Annals of Clinical Biochemistry, 1978, IS, 25-30
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay P. J. WOOD· AND V. MARKS From the Department of Biochemistry, University of Surrey, Guildford Antisera, raised against 2'-O-succinyl-cGMP conjugated to ovalbumin or kehole limpet haemocyanin, have been used to develop radioimmunoassays for the direct d~term~nation of uril":e and plasma cGMP levels. Details of these assays are presented, together with evidence for their validity.
SUMMARY
cGMPt has become the subject of considerable interest during the last five years as an alternative 'second messenger' to cAMPt for the actions of some hormones. There is now evidence that cholinergic and IX-adrenergic stimulation and the actions of certain peptide hormones-such as cholecystokinin, histamine, and possibly insulin-are mediated by anincrease in the intracellular level of cGMP rather than of cAMP (Nature, 1973). In some tissues, changes in the cAMP/cGMP ratio may determine the nature of the response to hormone stimulation (Goldberg et al., 1975), and it should, therefore, be of interest to be able to monitor both cyclic nucleotides in tissues and extracellular fluids. In mammalian tissues cGMP is present at approximately one-tenth to one-hundredth the concentration of cAMP. Human plasma and urine cGMP/cAMP ratios are approximately 1:2 and 1:10 respectively. Consequently, an assay system of high sensitivity as well as of high specificity is required to measure cGMP in the presence of cAMP and other nucleotides, Protein-binding assays for cGMP, which utilise intracellular receptor proteins from lobster tail muscle (Murad et al., 1971), silkmoth pupae (Fallon and Wyatt, 1975), or rat lung (Kobayashi and Fang, 1975) lack sensitivity and are not specific enough to be used for the analysis of tissue extracts without preliminary nucleotide fractionation. The measurement of cG MP by radioimmunoassay (Steiner et al., 1972b) has the advantages of greater sensitivity and specificity over protein-binding
methods and for this reason it was decided to develop such an assay system.
Materials and methods ANTISERUM PRODUCTION
2'-O-Succinyl cGMP (a gift from the Boehringer Corporation) was conjugated to ovalbumin (Sigma Fraction V) or to kehole limpet haemocyanin (CalBiochem) using the carbodiimide 'EDC' (l-ethyl-3 (3-dimethyl-amino-propyl)-earbodiimide Hel) in accordance with the method of Steiner et al. (1972b). The conjugates were estimated by making spectroscopic measurements of solutions of the conjugates, unconjugated 2'-O-succinyl cGMP, and proteins, to contain six molecules of cGMP per molecule of ovalbumin or 160 molecules of cGMP per molecule of kehole limpet haemocyanin (average mol. wt. 3 x 106 Daltons). The conjugates were stored freezedried in aliquots containing 1 mg protein. Five white half-lop rabbits (Ranch Rabbits Ltd) were each given a primary immunisation in six intramuscular sites with a total of 1 mg of the 2'-o-succinyl cG MP-ovalbumin conjugate per rabbit. The conjugate was dissolved in sterile water and emulsified in two volumes of 'Marcol 52' adjuvant (Marks et al., 1974) containing Bacillus Calmette Guerin (BCG; 3 mg/rng conjugate). The rabbits were given seven boosts at monthly intervals using the same regimen, but with the omission of BCG from the adjuvant. Blood samples were taken from an «:a:r vein 10 days after each boost. Because serum antibody titres remained very low on this regimen, the • Present address: Department of Chemical Pathology 2'-O-succinyl cGMP-kehole limpet haemocyanin Southampton General Hospital, Southampton conjugate, suspended in the same adjuv:ant system tcGMP is used as the abbreviation for guanosine-3'S'-cyclic as before, was then used to boost the rabbits (0'25 mg phosphate, cAMP as the abbreviation for adenosine-3'S'protein per rabbit given as six intramuscular cyclic phosphate 25
Downloaded from acb.sagepub.com by guest on June 4, 2016
26
P. J. Wood and V. Marks
injections). After the second monthly boost with this conjugate, antibody titres rose to between 1 in 50 to 1 in 100 (working dilution). Antisera from two of the rabbits, T2 and T5, proved highly specific for cGMP. These were diluted 1 in 5 with buffer (50 mmol/l Tris, pH 7'5, containing theophylline 8 rnmol/l and 2-mercaptoethanol 6 mmolfl) and stored freeze-dried in small aliquots. CHARACTERISATION OF ANTISERA T2 AND
T5 The influence of pH on the binding of 3H cGMP was studied over the pH range 3'5 to 9·0 using 50 mmol/l acetate, phosphate, or Tris buffers containing theophylline (8 rnmol/l) and 2-mercapto ethanol (6 mmol/l), The two antisera differed in their ability to bind the label at different pHs. Antiserum T2 showed a steady level of binding over the pH range 4'5 to 9,0, while antiserum T5 showed increased binding at pH 4·5. Further studies on the properties of both antisera were carried out at pH 7'5, using the assay protocol described below for the analysis of urine (Table 1). The length of incubation had little effect on the standard curve for either antiserum. Incubations of 5, 30, 60, 90, or 180 minutes yielded very similar standard curves, and the 5-minute and 3-hour incubation standard curves were not significantly different when compared by a paired Student's t test on the standard points. Test results for incubations of less than 30 minutes were unreliable, however. An incubation time of three hours was routinely used for the analysis of test samples.
Table 1 cGMP radioimmunoassay protocol: specimens. Incubations in duplicate polystyrene tubes (Luckham Ltd: type LP3) Additions (/"I)
Tests
Standards Non-specific hinding
Assay buffer (Tris 50 mrnoljl, pH "5, containing disodium EDTA,4 mmol/I) Test urine (Diluted 1 in 20 in assay buffer) Standards (0-10 pmol cGMP per 100 /"1 in assay buffer) (8·'HlcGMP (Amersham TRK 392; 1 pmol/50 /"1 in assay buffer) Antiserum (T2 or T5; working dilution 1 in SO in assay buffer) Mix, incubate for 3 hours at 4 "C, then add: Bovine y globulin (Sigma fraction II; 10 mg/ml assay buffer) (NH.),SO. Solution (lee-cold; 3'8 mol/I)
100
100
(NSB)
30
?:' :~
o
20
Q
"0
o ....
a ;§
10
"if!.
o o
0-1 1 10 Nucleotide level (pmol tube. log scale)
100
100 100 50
50
50
100
100
10
10
10
1'0 ml
1·0 ml
1'0 ml
Scatchard plots of the data from five standard curves using antiserum T5 were linear, and were interpreted in terms of a single antibody population with a mean dissociation constant Kn = 3·1 x 10- 9 rnol/l. A similar treatment of the data from two standard curves using antiserum T2 yielded curved Scatchard plots which could be interpretcrf- in terms
E u
300
Mix, leave at 4°C for 10 min, centrifuge for 30 min at 1800 g (Mistral4L Centrifuge). Aspirate supernatants and redissolve precipitates in 1'0 ml distilled water. Mix and take 0'5 ml aliquots for liquid scintillation counting.
