A non-chromatographic radioimmunoassay for plasma aldosterone

A non-Chromatographic radioimmunoassay has been developed for determining plasma aldosterone. Endogenous plasma proteins were used to bind corticosteroids while aldosterone was adsorbed to fuller's earth in an initial purification step. An antiserum to an aldosteronebovine serum albumin conjugate was used for a second purification step. Finally, radioimmunoassay of the purified material was performed using an antiserum prepared against a different aldosterone-bovine serum albumin conjugate. Two ml plasma samples were used for duplicate determinations. Accuracy and precision were satisfactory throughout the analytic range of the method (2.5 to 50 ng/100 ml). Comparison of results using this method with those obtained using an established radioimmunoassay containing a chromatographic purification step indicates that there is a small but tolerable degree of non-specific interference. Twenty-four samples can be assayed in 1 1/2 working days.     

Effect of atrial natriuretic factor on the plasma aldosterone response to potassium infusion in rats--in vivo study

Previous in vitro studies have shown that atrial natriuretic factor inhibits the secretion of aldosterone stimulated by AII, ACTH, and potassium in adrenal cell suspensions. The present study investigated the effects of atriopeptin II on the plasma aldosterone response to a potassium infusion in conscious unrestrained rats in vivo. The infusion of potassium chloride solution increased plasma aldosterone level from 20.4 +/- 3.7 to 168.4 +/- 27.3 ng/dl. The simultaneous administration of atriopeptin II reduced the increase in plasma aldosterone level (16.0 +/- 2.1 to 63.3 +/- 10.4 ng/dl). There was no significant difference in the plasma renin activity, corticosterone, or serum potassium levels between the two groups. These results suggest that atriopeptin II may be important in the regulation of aldosterone secretion.     

Aldosterone and corticosterone in amniotic fluid during various stages of pregnancy

Concentrations of unconjugated aldosterone and corticosterone were measured in amniotic fluid (AF) at different stages of pregnancy. At 9–20 weeks gestation the mean AF level of aldosterone was 14.4±0.7 ng/dl, and of corticosterone 82.9±6.4 ng/dl. Both showed the same pattern during pregnancy, with a rise in AF levels in the last few weeks. At 28–40 weeks gestation the mean AF aldosterone level was 25.5±2.0 ng/dl and the mean AF corticosterone was 218.3±26.6 ng/dl.     

Cortisol metabolism in the Bolivian squirrel monkey (Saimiri boliviensis boliviensis)

New World squirrel monkeys (Saimiri spp.) have high circulating cortisol levels but normal electrolytes and blood pressures. The goal of the present study was to gain insight into adaptive mechanisms used by Bolivian squirrel monkeys to minimize the effects of high cortisol on mineralocorticoid receptor (MR) activity and electrolyte and water balance. Aldosterone levels in serum from 10 squirrel monkeys were 17.7 +/- 3.4 ng/dl (normal range in humans, 4 to 31 ng/dl), suggesting that squirrel monkeys do not exhibit a compensatory increase in aldosterone. The squirrel monkey MR was cloned and expressed in COS-7 cells and found to have similar responsiveness to cortisol and aldosterone as human MR, suggesting that squirrel monkey MR is not inherently less responsive to cortisol. To determine whether altered metabolism of cortisol might contribute to MR protection in squirrel monkeys, serum and urinary cortisol and cortisone were measured, and a comprehensive urinary corticosteroid metabolite profile was performed in samples from anesthetized and awake squirrel monkeys. The levels of cortisone exceeded those of cortisol in serum and urine, suggesting increased peripheral 11beta-hydroxysteroid dehydrogenase 2 activity in squirrel monkeys. In addition, a significant fraction (approximately 20%) of total corticosteroids excreted in the urine of squirrel monkeys appeared as 6beta-hydroxycortisol, compared with that in man (1%). Therefore, changes in cortisol metabolism likely contribute to adaptive mechanisms used by Bolivian squirrel monkeys to minimize effects of high cortisol.     

