8 assay of albumin in serum based on specific ion-bindingand ion-selective electrode (ISE) potentiometry

We have used a new commercial TSH-IRMA (Serono Diagnost i c s , Randolp, MA) to evaluate the p r a c t i c a l s e n s i t i v i t y of the assay in order to c l e a r l y delineate normal subjects from hyperthyroid subjects. The r e s u l t s obtained in our l a b o r a t o r y show t h a t we had to modify the technique to improve the precision o f the assay at the 0.5 uU/ml l e v e l . With t h i s p r o t o c o l , we have been able to obtain an overa l l c o e f f i c i e n t of v a r i a t i o n (CV) o f 20% at the 0.5 uU/ml l e v e l . As the lower l i m i t of the reference range f o r normal patients is 0.5 uU/ml (Cobb et a l . ) i t means t h a t , in our conditions, we can achieve enough s e n s i t i v i t y and precision to d i s t i n g u i s h between normal and hyperthyroid subjects. A preliminary evaluation of hyperthyroid patients has permitted to v e r i f y that the TSH l e v e l s are under the detection l i m i t of the method. ?

A ONE STEP QUANTITATIVE METHOD FOR RIA OPTIMIZATION, Rousseau~ F., Forest, J.-C., Lemay, A. Service de biochimie et le Centre de recherche de l'hSpital Saint-Fran~ois-d'Assise, Quebec, Canada, GIL 3L5. We have developped a one step quantitative method for radioimmunoassay optimlzation, The method is rapid and necessitates only to perform a series of saturation curves with different titres of the antiserum, After calculating the saturation point and the affinity constant of the antiserum with the Scatchard plot, we have produced a table that predicts the main characteristics of the standard curve (B0/T, BO and T) that will prevail for any combination of antiserum titre and percentage of sites saturation. We have developped a microcomputer program able to interpolate all the data needed to produce such a table from the results of the saturation curves. This computer program permits also to predict the sensitivity potential o f the assay at any experimental conditions if the antibody does not discriminate between the labeled antigen and the native antigeno We have tested the accuracy of this optimization table with three in house RIA systems : B-estradiol, h-FSH and h-LH. The results obtained experimentally, including sensitivity determinations, were concordant with those predicted from the optimization table. Our method accelerates and improves greatly the process of optimization of radiolmmunoassays. 8 ASSAY OF ALBt~4IN IN SERUM BASED ON SPECIFIC ION-BINDING AND ION-SELECtIVE ~ D E (ISE) ~ O M E T R Y , T . K . C h r i stopoulos and E . P . D i ~ d i s (Lab.of Analyt.Chem.Univ.of Athens~ Athens 106 80 , Greece) T h e m e t h o d is b a s e d o n t h e b i n d i n g o f an i o n i c lig a n d to a l b u m i n a n d m o n i t o r i n g the free ligand with a ligand-ISE.Two mL of a buffered l i g a n d s o l u t i o n are p i p e t t e d in a b e a k e r a n d t h e e l e c t r o d e potential (EMF) is r e c o r d e d . F i f t y ~ L ( V ~ ) o f s e r u m is t h e n a d ded. T h e b i n d i n g of t h e l i g a n ~ to a l b u m i n c a u s e s an instantaneous potential s h i f t . S m a l l v o l u m e s of a c o n centrated l i g a n d s o l u t i o n a r e t h e n a d d e d to a d j u s t the E M F to i t s i n i t i a l v a l u e ( p o t e n t i o s t a t i c method). The volume,Vc,COnsumed for the EMF adjustment is p r o portional to t h e a l b u m i n c o n c e n t r a t i o n , [ A l b ] , o f the sample.The calibration ourve has the equation Y=A.[Alb]where Y=k V -V a n d k is a d i l u t i o n c o n s t • c, s ant. S l o p e , A , i s determined wlth standard albumin solutions. This linear relationship is d e r i v e d f r o m t h e Scatchard t h e o r y a n d is v a l i d f o r u n l i m i t e d binding sites. Advantages:It is n o t r e q u i r e d to k n o w t h e s l o p e and the constant term of the Nernst equation;standard ligand solutions are not used,precision is e x c e l lent and linearity extends from 5 g/L with no upper limit.The ligands bromcresol purple(BCP) and picrate(PIC) have been used.BCP binds to other serum prot e i n s as w e l l , w h e n b i n d i n g is monitored p o t e n t i o m e t r i cally. Picrate binding was specific.Correlation results were: Y (PIC) =0. 993X (BCP) -0.52, r 0. 984 and Y (PIC) =I. 12X (BCG) -8.58, r=0.974 (all values in g/L and n=26). 9

SOLID-STATE EXTRACTION OF DIGOXIN-LIKE SUBSTANCES FROM HUMAN URINE

A. Papanastasiou-Diamandi,E.P. Diamandis and A. Souvatzoglou, Lab. of Analyt. Chem., Univ. of Athens, Athens 106 80 and Clinic of Therapeutics, Medical School, Univ. of Athens, "Alexandra" Hospital, Athens 115 20 Greece Digoxin-like substances (DLS) have been extracted from human urine by use of Sep-Pak R (octadeeylsilane-bonded silica)

