Scand. J. din. Lab. Invest. 38, 393-396, 1978.
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
A semi-automated radioimmunoassay technique employing a nine-channel pipetting and sample handling PEKKA LAHTEENMAKI Steroid Research Laboratory, Department of Medical Chemistry, University of Helsinki, SF-00170 Helsinki 17, Finland
Lahteenmaki, P. A semi-automated radioimmunoassay technique employing a nine-channel pipetting and sample handling. Scand. J. d i n . Lab. Invest. 38, 393-396,1978. A semi-automated RIA-technique has been developed which employs simultaneous pipetting of nine aliquots of reagents with nine-channel pipettes and the processing of RIA samples in nine-tube units. This processing includes evaporation, centrifugation, and decantation of tubes at the end of separation step of the RIA. The nine-channel pipetting reduced the time necessary for pipetting to about one third as compared to the corresponding one channel pipetting, and made the pipetting less tiresome. The precision of pipetting was also improved. This nine-channel RIA-technique was applied to radioimmunoassays of plasma 178-oestradiol and progesterone, in which [' 251]iodohistamine-conjugated steroid-tracers were used. Key words: [ 251]iodohistamine; nine-channel technique; oestradiol RIA; progesterone RIA Pekka Lahteenmaki, M . D., Steroid Research Laboratory, Department of Medical Chemistry, University of Helsinki, SF-00170 Helsinki 17, Finland
There is an obvious need for increasing the degree of mechanization and automation of radioimmunoassay. If this were done, analysing capacity could be increased, reliability improved and, hopefully, costs reduced. This paper describes a semi-automated RIA technique, one which has been developed utilizing a ninechannel pipetting and sample handling technique, and which was subsequently applied to and progesRIA of plasma 178-oestradiol (E2) terone (P). In the radioimmunoassays of the steroids the properties of the antisera are of special importance, when the decision between 0036-5513/78/0600-0393$02.00
Fysiologisk Forenings Forlag
0 1978 Medisinsk
a direct RIA after extraction and a RIA after chromatographic purification of the extract is made. For many clinical purposes a direct RIA is good enough, if anantiserumwith highspecificity is available. When iodohistamine-conjugated ligands are used in steroid assays, homologous systems are rarely useful [2]. In the described oestradiol RIA, tracer and antisera were used, in which the couplings had beenmadewith thesame bridge (carboxymethyloxime) but to the adjacent carbon atoms (C-6 for tracer and C-7 for immunogen). This combination showed to be successful.The semi-automated technique described here increased the assay capacity of the laboratory and improved the reliability of the assays. 393
394
Pekka Lahteenmaki Procedure
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
MATERIAL A N D METHODS Petroleum ether (b.p. 40-55°C)or diethyl ether (Merck AG, Darmstadt, Germany) from freshly redistilled bottles were used for the extraction. Phosphate buffer, 0.1 mol/l, pH 7.4, containing 0.9% NaCl and 0.1 % sodium azide (PBS), was used for the preparation of assay buffers which contained 0.1 or 0.5% gelatin (Merck AG, Darmstadt, Germany). The dextran-coated charcoal (DCC) suspension contained 1.25 g Norit 'A' Amend Drug and Chemical Co.,Irvington, N.J., U.S.A.) and 0.125 g Dextran T-70 (AB Pharmacia, Uppsala, Sweden) in 500 ml PBS. Chloramine-T, sodium metabisulphite and histamine were purchased from Merck AG, Darmstadt, Germany. Dioxan (I ,6dioxane, FIuka AG, Buchs, Switzerland) was freshly redistilled. Isobutylchloroformate (Sigma Chemical Co.,St Louis, Mo., U.S.A.) and tri-n-butylamine (British Drug House Ltd, Poole, England) were used as supplied. The non-labelled steroids were obtained from Steraloids, Inc., Wilton, N.H., U.S.A., or Ikapharm, Ramat-Gan, Israel. Na' "I (IMS-30) was obtained from The Radiochemical Centre, Amershani, United Kingdom. The method of Nars & Hunter [3] was used to prepare [ I