|
|
||||||||
a Department of Obstetrics and Gynecology and
b Department of Anatomy, Medical College of Virginia, Richmond, Virginia 23298
c Department of Biology,
d Department of Internal Medicine, and
e NSF Center for Biological Timing, University of Virginia, Charlottesville, Virginia 22908
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
6 days) followed by an anovulatory condition described as persistent estrus [6]. Previous studies have revealed that proestrous LH surges are attenuated in some middle-aged rats [79]. Further, middle-aged female rats with attenuated proestrous LH surges cease to display regular estrous cyclicity within 2 mo, while rats of similar age that have normal LH surges continue to cycle regularly beyond 2 mo [10]. Despite these observable alterations in neuroendocrine function that precede changes in estrous cyclicity, the mechanisms involved are presently unknown. Age-related alterations in pulsatile LH release have been identified in female rats following ovariectomy [1113]. However, detailed investigations of the pulsatile characteristics of LH secretion have not been performed in middle-aged regularly cyclic females under physiologic, non-ovariectomized conditions. Previous studies have evaluated the proestrous LH surge in young cycling rats using frequent sampling techniques (23 min) combined with deconvolution analysis [14]. These studies have indicated that the proestrous LH surge could result from an increase in LH burst frequency and amplitude combined with increases in basal secretion and LH half-life as compared to episodic LH secretion at other times of the rat estrous cycle. Using this analytical approach, the current study was designed to determine whether the attenuated LH surges observed in middle-aged regularly cyclic female rats reflect decreased frequency, amplitude, basal secretion, half-life, and/or other parameters of LH secretion during the surge. Accordingly, we performed frequent (3 min) measurements of peripheral plasma LH concentrations during proestrus in regularly cyclic young and middle-aged rats and subjected the immunoreactive LH measurements to deconvolution analysis. A novel measure of the orderliness of serial hormone release, approximate entropy, was applied to quantify possible complementary changes in the (nonpulsatile) moment-to-moment regularity of LH secretion in the young versus middle-aged animals.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Young (4 mo) virgin and middle-aged (1012 mo) retired breeder female Long-Evans rats (Charles River Labs., Wilmington, MA) were maintained under standard laboratory conditions at room temperature (2325°C) with a lighting schedule of 14L:10D (lights-on from 0500 to 1900 h). Food and water were available ad libitum. Daily vaginal smears were taken to monitor estrous cyclicity. Only animals demonstrating at least three consecutive 4-day estrous cycles were used for these studies. Guidelines for the care and use of these animals were approved by the Virginia Commonwealth University Institutional Animal Care and Use Committee, and research was conducted in accordance with the Guiding Principles for the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction.
Experimental Design
Groups of young (n = 5) and middle-aged (n = 13) regularly cyclic rats were implanted with jugular vein catheters on diestrus Day 2 between 1000 and 1300 h according to the method of Harms and Ojeda [15]. Thereafter, the animals were housed in individual cages, and daily vaginal smears were continued; only those animals that maintained a regular 4-day estrous cycle were used on the subsequent day of proestrus. On the next proestrus (5 days after cannulation), blood samples (0.2 ml) were collected at 3-min intervals from 1600 to 2200 h for later plasma LH determination by RIA. Immediately after each blood sample was taken, an equal volume of replacement blood was reinfused through the catheter. The red blood cells for replacement blood were obtained from ovariectomized females and prepared as previously described [16]. After blood sampling on proestrus, daily estrous cycle monitoring was continued on all rats. Since many of the rats did not continue to cycle normally, only young and middle-aged rats that continued to demonstrate a 4-day estrous cycle were included in the plasma LH data analysis.
RIA
Concentrations of LH in each plasma sample were measured in triplicate (20 µl) by RIA using reagents kindly supplied by Dr. A.F. Parlow and the National Pituitary and Hormone Program of the NIDDK (Rockville, MD). Rat LH anti-serum (S-8) and rat LH reference preparation RP-2 were used. Intra- and interassay coefficients of variation calculated from pools of plasma were 8% and 12%, respectively.
Data Analysis
LH secretory burst detection The plasma LH concentration at any given moment was assumed to reflect the simultaneous operation of four secretory and half-life parameters including 1) the number and temporal locations, 2) the maximal release rate/secretory burst, and 3) the durations of all significant LH secretory events, acted upon by 4) single-component LH half-life as previously described [17]. Secretion was assumed to comprise a series of distinct bursts with statistically bounded amplitudes. A secretory event was defined as an algebraically Gaussian distribution of instantaneous molecular secretory rates with finite mass (calculated as the integral of the individual secretory bursts as a function of time), amplitude (maximal secretory rate), and half-duration (duration of the burst at half-maximal amplitude). A basal (time-invariant) hormone secretion rate was estimated concurrently from the lower 5% of the plasma LH concentrations. A single half-life of elimination was used to characterize LH elimination kinetics, with the assumption that neither half-life nor distribution volume changes during the interval of sampling.
