Pathophysiology of pulsatile and copulsatile release of thyroid-stimulating hormone, luteinizing hormone, follicle-stimulating hormone, and alpha-subunit.

0021-972X/90/7102-0425$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society

Vol. 71, No. 2 Printed in U.S.A.

Pathophysiology of Pulsatile and Copulsatile Release of Thyroid-Stimulating Hormone, Luteinizing Hormone, Follicle-Stimulating Hormone, and a-Subunit* M. H. SAMUELS, J. D. VELDHUIS, P. HENRY, AND E. C. RIDGWAY Departments of Internal Medicine (Divisions of Endocrinology and Metabolism), University of Texas Health Science Center (M.H.S.), San Antonio, Texas 78284; University of Virginia Health Sciences Center (J.D. V.), Charlottesville, Virginia 22908; and University of Colorado Health Sciences Center (P.H., E.C.R.), Denver, Colorado 80262

jects had increased a-subunit pulse amplitude. We then tested pulse concordance among the four simultaneous hormone series. a-Subunit and the gonadotropins were significantly coreleased (triple coincidence), suggesting that all three hormones are closely linked to processes that regulate GnRH secretion. a-Subunit bursts were also significantly coincident with those of TSH in men, postmenopausal women, and hypothyroid subjects. Interestingly, TSH pulses were significantly concordant with those of LH and FSH, and all four hormones were significantly concordant in men, postmenopausal women, and hypothyroid subjects. In conclusion, the present findings imply that an underlying unified signal coordinates pulsatile hormone secretion from both gonadotrophs and thyrotrophs. {J Clin Endocrinol Metab 7 1 : 425-432, 1990)

ABSTRACT. Under physiological conditions, TSH, LH, FSH, and a-subunit are released in discrete pulses. To further characterize their neuroregulation and to investigate possible copulsatile secretion of these glycoprotein hormones, we studied the 24-h pulse profiles of all four hormones in each of four subject groups: young men, young women, postmenopausal women, and subjects with untreated primary hypothyroidism. Gonadotropin pulse properties in euthyroid men and women were similar to those previously reported, and hypothyroid subjects had normal gonadotropin pulse patterns. TSH release was pulsatile in all groups; hypothyroid subjects had increased pulse amplitude, but loss of the usual nocturnal increases in pulse amplitude, aSubunit concentrations were pulsatile in all groups, with minimal circadian variation; postmenopausal and hypothyroid sub-

T

control TSH secretion. Similarly, pulsatile a-subunit secretion occurs in normal young men (7) or during pulsatile GnRH administration to men with hypogonadotropic hypogonadism (8). However, a-subunit pulses have not been well studied in other conditions, and it is not clear whether spontaneous a-subunit pulses are coupled to bursts of LH and FSH and/or to TSH release. Explicit statistical testing must be employed to answer this question, since purely random pulse coincidence between independently pulsing hormone can be substantial at high pulse frequencies (9). Application of appropriate algorithms can uncover temporal correlations between the release of a-subunit and the other glycoproteins, which may offer significant insights into mechanisms governing a-subunit secretion. To examine the pathophysiology of pulsatile and copulsatile release of LH, FSH, TSH, and free a-subunit, we measured all four pituitary glycoprotein levels over 24 h in each of four groups of subjects with a wide range of hormone secretory rates: normal young men, normal young women, postmenopausal women, and patients with primary hypothyroidism. Hormone levels were subjected to objective pulse analysis, followed by explicit

HE PITUITARY glycoproteins TSH, LH, FSH, and the free a-subunit are normally secreted in discrete pulses (1-7). Analysis of these pulses has important physiological implications regarding the nature of underlying mechanisms that control hormone secretion. In addition, deranged hormone pulse patterns are believed to cause target organ dysfunction, even when basal hormone levels are normal. Thus, pulsatile patterns of LH and FSH release have been extensively studied in health and disease (1, 2). However, relatively little is known about pulsatile TSH and a-subunit secretion. For example, TSH levels normally rise at night, due to increases in pulse frequency and amplitude (3, 4), but whether this is maintained in primary hypothyroidism is unknown. Study of TSH pulse patterns in such conditions can lead to better understanding of the factors that Received December 4, 1989. Address requests for reprints to: Dr. M. H. Samuels, Department of Medicine, Division of Endocrinology, University of Texas Health Sciences Center, San Antonio, TX 78284. * This work was supported in part by Adult General Clinical Research Center Grant M01-RR-00051 and NIH Grants DK-36843-03 (to M.H.S.), CA-47411-01 (to E.C.R.), and Research Career Development Award 1K04-HD-00634 (to J.D.V.). 425

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 19:47 For personal use only. No other uses without permission. . All rights reserved.

SAMUELS ET AL.

426

statistical tests of nonrandom pulse coincidence. We hypothesized that the following patterns would emerge from this analysis: 1) TSH pulse frequency and/or amplitude would be increased in primary hypothyroidism, due to loss of thyroid hormone negative feedback; 2) asubunit pulses would correlate with LH pulses, as seen in previous studies (7, 8), but would also correlate with TSH pulses; and 3) TSH pulses would occur independently of gonadotropin pulses.

