Radioimmunoassay of B2-Microglobulin T. PLESNER, B. NORGAARD-PEDERSEN & T. BOENISCH Dept. of Clinical Chemistry A, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Plesner, T., Norgaard-Pedersen, B. & Boenisch, T. Radioimmunoassay of Microglobulin. Scand. J . din. Lab. Invest. 35, 729-735, 1975.
p2-
A radioimmunoassay for quantitation of /?,-microglobulin in serum and urine is described. Polyethylene glycol 6000 was used to precipitate the antigen-antibody complex. The assay can be performed within 5 hr. In healthy human subjects (mean age, 30.6 years and 42.0 years for serum and urine determinations, respectively) the mean concentration of /?,-microglobulin in serum was 125 nmol/l (range, 49-190 nmol/l). In men the mean 24-hr urinary excretion was 12.9 nmol (range, 8.6-19.2 nmol), and in women it was 5.5 nmol (range, 2.9-8.7 nmol). The assay range was 0.23-19.40 nmol/l, and the detection limit was 0.042 nmol/l, using a prolonged incubation time. The coefficient of variation based on 20 determinations in serum dilutions with a concentration of /?,-microglobulin of 8.14 and 1.92 nmol/l was 8.1 per cent and 9.0 per cent, respectively. Key-words: B,-microglobulin; polyethylene glycol precipitation; radioimmunoassay T. Plesner, M.D., Dept. of Clinical Chemistry A , Rigshospitalet, University ol Copenhagen, 9, Blegdnnisvej, DK-2100 Copenhagen 0, Denmark
mic micro globulin was first described by Berggird & Bearn (1) in 1968. They isolated the protein and characterized it as being a singlechain polypeptide with a molecular weight of 11,800. The protein is found in all human body fluids examined. Quantitative determination of B,-rnicroglobulin has been used both as a renal function parameter (2, 8, 22, 23, 26) and to determine the degree of glomerular versus tubular damage (18). A growing field of considerable interest is the in vitro production of j32-microglobulin by normal and malignant cells (3, 15, 16) and the association with the HL-A system (10, 14, 19-21,24,25). /?,-microglobulin has been quantitated by single radial immunodiffusion (SRI) (13) and by solid-phase radioimmunoassay (6). The solidphase radioimmunoassay is now also commercially available as a Phadebas @C kit (Pharmacia, Uppsala, Sweden).
The purpose of this study was to develop a simple, sensitive, and rapid method for quantitation of ,&-microglobulin. MATERIAL AND METHODS Isolation of /?,-microglobulin. Urine from uraemic patients was concentrated by pressure dialysis through a P M 10 dialysis membrane (Amicon). The concentrated urinary proteins were dialysed in a Visking tube against a 0.01 mol/l tris buffer with 0.02 mol/l sodium chloride at p H 7.9 and chromatographed on Whatman DE52 cellulose equilibrated with the same buffer (column K26/40 from Pharmacia). The proteins were eluted with a linear salt gradient from 0.02 to 0.2 mol/l sodium chloride in 0.01 mol/l tris buffer at p H 7.9. Four-ml fractions were collected at a flow rate of 16 ml/hr, and the fractions containing p,-microglobulin were pooled and reconcentrated by pressure dialysis
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
730
T . Plesner, B. Norgaard-Pedersen& T . Boenisch
using a PM 10 dialysis membrane. The concentrated proteins were dialysed against a 0.05 mol/l tris buffer with 0.2 mol/l sodium chloride, pH 7.4, and chromatographed on Sephadex G75 superfine, specially prepared to obtain gel particles of uniform size (5). The gel particles were equilibrated with 0.05 mol/l tris buffer containing 0.2 mol/l sodium chloride, p H 7.4, and packed in column K26/100 from Pharmacia. Two-ml fractions were collected at a flow rate of 11 ml/hr. The fractions containing pure B,-microglobulin were pooled and concentrated in a Visking tube kept in a stream of air at room temperature. The isolated protein was quantitated by reading the optical density a t 280 nm, using a molar extinction coefficient of 19,850. It was stored at -20 OC after dilution with 0.1 mol/l phosphate buffer, pH 7.4, containing 14.5 pmol/l bovine serum albumin. The concentration of ~2-microglobulinin the solutions was 38.80 nmol/l. Standards with p2microglobulin concentrations from 0.23-19.40 nmol/l were prepared by further dilution with the albumin-containing phosphate buffer. Labelling of B2-microglobulin. P2-microglobulin was labelled by a modification of the chloramine-T method (12); 0.39 nmol of 8,microglobulin was labelled with 1 mCi [12SI]NaI (Amersham) to a specific activity of approximately 1.2 mCi/nmol. 3.55 nmol of chloramineT and a reaction time of 30 sec were used. Free and protein-bound lZsI were separated by chromatography on Sephadex G-25. Antibodies. Immunoglobulins produced in rabbits against human B2-microglobulin were obtained from Dakopatts, Copenhagen, Denmark. Lots No. 023 and 064 have been used. By crossed immunoelectrophoresis (9) of proteins from concentrated urine against the antisera, both lots were shown to react specifically with B,-microglobulin. The antibodies were titrated by incubating increasing dilutions of antibodies with a constant amount of labelled ,82-microglobulin (Fig. 1). Assay procedure. 200 p1 of standards or unknown samples were incubated with 200 pl of B,-microglobulin antibodies (lot No. 023 diluted 1:15,000, final dilution) and 100 pl of
ANTIBODY TITRATION
CURVE
C PM
1 x 10-6 250.0
62.5
15.6
3.9
0.975
0.244
ANTIBODY DILUTION
Fig. 1. Antibody-bound radioactivity when a constant amount of 1*51-labelledB,-microglobulin is incubated with increasing dilutions of antibodies
for 3 hr.