40
"0 C
urine
1000
Cross-reactivity of 17 nucleotides with antiserum T5 determined using the incubation procedure described in Table I. A: 2'-O-succinyl-3'5' cGMP B: 3'5' cGMP C: monobutyryI-3'5' cUMP D: dibutyryl-3'5'-eAMP E: dibutyryI-3'5'-eG MP F: 3'5'ciMP G: 3'5' cAMP H: 3'5' cTMP I: 3'5' cCMP J: 3'5' cUMP K: Mean result (± s.e.m.) for 5'-AMP, 5'-i\.TP, 5'GTP, 2'3' cAMP, 2'3' cGMP, AMP-PNP, GMP-PNP. i: Mean result (± s.e.m.j-nucleo... tides which are not plotted 10000 individually.
Fig. 1 Specificity of antiserum T5.
Downloaded from acb.sagepub.com by guest on June 4, 2016
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay full! 5
Blanks (Phosphodiesterase treated)
4 cGMP pmclltube
100 ...1 aliquots from II normal urine samples were incubated with 100 ...1 phosphate buffer (0,5 mol/I, pH 7'5) and 20 ...1 (0'03 U) ox-heart phosphodiesterase (Boehringer) for I hour at 37°C. Reaction tubes were then placed in a boiling water bath for 10 minutes. Incubation mixtures were made to 2·0 ml and their cGMP content was determined using antiserum T2.
•• • • •• •
3 2
• • 0 10
8 cGMP pmcl/ml
6 4
2
eeeuee e ce e e
1!m
(-ECTA)
Blanks (noEDTA)
• • • ••
cGMP levels in plasma from 18 normal subjects, collected with or without disodium EDTA. EDTA-plasma was stored at - 20°C, while 'blank' samples were left at room temperature overnight (for details see the text).
•• •• •• •• I
0
Fig. 2 (top)
27
DDDDDDDDCCCOCCCDDD
'Blank' urine levels;
(bottom) 'Blank' plasma levels.
of two antibody populations with mean dissociation constants Kni = 1·0 x 10- 9 mol/l and KD2 = 1·8 X 10- 8 mol/I. Figure 1 illustrates the crossreactivities of 17 nucleotides in the assay system using antiserum T5. Similar results were obtained with T2 (not shown). In particular cAMP was 10000 times less potent in displacing the [3H} cGMP label from the antiserum than was cGMP.
result of the change to the Tris/EDTA buffer. There was no evidence of rapid degradation of cGMP, and samples showed no loss of cGMP content after being left at either room temperature or at 4°C for periods of up to five days. Nevertheless, quality controls and urine specimens for clinical investigation were stored at - 20°C before analysis.
cGMP assay protocol-urine analysis
Recovery lof cGMP (2 ,umol/l) added to urine samples from 10 normal subjects was 95·2 ± 2·6 ~~ (mean ± s.e.m.) and 94·1 ± 2·1 % with antisera T5 and T2 respectively. The analysis of a high test urine gave results which were independent of the urine dilution, provided the individual assay values fell within the range 0'1-10 pmol cGMP per tube. Detection limits calculated by the method of Ekins (1974) for assays using either antiserum T2 or T5, were both in the region of 0'08 pmol cGMP per tube. Treatment of urine samples from 11 normal subjects with ox-heart phosphodiesterase reduced the measured cG MP levels to below the detection limit in every case (Fig. 2, top). A comparison of the results of analyses for cG MP in urine obtained using two radioimmunoassays employing antiserum T2 and antiserum T5, and a specific locust protein binding assay (Wood et al., 1976) showed close agreement between all three methods (Table 2). Interassay coefficients of variation for radioimmunoassay using either antiserum T2 or T5
VALIDATION OF THE URINE ASSAY
Details of the assay protocol for the analysis of urine and protein free tissue extracts are shown in Table 1. The addition of 10 ,ul of a solution of bovine ')I-globulin (Sigma Fraction II; 10 rng/ml) as a 'carrier' protein, before the precipitation of bound radioactivity with ammonium sulphate increased the precision of results for duplicate analyses. Half-millilitre aliquots from the dissolved precipitates were transferred to 'minivials' (G.D. Searle and Co.), mixed with 4·0 ml of toluene-'SynperonicNXP' scintillant (Wood et al., 1975), and the tritium content measured on an LKB 'Ultrabeta 1210' liquid scintillation counter with an efficiency of 35 %. Initially a Tris-mercaptoethanol buffer containing theophylline, a less potent inhibitor of cyclic nucleotide phosphodiesterase than EDTA, was used as the assay buffer. Standard curves for incubations with the two buffer systems were identical, and no change in the cG MP levels measured in four urine quality control pools was observed as a
Downloaded from acb.sagepub.com by guest on June 4, 2016
28
P. J. Wood and V. Marks
varied between 4·2 % and 8'7 % for test levels in the range 0·9 to 8'7 pmol cGMP per tube. Table 2 Correlation between cGMP levels measured by three different assay systems. Test urines were diluted 1 in 20 for analysis
x
y
pmolltub«
M C For rellrtss;on equation Y = MX + C
Locust CPB v. Radioimmunoassay T5 Locust CPB v, Radioimmunoassay T2 Radioimmunoassay T2 v. Radioimmunoassay T5
\'085 -0'218
0'969
28
0'999 +0'163
0'964
20
0'996-0,213
0'963
20
No. of Correlation dam coefficient point« N R
A DIRECT RADIOIMMUNOASSAY FOR PLASMA CGMP
blood which, because of age, was no longer suitable for transfusion, and stirred with Norit A charcoal (20 g/loo ml plasma) for 24 hours at 4°C. The mixture was centrifuged and the supernatant plasma vacuum filtered through a Seitz bacteriological filter. The clear plasma was mixed with 0·5 mol/l EDTA (pH 7'5) to give a final EDTA concentration of 5 mmol/l plasma, and small aliquots were stored at - 20°C until required. Blood samples for cGMP analysis were collected into lithium heparin tubes (10 iu lithium heparin/rnl blood) containing 1 % of the blood volume of 0'5 mol/I EDTA (pH 7'5) to prevent cyclic nucleotide degradation by endogenous cyclic nucleotide phosphodiesterase enzyme. Plasma was quickly separated and stored at - 20°C. VALIDATION OF THE PLASMA ASSAY
Ammonium sulphate precipitation of the bound counts created problems when plasma or other types of sample with a high protein content were analysed. Considerable amounts of the reagent were trapped in the bulky protein pellet and the addition of an aliquot of the resuspended pellet to the scintillation medium produced a dense white precipitate. To overcome these difficulties a direct plasma radioimmunoassay, using second antibody separation of bound counts, was developed. The protocol for this method is set out in Table 3. Table 3 Plasma cGMP radioimmunoassay protocol Addlt/on. (1'1)
Te., (1'1)
StandDrd. Non-specific (1'1) blndinK (NSB) (1'1)
Assay buffer (Tris 50 mmol/l pH 7'5 + disodium EDTA 4 mmol/I) Assay buffer + standard (0-10 pmol/tubc) 'Stripped' plasma Test plasma 18-'R] cGMP (Ameraham TRK 392) 1 pmol/50 1'1 Firat antibody (Antiserum T5, working dilution I in 50) Incubate for 3 houra at 4'C, then add: Second antibody (Donkey antirabbit LgG Guildhay HPD5-11F)
100
200 100 200
200
200 50
50
50
100
100
100
100
100
Incubate overnight at 4 "C. Centrifuge, aspirate supernatants, resuspend precipitates in 0·6 ml saline (9 g/I). mix. Remove 0·5 ml aliquot for liquid scintillation counting (as for the urine assay).