Subjects and Methods Patients Four groups of patients were studied at the University of Colorado Health Science Center Clinical Research Center (CRC): 1) five men, aged 18-30 yr, on no medications, with no history of thyroid or gonadal disease, who had normal physical examinations and serum thyroid hormone and testosterone concentrations; 2) seven women, aged 18-30 yr, on no medications, with no history of thyroid or gonadal disease, who had normal menstrual cycles, physical examinations, and serum thyroid hormone and estradiol levels (all women were studied on days 3-5 of their menstrual cycles, with day 0 defined as the onset of menses); 3) four women, aged 42-58 yr, with natural or surgical menopause at least 5 yr previously, on no medications, with no history of thyroid disease, who had normal physical examinations and serum thyroid hormone levels [all women had documented low serum estradiol (30 IU/L) before study] ; and 4) nine subjects, aged 17-55 yr, with primary hypothyroidism, documented by low serum thyroid hormone and elevated TSH levels; five of these patients had never been treated with thyroid hormones, three had discontinued thyroid hormones at least a year before the study, and one had discontinued T 3 treatment 2 weeks before the study. No subject received thyroid hormones during the study. Hypothyroidism was due to autoimmune thyroiditis, radioactive iodine treatment, or thyroidectomy for thyroid cancer. Patients with autoimmune thyroiditis had no evidence of other autoimmune processes, and patients with thyroid cancer had no evidence of metastatic disease on subsequent evaluation. One patient was postmenopausal, four were men, and four were premenopausal women studied on days 3-5 of their menstrual cycles. Patients were admitted to the CRC, and 3-mL blood samples were withdrawn via indwelling iv catheters every 15 min for 24 h, starting at 0800 h. Blood samples were centrifuged, and serum was frozen at —20 C. Hormone assays Basal blood samples were assayed for serum concentrations of T4, T 3 resin uptake, testosterone, and estradiol, using previously described methodologies (10-12). All blood samples were assayed for TSH, LH, and FSH by immunoradiometric assays (13-15), and a-subunit was measured by a double antibody RIA (16). Mean intra- and interassay coefficients of variation were, respectively, 6% and 7% for TSH, 6% and 11% for LH, 6% and 12% for FSH, and 12% and 20% for a-subunit at the hormone

JCE & M • 1990 Vol 71 • No 2

ranges present in the subjects. Detection limits were 0.08 mU/L for TSH, 0.5 IU/L for LH, 0.5 IU/L for FSH, and 0.1 ng/mL for a-subunit. The crossreactivity of the a-subunit assay with intact glycoproteins was 2.5-3.0% (16). TSH samples in hypothyroid patients that fell outside the assay range were rerun at 1:10 dilutions. All serum samples from an individual were run in the same assay. Statistical analaysis Significant hormone pulses were identified by Cluster analysis (17), using a variance model where dose-dependent coefficients of variation were calculated from all 97 sample replicates in each subject's hormone series. Cluster parameters were 2 points for test nadirs and 2 points for test peaks for all 4 hormones. T statistics were 1.0 for LH and FSH, and 2.0 for TSH and a-subunit. These parameters were defined for each hormone after extensive biophysical modeling of synthetic pulse series (18-20), and they reflect cluster sizes and t statistics that provide optimal sensitivity and positive accuracy for pulse detection. Significant differences in pulse parameters among groups were determined by analysis of variance, while significant differences between daytime and nocturnal pulse parameters within groups were determined by paired t tests. Nonrandom coincidence of hormone pulses was assessed by a previously described statistically based computer algorithm (9, 21). In this model, computer simulations were used to create synthetic endocrine time series pulsating randomly and independently at defined frequencies. The numbers of randomly coincident peaks in such unrelated synthetic series were used to estimate statistically expected values of random pulse concordance. Based upon the probability distributions of expected values, the significance of observed hormone copulsatility in each of the four groups of subjects could be determined. Nonrandom pulse concordance was, thus, determined for two, three, or four hormone series that were pulsing simultaneously, using the same criteria for the physiological and computer-simulated data. Since multiple comparisons were made for each hormone, a conservative P value of 0.01 or less was chosen to reject the null hypothesis of purely random pulse coincidence.

Results Hormone pulse parameters (Table 1) Gonadotropins. Mean 24-h LH pulse frequency was 13.4 in normal men, 15.7 in premenopausal women, 16.8 in postmenopausal women, 12.5 in hypothyroid men, and 14.5 in premenopausal hypothyroid women. LH pulse frequency was significantly higher in postmenopausal women than in hypothyroid men (P < 0.05), but there were no other differences in LH pulse frequency between groups. There were no differences between 12-h daytime (0800-1600 h) and nocturnal (1600-0800 h) pulse frequencies (data not shown). The mean 24-h LH pulse amplitude (maximal pulse height) was 7.5 IU/L in men, 9.3 IU/L in premenopausal women, 50.4 IU/L in postmenopausal women (P < 0.0001


