Two types of folate binding proteins are present in serum: a group of low-affinity binders. (K = i0 L/mol), the quanti- tatively most important representative.
systematically to underestimate the proportion should not be used for this purpose.
of Hb 5, it
moglobin by electrophoresis on cellulose acetate. Am. J. Clin. Pat hol. 50, 142-145 (1968). 6. Friedman, H. S., A rapid screening test for abnormal hemoglobins. The detection of hemoglobin A,. Gun. Chim. Acta 7, 100-107 (1962).
We thank Dr. James Eckman for providing several samples and Ms. Lorraine Bryan for assistance in obtaining these samples.
References 1. Bunn, H. F., Forget, B. G., and Ranney, H. M., Human Hemoglobins, W.B. Saunders Co., Philadelphia, PA, 1977, p 170. 2. Petrakis, N. L., Doherty, M. A., Grunbaum, B. W., eta)., Cellulose acetate membranes for the electrophoretic demonstration of hemoglobin A2. Acta Haematol. 27, 96-103 (1962). 3. Rozman, R. S., Sacks, R. P., and Kates, R., Rapid measurement of hemoglobin A2 by means of cellulose acetate membrane electrophoresis. J. Lab. Gun. Med. 62, 692-698 (1963). 4. Briere, R. 0., Golias, T., and Batsakis, J. G., Rapid qualitative and quantitative hemoglobin fractionation, cellulose acetate electrophoresis. Am. J. Clin. Pat hol. 44,695-701(1965). 5. Sheena, A. H., Fox, F. A., Bayha, M. et al., A simple microtechnic for screening abnormal hemoglobins and for quantitation of A, he-
CLIN. CHEM. 27/7,
1247-1249
7. Efremov, G. D., Huisman, T. H. J., Bowman, K., et al., Microchromatography of hemoglobins. II. A rapid method for the determination of hemoglobin A2. J. Lab. Clin. Med. 83,657-664 (1974). 8. Schmidt, R. M., Rucknagel, D. L., and Necheles, T. F., Comparison of methodologies for thalassemia screening by Hb A, quantitation. J. Lab. Clin. Med. 86, 873-882 (1975). 9. Brosious, B. M., Wright, J. M., Baine, R. M., et al., Microchromatographic methods for Hb A, quantitation compared. Clin. Chem. 24, 2196-2199 10.
Huisman,
(1978). W. A., Brodie, A. N., et a!., Miof hemoglobins. III. A simplified procedure for of hemoglobin A,. J. Lab. Clin. Med. 86,700-702
T. H. J., Schroeder,
crochromatography the determination (1975).
11. Abraham, E. C., Reese, A., Stallings, M., et a!., Separation of human hemoglobins by DEAE-cellulose chromatography using glycine-KCN-NaCI developers.Hemoglobin 1, 27-44 (1976). 12. Nie, N. H., Hull, C. H., Jenkins, J. G., et al., SPSS: Statistical Package for the SocialSciences,2nd ed., McGraw-Hill Book Co., New York, NY, 1975, pp 267-360.
(1981)
Evidence That the Low-Affinity Folate-Binding Protein in Erythrocyte Hemolysate Is Identical To Hemoglobin Steen lngemann Hansen, Jan HoIm, and Jprgen Lyngbye Gel filtration studies on erythrocyte hemolysate demonstrated the presence of a folate binding protein, apparently of the low-affinity type, that co-eluteS with hemoglobin. Further, the folate binder eluted with a low salt concentration after DEAESepharose#{174}CL-6B anion-exchange chromatography of erythrocyte hemolysate at pH 6.3. The chromatographic behavior of hemoglobin labeled with
The present study further substantiates this hypothesis. By the combined use of gel chromatography and anion-exchange chromatography we demonstrate that a low-affinity folate binder, probably identical to hemoglobin, is present in erythrocytes.
[3H]folate was so similar to that of the present binder as to suggest that the folate binder in erythrocytes is in fact hemoglobin.