To test the specificity of the assay, blood samples were taken from 18 normal subjects and each specimen was divided into two. Half was added to a lithium heparin tube containing EDTA, the plasma was separated and stored frozen; the other half was taken into heparin alone, and the plasma, after separation, was left overnight at room temperature. In each case the endogenous phosphodiesterase present in the samples left overnight at room temperature reduced cGMP levels to below the detection limit of the method, calculated to be 0'05 pmol per tube or 0'25 pmol per ml of plasma (Fig. 2, bottom). Recovery of cGMP (6 pmol/tube) added to 18 plasma samples was 98·1 ± 2'0% (mean ± s.e.m.). The method as described was also satisfactory with regard to parallelism of the results with the standard curve when high test plasmas were analysed at different dilutions in stripped plasma. Five plasma samples were extracted with ethanol, the extracts were dried, and the residues taken up in assay buffer according to the method of Brown et al. (1971). These extracts were then analysed for cGMP by radioimmunoassay, using standards made up in a similar extract of stripped plasma. There was good agreement between results obtained after ethanol extraction and by direct plasma assay (Table 4). Table 4 Results for the analysis offive plasma samples by direct radioimmunoassay and by radioimmunoassay of ethanol extracts ofplasma Plasma sample
cGMP-free plasma, used in standard tubes to compensate for protein effects on the assay, was 'stripping' plasma. by prepared charcoal Plasma was separated from acid-citrate-dextrose
Direct radioimmunoassay Ethanol extraction
Downloaded from acb.sagepub.com by guest on June 4, 2016
cGMP (pmol/ml plasma) 2 I J
4
J
1·0
4'9
9·0
11·3
28·0
1'J
5·0
9'3
10'3
26·0
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay COMP LEVELS IN NORMAL HUMAN URINE AND PLASMA
Concentrations of cAMP and cGMP found in the samespecimens of urine and of plasma from normal subjects are shown in Table 5. Urinary cAMP concentrations were determined by the method of Brown et al. (1971), plasma cAMP content was measured using the method of Tovey et al. (1974) (Amersham kits; TRK 432), and cGMP by the methods described above.
Table 5 Cyclic nucleotide levels in normal subjects, urinary cyclic nucleotides (morning urine specimens) 12 Normal subjects (6 men, 6 women)
,GMP cAMP cGMP
CAMP
Mean
Range
0'41 1-69
0·26-0·66 /olmol/s creatinine 0'83-3·23 /olmol/s creatinine
0'27
0-12-0'47
Pltuma cyclic nucleotides
No. of subjects Mean cGMP 30 cAMP 11 cGMP cAMP II
5'1 11'7 0·57
Range 1'5-10'6 pmol/ml 7'2-18·0 prnol/rnl 0'24-0,77
29
succinyl group, to protein before immunisation. Succinyl-cyclic-GMP was 10 times more potent in displacing labelled cGMP from antibody binding than was cGMP itself. There is, therefore, the potential to develop assays of even greater sensitivity than those described here by the introduction of initial succinylation for test samples. The values for normal urine and plasma cG MP levels obtained with the present method agree well with published data (Murad and Pak, 1972; Steiner et al., 1972a; Lebeau et al., 1975; Neethling and Shanley, 1976; Rosman et al., 1976; Rudman et al., 1976; Siddle et al., 1976). The high specificity of the antisera towards cGMP alone among naturally occurring compounds permits direct analysis of urine and plasma samples. This is in contrast to all previous methods for measuring plasma cGMP, which have required preliminary extraction of test samples. It may also prove possible to analyse tissue homogenates for cGMP without preliminary nucleotide fractionation, but this has still not been tested. The results indicate that there is little need to analyse individual 'blank' plasma samples, collected in the absence of a phosphodiesterase inhibitor, at least for normal subjects. It will be necessary in the future to establish whether there is a need to do so for pathological samples. We thank Brian Morris and Janet Smith for advice with this study, and Boehringer Limited for a gift of 2'-O-succinyl cGMP.
, Discussion The factors which influence the production of specific, high titre antisera are still poorly understood. There is some evidence from this study that 2'-o-succinyl cG MP conjugated to kehole limpet haemocyanin, itself a highly immunogenic protein, may have elicited a better immune response to the hapten cGMP than did the ovalbumin conjugate. Both the antisera, T2 and T5, chosen for investigation for their potential use in radioimmunoassay were found to be remarkably specific for cGMP. Of the four other cyclic nucleotides, cIMP cross-reacted , to the greatest extent. This is probably of little \ practical importance. Steiner et al. (1970) were I unable, using a specific radioimmunoassay, to , detect cIMP in tissue samples. cUMP and cCMP I have, however, recently been isolated from rat liver and other tissues (Bloch, 1975a,b). Synthetic butyrated cyclic nucleotides cross-reacted to a measurable degree in the assay procedure. This was . not altogether surprising, since one of the positions , of substitution of the butyryl group is the 2' position on the ribose ring, which is the point on the , cGMP molecule through which it was linked, via a
References Bloch, A. (1975a). Uridine 3',5'-monophosphate (cyclic U MP). I. Isolation from rat liver extracts. Biochemical and Biophysical Research Communications, 64, 210-218. Bloch, A. (I975b). Isolation of cytidine 3',5'-monophosphate from mammalian tissues and body fluids and its effects on leukemia L-1210 cell growth in culture. Advances in Cyclic Nucleotide Research,S, 331-338. Brown, B. L., Albano, J. D. M., Ekins, R. P., Sgherzi, A. M., and Tampion, W. (1971). A simple and sensitive saturation assay method for the measurement of adenosine 3' :5'-cyclic monophosphate, Biochemical Journal, 12.1, 561-562. Ekins, R. P. (1974). Radioimmunoassay and saturation analysis: Basic principles and theory. British Medical Bulletin, 30, 3-11. Fallon, A. M., and Wyatt, G. R. (1975). An improved assay for cyclic GMP using an insect binding protein. Annals of Biochemistry and Experimental Medicine, 63, 614-619. Goldberg, N. D., Haddox, M. K., Nicol, S. E., Glass, D. B., Sandford, C. H., Kuehl, F. A., Jr, and Estensen, R. (1975). Biologic regulation through opposing influences of cyclic G MP and cyclic AMP: The Yin Yang hypothesis. Advances in Cyclic Nucleotide Research,S, 307-330. Kobayashi, R., and Fang, V. S. (1975). A simple and sensitive competitive protein-binding assay for cyclic GMP.