COPULSATILITY OF PITUITARY GLYCOPROTEIN HORMONES

427

TABLE 1. TSH, LH, FSH, and a-subunit pulse parameters in males, premenopausal females, postmenopausal females, and hypothyroid subjects Males

Females (pre)

Females (post)

Hypothyroid

LH 24-h frequency

13.4 ± 0.7

15.7 ± 0.4

16.8 ± 1.5

24-h amplitude

7.5 ± 1.5

9.3 ± 2.1

50.4 ± 4.4°

FSH 24-h frequency

12.8 ± 0.7

14.1 ± 0.7

15.8 ± 1.7

24-h amplitude

5.2 ± 1.6

6.1 ± 1.3

72.9 ± 12.0°

8.8 ± 1.0 4.2 ± 0.4/4.6 ± 0.7 3.7 ± 1.1 2.5 ± 0.7/4.8 ± 1.4*

9.0 ± 0.7 4.3 ± 0.3/4.7 ± 0.4 2.4 ± 0.6 1.7 ± 0.4/3.1 ± 0.7*

9.0 ± 0 3.8 ± 0.8/5.3 ± 0.8 4.3 ± 0.9 3.2 ± 0.8/5.3 ± 1.1*

11.1 ± 1.0 5.6 ± 0.6/5.6 ± 0.4 134 ± 25° 130 ± 26/139 ± 25"

9.8 ± 0.9 4.6 ± 0.4/5.2 ± 0.6 1.0 ± 0.1 1.1 ± 0.1/1.0 ± 0.1

12.1 ± 0.8 5.9 ± 0.6/6.3 ± 0.6 1.1 ±0.1 1.0 ± 0.1/1.1 ± 0.1

16.5 ± 1.7° 8.3 ± 1.1/8.3 ± 0.6 3.3 ± 0.4° 3.4 ± 0.5/3.3 ± 0.4

11.1 ± 0.6 5.6 ± 0.5/5.5 ± 0.5 3.3 ± 0.6° 3.2 ± 0.5/3.2 ± 0.5

TSH 24-h frequency Day /nocturnal frequency 24-h amplitude Day/nocturnal amplitude

12.5 ± 0.6 (M) 14.5 ± 1.7 (F) 6.5 ± 1.7 (M) 5.4 ± 0.8 (F) 14.8 ± 0.9 (M) 14.8 ± 1.0 (F) 6.0 ± 2.5 (M) 5.1 ± 1.0 (F)

ct

24-h frequency Day/nocturnal frequency 24-h amplitude Day/nocturnal amplitude

Data are expressed as the mean ± SEM. Frequency, number of significant glycoprotein pulses per 24 h; amplitude, maximal height of the hormone pulse. Units are international units per L for LH and FSH, milliunits per L for TSH, and nanograms per mL for free a-subunit. The hypothyroid group was divided into male (M) and female (F) subjects to compare LH and FSH pulses to corresponding normal groups; one postmenopausal hypothyroid woman was not included in this analysis. " P < 0.05 compared to the same hormone in all other groups by analysis of variance. 6 P < 0.05 compared to daytime levels of the hormone in the same group by paired t test.

compared to other groups), 6.5 IU/L in hypothyroid men,

sal women, 4.3 mU/L in postmenopausal women, and

and 5.4 IU/L in hypothyroid premenopausal women.

134 mU/L in hypothyroid subjects (P < 0.0001 compared

There were no differences between mean daytime and nocturnal pulse amplitudes. The mean 24-h FSH pulse frequency was 12.8 in men, 14.1 in premenopausal women, 15.8 in postmenopausal women, 14.8 in hypothyroid men, and 14.8 in hypothyroid premenopausal women. None of these values was significantly different, and there were no differences between daytime and nocturnal pulse frequencies. The mean 24-h FSH pulse amplitude (maximal pulse height) was 5.2 IU/L in men, 6.1 IU/L in premenopausal women, 73 IU/L in postmenopausal women (P < 0.0001 compared to other groups), 6.0 IU/L in hypothyroid men, and 5.1 IU/L in hypothryoid premenopausal women. Daytime and nocturnal pulse amplitudes were not significantly different.

to other groups). The mean TSH pulse amplitude increased at night by 92% in men, 82% in premenopausal women, 66% in postmenopausal women, and 7% in hypothyroid subjects. All of these nocturnal increases were significant, but they were significantly blunted in the hypothyroid group. Figure 1 illustrates 24-h TSH pulse patterns that include one representative subject in each group.

TSH. The mean 24-h TSH pulse frequency was 8.8 in men, 9.0 in premenopausal women, 9.0 in postmenopausal women, and 11.1 in hypothyroid subjects. There were no differences among the groups, and there were no significant changes in pulse frequency at night compared to daytime pulse frequency. The mean 24-h TSH pulse amplitude (maximal pulse height) was 3.7 mU/L in men, 2.4 mU/L in premenopau-

a-Subunit. The mean 24-h a-subunit pulse frequency was 9.8 in men, 12.1 in premenopausal women, 16.5 in postmenopausal women (P < 0.004 compared to other groups), and 11.1 in hypothyroid subjects. There were no nocturnal increases in the mean a-subunit pulse frequency in any group. The mean 24-h a-subunit pulse amplitude was 1.0 ng/ mL in men, 1.1 ng/mL in premenopausal women, 3.3 ng/ mL in postmenopausal women, and 3.3 ng/mL in hypothyroid subjects. The latter two groups had higher amplitudes than those in men and premenopausal women (P < 0.0004). There were no significant nocturnal increases in mean pulse amplitude in any group. Figure 2 illustrates 24-h a-subunit pulse patterns in representative subjects, and Fig. 3 shows 24-h pulse patterns for all four hormones from a single individual.


SAMUELS ET AL.

428 Male

JCE & M • 1990 Vol 71 • No 2 Host-Menopousal Female

Time (h)

Pre-Menopausol Female

Primary Hypothyroidism

FIG. 1. Serial serum TSH concentrations over 24 h in a representative male (upper left), premenopausal female {lower left), postmenopausal female {upper right), and hypothyroid subject (lower right). Note the different scale for TSH values in the hypothyroid patient. Asterisks denote significant TSH pulses, detected by Cluster analysis (see Materials and Methods).

Coincidence of hormone pulses (Table 2)

Observed numbers of pulse coincidences are compared to expected numbers for two, three, or four hormones in Table 2. Two hormone pulses were considered to be concordant if they occurred within the same blood sample or at a fixed lag of 15 min, e.g. the peak maximum of hormone A lagged that of hormone B by 15 min or vice versa (but not both). Time lags for cosecretion of hormone pulses are also listed in Table 2. As expected, LH and FSH pulses were significantly concordant in most subjects, with approximately 25% of the total number of LH pulses coinciding with FSH pulses. FSH pulses occurred either simultaneously with or 15 min after LH pulses. LH and a-subunit as well as FSH and a-subunit were also cosecreted (P < 0.0001 in most cases), with either exact temporal correlation or with a pulses leading those of the intact gonadotropins by 15 min. Twenty-five to 40% of the total number of a pulses were concordant with LH pulses, depending on the group, and approximately 28% were concordant with FSH pulses. Although LH and a-subunit pulsations were not significantly concordant in the postmenopausal women (at least at P < 0.01 chosen in this study), a P value of 0.03 in this group suggests that a larger sample

size may have shown an even stronger correlation. Significant triple coincidence among release episodes of LH, FSH, and a-subunit were seen in men (P < 0.0001), premenopausal women (P < 0.0001), and hypothyroid subjects (P < 0.0001). Interestingly, TSH and LH pulses were coincident more often than expected randomly in all groups, although this was only significant at P < 0.01 in normal men (P < 0.002). Thirty percent of the total number of TSH pulses coincided with LH pulses in this group. Concordant TSH and FSH pulses were seen in normal men (27% of TSH pulses; P < 0.003), premenopausal women (29% of TSH pulses; P < 0.002), and hypothyroid subjects (23% of TSH pulses; P < 0.01), but did not reach significant levels in postmenopausal women (P < 0.05). Triple coincidence among TSH, LH, and FSH pulses was seen in men (P < 0.003) and approached significant rates in premenopausal women (P < 0.02) and hypothyroid subjects (P < 0.06). TSH and a-subunit were coreleased in men (20% of a-subunit pulses; P < 0.004), postmenopausal women (20% of a-subunit pulses; P < 0.002), and hypothyroid subjects (28% of a-subunit pulses; P < 0.0001), with exact temporal coincidence (no lag) in each case. Surprisingly, TSH, LH, FSH, and a-subunit were all



429

Post-Menopousol Female

Male 20

1.5

1.0

*

y

*

*

*

LA

05

o.oTim.

(h)

Pre-Menopausol Female

Time (h)

Time (h)

Primary Hypothyroidism

Time (h)

FIG. 2. Serial serum a-subunit concentrations over 24 h in a representative male (upper left), premenopausal female (lower left), postmenopausal female (upper right), and hypothyroid subject (lower right). Note the different scale for a-subunit values in the postmenopausal and hypothyroid subjects. Asterisks denote significant a-subunit pulses, detected by Cluster analysis.

quadruply coreleased in men (P < 0.001), postmenopausal women (P < 0.01), and hypothyroid subjects (P < 0.002). In addition, the coincidence rate for all four hormones reached P < 0.015 in premenopausal women.

Discussion The present study examined pituitary glycoprotein pulse parameters in four groups of subjects with a wide range of basal hormone levels. The groups with elevated hormone levels (postmenopausal and hypothyroid patients) were heterogeneous, and these subjects probably had variable defects in target organ function. Despite this, glycoprotein pulse parameters in the current study closely parallel those reported in previous studies of LH and FSH (1, 22-22), TSH (3-5), and a-subunit (6, 7) and provide a valid basis for analysis of hormone copulsatility. Patients with primary hypothyroidism had normal TSH pulse frequency and increased amplitude, suggesting that thyroid hormone exerts negative feedback on TSH mainly at the pituitary gland. Similar inferences have been reached in the past using exogenous thyroid hormone administration and TRH injections (5). Al-

though these subjects had nocturnal increases in TSH pulse amplitude, the increases were quite attenuated compared to those in normal subjects (6% us. 80-90%). Weeke and Laurberg (25) also reported loss of diurnal variation in TSH levels in hypothyroid patients; our data show that this reflects loss of the nocturnal augmentation of pulse amplitude. Whether this phenomenon represents a loss of direct thyroid hormone effects on circadian TSH secretion or changes in central mediators of TSH secretion, such as dopamine (4, 26, 27), remains to be determined. Hypothyroid patients frequently exhibit gonadal dysfunction and increased gonadotropin levels (28, 29), reportedly due to increased gonadotropin pulse amplitude (29). In contrast, no changes in gonadotropin pulse patterns were found in the current hypothyroid group. However, the sample size was small, and abnormalities in gonadotropin secretion may emerge from further study of hypothyroid patients under more intensive sampling conditions before and during thyroid hormone replacement (29). Close temporal coupling of LH and FSH pulses is confirmed in our study of men and premenopausal


SAMUELS ET AL.

430

JCE & M • 1990 Vol 71 • No 2

TSH

LH 8

* *

6

*

\

H

\ V

2-

0

FIG. 3. Twenty-four-hour profiles of pulsatile serum TSH (upper left), LH (upper right), FSH (lower left), and a-subunit (lower right) levels in a single representative subject (a premenopausal woman). Asterisks denote significant hormone pulses, detected by Cluster analysis.

TABLE 2. Coincident TSH, LH, FSH, and a-subunit pulses in males, premenopausal females, postmenopausal females, and hypothyroid subjects Females (pre)

Males 0 (E ± SD)

Pattern of lag (min)

LH/FSH

18 (8.2 ± 1.1)°

0,15

LH/a FSH/a

21 (6.3 ± 1.0)° 14 (5.9 ± 1.0)°

0,0 0,-15

LH/FSH/a TSH/LH TSH/FSH TSH/LH/FSH TSH/a TSH/LH/FSH/a

6 (0.8 ± 0.4)° 13 (6.0 ± 0.9)6 12 (5.4 ± 0.9)6 4 (0.7 ± 0.4)* 10(4.2±0.8) 6 2 (0.07 ± 0.4)*

0,15,0 0,-15 0,15 0,-15,0 0,0 -15,0,0,0 0,0,15,-15 0,15,15,0

O (E ± SD)


27 (16.1 ± 0,0 1.3)° 30 (14 ± 1.2)° 0,-15 22 (12.6 ± 0,0 1.2)* 10 (2.0 ± 0.5)° 0,0,-15 15 (10.3 ± 1.1) 0,0 18 (9.2 ± 1.0)* 0,0 5 (1.5 ± 0.4) 0,0,0 10 (8.0 ± 0.9) 0,0 2 (0.2 ± 0.2) 0,0,-15,-15

Females (post) O (E ± SD)


Hypothyroid O (E ± SD)


13 (10.8 ± 1.4)

0,15

29 (18.9 ± 1.2)*

17 (11.2 ± 1.4) 12 (10.7 ± 1.4)

0,0 0,0

29 (17.4 ± 1.2)° 0,0 29 (17.4 ± 1.2)° 0,0

4 (1.8 ± 10 (6.2 ± 10 (5.9 ± 0 (1.0 ± 13 (6.1 ± 2 (0.2 ±

0,0,15 0,0 0,0

11 (2.5 ± 0.5)° 21 (14.8 ± 1.1) 23 (14.8 ± 1.1)* 5 (2.2 ± 0.5) 29 (13.5 ± 1.1)° 3 (0.3 ± 0.2)*

0.7) 1.1) 1.1) 0.5) 1.1)* 0.2)*

0,0 0,0,0,15

0,0

0,0,0 0,0 0,0 0,0,0 0,0 0,0,0,0 0,15,15,15

Two, 3, or 4 hormone series were compared simultaneously, as designated in the first column. O denotes the observed numbers of coincident pulses, while E denotes the expected numbers of randomly coincident pulses (given in parentheses, ±SEM). Lag is the time lag (minutes) between pulses of the hormones in the order listed in the first column; a negative number indicates that the corresponding hormone pulse precedes that of the other hormone(s), whereas, a positive lag indicates that the peak of the corresponding hormone follows that of the other hormone(s). For example, in the male group, 13 concordant TSH and LH pulses were found, and the LH pulses occurred 15 min earlier than the TSH pulses. ° P < 0.0001 compared to the expected number of coincident pulses. * P < 0.01 compared to the expected number of coincident pulses.



women (9, 30) and is extended to hypothyroid patients. Gonadotropin pulse concordance is probably due to the underlying GnRH pulse generator, although discordant FSH control can be shown by altering exogenous GnRH pulse frequency (31) or by administering a GnRH antagonist (32). LH and FSH pulses were not concordant in our postmenopausal women; this may reflect the small sample size or may be another example of discordant gonadotropin control in states of altered GnRH pulse frequency. Our current data support earlier reports that a-subunit is cosecreted with the gonadotropins (6, 7, 9, 33, 34). In addition, using formal coincidence analysis, we have shown for the first time that a-subunit pulses are concordant with those of TSH in normal men, postmenopausal women, and hypothyroid subjects. In contrast, premenopausal women exhibited a degree of a-subunit and TSH pulse concordance expected by chance alone. This may reflect a greater contribution of estradiolstimulated a-subunit secretion from gonadotrophs in normal women (34). Surprisingly, we found that TSH pulses were significantly concordant with LH pulses in men and with FSH pulses in three of the groups. In addition, all three intact glycoproteins were coreleased in men, and all four hormones were copulsatile in three groups. Such temporal coordination has not been previously recognized and raises interesting possibilities regarding coupled or common central control mechanisms that regulate pulsatile glycoprotein secretion. For example, TSH (4, 26, 27), the gonadotropins (35-37), and free a-subunit (38) are all controlled to some extent by dopamine. One plausible consideration is that changes in hypothalamic dopamine levels coordinate the pulsatile release of all four glycoproteins. Another possible mediator of concordant glycoprotein pulses is the endogenous opiate system, since opiates may inhibit the release of GnRH (39-41) and TRH or TSH (42, 43). In the current study, concordant pulses comprised 40% or less of the total pulses for a specific hormone, rates similar to those reported for other hormone pulse series (9). Thus, although pulse concordance rates were highly significant, the majority of hormone pulses were not coincident with those of other hormones. This phenomenon may be due to the following. 1) Observed coincident pulse rates probably underestimate true physiological rates, since various sources of experimental uncertainty may decrease apparent coincidence. These sources include sample processing variations, measurement errors, errors in the peak detection program, and nonuniform secretory burst durations and/or half-lives of hormone disappearance among different subjects. 2) In addition to central control by factors such as dopamine or opiates,

431

each hormone is also under individual control by its specific target organ and hypothalamic factors. These independent modulators of pulsatile hormone release may tend to reduce the number of coincident pulses released in response to a common mediator. 3) Based on the present study, a-subunit pulses arise from both gonadotrophs and thyrotrophs and are, therefore, subject to control by both gonadotroph- and thyrotroph-specific factors. These divergent effects may decrease apparent a-subunit copulsatility with either LH/FSH or TSH. In summary, we have characterized pulsatile and copulsatile glycoprotein release in subjects harboring wide ranges of hormone levels. Hypothyroid patients continued to secrete TSH in pulses of increased amplitude and lost the normal TSH diurnal variation. These patients, however, did not have the expected abnormalities in LH or FSH pulse patterns. a-Subunit pulses correlated closely with gonadotropin pulses, as expected, and also with TSH pulses, a phenomenon not previously reported. Of considerable interest was the concordance of TSH and gonadotropin pulses in many instances. These results offer significant insights into coordinate mechanisms governing pulsatile endocrine activity.

Acknowledgments The authors would like to thank the staff of the University of Colorado Health Science Center Endocrinology and Clinical Research Center Core Laboratories for performing the glycoprotein assays. We also thank Douglas Robertson, PhD. for statistical assistance.

References 1. Urban RJ, Evans WS, Rogol AD, Kaiser DL, Johnson ML, Veldhuis JD. Contemporary aspects of discrete peak-detection algorithms. I. The paradigm of the luteinizing hormone pulse signal in men. Endocr Rev. 1988;9:3-37 2. Santoro N, Filicori M, Crowley WF. Hypogonadotropic disorders in men and women: diagnosis and therapy with pulsatile gonadotropin-releasing hormone. Endocr Rev. 1987;7:11-23 3. Greenspan SL, Klibanski A, Schoenfeld D, Ridgway EC. Pulsatile secretion of thyrotropin in man. J Clin Endocrinol Metab. 1987;63:661-8 4. Rossmanith WG, Mortola JF, Laughlin GA, Yen SSC. Dopaminergic control of circadian and pulsatile pituitary thyrotropin release in women. J Clin Endocrinol Metab. l988;67:560-4 5. Brabant G, Brabant A, Ranft U, et al. Circadian and pulsatile thyrotropin secretion in euthyroid man under the influence of thyroid hormone and glucocorticoid administration. J Clin Endocrinol Metab. 1987;65:83-8 6. Winters SJ, Troen P. Pulsatile secretion of immunoreactive alphasubunit in man. J Clin Endocrinol Metab. 1985;60:344-8 7. Veldhuis JD, Winters SJ. Nature of alpha subunit secretion in men: circadian rhythms, pulsatile release and secretory profiles. J Androl. 1989;10:248-58 8. Spratt DI, Chin WW, Ridgway EC, Crowley WF. Administration of low dose pulsatile gonadotropin-releasing hormone (GnRH) to GnRH-deficient men regulates free alpha-subunit secretion. J Clin Endocrinol Metab. l986;62:102-8 9. Veldhuis JD, Iranmanesh A, Clarke I, Kaiser DL, Johnson ML. Random and non-random coincidence between luteinizing hormone peaks and follicle-stimulating hormone, alpha subunit, pro-


432

10. 11. 12. 13.

14. 15.

16.

17. 18. 19. 20.

21.

22.

23. 24. 25. 26.