Z'I]iodohistamine conjugates of 1 1 a-hydroxyprogesterone-1 I -hemisuccinate and 17P-oestradiol-6-(O-carboxymethyl)oxime (Steraloids, Inc., Wilton, N.H., U.S.A.). The antiserum for progesterone was raised in rabbits against 11a-hydroxyprogesterone-I 1 hemisuccinate bovine serum albumin. The antiserum for 17 8-oestradiol was kindly donated by Dr J.P. Raynaud, Roussel-Uclaf, Romainville, France. The hapten for this anti-serum was 17~-oestradiol-7-(U-carboxymethyl)oxime. The equipment for nine-channel pipetting and sample handling was purchased from Labsystems Oy, Helsinki, Finland. The system is based on steplessly adjustable nine-channel pipettes. The sample handling proceeds in ninechannel units, including nitrogen stream evaporations, centrifugations and decantations. Disposable polypropylene tubes 55 x 11 mm, fixable into nine-tube plates (Labsystems Oy, Helsinki, Finland), were used for 1-2 week storage of steroid standard solutions. Polystyrene tubes were used as assay (70 x 1 1 mm) and counting(50 x 13mm) tubes. Thediameter of the counting tubes decreased towards the bottom.
-
Extraction. Plasma 0.25 ml for progesterone and 0.50 ml for oestradiol, was pipetted into disposable glass tubes and2mlof petroleum ether (b.p. 40-55°C)for progesterone and 4 ml of diethyl ether for oestradiol were added (A, see Fig. 1). The tubes were closed with plastic stoppers and shaken for 15 min in a three-dimensional mixer (Desaga, Heidelberg, Germany) in a horizontal position. After mixing the phases were allowed to separate and the tubes then frozen in a dry ice-ethanol bath, after which the organic phase was decanted into another test tube. The extraction was repeated once. The combined ether phases were evaporated under nitrogen using a nine-channel nitrogen stream dispenser (B). Progesterone extracts were redissolved in the original plasma volume and oestradiol extracts in half of the original plasma volume of ethanol (C).Portions of 0.1 ml of these were were dispensed in duplicate into the assay tubes inserted into the nine-channel test tube rack (D), and evaporated to dryness (c). Preparation of the standard curve. The blocks for the storage (at +4"C) of standard solutions ready for use were prepared by fixing seven polypropylene tubes into a nine-channel test tube plate (a, see Fig. 1). These tubes contained standard solutions in concentrations of either I , 2.5, 5 , 10,25, 50 and 100 pg of oestradiol/O.l ml ethanol or 10,25, 50, 100,250,500 and 1000 pg of progesterone/O.1 ml ethanol. Two additional tubes contained ethanol only. These tubes were used to calculate Bo ( = binding of the tracer to the antiserum in the absence of non-labelled steriod and T, ( = the radioactivity obtained after DCC treatment of the assay tubes without antiserum) for the assay standard curve. The standard solution blocks were kept on an ice bath and 0.1 ml of standards were dispensed with a nine-channel pipette into three blocks of assay tubes in a nine-channel test tube rack (b). The standard solutions were evaporated to dryness (c). The standards except T, were processed identically with the unknown samples. Linearization of the standard curve was performed by the logit plot method with a programmable calculator. Radioimmunoassay. Antisera were diluted with 0.1% gelatin-PBS to obtain 40-50% binding of
Nine-channel pipetting and sample handling in RIA EXTRACTION
395
STANDARD CURVE Estradiol Progesterone
U5ml plasma (P)or 0,5ml plasma (E,) -add 2ml of petroleum ether (PIor f m t drethyl ether (EJ -shake tor 15min -freeze. decant -repeat once
121
-evaporate the combined extracts into dryness under nitrogen
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
I
II
-redissolve into 025ml o t ethanol
-dispense in duplicate
P
I
into three blocks of 9-assay tubes
Q
-evaporate t o dryness
-evaporate t o dryness
I
I
B @ B @
-add 0,lmt antiserum -shake -30min raam temperature -add Qlmt tracer, 10000 cpm -shake -incubate over night at + A T
O X W E BATH'
n
- t a k e w i t h 9-channel-\ pipette 0,lml of standards.