Biochemical and Biophysical Research Communications. 67, 493-500.
Downloaded from acb.sagepub.com by guest on June 4, 2016
30
P. . I . Wood and V, Marks
Lebeau, M., Dumont, J. E., and Golstein, 5. (1975). Urinary cyclic AMP and cyclic GMP during the menstrual cycle. Hormone and Metabolic Research, I, 190-194. Marks, V., Morris, B. A,, and Teak, J. D. (1974). Radioimmunoassay and saturation analysis: Pharmacology. British Medical Bulletin, 30, 80-86. Murad, F., Manganiello, V., and Vaughan, M. (1971). A simple, sensitive protein-binding assay for guanosine 3’:5’monophosphate. Proceedings of the National Academy of Sciences of the United States of America, 68, 736-739. Murad, F., and Pak, C. Y. (1972). Urinary excretion of adenosine 3’,5’-monophosphate and guanosine 3’3‘ monophosphate. New England Journal of Medicine, 286, 1382-1287. Nature (1973). News and Views: From cyclic AMP to cyclic GMP. Nature, 246, 186-187. Neethling, A. C., and Shanley, B. C. (1976). Letter: Cyclic GMP excretion and hepatoma. Lancet, 2, 578. Rosman, P. M.,Agrawal, R., Goodman, A. D., and Steiner, A. L. (1976). Effect of angiotensin I1 on cyclic guanosine monophosphate and cyclic adenosine monophosphate in human plasma. Journal of Clinical Endocrinology and Metabolism, 42, 531-536. Rudman, D., Fleischer, A., and Kutner, M. H. (1976). Concentration of 3’,5’ cyclic adenosine monophosphate in ventricular cerebrospinal fluid of patients with prolonged coma after head trauma or intracranial hemorrhage. New England Journal of Medicine, 295, 635-638. Siddle, K., Davies, C. J., Shetty, K. J., and Elkeles, R. S. (1976). The effect of insulin on adenosine 3‘:5’-mono-
Book reviews
Blood Glucose Monitoring: Hormone and Metabolic Research Supplement Series No. 7. Guest editors: J. D. Kruse-Jarries and G. D. Molnar. (Pp. 157, 169 figures, 28 tables; D M 49.80). Georg Thieme, Stuttgart, 1977. Most symposia make poor books, partly because the papers prepared for oral delivery are not for permanent written record, and partly because of the inordinate delay between their presentation and their appearance in print. Neither of these criticisms is valid in the present case which represents the well-edited proceedings of a workshop sponsored by the German Society of Clinical Chemistry in March 1976. Great interest has been aroused by the invention of a n ‘artificial pancreas’ which depends upon the ability t o monitor continuously blood glucose concentration in vivo and adjust it by means of insulin or glucose infusions depending on whether it is too high or too low. Six of the contributions in this volume are concerned
phosphate and guanosine 3’5’-monophosphate concentrations in human plasma. Clinical Science and Molecular Medicine, 50, 487-491. Steiner, A. L., Pagliara, A. S., Chase, L. R., and Kipnis, D. M. (1972a). Radioimmunoassay for cyclic nucleotides. 11. Adenosine 3’,5’- monophosphate and guanosine 3’,5’monophosphate in mammalian tissues and body fluids. Journal of Biological Chemistry, 241, 11 14-1120. Steiner, A. L., Parker, C. W., and Kipnis, D. M. (1970). The measurement of cyclic nucleotides by radioimmunoassay. Advances in Biochemical Psychopharmacology, 3, 89-111. Steiner, A. L., Wehmann, R. E., Parker, C. W., and Kipnis, D. M. (1972b). Radioimmunoassay for the measurement of cyclic nucleotides. Advances in Cyclic Nucleotide Research, 2, 51-61. Tovey, K. C., Oldham, K. G., and Whelan, J. A. M. (1974). A -simple direct assay for cyclic AMP in plasma and other biological samples using an improved competitive protein binding technique. CIinica chimica acta, 56, 221-234. Wood, P., English, J., Chakraborty, J.,and Hinton, R. (1975). The use of industrial non-ionic detergents in liquid scintillation counting. Laboratory Practice, 24, 739-740. Wood, P., Hartman, G., and Marks, V. (1976). A simple and specific locust protein binding assay for cyclic 3’3’ GMP. Biochemical and Biophysical Research Communications, 71,1139-1 146. Accepted for publication 26 August 1977.
with the methodology of continuous in vivo glucose measurement and a further nine are concerned with its clinical and research applications and pathobiology. Blood glucose is presumably the first of many substances t o be capable of examination in this way. This volume well describes the progress that is being made in this direction and is certainly worthy of consultation by anyone interested in the future of clinical biochemistry as well as by all diabetologists. VINCENT MARKS
Drug Interference and Drug Measurement in Clinical Chemistry. Edited by G. Siest and D. S. Young. (Pp. 207; $37.75 approximately). Karger : Basel, 1976. This volume, which is the proceedings of the Third International Colloquium on Prospective Biology held in Pont-aMousson, France in October 1975, is concerned, as its title suggests, with two of the most important growth areas in clinical biochemistry-namely, the measurement of drugs in biological fluids and the interference by drugs with conventional clinical biochemistry tests. The two subjects have little in common, however, except their concern with drugs. The measurement of drugs in blood runs the risk of becoming
Downloaded from acb.sagepub.com by guest on June 4, 2016
the subject of demarcation disputes between clinical biochemists, pharmacists, and clinical pharmacologists but it is, nevertheless, likely t o become one of the most important and fruitful branches of clinical biochemistry. Drug interference with conventional laboratory tests, on the other hand, is unquestionably the responsibility of clinical biochemists and theirs alone. If clinical biochemists d o not know how particular drugs interfere with their analyses it is unlikely that anyone else will either. The conference a t which these subjects were discussed may have been as interesting and important as its contents deserved. Unfortunately, the proceedings, like those of so many other successful meetings, fail t o capture and impart the importance of the occasion. What may have made a first class lecture can too often become a third class paper which detracts from, rather than enhances, the value of the volume in which it appears. The topics dealt with are, in my opinion, too important for this generally unsatisfactory method of publication. VINCENT MARKS
Book reviews continued on p a g e 36
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay P. J. WOOD· AND V. MARKS From the Department of Biochemistry, University of Surrey, Guildford Antisera, raised against 2'-O-succinyl-cGMP conjugated to ovalbumin or kehole limpet haemocyanin, have been used to develop radioimmunoassays for the direct d~term~nation of uril":e and plasma cGMP levels. Details of these assays are presented, together with evidence for their validity.
