sam- ples while the absolute uptake data are not.

As shown, normalization of the uptake data for the amount of complexes present in the sera, re- suits in a normal uptake compared to control sera, for 13 out of the 14 cystic fibrosis sera tested. The deviating ratio of intracellular a2-macroglobulin to ct2-macroglobulin complexes in the supernatant observed with cystic fibrosis serum from patient N.N. indicates a discrepancy between the presence of the F2B 2 determinant on a2-macroglobulin com- plexes and uptake by human fibroblasts. Further- more, the decreased uptake and the decreased inhibition of binding of 125I-a2-macroglobulin- trypsin by human fibroblasts observed with iso- lated complexes demonstrate a deficient receptor-

binding of these complexes. Together with the 'normal' behavior of these complexes upon iso- electric focusing, rate electrophoresis or pore-limit electrophoresis, this indicates a minor modifica- tion of the receptor-recognition site at a site differ- ent from the F2B 2 epitope. In this context loss of the receptor-recognition site upon derivatisation of lysine residues of the a2-macroglobulin molecule [5] might be significant. In this model system however, parallel loss of the receptor-recognition site and the F2B 2 epitope occurs.

In the light of our previously reported data [1], we are tempted to speculate that the variation in relative amounts of a2-macroglobulin complexes demonstrated here could account for the inter- mediate values of uptake reported for some of the cystic fibrosis sera, while the sera showing ex- tremely low to absent endocytosis are probably similar to the N.N. serum investigated here.

In conclusion, a2-macroglobulin complexes, generated during clotting of cystic fibrosis sera, cannot be distinguished from complexes generated in control sera by the analytical techniques de- scribed above. The a2-macroglobulin complexes of the majority of cystic fibrosis sera are normally cleared by human fibroblasts. Therefore, use of this assay for detection of cystic fibrosis homo- zygotes or cystic fibrosis carriers cannot be en- visaged.

Although the abnormal complexes, lacking an intact receptor-recognition site, seem to be excep- tional and might or might not be related to the cystic fibrosis pathology, they raise important questions concerning the physiology of a2-macro- globulin. One might indeed wonder whether abnormal complexes are generated in vivo and how they are cleared. Therefore, the molecular nature of the defect in the receptor-recognition site of these complexes will be further investigated.

Acknowledgements

P.M. is a Research Assistent of the National Fund for Scientific Research, Belgium. The expert technical assistance of P. Servranckx and M. Wil- lems is gratefully acknowledged. This work was supported by Grant 3.0055.83 (National Fund for Scientific Research, Belgium), by a grant 'Gecon- certeerde Acties' from the Belgian Government

a n d b y a g r a n t f r o m t he A m e r i c a n C y s t i c F i b r o s i s

F o u n d a t i o n .

References

1 Van Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1979) J. Biol. Chem. 254, 5155-5160

2 Van Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1981) J. Biol. Chem. 256, 9016-9022

3 Van Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1979) Pediat. Res. 13, 1384-1386

4 Marynen, P., Van Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1981) J, Immunol. 127, 1782-1786

5 Marynen, P., Van Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1982) FEBS Lett. 137, 241-244

6 Van Leuven, F., Marynen, P., Cassiman, J.J. and Van den Berghe, H. (1982) Biochem. J. 203, 405-411

7 Van Leuven, F., Marynen, P., Cassiman, J.J. and Van den Berghe, H. (1983) Ann. New York Acad. Sci., in the press

8 Verlinden, J., Van .Leuven, F., Cassiman, J.J. and Van den Berghe, H. (1981) Cell. Mol. Biol. 27, 38-48

9 Cassiman, J.J., Verlinden, J., Vlietinck, R.F., Bellemans, J.,

193

Van Leuven, F., Deroover, J., Baro, F. and Van den Berghe, H. (1979) Hum. Genet. 53, 75-86

10 Hunter, N.M. (1978) in Handbook of Experimental Im- munology (Weir, D.M., ed.), Ch. 14, pp. 14.1-14.40, Blackwell Scientific Publication

11 Ohlson, K. (1974) Bayer Symposium V 'Proteinase Inhibi- tors', Springer Verlag, 94-105

12 Aubry, M. and Bieth, J. (1977) Clin. Chim. Acta 78, 371-380 13 Takada, A., Koide, T. and Takada, Y. (1979) Thromb. Res.

16, 59-68 14 Rao, G.J.S., Platt, W.M. and Nadler, H.M. (1978) Enzyme

23, 314-319 15 Parsons, M. and Romeo, G. (1980) Clin. Chim. Acta 100,

215-224 16 Bridges, M.A., Applegarth, D.A., Johannsen, J., Wung,

L.T.K. and Davidson, G.F. (1982) Clin. Chim. Acta 118, 33-43

17 Shapira, E., Ben-Yoseph, Y. and Nadler, H.M. (1977) Clin. Chim. Acta 78, 359-363

18 Shapira, E., Martin, C.L. and Nadler, H.L. (1977) J. Biol. Chem. 252, 7923-7929

19 Wilson, G.B. and Fudenbergh, H.H. (1976) Ped. Res. 10, 87-96