labelled ~2-microglobulinin assay buffer (approximately 0.052 pmol 12sI-labelled Ba-microglobulin). After 3 hr of incubation at room temperature 300 pl horse or bovine serum and 1,600 pl of a 15 per cent solution (w/v) of polyethylene glycol 6000 (PEG) in 0.1 mol/l phosphate buffer, pH 7.4, were added. The suspension was mixed on a Whirli-mixer and centrifuged at 3,000 g for 30 min at 4 OC. Subsequently the supernatant was discarded and the radioactivity in the precipitate was counted in a Selectronic gamma counter. The standard inhibition curve was plotted as counts per minute against the B,-microglobulin concentration of the standards (Fig. 2). All determinations were carried out in duplicate, and each assay set up included determinations on: 1) two controls with /&-microglobulin concentrations of 8.14 and 1.92 nmol/l, respectively, prepared by diluting serum in the
Radioimmunoassay of pz-Microglobulin
to be above 5.5 since it has been shown that enzymatic degradation of /&-microglobulin might take place below this value (7). Other reagents. Rabbit immunoglobulins against total normal human urinary proteins (Dakopatts, Copenhagen, Denmark) and antiserum against az-microglobulin (Behringwerke, Marburg/Lahn, Germany) were used to control the purity of isolated /&-microglobulin. Statistical evaluation. Standards and results were compared by linear regression analysis (the least squares method).
STANDARD INHIBITION CURVE
CPM 3OOOO
2oooo
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
131
\
t\
10000
RESULTS
NMOL/L 5
10
p2
15
20
MICROGLOBULIN
for radioFig. 2 Standard calibration diagram immunoassay of /&-microglobulin.
albumin-containing phosphate buffer, 2) one sample from yesterday’s determinations, and 3) two tubes containing 12sI-labelled p2-microglobulin with and without antibodies. The Phadebas P,-microglobulin test available from Pharmacia, Uppsala, Sweden, was used for comparison. The assay was carried out as described by the manufacturer. Samples from healthy individuals. Blood was drawn by venous puncture after minimal stasis during the day from 44 subjectively healthy, ambulant individuals, 24 men and 20 women aged 22-45 years (mean, 30.5 years). The blood was centrifuged after clotting and the serum stored undiluted at -20 OC until analysed. 24-hr urine samples were collected from 7 men and 8 women aged 23-66 years (mean, 42.0 years). The individuals were in usual daily activity during the collection period. The urine was collected in containers with sodium azide and stored at -20 OC until analysed. The pH values of the urine samples were controlled
Titration of antibodies. Fig. 1 shows the result of titration of anti-B,-microglobulin antisera, lot 023, from Dakopatts. Quality of isolated /&-microglobulin. The Pa-microglobulin preparation was found to be pure by agarose gel electrophoresis and Ouchterlony double diffusion against rabbit immunoglobulins to total normal human urinary proteins and antiserum to az-microglobulin. The isolated B,-microglobulin was homogeneous in prolonged electrophoresis and in crossed immunoelectrophoresis against anti-D2-microglobulin antibodies (Dakopatts, Copenhagen, Denmark). Standards from the Phadebas P,-microglobulin test and our standards of isolated p2microglobulin solutions were compared in our radioimmunoassay (PEG-RIA). The correlation between the standards was excellent: r = 0.996 (n = 12). To investigate whether purified p,-microglobulin behaved like the native protein, standard inhibition curves of serially increasing dilutions of urine, serum, and isolated p2-microglobulin in albumin-containing phosphate buffer solution were compared in three ways: first, graphically the curves appeared to have identical slopes (Fig. 3). Second, determination of /&-microglobulin in serum and urine at different dilutions gave identical results. Finally, the curves were transformed to straight lines with identical slopes after logarithmic transformation of the values of both axes. This result was
T . Plesner, B. Norgaard-Pedersen& T . Boenisch
732 CPM
CP
3oooO
2QyK)
30000
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
loo00
20000 .~ 5
10
l 2 MICROGLOBULIN
15
20
25 NMOLIL
Fig. 3. Dose-response curves for dilutions of isolated B2-microglobulin, serum, and urine. 10000
obtained by regression analyses using the least squares method. These findings indicate that buffer solutions of isolated B2-microglobulin can be used as standards when quantitating B2-microglobulin in serum and urine. Detection limit (defined as the smallest quantity of B2-microglobulin that can be distinguished from zero). It is possible to quantitate /32-microglobulin down to a concentration of 0.042 nmol/l if the antibodies are diluted 1:50,000 and the incubation time is increased to 18 hr. Reproducibility. The coefficient of variation, calculated from 20 different determinations on serum pool dilutions with a concentration of B2-microglobulin of 8.14 and 1.92 nmol/l, is 8.1 and 9.0 per cent, respectively. Accuracy. Addition of purified B2-microglobulin to serum or urine resulted in a recovery of 98 per cent, and a close correlation was found between calculated and measured ,!?2-microglobulinconcentrations when sera with known content of B2-microglobulin are mixed in different proportions. Optimal amount of added protein. The optimal amount of bovine or horse serum in the precipitation reaction was found by adding in-
(lo0 2.1
200 4.2
300 6.3
LOO
8.4
500 PI bov serum p 10.5 mg/rnl protein
Fig. 4. Amount of radioactivity precipitated with increasing amounts of bovine or horse serum. creasing amounts of serum and PEG to test tubes containing labelled /32-microglobulin, incubated with and without antibodies to B2microglobulin, and counting the radioactivity of the precipitate. The final concentration of PEG in all test tubes was 10 per cent. The result is shown in Fig. 4. Interference with the assay. The possibility of specific or nonspecific interference with the present PEG-RIA from substances in serum or urine was investigated by comparing the slopes of the inhibition curves obtained by incubation of serial dilutions of serum, urine, and solutions of isolated B2-microglobulin in the way described in ‘Quality of isolated p2microglobulin’. No such interference was found (Fig. 3). Influence of pH and temperature. The in-
Radioimmunoassay of P2-Microglobulin
733
Table I. &-microglobulin in serum and urine from healthy subjects /j,-rnicroglobulin in serum, nmol/l
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Women Men Total
P,-microglobulin in urine, nmo1/24 hr
No.
