Relationship of Bioassayable and Immunoas say able Plasma ACTH and Cortisol Concentrations in Normal Subjects and in Patients with Cushing's Disease D. T. KRIEGER AND W. ALLEN Division of Endocrinology and Metabolism, Department of Medicine, Mount Sinai School of Medicine and Department of Mathematics, Polytechnic Institute, New York, and ODAR Educational Testing Service ABSTRACT: Plasma ACTH and cortisol concentrations were determined at 5-min intervals over a 3or 4-h sampling period in 2 normal subjects. Time spans studied were 10:00 AM-1:00 PM, 4:00 PM8:00 PM, 8:00 PM-ll:00 PM, and 4:00 AM-8:00 AM. Similar sampling for 3 h, (onset 9:00-9:30 AM) was performed on 4 patients with Cushing's disease, 3 untreated and 1 in remission following pituitary irradiation. Two of these patients were studied on 2 separate occasions. Plasma ACTH was determined by both immunoassay (I) and bioassay (B). Although in general, these studies demonstrated significant correlation between I-ACTH or B-ACTH concentrations and those of plasma cortisol, a striking finding in both subject categories was the presence of 30- to 50-min episodes during which marked rises in both I- and B-ACTH concentrations occurred without concomittant, or markedly diminished, increments in plasma cortisol concentrations. This could not be explained by biological inactivity of the ACTH, since a highly significant correlation was present between I- and B-ACTH concentrations at all times; r values ranged between 0.86 and 0.98 for normal subjects, and 0.76 and 0.96 for patients with Cushing's disease. The lack of correlation in these episodes also does not appear to be secondary to an ll-/3-hydroxylase block, differences in the rate of change of plasma ACTH concentrations, lack of adrenal "priming" by prior ACTH or incapacity of the adrenal gland to further
increase secretion. I/B ACTH ratios were similar in the normal subjects (1.42-1.64) and in the patients with Cushing's disease (1.27-1.47). "Apparent" ACTH half lives calculated from "peaks" of ACTH secretion were 7-12 min for I-ACTH and 3-9 min for B-ACTH in the normal subjects; and 9-13 min and 7-9 min respectively, in the patients with Cushing's disease. Mean plasma ACTH I- and B-concentrations at comparable time periods were higher in patients with active Cushing's disease than in normal subjects. These studies also indicate that in Cushing's disease, the abnormality present resides in ACTH regulatory mechanisms, not in the nature of the ACTH secreted. Approximation of the total amount of immunoassayable ACTH secreted in one normal subject over a 24-h period yielded a value of 73 /xg. Total /xg/h secreted in the 2 normal subjects were highest in the hour preceding awakening (6:30-7:30 AM; 12.9 and 12.2 /xg/h); were 5.3 and 4.0 /xg/h between 10:00-11:00 AM, and 1.4 and 1.7 /xg/h between 9:00-10:00 PM. In the 3 patients with clinically active Cushing's disease, apparent ACTH secretion between 10:00-11:00 AM varied from 19.234.3 /xg/h, the magnitude of such secretion being positively correlated with the extent of increased adrenal cortical activity present. (J Clin Endocrinol Metah 10: 675, 1975)
AVAILABILITY of a sensitive radioimmunoassay (1) for the determination of plasma ACTH levels has enabled a more precise investigation of the interrelationships of the central nervous systempituitary-adrenal axis. Such investigations had been previously hampered by the availability only of a relatively insensitive
ACTH bioassay (2). Immunoassay studies have revealed a rough parallelism between spontaneous peaks of plasma ACTH and cortisol levels (3,4) and the responses of these paramenters to exogenous stimuli (5-10). The major biological activity of ACTH resides in its first 20 amino acids (11). It is recognized that various antisera recognize different portions of the ACTH molecule (12,13). There have been few studies in which simultaneous determinations of immunoassayable (I) and bioassayable (B)
JLJL
Received August 5, 1974. Supported by U.S.P.H.S. Service Grant FR-71 from the Division of Research Facilities and Grants, USPHS Grant NB-02893 and the Lita Annenberg Hazen Charitable Trust.