SUMMARY
cGMPt has become the subject of considerable interest during the last five years as an alternative 'second messenger' to cAMPt for the actions of some hormones. There is now evidence that cholinergic and IX-adrenergic stimulation and the actions of certain peptide hormones-such as cholecystokinin, histamine, and possibly insulin-are mediated by anincrease in the intracellular level of cGMP rather than of cAMP (Nature, 1973). In some tissues, changes in the cAMP/cGMP ratio may determine the nature of the response to hormone stimulation (Goldberg et al., 1975), and it should, therefore, be of interest to be able to monitor both cyclic nucleotides in tissues and extracellular fluids. In mammalian tissues cGMP is present at approximately one-tenth to one-hundredth the concentration of cAMP. Human plasma and urine cGMP/cAMP ratios are approximately 1:2 and 1:10 respectively. Consequently, an assay system of high sensitivity as well as of high specificity is required to measure cGMP in the presence of cAMP and other nucleotides, Protein-binding assays for cGMP, which utilise intracellular receptor proteins from lobster tail muscle (Murad et al., 1971), silkmoth pupae (Fallon and Wyatt, 1975), or rat lung (Kobayashi and Fang, 1975) lack sensitivity and are not specific enough to be used for the analysis of tissue extracts without preliminary nucleotide fractionation. The measurement of cG MP by radioimmunoassay (Steiner et al., 1972b) has the advantages of greater sensitivity and specificity over protein-binding
methods and for this reason it was decided to develop such an assay system.
Materials and methods ANTISERUM PRODUCTION
2'-O-Succinyl cGMP (a gift from the Boehringer Corporation) was conjugated to ovalbumin (Sigma Fraction V) or to kehole limpet haemocyanin (CalBiochem) using the carbodiimide 'EDC' (l-ethyl-3 (3-dimethyl-amino-propyl)-earbodiimide Hel) in accordance with the method of Steiner et al. (1972b). The conjugates were estimated by making spectroscopic measurements of solutions of the conjugates, unconjugated 2'-O-succinyl cGMP, and proteins, to contain six molecules of cGMP per molecule of ovalbumin or 160 molecules of cGMP per molecule of kehole limpet haemocyanin (average mol. wt. 3 x 106 Daltons). The conjugates were stored freezedried in aliquots containing 1 mg protein. Five white half-lop rabbits (Ranch Rabbits Ltd) were each given a primary immunisation in six intramuscular sites with a total of 1 mg of the 2'-o-succinyl cG MP-ovalbumin conjugate per rabbit. The conjugate was dissolved in sterile water and emulsified in two volumes of 'Marcol 52' adjuvant (Marks et al., 1974) containing Bacillus Calmette Guerin (BCG; 3 mg/rng conjugate). The rabbits were given seven boosts at monthly intervals using the same regimen, but with the omission of BCG from the adjuvant. Blood samples were taken from an «:a:r vein 10 days after each boost. Because serum antibody titres remained very low on this regimen, the • Present address: Department of Chemical Pathology 2'-O-succinyl cGMP-kehole limpet haemocyanin Southampton General Hospital, Southampton conjugate, suspended in the same adjuv:ant system tcGMP is used as the abbreviation for guanosine-3'S'-cyclic as before, was then used to boost the rabbits (0'25 mg phosphate, cAMP as the abbreviation for adenosine-3'S'protein per rabbit given as six intramuscular cyclic phosphate 25
Downloaded from acb.sagepub.com by guest on June 4, 2016
26
P. J. Wood and V. Marks
injections). After the second monthly boost with this conjugate, antibody titres rose to between 1 in 50 to 1 in 100 (working dilution). Antisera from two of the rabbits, T2 and T5, proved highly specific for cGMP. These were diluted 1 in 5 with buffer (50 mmol/l Tris, pH 7'5, containing theophylline 8 rnmol/l and 2-mercaptoethanol 6 mmolfl) and stored freeze-dried in small aliquots. CHARACTERISATION OF ANTISERA T2 AND
T5 The influence of pH on the binding of 3H cGMP was studied over the pH range 3'5 to 9·0 using 50 mmol/l acetate, phosphate, or Tris buffers containing theophylline (8 rnmol/l) and 2-mercapto ethanol (6 mmol/l), The two antisera differed in their ability to bind the label at different pHs. Antiserum T2 showed a steady level of binding over the pH range 4'5 to 9,0, while antiserum T5 showed increased binding at pH 4·5. Further studies on the properties of both antisera were carried out at pH 7'5, using the assay protocol described below for the analysis of urine (Table 1). The length of incubation had little effect on the standard curve for either antiserum. Incubations of 5, 30, 60, 90, or 180 minutes yielded very similar standard curves, and the 5-minute and 3-hour incubation standard curves were not significantly different when compared by a paired Student's t test on the standard points. Test results for incubations of less than 30 minutes were unreliable, however. An incubation time of three hours was routinely used for the analysis of test samples.
Table 1 cGMP radioimmunoassay protocol: specimens. Incubations in duplicate polystyrene tubes (Luckham Ltd: type LP3) Additions (/"I)
Tests
Standards Non-specific hinding
Assay buffer (Tris 50 mrnoljl, pH "5, containing disodium EDTA,4 mmol/I) Test urine (Diluted 1 in 20 in assay buffer) Standards (0-10 pmol cGMP per 100 /"1 in assay buffer) (8·'HlcGMP (Amersham TRK 392; 1 pmol/50 /"1 in assay buffer) Antiserum (T2 or T5; working dilution 1 in SO in assay buffer) Mix, incubate for 3 hours at 4 "C, then add: Bovine y globulin (Sigma fraction II; 10 mg/ml assay buffer) (NH.),SO. Solution (lee-cold; 3'8 mol/I)
100
100
(NSB)
30
?:' :~
o
20
Q
"0
o ....
a ;§
10
"if!.
o o
0-1 1 10 Nucleotide level (pmol tube. log scale)
100
100 100 50
50
50
100
100
10
10
10
1'0 ml
1·0 ml
1'0 ml
Scatchard plots of the data from five standard curves using antiserum T5 were linear, and were interpreted in terms of a single antibody population with a mean dissociation constant Kn = 3·1 x 10- 9 rnol/l. A similar treatment of the data from two standard curves using antiserum T2 yielded curved Scatchard plots which could be interpretcrf- in terms
E u
300
Mix, leave at 4°C for 10 min, centrifuge for 30 min at 1800 g (Mistral4L Centrifuge). Aspirate supernatants and redissolve precipitates in 1'0 ml distilled water. Mix and take 0'5 ml aliquots for liquid scintillation counting.