Mean
Range
No.
20 24 44
121 128 125
49-166 98-190 49-190
8 7 15
fluence of pH and temperature on the assay was investigated. At pH values above 8 and below 6.5, less antigen-antibody complex is precipitated, and below pH 5.5 free labelled /&-microglobulin is precipitated. No significant influence of temperature has been observed after incubation at 4,20, and 37 OC. Comparison with other methods. Comparison of determinations in serum and urine by single radial immunodiffusion, the Phadebas B,-rnicroglobulin test, and the PEG-RIA showed no significant difference. There was no systematic difference in the high or low range. PEG-RIA and Phadebas determinations gave r = 0.925 ( n = 16). PEG-RIA and single immunodiffusion determinations gave r = 0.992 (n = 16). Healthy individuals. The results of determinations in serum and urine from these subjects are listed in Table I. DISCUSSION A combination of immunochemical and physicochemical methods was used in the development of a sensitive radioimmunoassay for measurement of B,-microglobulin in human serum and urine. Separation of free and antibody-bound /?,microglobulin was accomplished by precipitation of immunoglobulins by PEG. The underlying mechanism of the precipitation seems to be a ‘solute exclusion’ of larger proteins such as immunoglobulins (4, 11). The amount of PEG used in the assay (final concentration, 10 per cent) was the minimal amount known to precipitate all rabbit immunoglobulins present. To obtain a complete precipitation, it was essential to add a rather large amount of horse
Mean
Range
5.5 12.9 8.7
2.9- 8.7 8.6-19.2 2.9-19.2
or bovine serum to the test tubes, because the precipitated serum globulins will help trap the antigen-antibody complexes in the precipitate. The amount of co-trapped free labelled bzmicroglobulin was small (Fig. 4). In this study the PEG-RIA is compared with two other quantitation methods: SRI and the Phadebas ,92-microglobulintest. The results obtained with the three methods were in good accordance. Compared to SRI, with a detection limit of about 80 nmol/l and an analysis time of 18 hr, the PEG-RIA has the advantage of greater sensitivity and more rapidly obtained results; compared to the Phadebas P,-microglobulin test, PEG-RIA is easier to perform and much cheaper. The lowest standards in both radioimmunoassays are just above 0.2 nmol/l, indicating an equal sensitivity, and the analysis time is about 5 hr in both. The coefficient of variation in the PEG-RIA compares favourably with the Phadebas kit. The simplicity of the PEG-RIA is a great advantage for automation of the assay. The detection limit is 0.042 nmol/l, and the coefficient of variation is 8.1 per cent and 9.0 per cent at microglo globulin concentrations of 8.14 and 1.92 nmol/l, respectively. The assay can be completed within 5 hr and requires only small amounts of serum or urine. The serum concentrations and 24-hr urinary excretions found in normal individuals (Table I) are in accordance with those reported by others (7). In our material, however, there is a sex difference in the 24-hr urinary output of µglobulin. Such a difference has not been reported previously and might be explained by the small number of subjects in our reference population.
734
T . Plesner, B. Norgaard-Pedersen & T . Boenisch
Table 11. Age distribution of healthy subjects in Table I Age of women Age of men (years) (years) Mean Range Mean Range
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Serum determinations Urine determinations
30.5 22-41 42.4 20-62
30.6 25-45 41.6 24-64
I t has been shown that the serum concentration of ~2-microglobulin increases with age, whereas the urinary excretion remains constant (7). The limited number of subjects an d the low mean age i n o u r reference population do not allow any conclusions about age dependence of the serum concentration and urinary output of ,!12-microglobulin (Table 11). The reference population included one woma n in early pregnancy and f o u r women taking oral contraceptives. The /32-microglobulin concentrations i n these subjects did not differ f r o m th e rest of the population.
A C K N O W L E D G E ME NTS
The technical assistance of Miss Laila Geertsen is gratefully acknowledged. ~2-microglobulinwas isolated as outlined by Bengt Johansson and Uffe Ravnskov, University Hospital of Lund, Lund, Sweden (personal communication). This study was supported by the Danish Cancer Society.