675
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
JCE & M • 1975 Vol 40 • No 4
KRIEGER AND ALLEN
676
plasma ACTH levels have been performed. Due to the insensitivity of the bioassay, such studies have been possible only in situations where high ACTH levels were present, such as: following infusion of ACTH to normal subjects (14); or conditions characterized by ACTH excess (ectopic tumor or pituitary tumor production of ACTH) (12) and Addison's disease (15). In these studies I-ACTH levels were usually higher than those of B-ACTH by a factor of 1.6 to >10. Recent studies have also indicated the presence in the circulation in some of these altered clinical states, of biologically inactive ACTH: either "Big" ACTH (16) or C-terminal, N-terminal or middle fragments (12). In view of such findings, the desirability of correlating Band I-ACTH concentrations with each other and those of plasma cortisol under physiological conditions in normal subjects and in diseases characterized by increased ACTH production, is apparent. Materials and Methods Blood samples were obtained every 5 min via indwelling intravenous catheters over separate 3- to 4-h spans during different portions of the 24-h day in 2 normal male subjects (age 25 and 33 y). Sampling spans were separated by inter-
vals of at least one month. Four patients with Cushing's disease were similarly studied (2 of these on two separate occasions) over a 3-h period beginning between 9:00-9:30 AM. Three of these patients were studied prior to any treatment (RD, WH, PR) (Table 1). One of these (WH) was restudied 4 months postproton beam therapy, while still manifesting active disease (although with evidence of improvement as compared with pretreatment parameters). The fourth patient (MB) was studied on two occasions when in clinical and laboratory remission 16 and 19 months following pituitary irradiation (cobalt). Subjects were supine throughout all sampling spans. All studies (save those between 4:00 AM-8:00 AM) were begun V2 h after insertion of a 21-gauge scalp vein needle which was connected via an adapter to a 5 foot length of PE 160 tubing (dead space 2.6 ml). Catheters were kept patent with a microdrip of isotonic saline. In the 4:00 AM-8:00 AM study, catheter insertion was at midnight; the subjects were.alseep at 12:30 AM. The experimenter was in the same room as the subjects in all studies save that from 4:00-8:00 AM. In that study blood samples were collected in an adjoining room (into which the catheters extended.) "Dead space" blood volume was returned to subjects via a 3-way stopcock. Maximum volume of blood removed for assay during a sampling period was 250 cc. Blood samples were immediately transferred to heparinized glass tubes which had been kept
TABLE 1. Parameters of adrenal cortical function in patients with Cushing's disease at time of 5-min sampling studies
17OHCS mg/24 h
17KS mg/24 h
Cortisol secretory rate mg/24 h
Active c. 4 yr
60.0
79.5
—
Active disease c. 15 mos.
39.1
22.1
90.7
Active disease 4 mos. p Proton beam therapy
17.5
30.6
Remission 16 mos. p completion pituitary (cobalt) irradiation
9.8
8.9
19 mos. p completion of Rx
6.4
9.6
32.2
15.1
Urinary Pt.
Age/ Sex
R.D.
22/M
W.H.
42/M
M.B.
P.R.
48/M
35/F
Clinical condition
Active disease c. 18 mos.