40
"0 C
urine
1000
Cross-reactivity of 17 nucleotides with antiserum T5 determined using the incubation procedure described in Table I. A: 2'-O-succinyl-3'5' cGMP B: 3'5' cGMP C: monobutyryI-3'5' cUMP D: dibutyryl-3'5'-eAMP E: dibutyryI-3'5'-eG MP F: 3'5'ciMP G: 3'5' cAMP H: 3'5' cTMP I: 3'5' cCMP J: 3'5' cUMP K: Mean result (± s.e.m.) for 5'-AMP, 5'-i\.TP, 5'GTP, 2'3' cAMP, 2'3' cGMP, AMP-PNP, GMP-PNP. i: Mean result (± s.e.m.j-nucleo... tides which are not plotted 10000 individually.
Fig. 1 Specificity of antiserum T5.
Downloaded from acb.sagepub.com by guest on June 4, 2016
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay full! 5
Blanks (Phosphodiesterase treated)
4 cGMP pmclltube
100 ...1 aliquots from II normal urine samples were incubated with 100 ...1 phosphate buffer (0,5 mol/I, pH 7'5) and 20 ...1 (0'03 U) ox-heart phosphodiesterase (Boehringer) for I hour at 37°C. Reaction tubes were then placed in a boiling water bath for 10 minutes. Incubation mixtures were made to 2·0 ml and their cGMP content was determined using antiserum T2.
•• • • •• •
3 2
• • 0 10
8 cGMP pmcl/ml
6 4
2
eeeuee e ce e e
1!m
(-ECTA)
Blanks (noEDTA)
• • • ••
cGMP levels in plasma from 18 normal subjects, collected with or without disodium EDTA. EDTA-plasma was stored at - 20°C, while 'blank' samples were left at room temperature overnight (for details see the text).
•• •• •• •• I
0
Fig. 2 (top)
27
DDDDDDDDCCCOCCCDDD
'Blank' urine levels;
(bottom) 'Blank' plasma levels.
of two antibody populations with mean dissociation constants Kni = 1·0 x 10- 9 mol/l and KD2 = 1·8 X 10- 8 mol/I. Figure 1 illustrates the crossreactivities of 17 nucleotides in the assay system using antiserum T5. Similar results were obtained with T2 (not shown). In particular cAMP was 10000 times less potent in displacing the [3H} cGMP label from the antiserum than was cGMP.
result of the change to the Tris/EDTA buffer. There was no evidence of rapid degradation of cGMP, and samples showed no loss of cGMP content after being left at either room temperature or at 4°C for periods of up to five days. Nevertheless, quality controls and urine specimens for clinical investigation were stored at - 20°C before analysis.
cGMP assay protocol-urine analysis
Recovery lof cGMP (2 ,umol/l) added to urine samples from 10 normal subjects was 95·2 ± 2·6 ~~ (mean ± s.e.m.) and 94·1 ± 2·1 % with antisera T5 and T2 respectively. The analysis of a high test urine gave results which were independent of the urine dilution, provided the individual assay values fell within the range 0'1-10 pmol cGMP per tube. Detection limits calculated by the method of Ekins (1974) for assays using either antiserum T2 or T5, were both in the region of 0'08 pmol cGMP per tube. Treatment of urine samples from 11 normal subjects with ox-heart phosphodiesterase reduced the measured cG MP levels to below the detection limit in every case (Fig. 2, top). A comparison of the results of analyses for cG MP in urine obtained using two radioimmunoassays employing antiserum T2 and antiserum T5, and a specific locust protein binding assay (Wood et al., 1976) showed close agreement between all three methods (Table 2). Interassay coefficients of variation for radioimmunoassay using either antiserum T2 or T5
VALIDATION OF THE URINE ASSAY
Details of the assay protocol for the analysis of urine and protein free tissue extracts are shown in Table 1. The addition of 10 ,ul of a solution of bovine ')I-globulin (Sigma Fraction II; 10 rng/ml) as a 'carrier' protein, before the precipitation of bound radioactivity with ammonium sulphate increased the precision of results for duplicate analyses. Half-millilitre aliquots from the dissolved precipitates were transferred to 'minivials' (G.D. Searle and Co.), mixed with 4·0 ml of toluene-'SynperonicNXP' scintillant (Wood et al., 1975), and the tritium content measured on an LKB 'Ultrabeta 1210' liquid scintillation counter with an efficiency of 35 %. Initially a Tris-mercaptoethanol buffer containing theophylline, a less potent inhibitor of cyclic nucleotide phosphodiesterase than EDTA, was used as the assay buffer. Standard curves for incubations with the two buffer systems were identical, and no change in the cG MP levels measured in four urine quality control pools was observed as a
Downloaded from acb.sagepub.com by guest on June 4, 2016
28
P. J. Wood and V. Marks
varied between 4·2 % and 8'7 % for test levels in the range 0·9 to 8'7 pmol cGMP per tube. Table 2 Correlation between cGMP levels measured by three different assay systems. Test urines were diluted 1 in 20 for analysis
x
y
pmolltub«
M C For rellrtss;on equation Y = MX + C
Locust CPB v. Radioimmunoassay T5 Locust CPB v, Radioimmunoassay T2 Radioimmunoassay T2 v. Radioimmunoassay T5
\'085 -0'218
0'969
28
0'999 +0'163
0'964
20
0'996-0,213
0'963
20
No. of Correlation dam coefficient point« N R
A DIRECT RADIOIMMUNOASSAY FOR PLASMA CGMP
blood which, because of age, was no longer suitable for transfusion, and stirred with Norit A charcoal (20 g/loo ml plasma) for 24 hours at 4°C. The mixture was centrifuged and the supernatant plasma vacuum filtered through a Seitz bacteriological filter. The clear plasma was mixed with 0·5 mol/l EDTA (pH 7'5) to give a final EDTA concentration of 5 mmol/l plasma, and small aliquots were stored at - 20°C until required. Blood samples for cGMP analysis were collected into lithium heparin tubes (10 iu lithium heparin/rnl blood) containing 1 % of the blood volume of 0'5 mol/I EDTA (pH 7'5) to prevent cyclic nucleotide degradation by endogenous cyclic nucleotide phosphodiesterase enzyme. Plasma was quickly separated and stored at - 20°C. VALIDATION OF THE PLASMA ASSAY
Ammonium sulphate precipitation of the bound counts created problems when plasma or other types of sample with a high protein content were analysed. Considerable amounts of the reagent were trapped in the bulky protein pellet and the addition of an aliquot of the resuspended pellet to the scintillation medium produced a dense white precipitate. To overcome these difficulties a direct plasma radioimmunoassay, using second antibody separation of bound counts, was developed. The protocol for this method is set out in Table 3. Table 3 Plasma cGMP radioimmunoassay protocol Addlt/on. (1'1)
Te., (1'1)
StandDrd. Non-specific (1'1) blndinK (NSB) (1'1)
Assay buffer (Tris 50 mmol/l pH 7'5 + disodium EDTA 4 mmol/I) Assay buffer + standard (0-10 pmol/tubc) 'Stripped' plasma Test plasma 18-'R] cGMP (Ameraham TRK 392) 1 pmol/50 1'1 Firat antibody (Antiserum T5, working dilution I in 50) Incubate for 3 houra at 4'C, then add: Second antibody (Donkey antirabbit LgG Guildhay HPD5-11F)
100
200 100 200
200
200 50
50
50
100
100
100
100
100
Incubate overnight at 4 "C. Centrifuge, aspirate supernatants, resuspend precipitates in 0·6 ml saline (9 g/I). mix. Remove 0·5 ml aliquot for liquid scintillation counting (as for the urine assay).