n
-odd 0,lml 0.5%
qelabn - PBS and 0,Sml DCC
-Vortex and rethe tubes into centrifuge adapters
-ircubate fur 15min a t +4'C -centrifuge I l m i n a t +LeC
fl
decant
i
n
p
d
0
cant
-count for 5min
FIG. 1. Flow sheet of the procedure. The letters are referred to in the text.
Bo in the assay after which 0.1 ml was dispensed into the assay tubes (E). In the case of the T,tube, 0.1 % gelatin-PBS was substituted for the antiserum. The racks were shaken and left at room temperature for 30 min. A total of 100 PI of [1251]-iodohistamineconjugated steroid tracer in 0.1 % gelatin-PBS were added (F) to give 10,000 cpm per tube. The tubes were vortexed and incubated overnight at
+4"C.After incubation, 0.1 ml of 0.5% gelatinPBS and 0.5 ml of DCC solution were added with nine-channel pipettes into the assay tubes on the ice bath (G).The tubes were vortexed and moved into centrifuge adapters which were kept at +4"C (H). The tubes were allowed to stand at +4"C for 15 min and then centrifuged at 3000 g for 10 min at +4"C. After the centrifugation, counting tubes fixed in a nine-tube block
396
Pekka Lahteeriniaki
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
were inserted on the top of the RIA-tubes in one adapter (I) and the whole system was decanted. The assay tubes containing the DCC pellets were picked up and discarded. Counting tubes were then closed with plastic stoppers and measured for radioactivity in a gammacounter. The results were calculated without recovery correction. RESULTS A N D DISCUSSION The main purpose of this study was to establish a semi-automatedRIA technique with increased assay capacity, improved reliability, and decreased expense. The weekly assay capacity of one technician increased three-fold compared to the traditional one-channel system used in our laboratory. The technique was applied to the radioimmunoassay of two clinically most important reproductive steroids, oestradiol and progesterone. In the assays the used combinations of iodohistamine-labelled ligands and antisera were successful. The specificities of the antisera used in the assays were tested employing both tritiated and iodinated tracers. These antisera showed high specificity against the most important steroids. The specificity improved in radioimmunoassays using iodinated tracers. The smallest amounts of oestradiol or progesterone in the standard curve which significantly differed from zero were 0.1 and 0.5 pg, respectively. However, the variations in the samples within the most sensitive part of the standard curves was high. Therefore, 5 pglml (1 pgltube) and 25 pglml (2.5 pgltube) of oestradiol and progesterone, respectively, were selected to be the practical detection limits of the assays. Blank values from samples obtained by the extraction of 0.1-1.0 ml of water never exceeded those detection limits. The accuracy of the assays was tested by recovery studies. When tritiated oestradiol or progesterone was added into plasma, the recoveries after extraction with diethyl ether or petroleum ether, respectively, were 93.8+1.2% (meankl SD, n = 10) for oestradiol and 80.4f3.2% (meant1 SD, n = 10) for progesterone. The recovery of non-labelled steroids was tested by assays of male plasma pools, which had known amounts of added non-labelled steroids. When 20 pglml and 200 pg/ml of
+