Metoclopramide,a dopamine antagonist,stimulates aldosterone secretion in rhesus monkeys but not in dogs or rabbits

This study was designed to investigate the role of dopamine in the control of aldosterone secretion in three frequently used laboratory animals. Five New Zealand rabbits, five mongrel dogs and five rhesus monkeys received metoclopramide (MCP) (200 μg/kg iv) and blood samples were collected at 0,5,15,30 and 45 minutes after drug administration. MCP had no effect on plasma aldosterone concentrations at any sampling time in the rabbits or dogs. However, MCP produced a rapid and marked increase in plasma aldosterone from 6.5±0.6 ng/dl to 18.1±2.8 ng/dl at 5 min. and a maximum level of 40.5±4.4 ng/dl at 10 min. after drug administration in the monkeys. MCP had no significant effect on plasma cortisol or plasma renin activity levels in the three species. Prolactin rose in the monkeys from 8.6±1.2 ng/ml to a maximum of 123.5±8.5 ng/ml at 15 min. after MCP. Administration of MCP resulted in a rise in plasma 18-hydroxycorticosterone in the monkeys from 12.5±1.4 ng/dl to a maximum concentration of 50.0±5.1 ng/dl 15 min. after drug administration. Plasma corticosterone, 11-deoxycorticosterone, and 18-hydroxydeoxycorticosterone were not altered by MCP. Although unlikely, it is possible that ketamine may have accounted for some of the changes in plasma aldosterone and 18-hydroxycorticosterone observed after metoclopramide in the monkeys. The findings suggest that dopamine modulates aldosterone biosynthesis in the monkey probably by regulating glomerulosa 18-hydroxylase activity.     

A sensitive sandwich enzyme-linked immunosorbent assay of rat apolipoprotein A-I: effect of various sample treatments on apolipoprotein A-I immunoreactivity and an application to young and aged rat sera.

A highly sensitive sandwich enzyme-linked immunosorbent assay for rat apo A-I was developed. Samples and standards were added to each well of microtiter plates precoated with immunoaffinity-purified IgG. Bound apo A-I was detected with immunoaffinity-purified Fab'-horseradish peroxidase conjugate by a colorimetric method. The sensitivity reached 2.5 pg/well, and the working range for the measurement of serum apo A-I concentration was 0.1 to 1.0 ng/well. The mean intra- and interassay coefficients of variation were 2.8 and 4.1%, respectively. The epitopes of apo A-I in serum were effectively exposed by the use of 6 mol/liter guanidine.HCl. Serum apo A-I concentrations in 36- to 40-week-old rats (62.3 +/- 8.6 mg/dl, mean +/- SD, n = 16) were significantly higher (P less than 0.05) than those in 8- to 12-week-old rats (55.1 +/- 4.3 mg/dl, n = 9). But the age-related change of serum apo A-I was much smaller than that of serum apo E. Apo A-I was contained in smaller HDL particles (or HDL2) in normal rat serum.     

Growth hormone activates renin-aldosterone system in children with idiopathic short stature and in a pseudohypoaldosteronism patient with a mutation in epithelial sodium channel alpha subunit

Growth hormone (GH) treatment causes salt and water retention, and this effect has been suggested to be mediated by activation of epithelial sodium channel (ENaC). Multi-system pseudohypoaldosteronism (PHA) is a salt wasting disease resulting from mutations in ENaC subunit genes. We examined effects of GH therapy for 12-21 months on the renin-angiotensin-aldosterone system (RAAS) in 12 children with idiopathic short stature (ISS) and a PHA patient with defective ENaC function and concomitant GH deficiency. On GH therapy (0.7 U/kg/week), plasma renin activity (PRA), serum aldosterone and insulin-like growth factor-I (IGF-I) levels were periodically determined every 1-3 months in all children. The PHA patient was studied for 6 yr during which time serum, urine, and sweat electrolytes and secretion rate were also examined before, on and off GH therapy. In the PHA patient, mean plasma aldosterone concentration, 7.7 nmol/l (278 ng/dl) before therapy (n=9) rose to 73 nmol/l (2650 ng/dl) 10 months after GH. PRA and IGF-I increased similarly, reaching a plateau between 8 and 12 months. Off GH, there was a decrease to pretreatment levels in 30 months. Aldosterone and PRA strongly correlated with IGF-I (r=0.66 and 0.67). GH therapy also improved the growth rate, and increased both sweat secretion rate and Na(+)/K(+) ratio. In children with ISS, aldosterone and IGF-I peaked 6-12 months after GH. Off GH their levels normalized in 3 months. These findings indicate that long-term GH activates the RAAS in both children with ISS and a PHA patient, and that this effect does not depend on a fully functional ENaC.     