202

cartridges. DLS were preconcentrated 200-fold or more using this fast,simple and reproducible technique. Mean excretion rates of Sep-Pak extractable DLS from human urine (24-h urine collections, 7 individuals)were (63±16)ng/g creatine when DLS were measured with a digoxin RIA kit (SPAC Digoxin, Mallinkrodt). Four individuals were acutely loaded on two occasions with i liter of water each time during a 24-h period. Sep-Pak extractable DLS from the 24-h urine collection, were not significantly increased. When waterload was performed and hourly urine collections were extracted, approximately 20% increase in DLS excretion rate was observed, 2-4 h after the waterload in all four individuals. Sep-Pak extractable DLS were fraetionated with Sephadex G-25 chromatography (i00 cm x 2.5 cm column, 3mL/min with phosphate buffer). DLS were eluted just after urea in at least three separate fractions. DLS were also extracted and preconcentrated 20-fold from human serum with Sep-Pak. Mean initial, Sep-Pak extractable, DLS concentration in serum was (16.0!5.0) pg/mL for five hypertensive patients and (14.6±5.4) pg/mL for five normotensive controls. l0 HPLC FRACTIONATION OF DIGOXIN-LIKE IMMUNOREACTIVE SUBSTANCES IN NEONATAL CORD-BLOOD AND MECONIUM AND ADULT URINE AND BILE, Morris R. Pudek, David W. Seccombe and Karin Humphries. Division of Clinical Chemistry, Vancouver General Hospital and Department of Pathology, University of British Columbia,Vancouver, B.C. V5Z IM9. Neonatal cord blood, adult urine and adult bile containing digoxin-like immunoreactive substances (DLIS) were passed through Sep-Pak R (Waters Scientific Inc.) cartridges as a preliminary purification step. DLIS was eluted from the Sep-Pak R with 100% Methanol and the eluates were dried down under nitrogen. Meconium was extracted using 100% ethanol and the extracts were also evaporated to dryness under nitrogen. Dried samples were reconstituted in 100% methanol and injected into a Varian Model 5000 HPLC system equipped with a Merck RP-18 25 cm column. The samples were eluted from the column with a 10-100% acetonitrile gradient in phosphate buffer (0.01M,pH 3.0). Fractions (30 sec) were collected using an LKB fraction collector, dried down under nitrogen, redissolved in a serum pool containing no DLIS. All fractions were then analyzed for DLIS using NML digoxin antisera lot DB-157. Several DLIS containing fractions were found in each of the extracts fraetionated. The elution volumes on our separation system and the cross-reactivity with the digoxin antisera (lot DB-157) of several steroid compounds were determined, including 14 different bile acids, digoxin and its metabolites, cortisone, estriol and progesterone. It was concluded that the bile acids could account for a large proportion of the DLIS found in bile and meconium but not for the DLIS found in cord blood. The ability of these fractions to inhibit Na+-K+-ATPase was also assessed using 86Rb uptake by red blood cells. ii CISPLATIN INDUCED HYPERURICEMIA. Nanji, A.A., Stewart, D.J., Mikhael, N.Z., Dept. of Pathology, Medicine and Laboratory Medicine, University of Ottawa and Ottawa General Hospital, Ottawa, Ontario KIH 8L6. Hyperuricemia has previously been noted in patients receiving cisplatin and is considered to be a consequence of cisplatin-induced nephrotoxicity. To investigate this matter further, we correlated changes in uric acid in patients receiving cisplatin with liver and kidney levels of platinum. A total of 15 patients with widespread malignancy and receiving cisplatin had serial measurement of serum uric acid. The change in serum uric acid was calculated for each patient from the time of the first dose to death. During the time of the study, the serum uric acid increased for 6.1 ± 1.0 mg/dL to 8.3 ± 1.3 mg/dL. The change in uric acid correlated with both a) total dose of cisplatin (r=0.41, p < 0.05) and liver platinum concentration measured at autopsy (r = 0.82, p < 0.01). There was no correlation with kidney platinum concentrations. The above correlation between change in uric acid and liver platinum and not kidney platinum suggests that overproduction of uric acid, possibly in the liver, may be more important than nephrotoxicity as a cause of hyperuricemia associated with cisplatin treatment. Supported by Health & Welfare, Canada. 12

PLASMA PROTEIN BINDING OF THYROXINE(T4)AND CORTISOL (F) IN PREGNANCY AND ORAL CONTRACEPTIVE THERAPY (BCP). P.M. Keane, Terisa Wong, W.H.C. Walker and Linda Thompson. Department of Laboratory Medicine, Foothills Hospital, Calgary, Alberta and M.U.M.C., Hamilton, Ontario. It is generally assumed that when alterations in thyroxine

CLINICAL BIOCHEMISTRY, VOLUME 18, JUNE 1985