oestradiol were added, 92.2 6.6% (mean f SD, n = 10) and 88.3 +7.0% (mean k 1 SD, n = lo), respectively, were recovered. The recovery of non-labelled progesterone was determined after petroleum ether extraction. When 5 ng/ml of progesterone was added into both the male plasma pool and water, the recovery from plasma was 85.0 k 5.8 ”/, (mean 1 SD, n = lo), and from water 84.7 f 5.9% (mean k SD, n = 10). The intra-assay coefficient of variation was determined according to Abraham et al. [I] from duplicate values in one assay. Within different parts of the standard curve, the intra-assay coefficients of variation for oestradiol were 9.7 % (5-50 pg/ml, n = 16), 8.8% (50-250 pg/ml, n = 25) and 8.1% (250-600 pglml, n = 13). The intra-assay coefficients for progesterone were 6.6% (0.05-0.5 nglml, n = 23), 6.4% (0.5-2.0 ng/ml, n = 51) and 7.8% (2.0-20.0 ng/ml, n = 27). The inter-assay coefficients of variation were determined from control plasma pools run in each assay (n = 6). For oestradiol they were 16.6% (low level, mean 64 pg/ml), 11.1% (medium level, mean 165 pglml) and 14.4% (high level, mean 415 pgiml). The inter-assay coefficients of variation for progesterone were 14.1% (mean 0.22 ng/ml) and 12.9% (mean 2.77 ng/ml). The technique described above is easily adapted into different separation techniques, and into radioimmunoassays using tritiated tracers as well. The counting can be performed from the aliquotes of the separated supernatant. If commercially available low volume high capacity scintillation liquids are used, the beta counting can be performed directly in the tubes after decantation.
REFERENCES 1 Abraham, G.E., Swerdloff, R., Tulchinsky D. &
Odell W.D. Radioimmunoassay of plasma progesterone. J . clin. Endocr. 32,619, 1971. 2 Cameron, E.H.D., Scarisbrick, J.J., Morris, S.E., Hillier S.G. & Read, G. Some aspects of the use of zsI-labelled ligands for steroid radioimmunoassay J. Steroid Biochem. 5, 749, 1974. 3 Nan, P.W.& Hunter W.M. A method for labelling oestradiol-l7b with radioiodine for radioimmunoassays J. Endocrinol. 57, 47, 1973. Received 30 November 1977 Accepted 29 December 1977
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
A semi-automated radioimmunoassay technique employing a nine-channel pipetting and sample handling PEKKA LAHTEENMAKI Steroid Research Laboratory, Department of Medical Chemistry, University of Helsinki, SF-00170 Helsinki 17, Finland
Lahteenmaki, P. A semi-automated radioimmunoassay technique employing a nine-channel pipetting and sample handling. Scand. J. d i n . Lab. Invest. 38, 393-396,1978. A semi-automated RIA-technique has been developed which employs simultaneous pipetting of nine aliquots of reagents with nine-channel pipettes and the processing of RIA samples in nine-tube units. This processing includes evaporation, centrifugation, and decantation of tubes at the end of separation step of the RIA. The nine-channel pipetting reduced the time necessary for pipetting to about one third as compared to the corresponding one channel pipetting, and made the pipetting less tiresome. The precision of pipetting was also improved. This nine-channel RIA-technique was applied to radioimmunoassays of plasma 178-oestradiol and progesterone, in which [' 251]iodohistamine-conjugated steroid-tracers were used. Key words: [ 251]iodohistamine; nine-channel technique; oestradiol RIA; progesterone RIA Pekka Lahteenmaki, M . D., Steroid Research Laboratory, Department of Medical Chemistry, University of Helsinki, SF-00170 Helsinki 17, Finland