Iodide-induced hypothyroidism in a patient with anorexia nervosa

A 39-year-old woman who had been suffering from anorexia nervosa was found to have hypothyroidism. Serum T4, free T4, T3, free T3 and TSH were 3.19 micrograms/dl, 0.5 ng/dl, 15.3 ng/dl, 1.2 pg/ml and 162.1 microU/ml, respectively. On careful questioning, she was found to have taken an iodine-rich diet. The serum iodine concentration was 122 micrograms/dl (normal: 4-9 micrograms/dl) and urinary iodide excretion was 13.05 mg/day (normal: less than 2 mg). After withdrawal of the iodine-rich diet, her serum T4 gradually increased and TSH returned to the normal range. She was diagnosed as having iodide-induced hypothyroidism. However, no significant elevation of serum T3 or free T3 was observed. Serum T4, free T4, T3, free T3 and TSH were 7.85 micrograms/dl, 0.8 ng/dl, 13.6 ng/dl, 4.3 pg/ml and 6.02 microU/ml, respectively. The iodide-perchlorate discharge test result was negative. These findings suggest that there exists some unknown mechanism by which a patient with anorexia nervosa may be sensitive to excess iodide. Furthermore, it is of interest to note that in a recovery phase from the hypothyroid state, normalization of serum T4 rather than T3 is well-correlated to TSH secretion.     

Effect of atrial natriuretic peptide on adrenal renin and aldosterone

The effect of atrial natriuretic peptide (ANP) on adrenal renin and aldosterone was investigated in anesthetized rats. Under pentobarbital anesthesia 40 mg/kg), intravenous infusion of ANP (0.25 micrograms/kg/min) for 45 min failed to alter the adrenal renin, adrenal aldosterone, and plasma aldosterone (PA). In this condition, intraperitoneal injection of ACTH (10 micrograms/kg) significantly increased the adrenal renin (from 2.4 +/- 0.1 to 5.0 +/- 0.08 ng/mg protein/h, P less than 0.05), adrenal aldosterone (from 13.6 +/- 1.3 to 22.7 +/- 2.3 ng/mg protein, P less than 0.01) and PA (from 59.8 +/- 5.8 to 75.5 +/- 7.4 ng/dl, P less than 0.05), respectively. Under ACTH stimulation, ANP infusion induced significant decreases in adrenal renin (from 5.0 +/- 0.08 to 2.8 +/- 0.2 ng/mg protein/h, P less than 0.05), adrenal aldosterone (from 22.7 +/- 2.3 to 16.2 +/- 1.8 ng/mg protein, P less than 0.05) and PA (from 75.5 +/- 7.4 to 61.6 +/- 4.9 ng/dl). These results suggest a possible role for adrenal renin in the mechanism underlying the inhibitory effect of ANP on aldosterone production in vivo.     

An enzyme-linked immunoassay for the cartilage proteoglycan

Proteoglycan was purified from a rat chondrosarcoma and antiserum prepared. An enzyme-linked immunoassay was designed using this serum. The assay detected rat and murine, but not chick, high-molecular-weight cartilage proteoglycan. It did not detect noncartilage proteoglycan nor the low-molecular-weight proteoglycans found in cartilage. As little as 100 ng/ml of rat cartilage proteoglycan could be detected.     

Effects of moderate short-term potassium depletion in normal humans the role of prostaglandins

The role of prostaglandins (PG) in the effects of potassium (K⁺)depletion was studied in six normal women. A mean K⁺-deficit of 220 mEq was induced with and without concomitant treatment with indomethacin (150 mg/day). Mean serum K⁺ concentration decreased from 4.2 ± (S.E.) 0.1 to 3.2 ± 0.1 mEq/L without indomethacin and from 4.1 ± 0.1 to 3.2 ± 0.1 mEq/L with indomethacin. “Supine” and “upright” plasma renin activity (PRA) and plasma norepinephrine concentration (NE) were unaltered by K⁺ -depletion alone but decreased with indomethacin. Plasma aldosterone (PA) was suppressed during K⁺-depletion (control: 7.2 ± 2.6 ng/dl supine, 19.3 ± 8.1 ng/dl upright; K⁺-depletion: 2.6 ± 0.3 ng/dl supine, 5.5 ± 1.3 ng/dl upright) and was paralleled by a decrease in urinary aldosterone. K⁺-depletion decreased urinary PGE₂ from 667 ± 133 to 343 ± 60 ng/day (P < 0.025) without a change in PGF₂. The dose of exogenous angiotensin II (A II) which increased diastolic blood pressure by 20 mm Hg (pressor dose) was 7.1 ± 1.4 ng/kg/min during control and increased to 11.0 ± 0.7 ng/kg/min during K⁺-depletion (P < 0.05). Indomethacin increased the sensitivity to A II both during control (pressor dose: 4.9 ± 0.6 ng/kg/min) and K⁺- depletion (pressor dose: 6.0 ± 1.0 ng/kg/min). These results indicate that in healthy subjects, moderate short-term K⁺-depletion does not affect PRA or NE but decreases production of aldosterone and PGE₂ by the kidney. The changes in vascular sensitivity to exogenous A II during K⁺-depletion and indomethacin and the decreases in plasma NE and PRA during indomethacin may be explained by changes in vascular vasodilator PG.     