22.8
—
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
ACTH (BIO-, IMMUNO-)CORTISOL RELATIONSHIP on crushed ice. They were immediately centrifuged in a refrigerated centrifuge and the plasma samples frozen immediately at — 20 C in separate tubes for I-ACTH, B-ACTH and cortisol determinations, so that each specimen was thawed only once prior to assay. Plasma I-ACTH concentrations were determined 1 on extracted plasma using an ACTH antiserum (Kendall) kindly provided by the N.I.H. B-ACTH concentrations were determined by a modification of the method of Sayers (17,18) which permits determination of such levels in plasma.2 Plasma cortisol, and in some instances, desoxycortisol, were determined by a competitive protein binding method (19). (The intra1
Plasma samples were extracted following acidification, by adsorption to silicic acid, a subsequent acid wash, and desorption of ACTH by aqueous acetone. The resultant extract was evaporated to dryness and reconstituted in a 0.5% human serum albumin phosphate buffer. Efficacy of extraction (monitored by 125 I-ACTH) varied from 72-75%TTo validate the use of 125I-ACTH as a marker for/recovery, in this and in the bioassay procedure, comparisons have been made of recovery of biologically active ACTH with that of 125 IhACTH to ascertain whether the hormone may be variably denatured during the extraction process. No significant loss of biological activity was seen. Antiserum was used at a dilution of 1:100,000 and 1:150,000. Minimum sensitivity was 10 pg/ml. Coefficient of variation was 6.13% (in intra-assay), 16.90% (interassay). For the dose range 2 to 5 pg, k = 0.079, for 5-50 pg, X = 0.091, and for 50-100 pg,
677
and interassay coefficients of variation for the cortisol assay are 3.1 and 3.5% respectively.) The time periods chosen for the clinically healthy subjects were 10:00 AM to 1:00 PM (subjects in a fasting state), 4:00 PM-8:00 PM, 8:00 PM to 11:00 PM (dinner hour at 6:00 PM) and 4:00 A M - 8 : 0 0 AM (observed sleep onset at 12:30 AM). During the first span plasma ACTH and cortisol concentrations are declining after their major circadian rise. During the second and third spans the least variation in plasma cortisol and immunoreactive ACTH concentrations has been noted in our previous studies (3). The last span encompasses the interval during which a major portion of pituitary and adrenal secretory activity occurs (3,4). Correlation coefficients were computed for log ACTH and plasma cortisol levels. The amount of ACTH secreted within given time periods was approximated as described by Gallagher et al.(3). Results 1. Normal subjects (Fig. 1, Table 2)
(a) Correlation of I and B-ACTH. There is excellent correlation between B and I-ACTH concentrations throughout all sampling spans for both subjects. The ratios of I-to-B-ACTH concentrations during different spans varied from 1.42-1.64. Both subjects displayed a major circadian X = 0.079. Characterization of the antibody was achieved by constructing displacement curves using rise in plasma ACTH concentrations at a-MSH, ah'-39ACTH, ap^ACTH, a^^ACTH, approximately 7:00 AM. Between 6:40-7:30 a'-10ACTH, a'-24ACTH, and a17"39ACTH. Only AM the ratios for the 2 subjects of I-to a'~24ACTH and a-MSH reacted fully with the anti- B-ACTH concentrations were 1.36 and body, with competitive binding curves almost indis- 1.37, which differed significantly (P < 0.05) tinguishable from that of authentic human ACTH. The other analogues demonstrated insignificant im- from the ratios obtained over the entire munoreactivity in the dose range employed in the 4:00 AM-8:00 AM span. Mean plasma assay. All samples from a given sampling period were ACTH and cortisol concentrations were run in duplicate in the same assay. significantly higher during this major cir2 Manuscript submitted for publication. Plasma cadian rise, when compared to means for samples