There is an obvious need for increasing the degree of mechanization and automation of radioimmunoassay. If this were done, analysing capacity could be increased, reliability improved and, hopefully, costs reduced. This paper describes a semi-automated RIA technique, one which has been developed utilizing a ninechannel pipetting and sample handling technique, and which was subsequently applied to and progesRIA of plasma 178-oestradiol (E2) terone (P). In the radioimmunoassays of the steroids the properties of the antisera are of special importance, when the decision between 0036-5513/78/0600-0393$02.00
Fysiologisk Forenings Forlag
0 1978 Medisinsk
a direct RIA after extraction and a RIA after chromatographic purification of the extract is made. For many clinical purposes a direct RIA is good enough, if anantiserumwith highspecificity is available. When iodohistamine-conjugated ligands are used in steroid assays, homologous systems are rarely useful [2]. In the described oestradiol RIA, tracer and antisera were used, in which the couplings had beenmadewith thesame bridge (carboxymethyloxime) but to the adjacent carbon atoms (C-6 for tracer and C-7 for immunogen). This combination showed to be successful.The semi-automated technique described here increased the assay capacity of the laboratory and improved the reliability of the assays. 393
394
Pekka Lahteenmaki Procedure
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
MATERIAL A N D METHODS Petroleum ether (b.p. 40-55°C)or diethyl ether (Merck AG, Darmstadt, Germany) from freshly redistilled bottles were used for the extraction. Phosphate buffer, 0.1 mol/l, pH 7.4, containing 0.9% NaCl and 0.1 % sodium azide (PBS), was used for the preparation of assay buffers which contained 0.1 or 0.5% gelatin (Merck AG, Darmstadt, Germany). The dextran-coated charcoal (DCC) suspension contained 1.25 g Norit 'A' Amend Drug and Chemical Co.,Irvington, N.J., U.S.A.) and 0.125 g Dextran T-70 (AB Pharmacia, Uppsala, Sweden) in 500 ml PBS. Chloramine-T, sodium metabisulphite and histamine were purchased from Merck AG, Darmstadt, Germany. Dioxan (I ,6dioxane, FIuka AG, Buchs, Switzerland) was freshly redistilled. Isobutylchloroformate (Sigma Chemical Co.,St Louis, Mo., U.S.A.) and tri-n-butylamine (British Drug House Ltd, Poole, England) were used as supplied. The non-labelled steroids were obtained from Steraloids, Inc., Wilton, N.H., U.S.A., or Ikapharm, Ramat-Gan, Israel. Na' "I (IMS-30) was obtained from The Radiochemical Centre, Amershani, United Kingdom. The method of Nars & Hunter [3] was used to prepare [ I Z'I]iodohistamine conjugates of 1 1 a-hydroxyprogesterone-1 I -hemisuccinate and 17P-oestradiol-6-(O-carboxymethyl)oxime (Steraloids, Inc., Wilton, N.H., U.S.A.). The antiserum for progesterone was raised in rabbits against 11a-hydroxyprogesterone-I 1 hemisuccinate bovine serum albumin. The antiserum for 17 8-oestradiol was kindly donated by Dr J.P. Raynaud, Roussel-Uclaf, Romainville, France. The hapten for this anti-serum was 17~-oestradiol-7-(U-carboxymethyl)oxime. The equipment for nine-channel pipetting and sample handling was purchased from Labsystems Oy, Helsinki, Finland. The system is based on steplessly adjustable nine-channel pipettes. The sample handling proceeds in ninechannel units, including nitrogen stream evaporations, centrifugations and decantations. Disposable polypropylene tubes 55 x 11 mm, fixable into nine-tube plates (Labsystems Oy, Helsinki, Finland), were used for 1-2 week storage of steroid standard solutions. Polystyrene tubes were used as assay (70 x 1 1 mm) and counting(50 x 13mm) tubes. Thediameter of the counting tubes decreased towards the bottom.