To test the specificity of the assay, blood samples were taken from 18 normal subjects and each specimen was divided into two. Half was added to a lithium heparin tube containing EDTA, the plasma was separated and stored frozen; the other half was taken into heparin alone, and the plasma, after separation, was left overnight at room temperature. In each case the endogenous phosphodiesterase present in the samples left overnight at room temperature reduced cGMP levels to below the detection limit of the method, calculated to be 0'05 pmol per tube or 0'25 pmol per ml of plasma (Fig. 2, bottom). Recovery of cGMP (6 pmol/tube) added to 18 plasma samples was 98·1 ± 2'0% (mean ± s.e.m.). The method as described was also satisfactory with regard to parallelism of the results with the standard curve when high test plasmas were analysed at different dilutions in stripped plasma. Five plasma samples were extracted with ethanol, the extracts were dried, and the residues taken up in assay buffer according to the method of Brown et al. (1971). These extracts were then analysed for cGMP by radioimmunoassay, using standards made up in a similar extract of stripped plasma. There was good agreement between results obtained after ethanol extraction and by direct plasma assay (Table 4). Table 4 Results for the analysis offive plasma samples by direct radioimmunoassay and by radioimmunoassay of ethanol extracts ofplasma Plasma sample
cGMP-free plasma, used in standard tubes to compensate for protein effects on the assay, was 'stripping' plasma. by prepared charcoal Plasma was separated from acid-citrate-dextrose
Direct radioimmunoassay Ethanol extraction
Downloaded from acb.sagepub.com by guest on June 4, 2016
cGMP (pmol/ml plasma) 2 I J
4
J
1·0
4'9
9·0
11·3
28·0
1'J
5·0
9'3
10'3
26·0
Direct measurement of cGMP in blood plasma and urine by radioimmunoassay COMP LEVELS IN NORMAL HUMAN URINE AND PLASMA
Concentrations of cAMP and cGMP found in the samespecimens of urine and of plasma from normal subjects are shown in Table 5. Urinary cAMP concentrations were determined by the method of Brown et al. (1971), plasma cAMP content was measured using the method of Tovey et al. (1974) (Amersham kits; TRK 432), and cGMP by the methods described above.
Table 5 Cyclic nucleotide levels in normal subjects, urinary cyclic nucleotides (morning urine specimens) 12 Normal subjects (6 men, 6 women)
,GMP cAMP cGMP
CAMP
Mean
Range
0'41 1-69
0·26-0·66 /olmol/s creatinine 0'83-3·23 /olmol/s creatinine
0'27
0-12-0'47
Pltuma cyclic nucleotides
No. of subjects Mean cGMP 30 cAMP 11 cGMP cAMP II
5'1 11'7 0·57
Range 1'5-10'6 pmol/ml 7'2-18·0 prnol/rnl 0'24-0,77
29
succinyl group, to protein before immunisation. Succinyl-cyclic-GMP was 10 times more potent in displacing labelled cGMP from antibody binding than was cGMP itself. There is, therefore, the potential to develop assays of even greater sensitivity than those described here by the introduction of initial succinylation for test samples. The values for normal urine and plasma cG MP levels obtained with the present method agree well with published data (Murad and Pak, 1972; Steiner et al., 1972a; Lebeau et al., 1975; Neethling and Shanley, 1976; Rosman et al., 1976; Rudman et al., 1976; Siddle et al., 1976). The high specificity of the antisera towards cGMP alone among naturally occurring compounds permits direct analysis of urine and plasma samples. This is in contrast to all previous methods for measuring plasma cGMP, which have required preliminary extraction of test samples. It may also prove possible to analyse tissue homogenates for cGMP without preliminary nucleotide fractionation, but this has still not been tested. The results indicate that there is little need to analyse individual 'blank' plasma samples, collected in the absence of a phosphodiesterase inhibitor, at least for normal subjects. It will be necessary in the future to establish whether there is a need to do so for pathological samples. We thank Brian Morris and Janet Smith for advice with this study, and Boehringer Limited for a gift of 2'-O-succinyl cGMP.
, Discussion The factors which influence the production of specific, high titre antisera are still poorly understood. There is some evidence from this study that 2'-o-succinyl cG MP conjugated to kehole limpet haemocyanin, itself a highly immunogenic protein, may have elicited a better immune response to the hapten cGMP than did the ovalbumin conjugate. Both the antisera, T2 and T5, chosen for investigation for their potential use in radioimmunoassay were found to be remarkably specific for cGMP. Of the four other cyclic nucleotides, cIMP cross-reacted , to the greatest extent. This is probably of little \ practical importance. Steiner et al. (1970) were I unable, using a specific radioimmunoassay, to , detect cIMP in tissue samples. cUMP and cCMP I have, however, recently been isolated from rat liver and other tissues (Bloch, 1975a,b). Synthetic butyrated cyclic nucleotides cross-reacted to a measurable degree in the assay procedure. This was . not altogether surprising, since one of the positions , of substitution of the butyryl group is the 2' position on the ribose ring, which is the point on the , cGMP molecule through which it was linked, via a
References Bloch, A. (1975a). Uridine 3',5'-monophosphate (cyclic U MP). I. Isolation from rat liver extracts. Biochemical and Biophysical Research Communications, 64, 210-218. Bloch, A. (I975b). Isolation of cytidine 3',5'-monophosphate from mammalian tissues and body fluids and its effects on leukemia L-1210 cell growth in culture. Advances in Cyclic Nucleotide Research,S, 331-338. Brown, B. L., Albano, J. D. M., Ekins, R. P., Sgherzi, A. M., and Tampion, W. (1971). A simple and sensitive saturation assay method for the measurement of adenosine 3' :5'-cyclic monophosphate, Biochemical Journal, 12.1, 561-562. Ekins, R. P. (1974). Radioimmunoassay and saturation analysis: Basic principles and theory. British Medical Bulletin, 30, 3-11. Fallon, A. M., and Wyatt, G. R. (1975). An improved assay for cyclic GMP using an insect binding protein. Annals of Biochemistry and Experimental Medicine, 63, 614-619. Goldberg, N. D., Haddox, M. K., Nicol, S. E., Glass, D. B., Sandford, C. H., Kuehl, F. A., Jr, and Estensen, R. (1975). Biologic regulation through opposing influences of cyclic G MP and cyclic AMP: The Yin Yang hypothesis. Advances in Cyclic Nucleotide Research,S, 307-330. Kobayashi, R., and Fang, V. S. (1975). A simple and sensitive competitive protein-binding assay for cyclic GMP.