were extracted in a manner similar to that each individual sampling span (P varied described for the immunoassay. The minimum sensitivity is 2.5 pg/ml. The coefficients of variation were: from < 0.001 to < 0.005). Intraassay variability 9.82%. Interassay variability 12.75%. a'~24ACTH in equimolar quantities to (b) Correlation between plasma ACTH and ap'-39ACTH, V-39ACTH was ca. 20%, more potent. cortisol concentration. The logarithm of Lysine vasopressin (0.01-1.0 IU), Pitressin® (0.2-2.0 plasma ACTH concentrations (bioassay) has IU) exhibited no biological activity. B-MSH 1 x 108 pg had potency equivalent to 2.5 pg ACTH. a-MSH been reported to have a linear relationship to plasma cortisol concentrations (20,21). 1 x 107 pg had potency equivalent to 10 pg ACTH.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
JCE & M • 1975 Vol 40 • No 4
KRIEGER AND ALLEN
678 200 180 160 140 120
S L • - • - • RlA ACTH Biocssoy ACTH ——• COHISOI
sioo 80 60 40 < 20 •
0
W B
200 180
— - • RIA A C T H
160 • - • - • Bioassay ACTH 140 ——
I 120
Cortisol
sioo i o < « Q-
_. 20 80 12 60 40 < ! 8 20 IQ- 4 n a n
/ \^VN
§l
5
6
7
8c
FIG. 1. Plasma ACTH (radioimmunoassay, bioassay) and cortisol concentrations, in 2 normal male subjects, sampled at 5-min intervals over 3-4 h spans. Sampling periods separated by intervals of at least 1 month. Ten AM-1:00 PM study performed with subjects in fasting state. Sleep onset was at 12:30 AM. Note parallelism between immunoassayable and bioassayable ACTH concentrations, as well as episodes in the 10:00 AM-1:00 PM and 4:00 A M - 8 : 0 0 AM intervals where major rises in ACTH concentrations are accompanied by none, or slight changes in cortisol concentrations.
Correlation coefficients between log I-ACTH and cortisol concentrations and log B-ACTH and cortisol concentrations are indicated in Table 2. It is apparent that both of these correlations are significant when considered for each entire sampling span (save for log B-ACTH/cortisol in WB from 10:00 AM to 1:00 PM) 3 and that such significant correlations persist in both of these subjects when each was restudied between 10:00 AM-1:00 PM. (Correlation coefficients (log I-ACTH and cortisol) in 4 other normal subjects studied half-hourly over a 24-h period (4) were 0.5272, 0.4892, 0.7678, and 0.7145, all significant at P < 0.001). Similar correlation coefficients were obtained when calculated for either log-I-ACTH or log-B-ACTH and the plasma cortisol concentration obtained 5 min subsequent to the ACTH determination. These 3
The almost even distribution of the number of B-ACTH determinations in the "peak" and "nonpeak" intervals (see text below) may explain the nonsignificance of this correlation, in contrast to the correlation seen between the I-ACTH and cortisol concentrations.
coefficients are not changed if one uses the plasma cortisol concentration obtained that would have been residual at each sampling time (utilizing a cortisol half life of 70 min) (20). However, in each subject, during similar spans (SL: 11:50 AM-noon, WB: 11:30 AM-12:05 PM, SL: 6:40 AM-7:30 AM, WB: 6:50 AM-7:30 AM) marked rises in both I-and B-ACTH concentrations occurred without concomitant, or with markedly diminished, increments in plasma cortisol concentrations. If one considers correlation coefficients for these cited episodes (Table 2), in subject WB there was no significant correlation between plasma ACTH concentrations (either I- or B-) and those of plasma cortisol. The rise in plasma cortisol concentrations between 6:50 and 7:05 AM was 64%, in contrast to rises of 646% and 900% respectively, in I-and B-ACTH concentrations (Table 3). In contrast, the immediately preceding rise (6:00-6:20 AM) of only 100-121% of I- and B-ACTH concentrations respectively, was accompanied by a plasma cortisol rise of 112%. Similar con-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
M/25
M/33
S.L.
W.B.