Labeled folate was supplied by the Radiochemical Centre, White Lion Road, Amersham, Buckinghamshire HP7 9LL, U.K. The following three types of radiochemical preparations were used, [2-14C]folic acid, potassium salt (cat. no. CFA. 333), with a specific activity of 55 Ci/mol and a radiochemical purity of 97-99%; [G-’H]folic acid, potassium salt (TRA. 34), with a spec. acty. of 5 kCi/mol and a radiochemical purity of 9597%; and [3’,5’,7,9-3Hjfolic acid, potassium salt (TRK. 212), with a spec. acty. of 29 kCi/mol and a radiochemical purity 93-97%. Hemoglobin was obtained from Sigma Chemical Co., St. Louis, MO 63178. Specimens of venous blood (EDTA stabilized), drawn from 10 healthy volunteers, were pooled. After centrifugation at 1500 X g for 15 mm, the plasma and buffy coat were discarded, and the erythrocytes were washed five times in isotonic NaCI solution (9 g/L) to remove contaminating plasma. In the last wash solution, no albumin was detectable by “rocket” immunoelectrophoresis (13) with monospecific antibody against albumin (Dakopatt, Copenhagen). The washed erythrocytes were diluted with an equal volume of distilled water and repeatedly frozen and thawed five to 10 times. The resulting hemolysate (hemolysis confirmed by microscopic examination) was dialyzed overnight at 4 #{176}C vs 0.17 mol/L Tris buffer,
Two types of folate binding proteins are present in serum: a group of low-affinity binders (K = i0 L/mol), the quantitatively most important representative of this category being albumin (1-3), and a high-affinity binder (K = 1011 L/mol) with characteristics similar to those of other specific folate binders in milk and leukocyte lysates (4-8). Because of its ionic properties, the latter trace protein can be separated from the low-affinity binders by anion-exchange chromatography of serum at near-neutral pH (6-11). However, we have recently reported the presence of a hitherto undescribed lowaffinity binder which, in contrast with other binders of this type, co-elutes with the specific folate binder during anionexchange chromatography of serum (12). The evidence presented suggested that this binder was identical to hemoglobin (12). Department of Clinical Chemistry, Research Division, Hospital, DK-3400 Hillerdd, Denmark. Received Dec. 19, 1980; accepted March 16, 1981.
Central
Materials and Methods
CLINICAL CHEMISTRY, Vol. 27, No. 7, 1981
1247
i0
dpm/g), both supplied by Packard. Hemoglobin concentration was measured absorbance of oxyhemoglobin at 540 nm.
by measuring
the
Results and DiscussIon
Q.
300
ELUTION
VOLUME
(mL)
Fig. 1. Gel filtration profile of erythrocyte hemolysate labeled with [‘l-l]folate, 100 nmol/L (A ) and 1250 nmol/L (U-
-
-),
Hemoglobin concentration, #{149}-. Abscissa, and hemoglobin concentration (right)
elution volume. Ordinate.
cpm (left)
pH 7.4, and stored at -18 #{176}C until use. No albumin was detectable in the hemolysate. For anion-exchange chromatography of erythrocyte hemolysate we used a 2.0 cm2 X 37 cm column of DEAE-Sepharose#{174} CL-6B (Pharmacia, Uppsala, Sweden), which we eluted (5 #{176}C, flow rate 20 mL/h) with 50 mmol/L imidazole buffer (pH 6.3, containing 30 mmol of NaCl per liter) as described earlier (10). Hemolysate and effluent obtained after this anion-exchange chromatography of hemolysate were incubated overnight at 25 #{176}C in the Tris buffer with labeled folate before gel filtration, which was performed on a 5.3 cm2 X 94 cm column of Ultrogel AcA 44 (LKB, Bromma, Sweden). The column was eluted (5 #{176}C, flow rate 50 mL/h) with the Tris buffer (10). For molecular-mass calibration we used substances and proteins of known molecular mass as before (10). Samples (400 tL) digested overnight in 1 mL of “Soluene350” (Packard-Becker B.V. Chemical Operations, Groningen, The Netherlands) were decolorized when they were allowed to stand for 30 mm at 37 #{176}C after addition of 200 isL of 2propanol and 200 jzL of a 300 g/L solution of hydrogen peroxide. Liquid scintillation counting was done as previously described (10) after addition of 10 mL of “Dimilume” (Packard) scintillation fluid. Counting efficiencies were controlled by internal standardization with [‘H]toluene (spec. acty., 2.71 X 106 dpm/g) and [14C]toluene (spec. acty., 5.09 x
Erythrocyte hemolysate labeled in the presence of increasing concentrations of [‘HI- or [14Clfolate was subjected to gel filtration. Two typical elution diagrams are shown in Figure 1. The peak eluted at 340 mL elution volume represents protein-bound [‘H]folate; the large peak at 500 mL elution volume represents unbound [‘Hifolate. As can be seen, hemoglobin co-elutes with the first peak. Other chromatographic runs showed that the relative protein binding of folate, as assessed by integration of the peaks, was rather constant (2-3%) over a wide folate concentration range (10 nmol/L to 1 mmol/L). Hemolysate was applied to an anion-exchange column and eluted with a low salt concentration (30 mmol/L NaC1) at pH 6.3. The effluent, which contained visible amounts of hemoglobin, was labeled with [‘H]folate before gel filtration. The elution profile (Figure 2) was similar to that of Figure 1. Solutions of hemoglobin and of effluent obtained after anion-exchange chromatography of hemoglobin were labeled with [‘H]folate before gel chromatography. The elution diagrams (not shown) were indistinguishable from those of Figures 1 and 2. From these experiments we conclude that the folate binding protein in erythrocytes seems to be identical to hemoglobin. The binding seemed to be of the low-affinity type, inasmuch as relative binding was constant over a wide folate concentration range. In contrast with other low-affinity folate binders such as albumin (12), the erythrocyte binder possessed a weak affinity for anion-exchange columns at near-neutral pH. This is in fair agreement with the supposed identity of the binder, because p1 values within the pH range 7-8 were reported for hemoglobin (14). The physiological role of folate binding in erythrocytes is obscure, but hemoglobin may serve as a transport or storage protein from which folate is easily mobilized.