Biochemical and Biophysical Research Communications. 67, 493-500.
Downloaded from acb.sagepub.com by guest on June 4, 2016
30
P. . I . Wood and V, Marks
Lebeau, M., Dumont, J. E., and Golstein, 5. (1975). Urinary cyclic AMP and cyclic GMP during the menstrual cycle. Hormone and Metabolic Research, I, 190-194. Marks, V., Morris, B. A,, and Teak, J. D. (1974). Radioimmunoassay and saturation analysis: Pharmacology. British Medical Bulletin, 30, 80-86. Murad, F., Manganiello, V., and Vaughan, M. (1971). A simple, sensitive protein-binding assay for guanosine 3’:5’monophosphate. Proceedings of the National Academy of Sciences of the United States of America, 68, 736-739. Murad, F., and Pak, C. Y. (1972). Urinary excretion of adenosine 3’,5’-monophosphate and guanosine 3’3‘ monophosphate. New England Journal of Medicine, 286, 1382-1287. Nature (1973). News and Views: From cyclic AMP to cyclic GMP. Nature, 246, 186-187. Neethling, A. C., and Shanley, B. C. (1976). Letter: Cyclic GMP excretion and hepatoma. Lancet, 2, 578. Rosman, P. M.,Agrawal, R., Goodman, A. D., and Steiner, A. L. (1976). Effect of angiotensin I1 on cyclic guanosine monophosphate and cyclic adenosine monophosphate in human plasma. Journal of Clinical Endocrinology and Metabolism, 42, 531-536. Rudman, D., Fleischer, A., and Kutner, M. H. (1976). Concentration of 3’,5’ cyclic adenosine monophosphate in ventricular cerebrospinal fluid of patients with prolonged coma after head trauma or intracranial hemorrhage. New England Journal of Medicine, 295, 635-638. Siddle, K., Davies, C. J., Shetty, K. J., and Elkeles, R. S. (1976). The effect of insulin on adenosine 3‘:5’-mono-
Book reviews
Blood Glucose Monitoring: Hormone and Metabolic Research Supplement Series No. 7. Guest editors: J. D. Kruse-Jarries and G. D. Molnar. (Pp. 157, 169 figures, 28 tables; D M 49.80). Georg Thieme, Stuttgart, 1977. Most symposia make poor books, partly because the papers prepared for oral delivery are not for permanent written record, and partly because of the inordinate delay between their presentation and their appearance in print. Neither of these criticisms is valid in the present case which represents the well-edited proceedings of a workshop sponsored by the German Society of Clinical Chemistry in March 1976. Great interest has been aroused by the invention of a n ‘artificial pancreas’ which depends upon the ability t o monitor continuously blood glucose concentration in vivo and adjust it by means of insulin or glucose infusions depending on whether it is too high or too low. Six of the contributions in this volume are concerned
phosphate and guanosine 3’5’-monophosphate concentrations in human plasma. Clinical Science and Molecular Medicine, 50, 487-491. Steiner, A. L., Pagliara, A. S., Chase, L. R., and Kipnis, D. M. (1972a). Radioimmunoassay for cyclic nucleotides. 11. Adenosine 3’,5’- monophosphate and guanosine 3’,5’monophosphate in mammalian tissues and body fluids. Journal of Biological Chemistry, 241, 11 14-1120. Steiner, A. L., Parker, C. W., and Kipnis, D. M. (1970). The measurement of cyclic nucleotides by radioimmunoassay. Advances in Biochemical Psychopharmacology, 3, 89-111. Steiner, A. L., Wehmann, R. E., Parker, C. W., and Kipnis, D. M. (1972b). Radioimmunoassay for the measurement of cyclic nucleotides. Advances in Cyclic Nucleotide Research, 2, 51-61. Tovey, K. C., Oldham, K. G., and Whelan, J. A. M. (1974). A -simple direct assay for cyclic AMP in plasma and other biological samples using an improved competitive protein binding technique. CIinica chimica acta, 56, 221-234. Wood, P., English, J., Chakraborty, J.,and Hinton, R. (1975). The use of industrial non-ionic detergents in liquid scintillation counting. Laboratory Practice, 24, 739-740. Wood, P., Hartman, G., and Marks, V. (1976). A simple and specific locust protein binding assay for cyclic 3’3’ GMP. Biochemical and Biophysical Research Communications, 71,1139-1 146. Accepted for publication 26 August 1977.
with the methodology of continuous in vivo glucose measurement and a further nine are concerned with its clinical and research applications and pathobiology. Blood glucose is presumably the first of many substances t o be capable of examination in this way. This volume well describes the progress that is being made in this direction and is certainly worthy of consultation by anyone interested in the future of clinical biochemistry as well as by all diabetologists. VINCENT MARKS
Drug Interference and Drug Measurement in Clinical Chemistry. Edited by G. Siest and D. S. Young. (Pp. 207; $37.75 approximately). Karger : Basel, 1976. This volume, which is the proceedings of the Third International Colloquium on Prospective Biology held in Pont-aMousson, France in October 1975, is concerned, as its title suggests, with two of the most important growth areas in clinical biochemistry-namely, the measurement of drugs in biological fluids and the interference by drugs with conventional clinical biochemistry tests. The two subjects have little in common, however, except their concern with drugs. The measurement of drugs in blood runs the risk of becoming
Downloaded from acb.sagepub.com by guest on June 4, 2016
the subject of demarcation disputes between clinical biochemists, pharmacists, and clinical pharmacologists but it is, nevertheless, likely t o become one of the most important and fruitful branches of clinical biochemistry. Drug interference with conventional laboratory tests, on the other hand, is unquestionably the responsibility of clinical biochemists and theirs alone. If clinical biochemists d o not know how particular drugs interfere with their analyses it is unlikely that anyone else will either. The conference a t which these subjects were discussed may have been as interesting and important as its contents deserved. Unfortunately, the proceedings, like those of so many other successful meetings, fail t o capture and impart the importance of the occasion. What may have made a first class lecture can too often become a third class paper which detracts from, rather than enhances, the value of the volume in which it appears. The topics dealt with are, in my opinion, too important for this generally unsatisfactory method of publication. VINCENT MARKS
Book reviews continued on p a g e 36