32
non "peak" specimens
46
35
non "peak" specimens 3.4
6.6
1.2
1.5
13.5 ±
36
39
8:00 PM-10:55 PM
4:00 AM-7:55 AM (all)
112.5 ± 19.9
"peak" specimens
2.3
34.8+
7.5
0.6
4.5
non "peak" specimens
50.7+
53.6 ±
13*
3.0
57.6 + 3.4
23*
"peak" specimens
55.8 ±
116.3+13.4
39.6 ±
non "peak" specimens
10:00 AM-1:00 PM (all)
37*
20.9 +
36
8:00 PM-10:55 PM
4:00 AM-7:55 AM (all) 57.9 ±
22.0 ±
48
"peak" specimens (6:40-7:30)
2.7
2.8
82.0 ± 14.1
80.6 ±
80.8 ±
Mean plasma I-ACTHf pg/ml + SE
87.0 ± 18.9
21.6+ 2.1
35.8+ 6.1
46.3+ 9.8
43.2 + 4.2
44.6 + 4.8
91.5 + 11.9
26.3 + 2.8
41.5 ± 5.5
15.1 ± 1.0
Mean plasma B-ACTHf pg/m! ± SE
11.2 ± 0.8
7.8 ± 0.4
0.3935
0.6040
0.6849
NS
increase secretion. I/B ACTH ratios were similar in the normal subjects (1.42-1.64) and in the patients with Cushing's disease (1.27-1.47). "Apparent" ACTH half lives calculated from "peaks" of ACTH secretion were 7-12 min for I-ACTH and 3-9 min for B-ACTH in the normal subjects; and 9-13 min and 7-9 min respectively, in the patients with Cushing's disease. Mean plasma ACTH I- and B-concentrations at comparable time periods were higher in patients with active Cushing's disease than in normal subjects. These studies also indicate that in Cushing's disease, the abnormality present resides in ACTH regulatory mechanisms, not in the nature of the ACTH secreted. Approximation of the total amount of immunoassayable ACTH secreted in one normal subject over a 24-h period yielded a value of 73 /xg. Total /xg/h secreted in the 2 normal subjects were highest in the hour preceding awakening (6:30-7:30 AM; 12.9 and 12.2 /xg/h); were 5.3 and 4.0 /xg/h between 10:00-11:00 AM, and 1.4 and 1.7 /xg/h between 9:00-10:00 PM. In the 3 patients with clinically active Cushing's disease, apparent ACTH secretion between 10:00-11:00 AM varied from 19.234.3 /xg/h, the magnitude of such secretion being positively correlated with the extent of increased adrenal cortical activity present. (J Clin Endocrinol Metah 10: 675, 1975)
AVAILABILITY of a sensitive radioimmunoassay (1) for the determination of plasma ACTH levels has enabled a more precise investigation of the interrelationships of the central nervous systempituitary-adrenal axis. Such investigations had been previously hampered by the availability only of a relatively insensitive
ACTH bioassay (2). Immunoassay studies have revealed a rough parallelism between spontaneous peaks of plasma ACTH and cortisol levels (3,4) and the responses of these paramenters to exogenous stimuli (5-10). The major biological activity of ACTH resides in its first 20 amino acids (11). It is recognized that various antisera recognize different portions of the ACTH molecule (12,13). There have been few studies in which simultaneous determinations of immunoassayable (I) and bioassayable (B)
JLJL
Received August 5, 1974. Supported by U.S.P.H.S. Service Grant FR-71 from the Division of Research Facilities and Grants, USPHS Grant NB-02893 and the Lita Annenberg Hazen Charitable Trust.