SAMUELS ET AL. lactin and gonadotropin-releasing hormone pulsations. J Neuroendocrinol. 1989;l:l-10 Gautuik KM, Tashjian AH, Kourides IA, et al. Thyrotropin-releasing hormone is not the sole physiologic mediator of prolactin release during suckling. N Engl J Med. 1974;290:1162-4 Ismail AA, Astley P, Burr WA, et al. The role of testosterone measurements in the investigation of androgen disorders. Ann Clin Biochem. 1986;23:113-34 Bergquist C, Nillius SJ, Wide L. Human gonadotropin therapy. I. Serum estradiol and progesterone patterns during conceptual cycles. Fertil Steril. 1983;39:761-5 van Heynigen V, Abbott SR, David SG, Anderson LJ, Ridgway EC. Development and utility of a monoclonal-antibody-based highly sensitive immunoradiometric assay of thyrotropin. Clin Chem. 1987;33:1387-90 Hunter WM, Bennie JG, Kellett HA et al. A monoclonal antibodybased immunoradiometric assay for h-LH. Ann Clin Biochem. 1984;21:275-83 Filicori M, Marseguerra M, Mimmi P, et al. The pattern of LH and FSH pulsatile release: physiological and clinical significance. In: Flamigni C, Givens JR, eds. The gonadotropins: basic science and clinical aspects in females. London: Academic Press; 1982:36576 Kourides IA, Weintraub BD, Ridgway EC, Maloof F. Pituitary secretion of free alpha and beta subunit of human thyrotropin in patients with thyroid disorders. J Clin Endocrinol Metab. 1975;40:872-85 Veldhuis JD, Johnson ML. Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol. 1986;250:E486-93 Urban RJ, Johnson ML, Veldhuis JD. Biophysical modeling of sensitivity and positive accuracy of detecting episodic endocrine signals. Am J Physiol. 1989;257:E88-94 Veldhuis JD, Johnson ML. A novel general biophysical model for simulating episodic endocrine gland signaling. Am J Physiol. 1988;255:E749-59 Urban RJ, Johnson ML, Veldhuis JD. In vitro biological validation and biophysical modeling of the sensitivity and positive accuracy of endocrine peak detection. I. The LH pulse signal. Endocrinology. 1989;124:2541-7 Johnson ML, Seneta E. Random versus non-random peak coincidence among 2, 3 or 4 pulsating endocrine time-series [Abstract 1719]. Proc of the 71st Annual Meet of The Endocrine Soc. 1989;452. Filicori M, Santoro N, Merriam GR, Crowley WF. Characteristics of the physiological pattern of episodic gonadotropin secretion throughout the human menstrual cycle. J Clin Endocrinol Metab. 1986;62:1136-44 Veldhuis JD, Johnson ML. In vivo dynamics of luteinizing hormone secretion and clearance in man: assessment by deconvolution mechanics. J Clin Endocrinol Metab. 1988;66:1291-300 Kazer RR, Liu CH, Yen SSC. Dependence of mean levels of circulating luteinizing hormone upon pulsatile amplitude and frequency. J Clin Endocrinol Metab. 1987;65:796-800 Weeke J, Laurberg P. Diurnal TSH variations in hypothyroidism. J Clin Endocrinol Metab. 1976;43:32-7 Kerr DJ, Singh VK, McConway MG, et al. Circadian variation of thyrotropin, determined by ultrasensitive immunoradiometric assay, and the effect of low dose nocturnal dopamine infusion. Clin

JCE & M • 1990 Vol 71 • No 2

Sci. 1987;72:737-41 27. Sowers JR, Catania RA, Hershman JM. Evidence for dopaminergic control of circadian variations in thyrotropin secretion. J Clin Endocrinol Metab. 1982;54:673-5 28. Drake TS, O'Brien WF, Tredway DR. Pituitary response to LHRH in hypothyroid women. Obstet Gynecol. 1980;56:488-91 29. Iranmanesh A, Lizarralde G. Primary hypothyroidism alters the pulsatile mode of LH and FSH release in men by amplitude modulation [Abstract 165]. Proc of the 71st Annual Meet of The Endocrine Soc. 1989;64. 30. Matsumoto AM, Bremner WJ. Modulation of pulsatile gonadotropin secretion by testosterone in man. J Clin Endocrinol Metab. 1984;58:609-14 31. Gross KM, Matsumoto AM, Bremner WJ. Differential control of luteinizing hormone and follicle-stimulating hormone secretion by luteinizing hormone-releasing hormone pulse frequency in man. J Clin Endocrinol Metab. 1987;64:675-80 32. Hall JE, Brodie TD, Badger TM, et al. Evidence of differential control of FSH and LH secretion by gonadotropin-releasing hormone (GnRH) from the use of a GnRH antagonist. J Clin Endocrinol Metab. 1988;67:524-31 33. Winters SJ, Troen P. Evidence for a role of endogenous estrogen in the hypothalamic control of gonadotropin secretion in men. J Clin Endocrinol Metab. 1985;61:842-5 34. Rosemberg E, Bulat G. Immunoreactive alpha and beta subunits of follicle stimulating and luteinizing hormones in peripheral blood throughout the menstrual cycle and following stimulation with synthetic gonadtropin releasing hormone (GnRH). J Endocrinol Invest. 1979;2:233-9 35. Matsubara M, Tango M, Nakagawa K. Effects of dopaminergic agonists on plasma luteinizing hormone-releasing hormone (LRH) and gonadotropins in man. Horm Metab Res. 1987;19:31-4 36. Andersen AN, Hagen C, Lange P, et al. Dopaminergic regulation of gonadotropin levels and pulsatility in normal women. Fertil Steril. 1987;47:391-7 37. Gambacciani M, Melis GB, Paoletti AM, et al. Pulsatile luteinizing hormone release in postmenopausal women: effect of chronic bromocriptine administration. J Clin Endocrinol Metab. 1987;65:4658 38. Cooper DS, Klibanski A, Ridgway EC. Dopaminergic modulation of TSH and its subunits: in vivo and in vitro studies. Clin Endocrinol (Oxf). 1983;18:265-75 39. Morley JE, Barenetsky NG, Winger TD, et al. Endocrine effects of naloxone-induced opiate receptor blockade. J Clin Endocrinol Metab. 1980;50:251-7 40. Ellingboe J, Veldhuis JD, Mendelson JH, Kuehnie SC, Mello NK. Effect of endogenous opioid blockade on the amplitude and frequency of pulsatile luteinzing hormone secretion in normal men. J Clin Endocrinol Metab. 1982;54:854-7 41. Delitala G, Giusti M, Matzocchi G, Granziera L, Tarditi W, Giordano G. Participation of endogenous opiates in regulation of the hypothalamic-pituitary-testicular axis in normal men. J Clin Endocrinol Metab. 1983;57:1277-81 42. Bruni JF, Van Vugt D, Marshall S, Meites J. Effects of naloxone, morphine and methionine enkephalin on serum prolactin, luteinizing hormone, follicle stimulating hormone, thyroid stimulating hormone and growth hormone. Life Sci. 1977;21:46l-6 43. Meites J, Bruni JF, Van Vugt DA, Smith AF. Relation of endogenous opioid peptides and morphine to neuroendocrine functions. Life Sci. 1979;24:1325-36


Reliability of estimates of pulsatile characteristics of luteinizing hormone and growth hormone release in women.

Preservation of pulsatile luteinizing hormone release during postpartum lactational amenorrhea.

Correlation between the serum luteinizing hormone to folliclestimulating hormone ratio and the anti-Müllerian hormone levels in normo-ovulatory women.

Interleukin 1 alpha inhibits prostaglandin E2 release to suppress pulsatile release of luteinizing hormone but not follicle-stimulating hormone.

Reduced luteinizing hormone release by synthetic luteinizing hormone-releasing hormone (LHRH) in postpartum lactating rats.

Interleukin-1 inhibits the ovarian steroid-induced luteinizing hormone surge and release of hypothalamic luteinizing hormone-releasing hormone in rats.

Luteinizing-hormone-releasing hormone.

Follicle stimulating hormone-secreting pituitary adenoma: inappropriate secretion and effect of pulsatile luteinizing hormone releasing hormone analogue (buserelin) administration.

Luteinizing hormone secretion in hypophysial stalk-transected gilts given hydrocortisone acetate and pulsatile gonadotropin-releasing hormone.

Neuropeptide Y is a neuromodulator of pulsatile luteinizing hormone-releasing hormone release in the gonadectomized rhesus monkey.

Thyroid hormone regulation of the pulsatile discharges of luteinizing hormone in ovariectomized rats.

Dissociation of the porcine anterior pituitary: the kinetics of luteinizing hormone release in response to luteinizing hormone-releasing hormone.

Effect of pulsatile luteinizing hormone-releasing hormone administration on pituitary-gonadal function in elderly man.

Neuroendocrine regulation of pulsatile luteinizing hormone secretion in elderly men.

Intrinsic pulsatile secretory activity of immortalized luteinizing hormone-releasing hormone-secreting neurons.

Regulation of pulsatile luteinizing hormone secretion in experimental uremia.

Blockage of estrogen-induced release of luteinizing hormone by reserpine and potentiation of synthetic gonadotropin-releasing hormone-induced release of luteinizing hormone by estrogen in the ovariectomized ewe.

Effect of melatonin on pulsatile release of luteinizing hormone in female lambs.

Luteinizing hormone and follicle stimulating hormone and the response to luteinizing hormone releasing hormone in relation to sex and age.

Differential control of luteinizing hormone and follicle-stimulating hormone by luteinizing hormone releasing hormone in the ram.

Delayed puberty in uremia: pituitary-gonadal function during short-term pulsatile luteinizing hormone-releasing hormone administration.

Release of luteinizing hormone releasing hormone and thyrotropin releasing hormone from a synaptosome-enriched fraction of hypothalamic homogenates.

Central estrogen action sites involved in prepubertal restraint of pulsatile luteinizing hormone release in female rats.

Release of luteinizing hormone and growth hormone after recovery from maximal exercise.