REFERENCES LBerggbrd, I. & Bearn, A. G. Isolation and properties of a low molecular weight /j,-globulin occurring in human biological fluids. J. biol. Cheni. 243, 4095, 1968. .2. Bernier, G. M., Cohen, R. J. & Conrad, M. E. Microglobulinaemia in renal faure. Nature (Lond.) 218, 598, 1968. 3. Bernier, G. M. & Fanger, M. W. Synthesis of P,-microglobulin by stimulated lymphocytes. J. Immunol. 109,407, 1972. 4. Desbuquois, B. & Aurbach, G. D. Use of polyethylene glycol to separate free and antibodybound peptide hormones in radioimmunoassays. J. clin. Endocr. 33, 732, 1971.
5.Ekman, R., Johansson, B. G. & Ravnskov, U. A simple method of increasing the resolution in Sephadex @ gel chromatography. Analyt. Biochem. In press. 6. Evrin, P. E., Peterson, P. A., Wide, L. & BerggPlrd, I. Radioimmunoassay of P,-microglobulin in human biological fluids. Scand. J . clin. Lab. Invest. 28, 439, 1971. 7. Evrin, P. E. & Wibell, L. The serum levels and urinary excretion of ,!l,-microglobulin in apparently healthy subjects. Scand. J. d i n . Lab. Invest. 29, 69, 1972. 8.Fermi1-1, E. A., Johnson, C. A., Eckel, R. E. & Bernier, G. M. Renal removal of low molecular weight proteins in myeloma and renal transplant patients. J . Lab. din. Med. 83, 681, 1974. 9. Ganrot, P. 0. Crossed immunoelectrophoresis. Scand. J. clin. Lab. Invest. 29, 39, 1972. 10. Grey, H. M., Kubo, R. T., Colon, S. M., Poulik, M.D., Cresswell, P., Springer, T., Turner, M. & Strominger, J. L. The small subunit of HL-A antigens is b,-microglobulin. J . exp. Med. 138, 1608, 1973. 11. Harrington, J. C., Fenton, J. W. & Pert, J. H. Polymer induced precipitation of antigen-antibody complexes: ‘precipiplex’ reactions. Immunochemistry 8, 413, 1971. 12.Hunter, W. M. pp. 3-23 in Kirkham, K. E. & Hunter, W. M. (ed.) Radioimmunoassay Methods. Churchill Livingstone, Edinburgh and London, 1971. 13.Mancini, G., Carbonara, A. 0. & Heremans, J. F. Immunochemical quantitation of antigens by single radial immunodiffusion. Imniunochemistry 2, 35, 1965. 14.Nakamur0, K., Tanigaki, N. & Pressman, D. Multiple common properties of human B2microglobulin and the common portion fragment derived from HL-A antigen molecules. Proc. nut. Acad. Sci. (Wash.) 70, 2863, 1973. 15. Nilsson, K., Ervin, P. E., Berggird, I. & Pontin, J. Involvement of lymphoid and nonlymphoid cells in the production of p,-microglobulin - a homologue of the constant domains of IgG. Nature (New Biol.) (Lond.) 244, 44, 1973. 16. Nilsson, K., Evrin, P. & Welsh, K. I. Production of /&-microglobulin by normal and malignant human cell lines and peripheral lymphocytes. Transplant. Rev. 21, 53, 1974. 17.Peterson, P. A., Cunningham, B. A., Berggird, I. & Edelman, G. M. P,-microglobulin - a free immunoglobulin domain. Proc. nut. Acad. Sci. (Wash.) 69, 1697, 1972. 18.Peterson, P. A., Evrin, P. E. & Berggird, I. Differentiation of glomerular, tubular and normal proteinuria: determinations of urinary
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Radioimmunoassay of B,-Microglobulin
excretion of B,-microglobulin, albumin, and total protein. J . d i n . Invest. 48, 1189, 1969. 19. Peterson, P. A., Rask, L. & Lindblom, J. B. Highly purified papain-solubilized HL-A antigens contain B,-microglobulin. Proc. nut. Acad. Sci. (Wash.) 71, 35, 1974. 20.Poulik, M. D., Bernoco, M., Bernoco, D. & Ceppellini, R. Aggregation of HL-A antigens at the lymphocyte surface induced by antiserum to B,-microglobulin. Science 182, 1352, 1973. 21. Poulik, M. D., Ferrone, S., Pellegrino, M. A., Sevier, D. E., Oh, S. K. & Reisfeld, R. A. Association of HL-A antigens and µglobulin: concepts and questions. Transplant. Rev. 21, 106, 1974. 22.Ravnskov, U., Johansson, B. G. & Gothlin, J. Received 24 February 1975 Accepted 10 July 1975
735
Renal extraction of P2-microglobulin. Scand. J . clin. Lab. Invest. 30, 71, 1972. 23. Ricanati, E. S. & Philip, W. H. Correlation of proteinuria and renal function in cadaveric transplants. pp. 357-361 in Peeters, H. (ed.) Protides of The Biological Fluids, 21st Colloquium. Pergamon Press, Oxford, 1973. 24. Smithies, 0. & Poulik, M. D. Initiation of protein synthesis at an unusual position in an immunoglobulin gene? Science 175, 186, 1972. 25. Solheim, B. G. & Thorsby, E. b2-microglobulin is part of the HL-A molecule in the lymphocyte membrane. Nature (Lond.) 249, 36, 1974. 26. Wibell, L., Evrin, P. & Bergglrd, I. Serum /j,-microglobulin in renal disease. Nephron 10, 320, 1973.