675
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
JCE & M • 1975 Vol 40 • No 4
KRIEGER AND ALLEN
676
plasma ACTH levels have been performed. Due to the insensitivity of the bioassay, such studies have been possible only in situations where high ACTH levels were present, such as: following infusion of ACTH to normal subjects (14); or conditions characterized by ACTH excess (ectopic tumor or pituitary tumor production of ACTH) (12) and Addison's disease (15). In these studies I-ACTH levels were usually higher than those of B-ACTH by a factor of 1.6 to >10. Recent studies have also indicated the presence in the circulation in some of these altered clinical states, of biologically inactive ACTH: either "Big" ACTH (16) or C-terminal, N-terminal or middle fragments (12). In view of such findings, the desirability of correlating Band I-ACTH concentrations with each other and those of plasma cortisol under physiological conditions in normal subjects and in diseases characterized by increased ACTH production, is apparent. Materials and Methods Blood samples were obtained every 5 min via indwelling intravenous catheters over separate 3- to 4-h spans during different portions of the 24-h day in 2 normal male subjects (age 25 and 33 y). Sampling spans were separated by inter-
vals of at least one month. Four patients with Cushing's disease were similarly studied (2 of these on two separate occasions) over a 3-h period beginning between 9:00-9:30 AM. Three of these patients were studied prior to any treatment (RD, WH, PR) (Table 1). One of these (WH) was restudied 4 months postproton beam therapy, while still manifesting active disease (although with evidence of improvement as compared with pretreatment parameters). The fourth patient (MB) was studied on two occasions when in clinical and laboratory remission 16 and 19 months following pituitary irradiation (cobalt). Subjects were supine throughout all sampling spans. All studies (save those between 4:00 AM-8:00 AM) were begun V2 h after insertion of a 21-gauge scalp vein needle which was connected via an adapter to a 5 foot length of PE 160 tubing (dead space 2.6 ml). Catheters were kept patent with a microdrip of isotonic saline. In the 4:00 AM-8:00 AM study, catheter insertion was at midnight; the subjects were.alseep at 12:30 AM. The experimenter was in the same room as the subjects in all studies save that from 4:00-8:00 AM. In that study blood samples were collected in an adjoining room (into which the catheters extended.) "Dead space" blood volume was returned to subjects via a 3-way stopcock. Maximum volume of blood removed for assay during a sampling period was 250 cc. Blood samples were immediately transferred to heparinized glass tubes which had been kept
TABLE 1. Parameters of adrenal cortical function in patients with Cushing's disease at time of 5-min sampling studies
17OHCS mg/24 h
17KS mg/24 h
Cortisol secretory rate mg/24 h
Active c. 4 yr
60.0
79.5
—
Active disease c. 15 mos.
39.1
22.1
90.7
Active disease 4 mos. p Proton beam therapy
17.5
30.6
Remission 16 mos. p completion pituitary (cobalt) irradiation
9.8
8.9
19 mos. p completion of Rx
6.4
9.6
32.2
15.1
Urinary Pt.
Age/ Sex
R.D.
22/M
W.H.
42/M
M.B.
P.R.
48/M
35/F
Clinical condition
Active disease c. 18 mos.
22.8
—
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
ACTH (BIO-, IMMUNO-)CORTISOL RELATIONSHIP on crushed ice. They were immediately centrifuged in a refrigerated centrifuge and the plasma samples frozen immediately at — 20 C in separate tubes for I-ACTH, B-ACTH and cortisol determinations, so that each specimen was thawed only once prior to assay. Plasma I-ACTH concentrations were determined 1 on extracted plasma using an ACTH antiserum (Kendall) kindly provided by the N.I.H. B-ACTH concentrations were determined by a modification of the method of Sayers (17,18) which permits determination of such levels in plasma.2 Plasma cortisol, and in some instances, desoxycortisol, were determined by a competitive protein binding method (19). (The intra1