We acknowledge the valuable technical assistance of Mrs. Jytte Rasmussen and Mrs. Solveig Nordlunde. The study was supported by grants from the Danish Medical Research Council (512-9140 and 5 12-15528), the Bryde Nielsen Foundation, and the Danish Medical Research Foundation Region 3.
References 1. Zettner, A., and Duly, P. E., The weak binding reaction folate and human serum proteins. Ann. Gun. Lab. Science (1978). 2. Elsborg, L., Binding Haematol. 48, 207-212 3. Soliman, H. A., and plasma albumin. Scand.
0
4. Waxman, folate binding (1977).
I. 0
between 8,57-63
of folic acid to human plasma proteins. Acta (1972). Olesen, H., Fo!ic acid binding by human J. Gun. Lab. Invest. 36, 299-308 (1976).
S., Schreiber, C., and Rubinoff, proteins on folate metabolism.
M., The significance of Adv. Nutr. Res. 1,55-75
5. Lyngbye, J., Hansen, S. I., and Ho!m, J., Kinetics of folate-protein binding. In Methods in Enzymology 66, Part E, S. P. Colowick and
N. 0. Kaplan, Eds., Academic Press, New York, NY, 1980, pp 694709.
HoIm, J., Hansen, S. I., binding of folate to proteins Biophys. Acta 629, 539-545 7. Ho!m, J., Hansen, S. I., 6.
ELUTION
VOLUME
(mL)
Fig. 2. Gel chromatography
of effluent obtained on DEAESepharose CL-6B anion-exchange chromatography of erythrocyte hemolysate [‘H]folate
before gel
Effluent
labeled with 1250 nmol/L
1248
CLINICAL CHEMISTRY, Vol. 27, No. 7, 1981
filtration
binding
protein
40, 523-527
in umbilical
and Lyngbye, J., High and low affinity in serum of pregnant women. Biochim. (1980). and Lyngbye, J., A high-affinity folate cord serum. Scand. J. Gun. Lab. Invest.
(1980).
8. Holm, J., Hansen, S. I., and Lyngbye, J., High affinity binding of
folate to a protein in serum of male subjects. Gun. Chim. Acta 100, 113-119 (1980). 9. Waxman, S., and Schreiber, C., Characteristics of folic acid-binding protein in folate-deficient serum. Blood 42, 291-301 (1973). 10. Fischer, C. D., Da Costa, M., and Rothenberg, S. P., The heterogeneity and properties of folate binding proteins from chronic myelogenous leukemic cells. Blood 46, 855-867 (1975). 11. Zettner, A., and Duly, P. E., Separation of folate binding protein from human serum by DEAE-cellulose column chromatography. Clin. Chem. 22, 1047-1052 (1976).
12. HoIm, J., Hansen, S. I., and Lyngbye, J., Characterization and identification of low-affinity folate binding proteins in human serum. IRCS Med. Sci. 8,840-841(1980). 13. Weeke, B., Rocket immuno-electrophoresis. In A Manuau of Quantitative Immuno-electrophoresis Methods and Applications, N. H. Axelsen, J. Kroell, and B. Weeke, Eds., Universitetsforlaget, Oslo, 1973, pp 37-46. 14. Righetti, P. G., and Caravaggio, T., Isoelectric points and molecular weights of proteins. A table. J. Ghromatogr. 127, 1-28 (1976).