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Plesner, T., Norgaard-Pedersen, B. & Boenisch, T. Radioimmunoassay of Microglobulin. Scand. J . din. Lab. Invest. 35, 729-735, 1975.
p2-
A radioimmunoassay for quantitation of /?,-microglobulin in serum and urine is described. Polyethylene glycol 6000 was used to precipitate the antigen-antibody complex. The assay can be performed within 5 hr. In healthy human subjects (mean age, 30.6 years and 42.0 years for serum and urine determinations, respectively) the mean concentration of /?,-microglobulin in serum was 125 nmol/l (range, 49-190 nmol/l). In men the mean 24-hr urinary excretion was 12.9 nmol (range, 8.6-19.2 nmol), and in women it was 5.5 nmol (range, 2.9-8.7 nmol). The assay range was 0.23-19.40 nmol/l, and the detection limit was 0.042 nmol/l, using a prolonged incubation time. The coefficient of variation based on 20 determinations in serum dilutions with a concentration of /?,-microglobulin of 8.14 and 1.92 nmol/l was 8.1 per cent and 9.0 per cent, respectively. Key-words: B,-microglobulin; polyethylene glycol precipitation; radioimmunoassay T. Plesner, M.D., Dept. of Clinical Chemistry A , Rigshospitalet, University ol Copenhagen, 9, Blegdnnisvej, DK-2100 Copenhagen 0, Denmark
mic micro globulin was first described by Berggird & Bearn (1) in 1968. They isolated the protein and characterized it as being a singlechain polypeptide with a molecular weight of 11,800. The protein is found in all human body fluids examined. Quantitative determination of B,-rnicroglobulin has been used both as a renal function parameter (2, 8, 22, 23, 26) and to determine the degree of glomerular versus tubular damage (18). A growing field of considerable interest is the in vitro production of j32-microglobulin by normal and malignant cells (3, 15, 16) and the association with the HL-A system (10, 14, 19-21,24,25). /?,-microglobulin has been quantitated by single radial immunodiffusion (SRI) (13) and by solid-phase radioimmunoassay (6). The solidphase radioimmunoassay is now also commercially available as a Phadebas @C kit (Pharmacia, Uppsala, Sweden).
The purpose of this study was to develop a simple, sensitive, and rapid method for quantitation of ,&-microglobulin. MATERIAL AND METHODS Isolation of /?,-microglobulin. Urine from uraemic patients was concentrated by pressure dialysis through a P M 10 dialysis membrane (Amicon). The concentrated urinary proteins were dialysed in a Visking tube against a 0.01 mol/l tris buffer with 0.02 mol/l sodium chloride at p H 7.9 and chromatographed on Whatman DE52 cellulose equilibrated with the same buffer (column K26/40 from Pharmacia). The proteins were eluted with a linear salt gradient from 0.02 to 0.2 mol/l sodium chloride in 0.01 mol/l tris buffer at p H 7.9. Four-ml fractions were collected at a flow rate of 16 ml/hr, and the fractions containing p,-microglobulin were pooled and reconcentrated by pressure dialysis
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
730
T . Plesner, B. Norgaard-Pedersen& T . Boenisch
using a PM 10 dialysis membrane. The concentrated proteins were dialysed against a 0.05 mol/l tris buffer with 0.2 mol/l sodium chloride, pH 7.4, and chromatographed on Sephadex G75 superfine, specially prepared to obtain gel particles of uniform size (5). The gel particles were equilibrated with 0.05 mol/l tris buffer containing 0.2 mol/l sodium chloride, p H 7.4, and packed in column K26/100 from Pharmacia. Two-ml fractions were collected at a flow rate of 11 ml/hr. The fractions containing pure B,-microglobulin were pooled and concentrated in a Visking tube kept in a stream of air at room temperature. The isolated protein was quantitated by reading the optical density a t 280 nm, using a molar extinction coefficient of 19,850. It was stored at -20 OC after dilution with 0.1 mol/l phosphate buffer, pH 7.4, containing 14.5 pmol/l bovine serum albumin. The concentration of ~2-microglobulinin the solutions was 38.80 nmol/l. Standards with p2microglobulin concentrations from 0.23-19.40 nmol/l were prepared by further dilution with the albumin-containing phosphate buffer. Labelling of B2-microglobulin. P2-microglobulin was labelled by a modification of the chloramine-T method (12); 0.39 nmol of 8,microglobulin was labelled with 1 mCi [12SI]NaI (Amersham) to a specific activity of approximately 1.2 mCi/nmol. 3.55 nmol of chloramineT and a reaction time of 30 sec were used. Free and protein-bound lZsI were separated by chromatography on Sephadex G-25. Antibodies. Immunoglobulins produced in rabbits against human B2-microglobulin were obtained from Dakopatts, Copenhagen, Denmark. Lots No. 023 and 064 have been used. By crossed immunoelectrophoresis (9) of proteins from concentrated urine against the antisera, both lots were shown to react specifically with B,-microglobulin. The antibodies were titrated by incubating increasing dilutions of antibodies with a constant amount of labelled ,82-microglobulin (Fig. 1). Assay procedure. 200 p1 of standards or unknown samples were incubated with 200 pl of B,-microglobulin antibodies (lot No. 023 diluted 1:15,000, final dilution) and 100 pl of
ANTIBODY TITRATION
CURVE
C PM
1 x 10-6 250.0
62.5
15.6
3.9
0.975
0.244
ANTIBODY DILUTION
Fig. 1. Antibody-bound radioactivity when a constant amount of 1*51-labelledB,-microglobulin is incubated with increasing dilutions of antibodies
for 3 hr.