-
Extraction. Plasma 0.25 ml for progesterone and 0.50 ml for oestradiol, was pipetted into disposable glass tubes and2mlof petroleum ether (b.p. 40-55°C)for progesterone and 4 ml of diethyl ether for oestradiol were added (A, see Fig. 1). The tubes were closed with plastic stoppers and shaken for 15 min in a three-dimensional mixer (Desaga, Heidelberg, Germany) in a horizontal position. After mixing the phases were allowed to separate and the tubes then frozen in a dry ice-ethanol bath, after which the organic phase was decanted into another test tube. The extraction was repeated once. The combined ether phases were evaporated under nitrogen using a nine-channel nitrogen stream dispenser (B). Progesterone extracts were redissolved in the original plasma volume and oestradiol extracts in half of the original plasma volume of ethanol (C).Portions of 0.1 ml of these were were dispensed in duplicate into the assay tubes inserted into the nine-channel test tube rack (D), and evaporated to dryness (c). Preparation of the standard curve. The blocks for the storage (at +4"C) of standard solutions ready for use were prepared by fixing seven polypropylene tubes into a nine-channel test tube plate (a, see Fig. 1). These tubes contained standard solutions in concentrations of either I , 2.5, 5 , 10,25, 50 and 100 pg of oestradiol/O.l ml ethanol or 10,25, 50, 100,250,500 and 1000 pg of progesterone/O.1 ml ethanol. Two additional tubes contained ethanol only. These tubes were used to calculate Bo ( = binding of the tracer to the antiserum in the absence of non-labelled steriod and T, ( = the radioactivity obtained after DCC treatment of the assay tubes without antiserum) for the assay standard curve. The standard solution blocks were kept on an ice bath and 0.1 ml of standards were dispensed with a nine-channel pipette into three blocks of assay tubes in a nine-channel test tube rack (b). The standard solutions were evaporated to dryness (c). The standards except T, were processed identically with the unknown samples. Linearization of the standard curve was performed by the logit plot method with a programmable calculator. Radioimmunoassay. Antisera were diluted with 0.1% gelatin-PBS to obtain 40-50% binding of
Nine-channel pipetting and sample handling in RIA EXTRACTION
395
STANDARD CURVE Estradiol Progesterone
U5ml plasma (P)or 0,5ml plasma (E,) -add 2ml of petroleum ether (PIor f m t drethyl ether (EJ -shake tor 15min -freeze. decant -repeat once
121
-evaporate the combined extracts into dryness under nitrogen
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
I
II
-redissolve into 025ml o t ethanol
-dispense in duplicate
P
I
into three blocks of 9-assay tubes
Q
-evaporate t o dryness
-evaporate t o dryness
I
I
B @ B @
-add 0,lmt antiserum -shake -30min raam temperature -add Qlmt tracer, 10000 cpm -shake -incubate over night at + A T
O X W E BATH'
n
- t a k e w i t h 9-channel-\ pipette 0,lml of standards.
n
-odd 0,lml 0.5%
qelabn - PBS and 0,Sml DCC
-Vortex and rethe tubes into centrifuge adapters
-ircubate fur 15min a t +4'C -centrifuge I l m i n a t +LeC
fl
decant
i
n
p
d
0
cant
-count for 5min
FIG. 1. Flow sheet of the procedure. The letters are referred to in the text.
Bo in the assay after which 0.1 ml was dispensed into the assay tubes (E). In the case of the T,tube, 0.1 % gelatin-PBS was substituted for the antiserum. The racks were shaken and left at room temperature for 30 min. A total of 100 PI of [1251]-iodohistamineconjugated steroid tracer in 0.1 % gelatin-PBS were added (F) to give 10,000 cpm per tube. The tubes were vortexed and incubated overnight at
+4"C.After incubation, 0.1 ml of 0.5% gelatinPBS and 0.5 ml of DCC solution were added with nine-channel pipettes into the assay tubes on the ice bath (G).The tubes were vortexed and moved into centrifuge adapters which were kept at +4"C (H). The tubes were allowed to stand at +4"C for 15 min and then centrifuged at 3000 g for 10 min at +4"C. After the centrifugation, counting tubes fixed in a nine-tube block
396
Pekka Lahteeriniaki
Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/27/14 For personal use only.