CLIN. CHEM. 27/7, 1249-1254 (1981)
3H and 1251 Radioimmunoassays of Haloperidol Compared with Fluoroimmunoassay Involving Antibody Coupled to Magnetizable Solid
Phase F. J. Rowell,’ S. M. Hui,1 and S. R. KameI2 for haloperidol are described, inor 125l-lled drug or tritium-labeled spiroperidol, and a rabbit antiserum to a drug/bovine serum albumin conjugate. The ‘25l-labeled drug was prepared by Radioimmunoassays volving use of tritium-
apy that is often required; such an approach would be expected to reduce the incidence of dyskinesias and other side effects. Specific, precise, and routine assays for neuroleptics should therefore be available. Existing techniques for haloperidol assay include gas chromatography (5), “high-pressure” liquid chromatography (6), and gas chromatography-mass spectrometry (7), techniques not suited for use in routine analysis of many samples. A radioimmunoassay for haloperidol involving tritiated haloperidol (8) and a radioreceptor assay for antischizophrenic drugs, again involving a tritiated label (9), appear to be suitable for monitoring neuroleptic drugs, but tritiated tracers are relatively costly, and their use is time consuming and not
the Chloramine T iodination technique. A fluoroimmunoassay for haloperidol is also described in which the antiserum is coupled to magnetizable solid-phase medium, and fluorescein-labeled haloperidol is used. The assays have acceptable accuracy, precision, and reproducibility, and are specific for haloperidol and similar butyrophenones, with no significant interference from known metabolites and other drugs. Only the radioimmunoassays have sufficient sensitivity to cover the whole range of readily applicable to rapid analysis and automation. haloperidol concentrations in serum. The fluoroimmuWe report here the development and evaluation of a ranoassay can be used to monitor high concentrations of dioimmunoassay for haloperidol with use of 1251, which allows haloperidol
in 15O-i.tL samples or the complete
concen-
tration range of 1-mL serum samples that are extracted and concentrated
before assay.
AddItionalKeyphrases: schizophrenia
‘
monitoring
drug assay
.
therapy treatment neuroleptic drugs
of
Haloperidol, a neuroleptic, is widely used as an intramuscular depot injection or orally, to suppress psychiatric disorders. Its use, however, frequently induces extrapyramidal dysfunction, including persistent and sometimes irreversible dyskinesias, which occur most frequently with high-dose regimens or prolonged treatment (1). Individual variations in steady-state concentrations of total drug (2-4) and “free” drug (4) in sera of patients on identical doses of haloperidol have been reported, but these concentrations, total and “free,” significantly correlate with daily dosage (3, 4). The optimal
therapeutic effect in each patient therefore depends on individualized dosage and drug monitoring during the long ther-
1 School of Pharmacy, Sunderland Polytechnic, Sunderland, SRi 3SD, U.K. 2 Department of Chemical Pathology, St. Bartholomew’s Hospital 51, Bartholomew Close, London, EC1A 7HL, U.K. Received Aug. 28, 1980; accepted March 5, 1981.
semi-automation of the method. In addition, we examined the potential of a fluoroimmunoassay for haloperidol in serum or plasma, in which antibody to haloperidol is coupled to the magnetizable solid phase and a fluorescein-labeled haloperidol analog is the tracer. This technique enables rapid separation in immunoassays, obviates the need for centrifugation (1013), and removes endogenous serum fluorophores and other interfering components in serum or plasma samples before the endpoint determination is made (14).
Materials and Methods Materials Haloperidol hydrochloride was from Searle Lab., Morpeth, Tyne and Wear, U.K.; trifluperidol hydrochloride, spiroperidol hydrochloride, spirilene, moperone hydrochloride, and pipamperone hydrochloride were from Janssen Pharmaceutics N.y., Beerse, Belgium. The haloperidol metabolites were gifts from McNeil Laboratories, Inc., Fort Washington, PA 19034. The other drugs used in the specificity studies were gifts from their respective manufacturers. The [‘H]spiroperidol (spec. acty. 22.1 kCi/mol) was from the Radiochemical Centre, Amersham, U.K., and the [‘H]haloperidol (spec. acty. 10.5 kCi/mol) was from I.R.E., Brussels, Belgium. CLINICAL CHEMISTRY, Vol. 27, No. 7, 1981
1249
