We now describe an RIA, developed at the Littlemore. Hospital. Research Unit, for which extraction is not required. The reagents for the radioimmurio- assay.
tion.
Peter J. Howanitz Joan H. Howanitz
Both the patient’s specimen and
metrizamide solution exhibited negligible absorbance at 540 nm, but both
showed substantial absorbance at 340 nm. Metrizamide has a major absorption peak at 230 nm, a minor peak at 280 nm. Thus, if CSF specimens contain metrizamide, the ultraviolet absorbance peaks are so broad that they contribute to measurements made at 340 nm, thereby spuriously increasing CSF protein values measured with the aca. Metrizamide added to a pooled specimen of CSF increased linearly the absorbances obtained with the aca method (y = 4.71x + 20.3). Methicillin similarly interferes with the aca determination of protein in CSF (3). Ordinarily, spinal fluid for use in laboratory determinations is withdrawn before metrizamide is administered, but on rare occasions physicians may obtain a specimen after its administration. It takes about 24 h for the metrizamide to be absorbed from the CSF, so specimens obtained during this interval will potentially show this interference. In the past, myelography contrast material used was insoluble in CSF and was removed by aspiration after the procedure. Any insoluble contrast material remaining in a CSF specimen was easily recognized because the specimen was cloudy. In contrast, the increased use of water-soluble radiographic contrast material will result in CSF specimens that, even though clear, contain material that can interfere with measurements in the ultraviolet range, leading to spurious results. If the usual amounts of metrizamide are administered with CSF (-150
and uniform mixing mL) occurs, a metriza-
mide concentration in CSF of about 50 mmollL would result. However, the relative density of metrizamide solution is 1.184, whereas that of CSF is only 1.007. Thus metrizamide may stratify in the CSF and, if the patient is upright, its concentration would be greatest in the lumbar region, from which CSF is sampled. References 1. CFP cerebrospinal fluid protein. Chemisty Instruction Manual, Du Pont, Clinical Systems Division, Wilmington, DE, 1982. 2. Meulemans 0. Determination of total protein in spinal fluid with sulphosalicylic acid and trichloroacetic acid. Clin Chim Acta 5, 757-761 (1960). 3. Holmes EW, Kahn SE, Bermes EW. Methicillin interference with the Du Pont aca cerebrospinal fluid protein assay. Clin Chem 28, 1235-1236 (1982). Robert L. Sunheimer Lawrence P. Gordon Pathology Crouse-Irving Syracuse, NY
Memorial 13210
Hosp.
00 #{149}0
Clin. Pat hol. Upstate Med. Center Syracuse, NY 13210
50 70
-
so
-
40
00
Direct Radioimmunoassay
for
20
Melatoninin Plasma
2
To the Editor:
Ltd.,
London,
U.K.);
and
other common reagents, all “Analar” grade (British Drug Houses, Poole, Dorset, U.K.). The antiserum (Guildhay batch no. 704/189), supplied by the Department of Biochemistry, University of Surrey, was raised against Nacetyl-5-methoxytryptophan, with thyroglobulin or bovine serum albumin as the protein carrier, and conjugated with carbodiimide (2). Blood samples for analysis were collected in tubes containing lithium hep-
Table 1. Precision of the RIA for Melatonln Pool
1
Pool 2
Pool
3
Within-assay Mean, pmol/L
56 6 13.8
1
6.5
6 7.2
189 25.8
391 25.8
n
60 8.6 13
CV, %
14.3
17 13.6
16 6.6
SD, pmol/L n
CV, %
417 0 6
Between-assay
Mean, pmol/L SD, pmol/L
396 CLINICAL CHEMISTRY, Vol. 29, No. 2, 1983
5
2 5
25
50
00
250
Fig. 1. Parallelism of dilutions of three different
Many of the physiological functions of the pineal gland are mediated by its hormone melatonin (1). Investigations into the endocrine function of the pineal in humans rely heavily on the quantitative estimation of melatonin, especially in plasma. Radioimmunoassay (RIA) is the most widely used means of measuring melatonin (2); where possible, however, gas chromatographymass spectrometry (GC-MS) is used (3). Both RIA and CC-MS involve at least one extraction stage before assay. This being both time consuming and expensive, a direct method has obvious advantages. We now describe an RIA, developed at the Littlemore Hospital Research Unit, for which extraction is not required. The reagents for the radioimmurioassay were obtained as follows: activated charcoal; N-[tris(hydroxymethyl) methyliglycine (Tricine); dextran; and melatonin (Sigma Chemical Co., St. Louis, MO 63178); [3H]melatonin (specific activity 30.4 kCilmol (New England Nuclear, Boston, MA 02118); ethanol, distilled before use (James Burroughs
5
plasma samples (U,A, +) with the duplicatestandard curve (#{149}) arm, centrifuged without delay at 1500 x g for 15 mm at 4 #{176}C, and stored
frozen at -20 #{176}C until assayed. A melatonin-free plasma pool, used for making up standards and for diluting unknown plasma samples, was obtained from volunteers within the laboratory. Blood samples were taken in mid-afternoon, when melatonin concentrations have been shown by CCMS usually to be below the concentration detectable by present RIAs (3). Before use, all plasma samples were individually screened and any apparently containing melatonin were rejected.
Solutions
of working
antibody
and [3Hlmelatonin and suspensions of activated charcoal were made up in the assay buffer (Tricine,0.1 mol/L, pH 8.0, containing 9 g of NaCl and 1 g of gelatin per liter). These solutions and the standards were made up freshly for each assay. Where possible, we used disposable plastic apparatus, and cleaned all essential glassware in chro-
mic acid before use, to eliminate
con-
tamination.
To 500 L of each sample or standard (2.5, 5, 12.5, 25, 50, 100, and 250 pg per tube), add 200 L of antibody (initial dilution 1000-fold), vortex-mix, and leave at room temperature for 30 mm. Add 100 L of [3Hlmelatonin (4000 cpm, per tube), mix again, and incubate for 18 h at 4#{176}C. Separate the antibody-bound melatonin from the free fraction by incubation with 0.5 mL of dextran-coated charcoal (0.1 g of dextran 75 plus 10 g of charcoal per 500 mL of buffer). Centrifuge at 1500 x g for 15 mm at 4 #{176}C and decant the supernate into 10 mL of scintillation fluid. Count the radioactivity in all tubes with a beta scintillation counter and determine the melatonin concentrations from the dose-response curve. Eight replicate assays showed that the precision throughout the dose curve was satisfactory (CV 1.7% to 4.9%) with a sensitivity (defined as twice the standard deviation of maximum binding) of 5.5 pg per tube (47 pmoWLI.
ably well with
Table 2. Concentrations of Melatonin in 13 Plasma as Determined Present
by Two RIAs
RIA
Comparison
RIA
(sensitIvity
(sensitivity
47 pmol/L)
129 pmol/L)
202
149 ND 173 469 ND ND 260 223 248 229 198
112 211 460 86 103 172
288 366 187
224
using
and
The only measurable cross reactivity, expressed relative to the quantity of melatonin that displaced 50% of antibody-bound [3Hlmelatonin, was from 6-hydroxymelatonmn (0.4%), N-acetyltryptamine (0.35%), and N-acetylserotonin (0.03%). Cross reactivity of the other compounds tested was