labelled ~2-microglobulinin assay buffer (approximately 0.052 pmol 12sI-labelled Ba-microglobulin). After 3 hr of incubation at room temperature 300 pl horse or bovine serum and 1,600 pl of a 15 per cent solution (w/v) of polyethylene glycol 6000 (PEG) in 0.1 mol/l phosphate buffer, pH 7.4, were added. The suspension was mixed on a Whirli-mixer and centrifuged at 3,000 g for 30 min at 4 OC. Subsequently the supernatant was discarded and the radioactivity in the precipitate was counted in a Selectronic gamma counter. The standard inhibition curve was plotted as counts per minute against the B,-microglobulin concentration of the standards (Fig. 2). All determinations were carried out in duplicate, and each assay set up included determinations on: 1) two controls with /&-microglobulin concentrations of 8.14 and 1.92 nmol/l, respectively, prepared by diluting serum in the
Radioimmunoassay of pz-Microglobulin
to be above 5.5 since it has been shown that enzymatic degradation of /&-microglobulin might take place below this value (7). Other reagents. Rabbit immunoglobulins against total normal human urinary proteins (Dakopatts, Copenhagen, Denmark) and antiserum against az-microglobulin (Behringwerke, Marburg/Lahn, Germany) were used to control the purity of isolated /&-microglobulin. Statistical evaluation. Standards and results were compared by linear regression analysis (the least squares method).
STANDARD INHIBITION CURVE
CPM 3OOOO
2oooo
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
131
\
t\
10000
RESULTS
NMOL/L 5
10
p2
15
20
MICROGLOBULIN
for radioFig. 2 Standard calibration diagram immunoassay of /&-microglobulin.
albumin-containing phosphate buffer, 2) one sample from yesterday’s determinations, and 3) two tubes containing 12sI-labelled p2-microglobulin with and without antibodies. The Phadebas P,-microglobulin test available from Pharmacia, Uppsala, Sweden, was used for comparison. The assay was carried out as described by the manufacturer. Samples from healthy individuals. Blood was drawn by venous puncture after minimal stasis during the day from 44 subjectively healthy, ambulant individuals, 24 men and 20 women aged 22-45 years (mean, 30.5 years). The blood was centrifuged after clotting and the serum stored undiluted at -20 OC until analysed. 24-hr urine samples were collected from 7 men and 8 women aged 23-66 years (mean, 42.0 years). The individuals were in usual daily activity during the collection period. The urine was collected in containers with sodium azide and stored at -20 OC until analysed. The pH values of the urine samples were controlled
Titration of antibodies. Fig. 1 shows the result of titration of anti-B,-microglobulin antisera, lot 023, from Dakopatts. Quality of isolated /&-microglobulin. The Pa-microglobulin preparation was found to be pure by agarose gel electrophoresis and Ouchterlony double diffusion against rabbit immunoglobulins to total normal human urinary proteins and antiserum to az-microglobulin. The isolated B,-microglobulin was homogeneous in prolonged electrophoresis and in crossed immunoelectrophoresis against anti-D2-microglobulin antibodies (Dakopatts, Copenhagen, Denmark). Standards from the Phadebas P,-microglobulin test and our standards of isolated p2microglobulin solutions were compared in our radioimmunoassay (PEG-RIA). The correlation between the standards was excellent: r = 0.996 (n = 12). To investigate whether purified p,-microglobulin behaved like the native protein, standard inhibition curves of serially increasing dilutions of urine, serum, and isolated p2-microglobulin in albumin-containing phosphate buffer solution were compared in three ways: first, graphically the curves appeared to have identical slopes (Fig. 3). Second, determination of /&-microglobulin in serum and urine at different dilutions gave identical results. Finally, the curves were transformed to straight lines with identical slopes after logarithmic transformation of the values of both axes. This result was
T . Plesner, B. Norgaard-Pedersen& T . Boenisch
732 CPM
CP
3oooO
2QyK)
30000
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
loo00
20000 .~ 5
10
l 2 MICROGLOBULIN
15
20
25 NMOLIL
Fig. 3. Dose-response curves for dilutions of isolated B2-microglobulin, serum, and urine. 10000
obtained by regression analyses using the least squares method. These findings indicate that buffer solutions of isolated B2-microglobulin can be used as standards when quantitating B2-microglobulin in serum and urine. Detection limit (defined as the smallest quantity of B2-microglobulin that can be distinguished from zero). It is possible to quantitate /32-microglobulin down to a concentration of 0.042 nmol/l if the antibodies are diluted 1:50,000 and the incubation time is increased to 18 hr. Reproducibility. The coefficient of variation, calculated from 20 different determinations on serum pool dilutions with a concentration of B2-microglobulin of 8.14 and 1.92 nmol/l, is 8.1 and 9.0 per cent, respectively. Accuracy. Addition of purified B2-microglobulin to serum or urine resulted in a recovery of 98 per cent, and a close correlation was found between calculated and measured ,!?2-microglobulinconcentrations when sera with known content of B2-microglobulin are mixed in different proportions. Optimal amount of added protein. The optimal amount of bovine or horse serum in the precipitation reaction was found by adding in-
(lo0 2.1
200 4.2
300 6.3
LOO
8.4
500 PI bov serum p 10.5 mg/rnl protein
Fig. 4. Amount of radioactivity precipitated with increasing amounts of bovine or horse serum. creasing amounts of serum and PEG to test tubes containing labelled /32-microglobulin, incubated with and without antibodies to B2microglobulin, and counting the radioactivity of the precipitate. The final concentration of PEG in all test tubes was 10 per cent. The result is shown in Fig. 4. Interference with the assay. The possibility of specific or nonspecific interference with the present PEG-RIA from substances in serum or urine was investigated by comparing the slopes of the inhibition curves obtained by incubation of serial dilutions of serum, urine, and solutions of isolated B2-microglobulin in the way described in ‘Quality of isolated p2microglobulin’. No such interference was found (Fig. 3). Influence of pH and temperature. The in-
Radioimmunoassay of P2-Microglobulin
733
Table I. &-microglobulin in serum and urine from healthy subjects /j,-rnicroglobulin in serum, nmol/l
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Women Men Total
P,-microglobulin in urine, nmo1/24 hr
No.