Serodiagnosis of microbiosis in mice with a quantitative assay of immunoglobulins by a microcomputer-introduced ELISA system. 1. Anti-Sendai virus antibodies in naturally infected mouse sera

An enzyme-linked immunosorbent assay system combined with microcomputer data analysis was established as a quantitative assay method of immunoglobulins. The assay system was applied to measure IgG and IgM levels of anti-microbe antibodies in animals, especially mouse and rat. And now the measurement of IgG and IgM levels (ng/ml) of anti-Sendai virus (HVJ) antibodies in naturally infected mice is available. The assay system could improve serodiagnosis in the specificity and sensitivity and in the rapid treatment of many serum samples. The operation of this system was performed by a microcomputer, FM 8 connected Titertek Multiskan MC. The limited sensitivity of this assay for IgG and IgM was 10 ng/ml and 30 ng/ml, respectively. Ninety-one of serum samples were positive for IgG and/or IgM (45 samples for IgG and IgM, 44 samples for IgG, 2 samples for IgM) to Sendai virus in the tested 279 mouse sera, and serum titers were ranged from 1: 10 to 1: 12,800 in the IgG, and from 1: 20 to 1: 160 in the IgM. In these titers, serum IgG and IgM amounts were estimated to be 0.1 to 154 micrograms/ml and 0.5 to 4.8 micrograms/ml, respectively. Relationships of serum titers and antibody amounts were almost consisted, being judged like that approximately 10 micrograms/ml is 1: 400, 30 micrograms/ml is 1: 1,600 in IgG, and 2.4 micrograms/ml is 1: 80, 4 micrograms/ml is 1: 160 in IgM.     

Association of IGF-I and the IGF-I/IGFBP-3 ratio with plasma aldosterone levels in the general population

Several studies in patients with acromegaly or growth hormone (GH) deficiency suggest a stimulatory effect of the growth hormone (GH)/insulin-like growth factor I (IGF-I) axis on the renin-angiotensin-aldosterone system (RAAS). We analyzed the association of serum IGF-I with plasma aldosterone and the aldosterone-to-renin ratio in a large sample from the general population. In addition to serum IGF-I levels, we also considered the IGF-I-to-IGF binding protein (IGFBP)-3 ratio. A total of 1 504 men and 1 566 women aged 25-88 were selected from the first follow-up of the population-based Study of Health in Pomerania. Plasma aldosterone and renin concentrations, as well as serum IGF-I and IGFBP-3 levels were determined with immunoassays. Analyses of variance and linear regression analyses were performed. We found positive associations between serum IGF-I or the IGF-I/IGFBP-3 ratio with plasma aldosterone in women but not in men. Plasma aldosterone levels increased by 2.91 ng/l per IGF-I standard deviation (SD) and by 2.17 ng/l per IGF-I/IGFBP-3 SD. The associations remained significant after exclusion of subjects taking RAAS-altering medication and of subjects with serum IGF-I levels and aldosterone-to-renin ratios outside the reference range. We conclude that, serum IGF-I and the IGF-I/IGFBP-3 ratio are associated with plasma aldosterone levels in women but not in men from the general population.     

Effect of aldosterone on vascular angiotensin II receptors in the rat

The effect of aldosterone on the density and affinity of binding sites for 125I-labelled angiotensin II was investigated in a particulate fraction prepared from the rat mesenteric arteriolar arcades. The infusion of aldosterone 6.6 micrograms/h intraperitoneally via Alzet osmotic minipumps for 6 d produced an increase in the density of binding sites for 125I-labelled angiotensin II without change in affinity. After sodium depletion, mesenteric artery angiotensin II receptors were down-regulated as expected. An increase in the number of binding sites could be found when aldosterone was infused into sodium-depleted rats with no change in the elevated plasma renin activity. The intraperitoneal infusion of angiotensin II (200 ng X kg-1 X min-1 for 6 d) simultaneously with aldosterone resulted in down-regulation of vascular angiotensin II receptors, whereas after intravenous angiotensin II infusion (at 60 ng X kg-1 X min-1) the density of angiotensin II binding sites rose with aldosterone infusion. Plasma renin activity (PRA) was reduced and plasma angiotensin II increased in a dose-dependent fashion after angiotensin II infusion. An aldosterone concentration of 3 ng/mL for 18 h produced an increase in the number of angiotensin II binding sites in rat mesenteric artery smooth muscle cells in culture. We conclude that increased plasma aldosterone may result in up-regulation of vascular angiotensin II receptors independently of changes in plasma renin activity, and may in certain physiological states effectively antagonize the down-regulating action of angiotensin II.     

Estimation of total and direct serum bilirubin using modified micro assay method

Estimation of total and direct bilirubin in serum plays an important role in differential diagnosis of hyperbilirubinemia. Several direct spectrophotometric methods are commercially available for total and direct bilirubin estimation in which the amount of the sample (serum) varies from 200 ml to 800 ml. It is difficult to collect such amount of serum from infants, as neonatal jaundice is the most common problem in this age group. To overcome this problem modified micro assay method was developed using dimethylsulfoxide (DMSO). The amount of the serum sample is reduced from 100 ml to 20 ml per test for both total and direct bilirubin. A method comparison study was performed using 100 consecutive serum samples, by modified micro assay method and a reference Jendrassik-Grof method. Total bilirubin in these human serum samples ranged from 0.4-15.0 mg/dl and direct bilirubin ranged from 0.05-12.0 mg/dl. The results conclude that modified micro assay method had significant correlation with r-value of 0.99989 for total serum bilirubin and with r-value of 0.99971 for direct serum bilirubin. Linearity of the method is 20 mg/dl and 15 mg/dl for total and direct bilirubin, respectively. Monoreagent used during the assay is stable for 24 hours at 2-8 degrees C while the kit is stable for one year at 2-8 degrees C. In conclusion this micro assay method is rapid, reliable, simple and accurate for the estimation of total and direct bilirubin with small serum quantities. It is equally reliable for manual; semi automated and automated chemistry analyzers.     

A bispecific antibody based assay shows potential for detecting tuberculosis in resource constrained laboratory settings

The re-emergence of tuberculosis (TB) as a global public health threat highlights the necessity of rapid, simple and inexpensive point-of-care detection of the disease. Early diagnosis of TB is vital not only for preventing the spread of the disease but also for timely initiation of treatment. The later in turn will reduce the possible emergence of multi-drug resistant strains of Mycobacterium tuberculosis. Lipoarabinomannan (LAM) is an important non-protein antigen of the bacterial cell wall, which is found to be present in different body fluids of infected patients including blood, urine and sputum. We have developed a bispecific monoclonal antibody with predetermined specificities towards the LAM antigen and a reporter molecule horseradish peroxidase (HRPO). The developed antibody was subsequently used to design a simple low cost immunoswab based assay to detect LAM antigen. The limit of detection for spiked synthetic LAM was found to be 5.0 ng/ml (bovine urine), 0.5 ng/ml (rabbit serum) and 0.005 ng/ml (saline) and that for bacterial LAM from M. tuberculosis H37Rv was found to be 0.5 ng/ml (rabbit serum). The assay was evaluated with 21 stored clinical serum samples (14 were positive and 7 were negative in terms of anti-LAM titer). In addition, all 14 positive samples were culture positive. The assay showed 100% specificity and 64% sensitivity (95% confidence interval). In addition to good specificity, the end point could be read visually within two hours of sample collection. The reported assay might be used as a rapid tool for detecting TB in resource constrained laboratory settings.     

Radioimmunoassay of apolipoprotein A-I of rat serum.