Plasma samples were extracted following acidification, by adsorption to silicic acid, a subsequent acid wash, and desorption of ACTH by aqueous acetone. The resultant extract was evaporated to dryness and reconstituted in a 0.5% human serum albumin phosphate buffer. Efficacy of extraction (monitored by 125 I-ACTH) varied from 72-75%TTo validate the use of 125I-ACTH as a marker for/recovery, in this and in the bioassay procedure, comparisons have been made of recovery of biologically active ACTH with that of 125 IhACTH to ascertain whether the hormone may be variably denatured during the extraction process. No significant loss of biological activity was seen. Antiserum was used at a dilution of 1:100,000 and 1:150,000. Minimum sensitivity was 10 pg/ml. Coefficient of variation was 6.13% (in intra-assay), 16.90% (interassay). For the dose range 2 to 5 pg, k = 0.079, for 5-50 pg, X = 0.091, and for 50-100 pg,
677
and interassay coefficients of variation for the cortisol assay are 3.1 and 3.5% respectively.) The time periods chosen for the clinically healthy subjects were 10:00 AM to 1:00 PM (subjects in a fasting state), 4:00 PM-8:00 PM, 8:00 PM to 11:00 PM (dinner hour at 6:00 PM) and 4:00 A M - 8 : 0 0 AM (observed sleep onset at 12:30 AM). During the first span plasma ACTH and cortisol concentrations are declining after their major circadian rise. During the second and third spans the least variation in plasma cortisol and immunoreactive ACTH concentrations has been noted in our previous studies (3). The last span encompasses the interval during which a major portion of pituitary and adrenal secretory activity occurs (3,4). Correlation coefficients were computed for log ACTH and plasma cortisol levels. The amount of ACTH secreted within given time periods was approximated as described by Gallagher et al.(3). Results 1. Normal subjects (Fig. 1, Table 2)
(a) Correlation of I and B-ACTH. There is excellent correlation between B and I-ACTH concentrations throughout all sampling spans for both subjects. The ratios of I-to-B-ACTH concentrations during different spans varied from 1.42-1.64. Both subjects displayed a major circadian X = 0.079. Characterization of the antibody was achieved by constructing displacement curves using rise in plasma ACTH concentrations at a-MSH, ah'-39ACTH, ap^ACTH, a^^ACTH, approximately 7:00 AM. Between 6:40-7:30 a'-10ACTH, a'-24ACTH, and a17"39ACTH. Only AM the ratios for the 2 subjects of I-to a'~24ACTH and a-MSH reacted fully with the anti- B-ACTH concentrations were 1.36 and body, with competitive binding curves almost indis- 1.37, which differed significantly (P < 0.05) tinguishable from that of authentic human ACTH. The other analogues demonstrated insignificant im- from the ratios obtained over the entire munoreactivity in the dose range employed in the 4:00 AM-8:00 AM span. Mean plasma assay. All samples from a given sampling period were ACTH and cortisol concentrations were run in duplicate in the same assay. significantly higher during this major cir2 Manuscript submitted for publication. Plasma cadian rise, when compared to means for samples were extracted in a manner similar to that each individual sampling span (P varied described for the immunoassay. The minimum sensitivity is 2.5 pg/ml. The coefficients of variation were: from < 0.001 to < 0.005). Intraassay variability 9.82%. Interassay variability 12.75%. a'~24ACTH in equimolar quantities to (b) Correlation between plasma ACTH and ap'-39ACTH, V-39ACTH was ca. 20%, more potent. cortisol concentration. The logarithm of Lysine vasopressin (0.01-1.0 IU), Pitressin® (0.2-2.0 plasma ACTH concentrations (bioassay) has IU) exhibited no biological activity. B-MSH 1 x 108 pg had potency equivalent to 2.5 pg ACTH. a-MSH been reported to have a linear relationship to plasma cortisol concentrations (20,21). 1 x 107 pg had potency equivalent to 10 pg ACTH.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
JCE & M • 1975 Vol 40 • No 4
KRIEGER AND ALLEN
678 200 180 160 140 120
S L • - • - • RlA ACTH Biocssoy ACTH ——• COHISOI
sioo 80 60 40 < 20 •
0
W B
200 180
— - • RIA A C T H
160 • - • - • Bioassay ACTH 140 ——
I 120
Cortisol
sioo i o < « Q-
_. 20 80 12 60 40 < ! 8 20 IQ- 4 n a n
/ \^VN
§l
5
6
7
8c
FIG. 1. Plasma ACTH (radioimmunoassay, bioassay) and cortisol concentrations, in 2 normal male subjects, sampled at 5-min intervals over 3-4 h spans. Sampling periods separated by intervals of at least 1 month. Ten AM-1:00 PM study performed with subjects in fasting state. Sleep onset was at 12:30 AM. Note parallelism between immunoassayable and bioassayable ACTH concentrations, as well as episodes in the 10:00 AM-1:00 PM and 4:00 A M - 8 : 0 0 AM intervals where major rises in ACTH concentrations are accompanied by none, or slight changes in cortisol concentrations.
Correlation coefficients between log I-ACTH and cortisol concentrations and log B-ACTH and cortisol concentrations are indicated in Table 2. It is apparent that both of these correlations are significant when considered for each entire sampling span (save for log B-ACTH/cortisol in WB from 10:00 AM to 1:00 PM) 3 and that such significant correlations persist in both of these subjects when each was restudied between 10:00 AM-1:00 PM. (Correlation coefficients (log I-ACTH and cortisol) in 4 other normal subjects studied half-hourly over a 24-h period (4) were 0.5272, 0.4892, 0.7678, and 0.7145, all significant at P < 0.001). Similar correlation coefficients were obtained when calculated for either log-I-ACTH or log-B-ACTH and the plasma cortisol concentration obtained 5 min subsequent to the ACTH determination. These 3
The almost even distribution of the number of B-ACTH determinations in the "peak" and "nonpeak" intervals (see text below) may explain the nonsignificance of this correlation, in contrast to the correlation seen between the I-ACTH and cortisol concentrations.
coefficients are not changed if one uses the plasma cortisol concentration obtained that would have been residual at each sampling time (utilizing a cortisol half life of 70 min) (20). However, in each subject, during similar spans (SL: 11:50 AM-noon, WB: 11:30 AM-12:05 PM, SL: 6:40 AM-7:30 AM, WB: 6:50 AM-7:30 AM) marked rises in both I-and B-ACTH concentrations occurred without concomitant, or with markedly diminished, increments in plasma cortisol concentrations. If one considers correlation coefficients for these cited episodes (Table 2), in subject WB there was no significant correlation between plasma ACTH concentrations (either I- or B-) and those of plasma cortisol. The rise in plasma cortisol concentrations between 6:50 and 7:05 AM was 64%, in contrast to rises of 646% and 900% respectively, in I-and B-ACTH concentrations (Table 3). In contrast, the immediately preceding rise (6:00-6:20 AM) of only 100-121% of I- and B-ACTH concentrations respectively, was accompanied by a plasma cortisol rise of 112%. Similar con-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 22:25 For personal use only. No other uses without permission. . All rights reserved.
M/25
M/33
S.L.
W.B.
32
non "peak" specimens
46
35
non "peak" specimens 3.4
6.6
1.2
1.5
13.5 ±
36
39
8:00 PM-10:55 PM
4:00 AM-7:55 AM (all)
112.5 ± 19.9
"peak" specimens
2.3
34.8+
7.5
0.6
4.5
non "peak" specimens
50.7+
53.6 ±
13*
3.0
57.6 + 3.4
23*
"peak" specimens
55.8 ±
116.3+13.4
39.6 ±
non "peak" specimens
10:00 AM-1:00 PM (all)
37*
20.9 +
36
8:00 PM-10:55 PM
4:00 AM-7:55 AM (all) 57.9 ±
22.0 ±
48
"peak" specimens (6:40-7:30)
2.7
2.8
82.0 ± 14.1
80.6 ±
80.8 ±
Mean plasma I-ACTHf pg/ml + SE
87.0 ± 18.9
21.6+ 2.1
35.8+ 6.1
46.3+ 9.8
43.2 + 4.2
44.6 + 4.8
91.5 + 11.9
26.3 + 2.8
41.5 ± 5.5
15.1 ± 1.0
Mean plasma B-ACTHf pg/m! ± SE
11.2 ± 0.8
7.8 ± 0.4
0.3935
0.6040
0.6849
NS