Mean
Range
No.
20 24 44
121 128 125
49-166 98-190 49-190
8 7 15
fluence of pH and temperature on the assay was investigated. At pH values above 8 and below 6.5, less antigen-antibody complex is precipitated, and below pH 5.5 free labelled /&-microglobulin is precipitated. No significant influence of temperature has been observed after incubation at 4,20, and 37 OC. Comparison with other methods. Comparison of determinations in serum and urine by single radial immunodiffusion, the Phadebas B,-rnicroglobulin test, and the PEG-RIA showed no significant difference. There was no systematic difference in the high or low range. PEG-RIA and Phadebas determinations gave r = 0.925 ( n = 16). PEG-RIA and single immunodiffusion determinations gave r = 0.992 (n = 16). Healthy individuals. The results of determinations in serum and urine from these subjects are listed in Table I. DISCUSSION A combination of immunochemical and physicochemical methods was used in the development of a sensitive radioimmunoassay for measurement of B,-microglobulin in human serum and urine. Separation of free and antibody-bound /?,microglobulin was accomplished by precipitation of immunoglobulins by PEG. The underlying mechanism of the precipitation seems to be a ‘solute exclusion’ of larger proteins such as immunoglobulins (4, 11). The amount of PEG used in the assay (final concentration, 10 per cent) was the minimal amount known to precipitate all rabbit immunoglobulins present. To obtain a complete precipitation, it was essential to add a rather large amount of horse
Mean
Range
5.5 12.9 8.7
2.9- 8.7 8.6-19.2 2.9-19.2
or bovine serum to the test tubes, because the precipitated serum globulins will help trap the antigen-antibody complexes in the precipitate. The amount of co-trapped free labelled bzmicroglobulin was small (Fig. 4). In this study the PEG-RIA is compared with two other quantitation methods: SRI and the Phadebas ,92-microglobulintest. The results obtained with the three methods were in good accordance. Compared to SRI, with a detection limit of about 80 nmol/l and an analysis time of 18 hr, the PEG-RIA has the advantage of greater sensitivity and more rapidly obtained results; compared to the Phadebas P,-microglobulin test, PEG-RIA is easier to perform and much cheaper. The lowest standards in both radioimmunoassays are just above 0.2 nmol/l, indicating an equal sensitivity, and the analysis time is about 5 hr in both. The coefficient of variation in the PEG-RIA compares favourably with the Phadebas kit. The simplicity of the PEG-RIA is a great advantage for automation of the assay. The detection limit is 0.042 nmol/l, and the coefficient of variation is 8.1 per cent and 9.0 per cent at microglo globulin concentrations of 8.14 and 1.92 nmol/l, respectively. The assay can be completed within 5 hr and requires only small amounts of serum or urine. The serum concentrations and 24-hr urinary excretions found in normal individuals (Table I) are in accordance with those reported by others (7). In our material, however, there is a sex difference in the 24-hr urinary output of µglobulin. Such a difference has not been reported previously and might be explained by the small number of subjects in our reference population.
734
T . Plesner, B. Norgaard-Pedersen & T . Boenisch
Table 11. Age distribution of healthy subjects in Table I Age of women Age of men (years) (years) Mean Range Mean Range
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Serum determinations Urine determinations
30.5 22-41 42.4 20-62
30.6 25-45 41.6 24-64
I t has been shown that the serum concentration of ~2-microglobulin increases with age, whereas the urinary excretion remains constant (7). The limited number of subjects an d the low mean age i n o u r reference population do not allow any conclusions about age dependence of the serum concentration and urinary output of ,!12-microglobulin (Table 11). The reference population included one woma n in early pregnancy and f o u r women taking oral contraceptives. The /32-microglobulin concentrations i n these subjects did not differ f r o m th e rest of the population.
A C K N O W L E D G E ME NTS
The technical assistance of Miss Laila Geertsen is gratefully acknowledged. ~2-microglobulinwas isolated as outlined by Bengt Johansson and Uffe Ravnskov, University Hospital of Lund, Lund, Sweden (personal communication). This study was supported by the Danish Cancer Society.