were inserted on the top of the RIA-tubes in one adapter (I) and the whole system was decanted. The assay tubes containing the DCC pellets were picked up and discarded. Counting tubes were then closed with plastic stoppers and measured for radioactivity in a gammacounter. The results were calculated without recovery correction. RESULTS A N D DISCUSSION The main purpose of this study was to establish a semi-automatedRIA technique with increased assay capacity, improved reliability, and decreased expense. The weekly assay capacity of one technician increased three-fold compared to the traditional one-channel system used in our laboratory. The technique was applied to the radioimmunoassay of two clinically most important reproductive steroids, oestradiol and progesterone. In the assays the used combinations of iodohistamine-labelled ligands and antisera were successful. The specificities of the antisera used in the assays were tested employing both tritiated and iodinated tracers. These antisera showed high specificity against the most important steroids. The specificity improved in radioimmunoassays using iodinated tracers. The smallest amounts of oestradiol or progesterone in the standard curve which significantly differed from zero were 0.1 and 0.5 pg, respectively. However, the variations in the samples within the most sensitive part of the standard curves was high. Therefore, 5 pglml (1 pgltube) and 25 pglml (2.5 pgltube) of oestradiol and progesterone, respectively, were selected to be the practical detection limits of the assays. Blank values from samples obtained by the extraction of 0.1-1.0 ml of water never exceeded those detection limits. The accuracy of the assays was tested by recovery studies. When tritiated oestradiol or progesterone was added into plasma, the recoveries after extraction with diethyl ether or petroleum ether, respectively, were 93.8+1.2% (meankl SD, n = 10) for oestradiol and 80.4f3.2% (meant1 SD, n = 10) for progesterone. The recovery of non-labelled steroids was tested by assays of male plasma pools, which had known amounts of added non-labelled steroids. When 20 pglml and 200 pg/ml of
+
oestradiol were added, 92.2 6.6% (mean f SD, n = 10) and 88.3 +7.0% (mean k 1 SD, n = lo), respectively, were recovered. The recovery of non-labelled progesterone was determined after petroleum ether extraction. When 5 ng/ml of progesterone was added into both the male plasma pool and water, the recovery from plasma was 85.0 k 5.8 ”/, (mean 1 SD, n = lo), and from water 84.7 f 5.9% (mean k SD, n = 10). The intra-assay coefficient of variation was determined according to Abraham et al. [I] from duplicate values in one assay. Within different parts of the standard curve, the intra-assay coefficients of variation for oestradiol were 9.7 % (5-50 pg/ml, n = 16), 8.8% (50-250 pg/ml, n = 25) and 8.1% (250-600 pglml, n = 13). The intra-assay coefficients for progesterone were 6.6% (0.05-0.5 nglml, n = 23), 6.4% (0.5-2.0 ng/ml, n = 51) and 7.8% (2.0-20.0 ng/ml, n = 27). The inter-assay coefficients of variation were determined from control plasma pools run in each assay (n = 6). For oestradiol they were 16.6% (low level, mean 64 pg/ml), 11.1% (medium level, mean 165 pglml) and 14.4% (high level, mean 415 pgiml). The inter-assay coefficients of variation for progesterone were 14.1% (mean 0.22 ng/ml) and 12.9% (mean 2.77 ng/ml). The technique described above is easily adapted into different separation techniques, and into radioimmunoassays using tritiated tracers as well. The counting can be performed from the aliquotes of the separated supernatant. If commercially available low volume high capacity scintillation liquids are used, the beta counting can be performed directly in the tubes after decantation.
REFERENCES 1 Abraham, G.E., Swerdloff, R., Tulchinsky D. &
Odell W.D. Radioimmunoassay of plasma progesterone. J . clin. Endocr. 32,619, 1971. 2 Cameron, E.H.D., Scarisbrick, J.J., Morris, S.E., Hillier S.G. & Read, G. Some aspects of the use of zsI-labelled ligands for steroid radioimmunoassay J. Steroid Biochem. 5, 749, 1974. 3 Nan, P.W.& Hunter W.M. A method for labelling oestradiol-l7b with radioiodine for radioimmunoassays J. Endocrinol. 57, 47, 1973. Received 30 November 1977 Accepted 29 December 1977