A double antibody radioimmunoassay technique was developed for quantification of apolipoprotein A-I, the major apoprotein of rat high density lipoprotein. Apo A-I was labeled with 125I by the chloramine-T method. 125I-labeled apo A-I had the same electrophoretic mobility as unlabeled apo A-I and more than 80% of the 125I was precipitated by rabbit anti apo A-I antibodies. The assay is sensitive at the level of 0.5-5 ng, and has intraassay and interassay coefficients of variation of 4.5 and 6.5% respectively. The specificity of the assay was established by competitive displacement of 125I-labeled apo A-I from its antibody by apo A-I and lipoproteins containing apo A-I, but not by rat albumin and other apoproteins. Immunoreactivity of high density lipoprotein and serum was only about 35% of that of their delipidated forms when Veronal buffer was used as a diluent. Inclusion of 5 mM sodium decyl sulfate in the incubation mixture brought out reactivity equivalent to that found after delipidation. Completeness of the reaction was verified by comparison with the amount of apo A-I in chromatographic fractions of the total apoprotein of high density lipoprotein. Content (weight %, mean values +/- S.D.) of immunoassayable apo A-I was: 62.3 +/- 5.9 in high density lipoprotein; 1.7 +/- 0.3 in low density lipoprotein; 0.09 +/- 0.03 in very low density lipoprotein and 25.0 +/- 5.0 in lymp chylomicrons. Concentration in whole serum was 51.4 +/- 8.9 mg/dl and 33.6 +/- 4.1 mg/dl for female and male rats, respectively (p less than 0.002), equivalent to the sex difference in concentration of high density lipoprotein. 95% of the apo A-I in serum was in high density lipoprotein, 5% in proteins of d greater than 1.21 g/ml and less than 1% in lipoproteins of d less than 1.063 g/ml.     

Decreased serum 19-nor-deoxycorticosterone (21-hydroxy-19-nor-4-pregnene-3, 20-dione) level in adrenal regeneration hypertensive rats

An enzyme immunoassay of 19-nor-deoxycorticosterone in rat serum was established. The normal value of 19-nor-DOC in rat serum obtained from 9:00 am to 10:00 am was 148 +/- 30 ng/dl (mean +/- SE,n = 10). Serum levels of this steroid decreased in rats with adrenal regeneration hypertension during the course of the experiment up to 8 weeks, while systolic blood pressure rose progressively. We concluded that this mineralocorticoid is not involved at least as a circulating hormone in the pathogenesis of adrenal regeneration hypertension in rats.     

Localization of aldosterone-producing adenoma: venous sampling in primary aldosteronism

Adrenal venous sampling of blood was performed for nine patients with aldosterone-producing adenoma (APA). Measurement of adrenal venous aldosterone is useful for localization of APA but difficult, because catheterization of the right adrenal vein is not easy, and the blood is diluted by nonadrenal flow. To solve these problems, levels of aldosterone (A; ng/dl) and cortisol (C; micrograms/dl) were measured in samples from the left adrenal vein (LAV) and the inferior vena cava (IVC), and the LAV A/C and (LAV A/C)/(IVC A/C) ratios were calculated. These ratios were also obtained for 16 patients with essential hypertension. The adenoma could be localized in three of the nine cases by the measurement of aldosterone alone, but the use of a LAV A/C ratio greater than 5 x 10(-3) and a (LAV A/C)/(IVA A/C) ratio less than 1.0 as criteria separated the patients into those with a left APA, right APA, or essential hypertension. Consequently, adrenal venous sampling and the calculation of these ratios enables preoperative localization of APA with more accuracy, especially when the tumor is small or the result of CT and adrenal scintigraphy is not consistent.     

Sandwich Enzyme Immunoassay for Rat Retinol-Binding Protein Using Antibody against Recombinant Antigen and Its Application

The development of a sandwich enzyme immunoassay for rat retinol-binding protein using molecular biological techniques was described. Rat retinol-binding protein gene cloned by the PCR method was expressed by a fusion vector pEZZI8 in Escherichia coli strain HB101. A recombinant retinol-binding protein fused with IgG-binding domain ZZ of protein A was purified with IgG-Sepharose. Antibody against the recombinant protein was found to be specific to rat retinol-binding protein in plasma by immunoblot analysis. Affinity-purified anti-recombinant protein IgG was biotinylated and used for the sandwich enzyme immunoassay. In this assay, the measurable range is 1.9-60 ng/ml and the coefficients of variation within and between the assay series (assay range: 4-30 ng/ml) are 4.30 ± 4.33 and 5.32 ± 1.45%, respectively. Cross-reactivity of the immunoassay was examined using bovine, human, and mouse serum. There was a cross-reaction only with mouse serum. In an in vitro experiment, retinol-binding protein produced by rat hepatocytes could be measured by the sandwich enzyme immunoassay.