REFERENCES LBerggbrd, I. & Bearn, A. G. Isolation and properties of a low molecular weight /j,-globulin occurring in human biological fluids. J. biol. Cheni. 243, 4095, 1968. .2. Bernier, G. M., Cohen, R. J. & Conrad, M. E. Microglobulinaemia in renal faure. Nature (Lond.) 218, 598, 1968. 3. Bernier, G. M. & Fanger, M. W. Synthesis of P,-microglobulin by stimulated lymphocytes. J. Immunol. 109,407, 1972. 4. Desbuquois, B. & Aurbach, G. D. Use of polyethylene glycol to separate free and antibodybound peptide hormones in radioimmunoassays. J. clin. Endocr. 33, 732, 1971.
5.Ekman, R., Johansson, B. G. & Ravnskov, U. A simple method of increasing the resolution in Sephadex @ gel chromatography. Analyt. Biochem. In press. 6. Evrin, P. E., Peterson, P. A., Wide, L. & BerggPlrd, I. Radioimmunoassay of P,-microglobulin in human biological fluids. Scand. J . clin. Lab. Invest. 28, 439, 1971. 7. Evrin, P. E. & Wibell, L. The serum levels and urinary excretion of ,!l,-microglobulin in apparently healthy subjects. Scand. J. d i n . Lab. Invest. 29, 69, 1972. 8.Fermi1-1, E. A., Johnson, C. A., Eckel, R. E. & Bernier, G. M. Renal removal of low molecular weight proteins in myeloma and renal transplant patients. J . Lab. din. Med. 83, 681, 1974. 9. Ganrot, P. 0. Crossed immunoelectrophoresis. Scand. J. clin. Lab. Invest. 29, 39, 1972. 10. Grey, H. M., Kubo, R. T., Colon, S. M., Poulik, M.D., Cresswell, P., Springer, T., Turner, M. & Strominger, J. L. The small subunit of HL-A antigens is b,-microglobulin. J . exp. Med. 138, 1608, 1973. 11. Harrington, J. C., Fenton, J. W. & Pert, J. H. Polymer induced precipitation of antigen-antibody complexes: ‘precipiplex’ reactions. Immunochemistry 8, 413, 1971. 12.Hunter, W. M. pp. 3-23 in Kirkham, K. E. & Hunter, W. M. (ed.) Radioimmunoassay Methods. Churchill Livingstone, Edinburgh and London, 1971. 13.Mancini, G., Carbonara, A. 0. & Heremans, J. F. Immunochemical quantitation of antigens by single radial immunodiffusion. Imniunochemistry 2, 35, 1965. 14.Nakamur0, K., Tanigaki, N. & Pressman, D. Multiple common properties of human B2microglobulin and the common portion fragment derived from HL-A antigen molecules. Proc. nut. Acad. Sci. (Wash.) 70, 2863, 1973. 15. Nilsson, K., Ervin, P. E., Berggird, I. & Pontin, J. Involvement of lymphoid and nonlymphoid cells in the production of p,-microglobulin - a homologue of the constant domains of IgG. Nature (New Biol.) (Lond.) 244, 44, 1973. 16. Nilsson, K., Evrin, P. & Welsh, K. I. Production of /&-microglobulin by normal and malignant human cell lines and peripheral lymphocytes. Transplant. Rev. 21, 53, 1974. 17.Peterson, P. A., Cunningham, B. A., Berggird, I. & Edelman, G. M. P,-microglobulin - a free immunoglobulin domain. Proc. nut. Acad. Sci. (Wash.) 69, 1697, 1972. 18.Peterson, P. A., Evrin, P. E. & Berggird, I. Differentiation of glomerular, tubular and normal proteinuria: determinations of urinary
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Kainan University on 04/19/15 For personal use only.
Radioimmunoassay of B,-Microglobulin
excretion of B,-microglobulin, albumin, and total protein. J . d i n . Invest. 48, 1189, 1969. 19. Peterson, P. A., Rask, L. & Lindblom, J. B. Highly purified papain-solubilized HL-A antigens contain B,-microglobulin. Proc. nut. Acad. Sci. (Wash.) 71, 35, 1974. 20.Poulik, M. D., Bernoco, M., Bernoco, D. & Ceppellini, R. Aggregation of HL-A antigens at the lymphocyte surface induced by antiserum to B,-microglobulin. Science 182, 1352, 1973. 21. Poulik, M. D., Ferrone, S., Pellegrino, M. A., Sevier, D. E., Oh, S. K. & Reisfeld, R. A. Association of HL-A antigens and µglobulin: concepts and questions. Transplant. Rev. 21, 106, 1974. 22.Ravnskov, U., Johansson, B. G. & Gothlin, J. Received 24 February 1975 Accepted 10 July 1975
735
Renal extraction of P2-microglobulin. Scand. J . clin. Lab. Invest. 30, 71, 1972. 23. Ricanati, E. S. & Philip, W. H. Correlation of proteinuria and renal function in cadaveric transplants. pp. 357-361 in Peeters, H. (ed.) Protides of The Biological Fluids, 21st Colloquium. Pergamon Press, Oxford, 1973. 24. Smithies, 0. & Poulik, M. D. Initiation of protein synthesis at an unusual position in an immunoglobulin gene? Science 175, 186, 1972. 25. Solheim, B. G. & Thorsby, E. b2-microglobulin is part of the HL-A molecule in the lymphocyte membrane. Nature (Lond.) 249, 36, 1974. 26. Wibell, L., Evrin, P. & Bergglrd, I. Serum /j,-microglobulin in renal disease. Nephron 10, 320, 1973.