Estimation of approximate entropy The approximate entropy (ApEn) statistic has provided new insights into the degree of pattern orderliness or replicability for moderately noisy and short time series [1820]. ApEn was thus used here to quantify serial irregularity in LH concentration-time series. ApEn is complementary to pulse and secretory burst detection procedures in that it evaluates both dominant (pulsatile) and subordinate sample-by-sample subpatterns in data. Mechanistically, pulse detection identifies underlying discrete secretory events whereas ApEn reflects within-axis organization and feedback control. ApEn assigns a non-negative number to a time series, with larger values corresponding to greater apparent serial irregularity or disorderliness in hormone release patterns over time. Thus significantly higher ApEn values suggest increased complexity in regulatory inputs or decreased coordination of feedback controls. Two input parameters, m and r, must be specified to compute ApEn, which technically measures the logarithmic likelihood that runs of patterns that are close (within r) for m contiguous observations remain close (within the same tolerance width r) on next incremental comparisons. In this study, we calculated ApEn (m, r) values for all data sets to be m = 1 and r = 20% of the SD of the individual rat's LH surge concentration-time series. Normalizing r to each time-series SD gives ApEn a translation- and scale-invariance to absolute plasma (LH) concentration levels. ApEn values were also calculated for m = 2 and r = 0.2, which yielded similar inferences, thus illustrating the generality of the interpretation. ApEn was calculated after detrending the LH profiles by first-differencing of the data. Three hundred Monte Carlo simulations were used for each series in order to estimate the within-assay-dependent SD of ApEn in each rat by perturbing each original LH concentration-time series with random experimental variation defined by the dose-dependent within-sample SD [21].
Statistical Analysis
Comparisons of parameters were performed by Kruskal-Wallis one-way ANOVA and the post hoc comparisons by Dunn's method. A confidence level of p < 0.05 was considered statistically significant. Data are expressed as means ± SEM.
| RESULTS |
|---|
|
|
|---|
|
Table 1 presents the deconvolution analysis results of LH secretion and half-life on proestrus (16002200 h) in young females (n = 5), middle-aged females (n = 7) with LH surges of normal magnitude, and middle-aged females (n = 6) with attenuated LH surges. There were no differences in the mean LH burst frequencies during proestrus in young (3.6 ± 0.9 bursts/h) and middle-aged rats with normal (3.4 ± 0.5 bursts/h) or attenuated (3.4 ± 0.7 bursts/h) LH surges. Furthermore, no differences in the time interval between bursts, the calculated LH secretory burst half-duration (the duration of the secretory burst at the half-maximal release rate), the basal secretion rate, or the LH half-life were seen between the young and middle-aged groups. However, middle-aged rats with attenuated LH surges had a 4-fold decrease (p < 0.01) in the estimated mass of LH secreted per burst and a 4-fold decrease (p < 0.01) in the computed maximal rate of LH secretion per burst (amplitude) in comparison to young rats and middle-aged rats with LH surges of normal magnitude. The decrease in the mass of LH secreted per burst resulted in a 4-fold decrease (p < 0.01) in the total amount of LH released per surge in the middle-aged rats with attenuated LH surges.
|
Figure 2 depicts the ApEn estimates for the young female rats (1.272 ± 0.123) and for middle-aged females with normal (1.202 ± 0.139) and attenuated (1.327 ± 0.142) LH surges. As can be appreciated, there were no significant differences by ANOVA in the mean ApEn (1, 20%) values.
|
| DISCUSSION |
|---|
|
|
|---|
Neuroendocrine studies of middle-aged female rats suggest that the pituitary gland becomes less responsive to GnRH with age. Pituitary release of LH in response to exogenous GnRH or agonists is less in middle-aged cyclic rats than in young rats both in vivo and in vitro [13, 2527]. Naloxone, an opioid-receptor antagonist that stimulates GnRH release, also elicits a greater release of LH from young than from middle-aged females [28, 29], although Wise et al. [12] found no difference in LH concentration or pulse frequency between young and middle-aged ovariectomized rats given naloxone. Some of the decrease in pituitary responsiveness to GnRH may be the result of reduced pituitary GnRH receptors: middle-aged rats have fewer GnRH receptors during diestrus, and they show an attenuated increase in receptors in response to electrical stimulation of the medial preoptic area compared to young rats [30]. In addition, it is entirely possible that factors that are not intrinsic to the GnRH-LH secretory pathway, but that could modulate this pathway, could be responsible for diminished LH secretion in response to GnRH.
The observation that LH secretory burst duration is not compromised in middle-aged rats with attenuated LH surges suggests that the amount of LH available for secretion in response to an appropriate GnRH signal may not necessarily be reduced in these animals. We did not measure pituitary content of LH in the current study. However, even when this measurement has been made, it is not known whether normal LH content implies that a normal amount of LH is available for release. Moreover, studies in which the aim has been to document whether pituitary LH production and content are altered as a function of age are not entirely consistent. Thus, although some reports suggest that pituitary content and basal release of LH are the same in young and middle-aged rats [31, 32], middle-aged rats with attenuated LH surges show a lower expression of LHß mRNA during proestrus [31]. Further studies are clearly required to clarify this issue.
The decrease in LH burst mass observed in the present study could also be the result of a change in the GnRH signal reaching the pituitary gonadotrophs. It is not clear whether there is an alteration in hypothalamic content of GnRH with age. Young and middle-aged rats have a similar number of GnRH neurons as identified by immunoreactivity on proestrus [33]. Basal release of GnRH from hypothalamic fragments may be higher in cyclic middle-aged rats [32], or the same as in young rats [34]. Stimulated release of GnRH may be lower in middle-aged rats [34] or similar [32] between the two ages. A comparison of young vs. middle-aged rats over the course of proestrus does reveal age-related differences. In young rats, the content of GnRH in the median eminence decreases [35], the number of GnRH-immunopositive cells increases [36], and the colocalization of c-fos and Jun in GnRH cells increases [33, 37] throughout the day of proestrus. In middle-aged rats, these changes are attenuated or nonexistent.
There is some evidence that the control of GnRH release may also be affected by age. Periventricular preoptic area neurons that influence GnRH release contain less Fos in middle-aged cyclic rats than in young rats [38]. Young ovariectomized rats respond to estradiol with a longer period of elevated c-fos expression in GnRH neurons than middle-aged rats [39]. Ovariectomized middle-aged rats given estradiol also lose the diurnal pattern of hypothalamic norepinephrine turnover [25],
1-adrenergic receptor number [40], and proopiomelanocortin mRNA content [41] seen in young rats treated in the same manner. Cyclic middle-aged rats have a higher basal release of norepinephrine and higher basal activity of tyrosine beta-hydroxylase, the rate-limiting enzyme of catecholamine production, in the medial preoptic area during proestrus. The amplitude and timing of the proestrous norepinephrine peak are more variable among middle-aged rats than young ones [42, 43], and the timing of GnRH pulses in middle-aged ovariectomized rats is altered: LH pulse frequency (which typically reflects the frequency of GnRH pulsatile release) begins to decrease in cyclic middle-aged rats as compared to young ones [13].
The reproductive status of an individual cannot be determined by chronological age alone. Clearly, alterations in the hypothalamic-pituitary-gonadal axis can occur before signs of changes in the estrous cycle. The data presented here demonstrate that the attenuation of the proestrous LH surge occurs in some, but not all, middle-aged cyclic rats. Approximately half of the 10- to 12-mo-old rats had LH surges of normal magnitude that did not differ from the LH surges of young rats in any of the parameters measured. The other half of the middle-aged rats exhibited attenuated LH surges, which differed in the mass and rate of LH secretion (Table 1). Thus, one must use caution in interpreting data collected from heterogeneous middle-aged cyclic rats, as the group may be made up of individuals that vary in their reproductive age [30, 44]. It is important to indicate that in the present studies, "normal cyclicity" was determined by observing the expected changes in the vaginal epithelium. It is not certain whether such changes correlate with ovulation, particularly in females with profoundly attenuated LH surges. It is possible that the threshold for ovulation may be greater than that necessary to induce changes in the vaginal epithelium, and thus some normally cycling rats with attenuated LH surges may indeed be anovulatory.
In summary, we have demonstrated that the attenuated proestrous LH surges observed in middle-aged cyclic rats result from a decrease in the amount of LH secreted per LH burst, which in turn reflects a decrease in the rate of LH secretion within each burst. No alterations in the frequency of LH secretory bursts, in burst duration, in the estimated LH half-life, or in the orderliness of LH secretion were documented. Taken as a group, these observations are consistent with the notion that the attenuation of the proestrous LH surge in middle-aged rats reflects a decrease in the mass of GnRH secreted and/or a decrease in pituitary responsiveness to GnRH. Further studies are necessary to clarify the potential mechanisms as well as to elucidate further whether the amount of LH available for secretion is normal or reduced in gonadotrophs from middle-aged rats.
| FOOTNOTES |
|---|
2 Correspondence: Dennis W. Matt, Department of Obstetrics and Gynecology, Medical College of Virginia, Box 34, Richmond, VA 23298. FAX: 804 828 0573; matt{at}hsc.vcu.edu ![]()
Accepted: August 6, 1998.
Received: January 30, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. R. Chakraborty and A. C. Gore Aging-Related Changes in Ovarian Hormones, Their Receptors, and Neuroendocrine Function Experimental Biology and Medicine, November 1, 2004; 229(10): 977 - 987. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |