- Email: [email protected]
Anterior pituitary luteinizing hormone secretion during continuous perifusion in aging male rats
Mechanisms of Ageing and Development, 25 (1984) 103--115
103
Elsevier ScientificPublishersIreland Ltd.
ANTERIOR PITUITARY LUTEINIZING HORMONE SECRETION DURING CONTINUOUS PERIFUSION IN AGING MALE RATS
LAWRENCE W. K A L E R and VAUGHN CRITCHLOW Division of Reproductive Biology and Behavior, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue, Beaverton, Oregon 97006 (U.S.A.)
(Received April 26th, 1983) (Revision received October 3rd, 1983) SUMMARY The possibility that the gonadotrope population of the anterior pituitary experiences an age-related alteration in function was investigated by in vivo and in vitro methods in young (4- to 6-month) and old (18- to 20-month) male Sprague-Dawley rats. Plasma concentrations of testosterone were 50% lower in the old rats, but resting concentrations of luteinizing hormone (LH) were similar in the two age groups. After leg-restraint and blood-withdrawal stress, plasma LH levels were significantly elevated in both young and old males; however, LH levels achieved by aged males were 39% less than those achieved by young males. During perifusion of anterior pituitary, release of LH (ng/ml per 10 min) was stable for 7 h; old anterior pituitary released only 52% as much LH as young anterior pituitaries. The anterior pituitary LH content after perifusion was not altered with age. Castration 2 weeks prior to perifusion caused elevations in plasma LH and in LH released from anterior pituitary during perifusion that were similar in the two age groups. Implantation of testosterone-filled Silastic capsules 2 weeks prior to perifusion elevated plasma testosterone and reduced plasma LH levels in both age groups. The in vitro release of LH from anterior pituitaries was similarly reduced in both age groups. Administration of varying doses of LH-releasing hormone (LHRH) during perifusion caused similar releases of LH above baseline levels in young and old rats. These in vitro results show that aged male rat anterior pituitaries release less LH than anterior pituitaries from young males. However, the magnitude of the LH response of anterior pituitaries to L H R H is not reduced with aging. These findings suggest that the decline in androgen status in old rats is not attributable to a deficit in pituitary responsiveness to LHRH. The effects of manipulating testosterone levels failed to implicate a change in anterior pituitary sensitivity to feedback as a cause for the hormonal status of aged male rats. K e y words: Pituitary luteinizing hormone; Perifusion; Aging male rat
0047-6374/84/$3.00 Printed and Publishedin Ireland
© 1984Elsevier ScientificPublishers Ireland Ltd.
104 INTRODUCTION Available evidence indicates that a decrease in plasma luteinizing hormone (LH) levels is an important factor in the decline of circulating testosterone (T) concentrations that occurs in aging rats [1-14]. Whether this reduction in androgen status stems from changes in the function of pituitary gonadotropes, the hypothalamus, or both is unclear. That functions of pituitary gonadotropes are reduced during advancing age is suggested by the finding that aged male rats are less responsive than young males to a single injection of LH-releasing hormone (LHRH) [11]. Similarly, Bruni et al. [12] observed smaller increases in serum LH in aged than in young male rats after multiple injections of LHRH, and suggested that there is a smaller releasable gonadotropin pool in aged pituitaries. Hypothalamic function may change with age in rats, because LH-releasing activities [15] and L H R H immunocytochemical fiber staining [16] are reduced in the hypothalamus of aged male rats. To obviate the many problems inherent in studying age-related changes in pituitary function in vivo, Riegle et al. [15] used an in vitro approach. They observed an age-associated diminution of LH secretion after administration of L H R H during static incubations of rat pituitaries. Our objective was to use perifusion of the anterior pituitary (AP) to study further the effects of age on LH release in vitro. Perifusion precludes interference by accumulating secretory products and permits investigation of the dynamics of secretory responses. MA'I~ERIALSAND METHODS Male Sprague-Dawley rats (Charles River) were received 4 a n d 18 months after birth. During the 2-month period of study, the rats were housed under conditions of controlled light (0500-1900 h) and temperature (25 - 2°C). Rat chow and water were available at all times. The rats were examined daily, and those with respiratory or other obvious health problems were removed from the study. Body weights (shown in Table I) were obtained 2 weeks prior to, and the rats were housed individually 1 week prior to, necropsy. Five young and 5 old male rats were assigned to each of five treatment groups: intact control, sham-implanted, sham-castrated, T-implanted (intact), and castrated. Treatments were assigned and performed according to a completely randomized design. Two weeks prior to death, all rats were anesthetized with ether, and the designated treatments were performed. To determine the effects of a 2-week exposure to high circulating T levels in vivo on AP function during perifusion, two 30-mm T-filled Silastic capsules were implanted subcutaneously in young males (one capsule per 200 g of body weight); old males received three such capsules. The T implants were prepared according to the methods of Legan et al. [17].
105
Sham-implanted males received empty Silastic capsules. The effects of long-term low circulating T levels in vivo on AP function during perifusion were assessed after castrating both young and old males 2 weeks before the in vitro experiments. Epididymal fat, instead of the testes, was removed from sham-castrated controls.
Effects o[ aging on plasma testosterone and luteinizing hormone responses to stress Two weeks after the designated treatments, each rat was removed from its individual cage and transported to an adjoining room at 0900 to 0930 h. The animal room had been locked at 1500h on the previous day to minimize environmental disturbances. A nonstress blood sample was collected from a tail vein under vacuum within 60 sec of cage opening, as described previously [18]. Additional blood was collected after decapitation at 15 rain to assess the effects of leg-restraint stress on plasma T and LH concentrations. With these procedures, stress is considered to be a composite of stimuli, i.e. handling, bleeding and 15-rain leg restraint. Leg restraint involved binding the hind legs together with a pipe cleaner. It should be noted that all rats were in a stressed condition prior to removal of the APs for perifusion. Tissue preparation After decapitation, each pituitary was removed and the AP was dissected and placed in a vial containing a 1000-fold (v/w) ratio of the oxygenated perifusion medium (described below). The AP was then taken to another laboratory, where it was weighed to the nearest 0.01 mg (wet weight), sectioned into eighths, and placed in a perifusion chamber. Perifusion was started within 6 rain from the time of decapitation. Preliminary studies were undertaken to determine the optimal size of pituitary fragments for perifusion. These studies included histological examination of formalin-fixed and methacrylate-embedded sections of AP fragments of varying sizes that were perifused for up to 8 h. In addition, various sized AP fragments were studied in vitro to examine baseline LH release and LH secretion in response to L H R H or 50mM potassium. The results of these preliminary experiments indicated that sectioning the AP into eighths is optimal for perifusion. Perifusion system Each AP was placed into a separate polyethylene incubation chamber of 0.1 ml total volume (prepared from Microeppendorf conicalcentrifuge tubes). The inlet port of each chamber was connected with polyethylene tubing to an infusion medium reservoir, and the outlet port was drained by polyethylene tubing through an infusion pump (Sage four-channel) and into a fraction collector. Rat L H R H (provided by the NIADDK) was administered to each incubation chamber through side channels without interruption of the flow of incubation medium to the tissue chambers.
106 Incubation medium reservoir and individual perifusion chambers were submerged in a constant-temperature water bath at 37°C. Effluent fraction collector tubes were chilled on ice and subsequently stored at -20°C until the contents were assayed. The APs were perifused with Medium-199 (Grand Island Biological) that was modified to contain 0.15% bovine serum albumin (radioimmunoassay [RIA] grade), 250mg/l ampicillin, 0.01 mM bacitracin, and 0.1 mM ascorbate. The pH was adjusted to 7.35 after filtration through a 0.45-/zm Millipore filter. Medium was stored in sterile containers prior to and during continuous perifusion. Bovine serum albumin prevents significant nonspecific adsorption of hormone to polyethylene tubing [19]. Filtration through a Millipore filter, plus the addition of ampicillin to the infusion medium, reduced bacterial contamination below histologically detectable levels during prolonged perifusion periods. Bacitracin was added to inhibit protease degradation of peptide hormone [20]. Finally, 0.1 mM ascorbate was added to the medium to make it similar to that which we are using in concurrent studies that involve the in-series perifusion of hypothalamic and pituitary tissues; ascorbate inhibits dopamine degradation [21]. The medium was prewarmed to 37°C prior to initiation of perifusion and saturated with 02 and CO2 by constant bubbling with a 95% 02-5% CO2 gas mixture. Measurements with a blood gas/pH monitor verified that the 02 content of the perifusion medium before and after passage through the incubation chambers remained at saturation levels and that CO2 production did not increase measurably. The pH of the perifusion medium remained stable after passage through the incubation chambers. The flow rate for perifusion was 0.1 ml/min, and effluent fractions were collected at 10-min intervals. The total clearance time at this flow rate, i.e. the time for a known volume of 12SI-labeled LH to be withdrawn from a side-channel reservoir and passed through the perifusion system into collection tubes, was approximately 40min; more than 37% of the total amount of radiolabeled hormone appeared in the effluent within 20 min, 86% within 30 min, and 95% within 40 min.
Luteinizing hormone release from perifused young and old anterior pituitaries in response to L H R H The APs obtained from each of the treatment groups were used to evaluate the effects of aging on baseline release of LH and LH release in response to varying doses of LHRH. After a 1-h equilibration poriod, i.e. the amount of time needed for LH levels in effluent fractions to reach a near-constant rate of release, the APs were challenged with pulses of increasing L H R H concentrations. These pulses were given in 1.0 ml for 10 rain at concentrations that ranged from 10 -9 M to 10-s M, at hourly intervals. Preliminary studies (unpublished data) indicated that APs from young males respond to L H R H of varying concentrations by releasing LH above baseline levels, and the responses are related to the concentration of
107 L H R H administered, regardless of the sequence in which different concentrations of L H R H are given. The concentration of LH in each 1 ml of effluent fraction was determined with RIA kits obtained from Dr. A. Parlow through the NIADDK. Assays were performed on four dilutions of either unextracted blood plasma or perifusion effluents. The LH concentrations in unknown samples were expressed in terms of rat LH-RP-1. The sensitivity of the rat LH RIA was 1.9 ng/ml, and the intra- and interassay coefficients of variation were 5% and 4.5%, respectively. Baseline release of LH for each age group was defined as the line which best fit the ten lowest mean LH values during the perifusion period. Amounts of LH released in response to L H R H were determined by integration of the areas beneath each peak from the time of administration of L H R H until the time that LH levels returned to baseline. The T RIA was performed on duplicate aliquots of rat plasma in a single assay by Dr. David Hess of the Oregon Regional Primate Research Center as described elsewhere [22]. The sensitivity of this RIA was 0.4ng/ml and the intraassay coefficient of variation was 4%.
Statistical methods In order to assess differences in baseline LH release from the perifused APs, the differences in elevations of multiple regression lines were determined b y analysis of covariance. If significant differences were found between treatment groups, an analysis of pairwise differences of means was performed with the Student-Newman-Keuls test [23]. The independent or interactive effects of age and L H R H dose on amounts of LH released during perifusion were determined by two-tailed, two-factor analysis of variance with repeated measures [24]. Probabilities less than 0.05 were considered significant. EXPERIMENTALRESULTS Table I summarizes body and AP weights, hormonal data (plasma T and LH) obtained in vivo, and pituitary concentrations of LH for the young and old rats in the several experimental groups. Because the data for the control groups (intact, sham-T-implanted, .and sham-castrated) were similar, they were pooled. As expected, old males weighed more (P <0.0001) than young males and their pituitaries were slightly, but not significantly, heavier; neither T implants nor castration affected AP weight differences between age groups. Plasma T levels under nonstress conditions were higher (P < 0.001) in the young than in the old control males. Plasma T levels in the T-implanted rats were much higher (P < 0.0001) than in the controls and, although attempts were made to correct for differences in body weight and blood volume, the levels were higher (P < 0.025) in the old than in the young T-implanted rats. Conversely, castration produced a marked (P < 0.0001) depression in plasma T concentrations, and the levels found
108
0 0
z
+I
+1
+1 +1
+1 +1
.= +1
e.i -I-I
+1 +i
+1 +1
+1
+1
+1 +1
+1 +1
z M < < ,-1 Z ©
e
t-
t-
< < M
÷I
+1
+1 +1
÷1 +1 ¢~
('4
÷1
÷1
+1 +1
41 +1
¢xl
~
~
ee
+1
+1
"H +l
+1 +1
~
c,,I
V
~V
zI
0
<
ze_
,-1 <
"H
+1
+1 t-I
+1 +1
~.-~ v~ ~ v
109
in the young and old castrated groups were similar. Circulating nonstress levels of LH tended to be higher in the young than in the old intact males, although this difference was not significant. The young intact males showed an approximate 100% increase (P < 0.005) in plasma LH concentrations in response to stress; the old intact males also showed a response to stress (P < 0.025), but it was smaller ( P < 0 . 0 2 5 ) than that in the young rats. Nonstress plasma LH concentrations were similar in the young and old T-implanted and castrated groups, and significant L H responses to stress were not obtained and either the young or the old rats after these manipulations. Pituitary concentrations of LH were similar in glands from young and old intact rats. As expected, castration elevated (P < 0.05) pituitary concentrations of LH in both young and old rats, but T implants did not significantly suppress LH concentrations in either age group. These results were the same when L H was expressed as/zg per AP. Data on the release of L H from the APs of young and old intact males during perifusion are presented in Fig. 1. Comparison of the slopes and elevations of the regression lines indicated that baseline LH release was higher (P < 0.0008) from APs of young rats than that from APs of old rats. Anterior pituitaries from both young and old intact males showed dose-related integrated LH responses (P < 0.025) to increasing concentrations of LHRH, and there was no effect of age on these responses. In addition to the effects noted in vivo (Table I), castration caused elevations (P <0.0008) in baseline levels of LH released during perifusion in both age groups, and there was no effect of age on the amplitude of these elevations (Fig. 2). Although APs of both young and old castrated males released large amounts of LH in response to L H R H , the responses were not dose-dependent. Unlike the LH .
.
.
.
.
.
.
Aged Intact Controls Young Intact Controls
.
--
2'01 m
"~ 1'Ok
t ~ m
~ ~
~
L.=:b~..~_-L~J ,~_r '-I j
-J
/
0
f~L_
-I_1
10-1M
I~ 120
10-8M
I~ 180
f~
I0-7M
I~ 240
10-6M LHRH
I~l 300
I 360
Time (rain) Fig. 1. Mean luteinizing hormone (LH) concentrations (/zg/mi) released from anterior pituitaries (APs) of young and aged intact males during continuous perifusion. Baseline LH concentrations per group were defined as the line that best fit the ten lowest mean LH values during the perifusion period. Amounts of LH released in response to 10-min pulses of increasing concentrations of LH-releasing hormone (LHRH) were determined by integration of the areas beneath each peak. Baseline release of LH was higher (P < 0.0008) from APs of young rats than from those of old rats.
110
A
B
4.0 --
4.0 -
......
Young C a s t r a t e Young Sham C a s t r a t e
.....
3.0 -
Old C a s t r a t e Sham C a s t r a t e
3.0
E ,,,~ 2.0
2.0-
'
"1" .-I
,.o
,o
[
r-
L r~ !i YT '-, ~: L~. r ~ :__~r ~-'l. -
0
~
iO-IM I~ 120
_
_
I
.
tO-OM ~ 180
~
~
tO-TM 1~ 240
10-$M LHRH ~ I 300 360
I~ 120
~ 180
~ 240
I~ 300
I 360
Time (rain)
Fig. 2. Mean hteinizinghormone (LH) concentrations(/zgJml)released from anterior pituitaries of youngand aged castratedmalesduringcontinuousperifusion.Mean LH concentrationsreleased from APs of young and aged sham-castrated controls are shown for comparison. Castration caused elevations (P < 0.0008) in baselineLH levels released in both age groups.
release rates observed in intact controls (see Fig. 1), baseline rates were similar from APs of young and old sham-castrated rats. Figure 3 shows the effects of implanting T-filled Silastic tubing 2 weeks prior to removal of APs for perifusion. Baseline LH release from glands of young rats was higher (P < 0.001) than that from old rats after implantation with the control, unfilled Silastic tubing. The high circulating T levels that occurred in vivo after T implantation suppressed (P < 0.0008) the in vitro release of LH from APs of both young and old T-implanted males, and the baseline LH levels produced by the young and old APs during perifusion were similar. Comparison of the data from T- and sham-implanted rats indicated that high circulating T levels in vivo reduced (P < 0.025) young and old AP responses to the several doses of L H R H in vitro. DISCUSSION
Despite the accumulated evidence that the decline in circulating T concentration in aging male rodents is in part attributable to extratesticular events [5,7,8,14,25-27], surprisingly few investigations have been carried out to determine whether this age-related change is reflected in AP function in vitro. In one such study, AP cell cultures derived from aging male mice released more LH into
111
A
B
3"5 I [ ] 1"1 I I t I
3.0
--Young T-Implant ........ Young Sham Implant
"TTT
1.~
t.5,
'!i i=
il i L~= I'
E 1.0
ri ' i
'
~L
II I:
',
r-J
,
r-~
LJ
i
~L I 17.
L~
2; ..J
L_
r~ I F l
/ I
- Old T-Implant ........ Old Sham Implant
] (
!,
1.0
L.-.
~J ..-----~
Fi =,
.~ ~ ' J ] " ~ ' ~
L,I
..
r'l
1 I
I •
i,
, L~
,
= l ~
LZ'~
0.5 t___a
10-I)M
120
10"8M
~ 180
tO'?M
~ 240
tO-|M LHflH
r~ 300
I 360
0
L.,_J
I@-tM
tO-SM
io-TM
10-11MLHRH
I~ 120
I~ 180
I~ 240
I~1 300
I 360
Time (rain)
Fig. 3. Mean luteinizing hormone (LH) concentrations ~g/ml) released from anterior pituitaries of young and aged testosterone-implanted males during continuous perifusion. Mean LH concentrations released from APs of young and aged sham-implanted controls are shown for comparison. Testosterone implants suppressed the release of LH (P < 0.001) and the responses to the several doses of LH-releasing hormone (LHRH) (P < 0.025) in both age groups. the culture medium than did cells obtained from young males [25], and the investigators concluded that L H secretion is not compromised in aging male mice. In contrast, Riegle et al. [15] observed that less L H was released from aged than from young rat hemipituitaries during static in vitro incubation. In our study, a perifusion approach also demonstrated that APs from aged male rats release less L H per unit of time under baseline conditions than do pituitaries from young males. The amounts released by young APs during our perifusion study are consistent with those previously reported by others [19,28,29]. Our results are consistent with those reported previously [14] in demonstrating that plasma L H levels are similar in intact nonstressed aged and young male rats. Likewise, the attenuation of the stress-induced elevation of plasma L H that we observed in intact aging male Sprague-Dawley rats is similar to the findings in Long-Evans rats [11]. Although the basis and significance of these findings are uncertain, they could reflect an age-associated decline in the amount of L H R H released in response to stress. Despite the absence of an in vivo age-related difference in circulating L H concentrations under nonstress conditions, a difference in baseline L H release was conspicuous during perifusion; APs from intact aged males released only 48% (P < 0.025) of that released by the glands from young intact rats. The absence of an age-related difference in circulating L H
112 in vivo in the presence of a marked age difference in the amount of LH released in vitro is unexplained but may reflect the effect of variations in metabolic clearance rates in vivo or the effects of complete removal of hypothalamic influences on pituitary function during perifusion. The finding that the in vitro
secretion rate of LH under baseline conditions is reduced in pituitaries from old rats supports the concept that the decline in androgen status experienced by aging male rats is the result of long-term understimulation of relatively normal testes by LH [5,14,261. The possibility exists that APs of aged male rats release less LH because of a reduced sensitivity to LHRH. The finding of Riegle et al. [15] that incubated hemipituitaries from aged male Long-Evans rats release less LH in response to L H R H than d/0 those from young rats is consistent with this possibility. However, our study, involving perifusion methods that obviated feedback effects on the AP, demonstrated that old and young rat APs release equivalent amounts of LH in response to L H R H stimulation. This discrepancy in results may be due to one or more methodological differences. For example, Riegle et ai. [15] used a static in vitro system, with the potential for complications due to the effects of feedback, hemipituitaries and a 4-h exposure to LHRH. The latter could have produced AP desensitization to L H R H [30]. In contrast, our perifusion approach probably precluded feedback effects, utilized smaller fragments of pituitary tissue and involved a series of 10-rain exposures to LHRH. Our results suggest that the age-related decline in pituitary release of LH is not due to reduced sensitivity of the AP to LHRH. The LH content in APs of the young intact males in our study is in agreement with that found in Long-Evans rats by Riegle et al. [15]. Unlike these investigators, however, we observed only a slight and nonsignificant loss of LH content in APs of old intact Sprague-Dawley males. The basis for this discrepancy in results is unknown, but it may reflect a strain difference. That APs from old intact males released 48% less LH per unit of time during perifusion with no detectable change in AP LH content implies a reduction in the rate of pituitary synthesis of LH. Such a reduction might reflect an intrinsic decrease in pituitary biosynthetic capacity or a decrease in stimulation by the hypothalamus. The ability of APs from aged castrated males to release high baseline levels of LH in vitro and the similarity of the responses to L H R H by the APs from the two age groups favor the latter alternative. Although castration produced changes in LH release from old and young APs that were similar, the baseline LH levels declined during perifusion of the old APs. This was the only decline noted in this series of experiments, and the basis for it is unknown. Because AP LH content of aged castrated males was less (P < 0.025) than that of young castrated males, a deficit in physiological reserve cannot be discounted. However, if this were the case, the potential deficit did not limit the LH responses to LHRH. Testosterone and its metabolites exert negative feedback effects on circulating levels of LH, and recent findings suggest that old men are more sensitive than
113 young men to such feedback [31]. Likewise, middle-aged rats show a greater and more prolonged suppression of L H than do young rats when exposed to the same circulating T levels [32]. We did not detect such an age-dependent difference in sensitivity to the negative feedback effects of T, perhaps owing to the supraphysiologic levels of T that were used. Plasma T levels in the T-implanted males were approximately eight times higher than normal; these high levels probably maximally inhibited L H release in both age groups. The administration of T in vivo for 2 weeks before perifusion clearly inhibited the release of L H from the isolated APs of both young and old rats in vitro, an indication that this feedback effect is due, at least in part, to changes in the AP that persist in the absence of continued exposure to circulating T. Treatment with T in vivo also reduced the magnitude of the AP responses to L H R H , and the reductions were similar in the APs of young and old males. The similarity of the responses to L H R H in the young and old T-treated groups is consistent with the similarity of responses obtained with APs of intact and castrated rats in suggesting that the reduced L H release in aged males does not involve pituitary hyporesponsiveness to LHRH. In summary, these results support the concept that the decline in plasma T levels experienced by aging male rats reflects in part the long-term understimulation of relatively normal testes due to a reduction in LH release. Because LH responses to L H R H were consistently similar for APs from young and old rats, it is likely that the onus of declining androgen status in old rats rests with functional deficits at the level of the hypothalamus. This conclusion is consistent with the reported reductions of hypothalamic L H R H activity [15] and immunocytochemically stained L H R H fibers [16] in old male rats. ACKNOWLEDGEMENTS Special thanks are due to Dr. Robert Mioduszweski for assistance in the development of the perifusion system. We thank Jessie Kroning and Paul Ballinger for their excellent technical assistance. The materials used in the protein hormone assays were kindly provided by the N I A D D K through Dr. A. Parlow. The work described in this article, Publication No. 1295 of the Oregon Regional Primate Research Center, was supported in part by the Animal Resources Branch Grant RR-00163 and by a National Research Service Award AG05220 (LWK), the Medical Research Foundation of Oregon Grant 8130--882 (LWK), and the National Institutes of Health Research Grant AM32442 (VC). REFERENCES 1 R. Ghanadian, J.G. Lewis and G.D. Chisholm, Serum testosterone and dihydrotestosterone changes with age in rats. Steroids, 25 (1975)753-762. 2 A.E. Miller and G.D. Riegle, Aging effects on hypothalamic catecholamine and testosterone secretion in the male rat. Fed. Proc., 34 (1975)303. 3 S.W.C. Chan, J.H. Leathernand T. Esashi, Testieularmetabolism and serum testosteronein aging male rats. Endocrinology, 101 (1977) 128-133.
114 4 G.D. Gray, Changes in the levels of luteinizing hormone and testosterone in the circulation of ageing male rats. ,L Endocrinol., 76 (1978) 551-552. 5 S.M. Harman, R.L. Danner and G.S. Roth, Testosterone secretion in the rat in response to chorionic gonadotropir): alterations with age. Endocrinology, 102 (1978) 540-544. 6 A.E. Miller and G.D. Riegle, Serum testosterone and testicular response to HCG in young and aged male rats. J. Gerontol., 33 (1978) 197-203. 7 K.M. Pirke, M. Geiss and H.J. Vogt, In vitro and in vivo studies on Leydig cell function in old rats. Acta Endocrinol. Suppl., 215 (1978) 37 (Abstract). 8 K.M. Pirke, H.J. Vogt and M. Geiss, In vitro and in vivo studies on Leydig cell function in old rats. Acta Endocrinol., 89 (1978) 393--403. 9 C.L. Bethea and R.F. Walker, Age-related changes in reproductive hormones and in Leydig cell respomivity in the male Fischer 344 rat. J. Gerontol., 34 (1979) 21-27. 10 C.J. Shaar, J.S. Euker, G.D. Riegle and J. Meites, Effects of castration and gonadal steroids on serum luteinizing hormone and prolactin in old and young rats. Y. Endocrinol., 66 (1975) 45-51. 11 G.D. Riegle and J. Meites, Effects of aging on LH and prolactin after LHRH, L-dopa, methyldopa, and stress in male rats. Proc. Soc. Exp. Biol. Med., 151 (1976) 507-511. 12 J.F. Bruni, H. Huang, S. Marshall and J. Meites, Effects of single and multiple injections of synthetic CmRH on serum LH, FSH, and testosterone in young and old male rats. Biol. Reprod., 17 (1977) 309-312. 13 P.D. Tsitouras, M.A. Kowatch and S.M. Harman, Age-related alterations of isolated rat Leydig cell function: gonadotropin receptors, adenosine 3',5'-monophosphate response, and testosterone secretion. Endocrinology, 105 (1979) 1400-1405. 14 L.W. Kaler and W.B. Neaves, The androgen status of aging male rats. Endocrinology, 108 (1981) 712-719. 15 G.D. Riegle, J. Meites, A.E. Miller and S.M. Wood, Effect of aging on hypothalamic LH-releasing and prolactin inhibiting activities and pituitary responsiveness to LHRH in the male laboratory rat. J. Gerontol., 32 (1977) 13-18. 16 G.E. Hoffman and J.R. Sladek, Age-related changes in dopamine, LHRH and somatostatin in the rat hypothalamus. Neurobiol. Aging, I (1980) 27-37. 17 S.J. Legan, G.A. Coon and F.J. Karsch, Role of estrogen as initiator of daily LH surges in the ovariectomized rat. Endocrinology, 96 (1975) 50-56. 18 R. Mioduszewski and V. Critchlow, Marked and selective suppression of prolactin secretion after interruption of neural connections of the medial basal hypothalamus of the rat. Brain Res., 216 (1981) 35-44. 19 A.J. Dowd, A.L. Barofsky, N. Chaudhuri, C.W. Lloyd and J. Weisz, Patterns of LH and FSH release from perifused rat pituitaries in response to infusions of hypothalamic extract. Endocrinology, 96 (1975) 243-252. 20 K. Maeda and L.A. Frohman, Release of somatostatin and thyrotropin-releasing hormone from rat hypothalamic fragments/n vitro. Endocrinology, 106 (1980) 1837-1842. 21 A. Negro-Vilar, S.R. Ojeda and S.M. McCann, Catecholaminergic modulation of luteinizing hormone-releasing hormone release by median eminence terminals in vitro. Endocrinology, 104 (1979) 1749-1757. 22 J.A. Resko, A. Malley, D. Begley and D.L. Hess, Radioimmunoassay of testosterone during fetal development of the rhesus monkey. Endocdno/ogy, 93 (1973) 156-161. 23 J.H. Zar, BiostatJstical Analysis, Prentice-Hall, Englewood Cliffs, New Jersey, 1974. 24 B.J. Winer, Stat~tical Principles in Exper~ental Design, 2nd edn., McGraw-HiE, New York, 1971. 25 C.E. Finch, V. Jones, J.R. Wisher, Y.N. Sinha, J.S. de Vellis and R.S. Swerdloff, Hormone production by the pituitary and testes of male C57BL/6J mice during aging. Endocrinology, I01 (1977) 1310-1317. 26 L.W. Kaler and W.B. Neaves, The steroidogenic capacity of the aging rat testis. J. Gerontol., 36
(1981) 398-404. 27 G. Schafer, A.F. Holstein and H. Hilz, Steroidogenic capacity of isolated adult mouse Leydig cells does not decrease with age. Endocrinology, 110 (1982) 1362-1366. 28 M.J. Kilpatrick, W.P. Collins and J.R. Newton, Studies on the release of gonadotrophins during the superfusion of isolated rat pituitaries in a continuous flow system. Y. Reprod. Fer61., 46 (1976) 25-30.
115 29 M..l. Kilpatrick, J.R. Newton and W.P. Collins, Studies on factors affecting release of gonadotrophins during the superfusion of isolated rat pituitaries. J. Reprod. Fertil., 49 (1977) 347-351. 30 M.A. Smith, M.H. Pert'in and W.W. Vale, Desensitization of cultured pituitary cells to gonadotropin-releasing hormone: evidence for a post-receptor mechanism. Moi. cell. Endocrinol., 30 (1983) 85-96. 31 S.J. Winters and P. Troen, The gonadotropin-suppressive potency of androgen is increased in elderly men. Endocrinology, (1982) 222 (Abstract No. 572). 32 G.D. Gray, E.R. Smith and J.M. Davidson, Oonadotropin regulation in middle-aged male rats. Endocrinology, 107 (1980) 2021-2026.
103
Elsevier ScientificPublishersIreland Ltd.
ANTERIOR PITUITARY LUTEINIZING HORMONE SECRETION DURING CONTINUOUS PERIFUSION IN AGING MALE RATS
LAWRENCE W. K A L E R and VAUGHN CRITCHLOW Division of Reproductive Biology and Behavior, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue, Beaverton, Oregon 97006 (U.S.A.)
(Received April 26th, 1983) (Revision received October 3rd, 1983) SUMMARY The possibility that the gonadotrope population of the anterior pituitary experiences an age-related alteration in function was investigated by in vivo and in vitro methods in young (4- to 6-month) and old (18- to 20-month) male Sprague-Dawley rats. Plasma concentrations of testosterone were 50% lower in the old rats, but resting concentrations of luteinizing hormone (LH) were similar in the two age groups. After leg-restraint and blood-withdrawal stress, plasma LH levels were significantly elevated in both young and old males; however, LH levels achieved by aged males were 39% less than those achieved by young males. During perifusion of anterior pituitary, release of LH (ng/ml per 10 min) was stable for 7 h; old anterior pituitary released only 52% as much LH as young anterior pituitaries. The anterior pituitary LH content after perifusion was not altered with age. Castration 2 weeks prior to perifusion caused elevations in plasma LH and in LH released from anterior pituitary during perifusion that were similar in the two age groups. Implantation of testosterone-filled Silastic capsules 2 weeks prior to perifusion elevated plasma testosterone and reduced plasma LH levels in both age groups. The in vitro release of LH from anterior pituitaries was similarly reduced in both age groups. Administration of varying doses of LH-releasing hormone (LHRH) during perifusion caused similar releases of LH above baseline levels in young and old rats. These in vitro results show that aged male rat anterior pituitaries release less LH than anterior pituitaries from young males. However, the magnitude of the LH response of anterior pituitaries to L H R H is not reduced with aging. These findings suggest that the decline in androgen status in old rats is not attributable to a deficit in pituitary responsiveness to LHRH. The effects of manipulating testosterone levels failed to implicate a change in anterior pituitary sensitivity to feedback as a cause for the hormonal status of aged male rats. K e y words: Pituitary luteinizing hormone; Perifusion; Aging male rat
0047-6374/84/$3.00 Printed and Publishedin Ireland
© 1984Elsevier ScientificPublishers Ireland Ltd.
104 INTRODUCTION Available evidence indicates that a decrease in plasma luteinizing hormone (LH) levels is an important factor in the decline of circulating testosterone (T) concentrations that occurs in aging rats [1-14]. Whether this reduction in androgen status stems from changes in the function of pituitary gonadotropes, the hypothalamus, or both is unclear. That functions of pituitary gonadotropes are reduced during advancing age is suggested by the finding that aged male rats are less responsive than young males to a single injection of LH-releasing hormone (LHRH) [11]. Similarly, Bruni et al. [12] observed smaller increases in serum LH in aged than in young male rats after multiple injections of LHRH, and suggested that there is a smaller releasable gonadotropin pool in aged pituitaries. Hypothalamic function may change with age in rats, because LH-releasing activities [15] and L H R H immunocytochemical fiber staining [16] are reduced in the hypothalamus of aged male rats. To obviate the many problems inherent in studying age-related changes in pituitary function in vivo, Riegle et al. [15] used an in vitro approach. They observed an age-associated diminution of LH secretion after administration of L H R H during static incubations of rat pituitaries. Our objective was to use perifusion of the anterior pituitary (AP) to study further the effects of age on LH release in vitro. Perifusion precludes interference by accumulating secretory products and permits investigation of the dynamics of secretory responses. MA'I~ERIALSAND METHODS Male Sprague-Dawley rats (Charles River) were received 4 a n d 18 months after birth. During the 2-month period of study, the rats were housed under conditions of controlled light (0500-1900 h) and temperature (25 - 2°C). Rat chow and water were available at all times. The rats were examined daily, and those with respiratory or other obvious health problems were removed from the study. Body weights (shown in Table I) were obtained 2 weeks prior to, and the rats were housed individually 1 week prior to, necropsy. Five young and 5 old male rats were assigned to each of five treatment groups: intact control, sham-implanted, sham-castrated, T-implanted (intact), and castrated. Treatments were assigned and performed according to a completely randomized design. Two weeks prior to death, all rats were anesthetized with ether, and the designated treatments were performed. To determine the effects of a 2-week exposure to high circulating T levels in vivo on AP function during perifusion, two 30-mm T-filled Silastic capsules were implanted subcutaneously in young males (one capsule per 200 g of body weight); old males received three such capsules. The T implants were prepared according to the methods of Legan et al. [17].
105
Sham-implanted males received empty Silastic capsules. The effects of long-term low circulating T levels in vivo on AP function during perifusion were assessed after castrating both young and old males 2 weeks before the in vitro experiments. Epididymal fat, instead of the testes, was removed from sham-castrated controls.
Effects o[ aging on plasma testosterone and luteinizing hormone responses to stress Two weeks after the designated treatments, each rat was removed from its individual cage and transported to an adjoining room at 0900 to 0930 h. The animal room had been locked at 1500h on the previous day to minimize environmental disturbances. A nonstress blood sample was collected from a tail vein under vacuum within 60 sec of cage opening, as described previously [18]. Additional blood was collected after decapitation at 15 rain to assess the effects of leg-restraint stress on plasma T and LH concentrations. With these procedures, stress is considered to be a composite of stimuli, i.e. handling, bleeding and 15-rain leg restraint. Leg restraint involved binding the hind legs together with a pipe cleaner. It should be noted that all rats were in a stressed condition prior to removal of the APs for perifusion. Tissue preparation After decapitation, each pituitary was removed and the AP was dissected and placed in a vial containing a 1000-fold (v/w) ratio of the oxygenated perifusion medium (described below). The AP was then taken to another laboratory, where it was weighed to the nearest 0.01 mg (wet weight), sectioned into eighths, and placed in a perifusion chamber. Perifusion was started within 6 rain from the time of decapitation. Preliminary studies were undertaken to determine the optimal size of pituitary fragments for perifusion. These studies included histological examination of formalin-fixed and methacrylate-embedded sections of AP fragments of varying sizes that were perifused for up to 8 h. In addition, various sized AP fragments were studied in vitro to examine baseline LH release and LH secretion in response to L H R H or 50mM potassium. The results of these preliminary experiments indicated that sectioning the AP into eighths is optimal for perifusion. Perifusion system Each AP was placed into a separate polyethylene incubation chamber of 0.1 ml total volume (prepared from Microeppendorf conicalcentrifuge tubes). The inlet port of each chamber was connected with polyethylene tubing to an infusion medium reservoir, and the outlet port was drained by polyethylene tubing through an infusion pump (Sage four-channel) and into a fraction collector. Rat L H R H (provided by the NIADDK) was administered to each incubation chamber through side channels without interruption of the flow of incubation medium to the tissue chambers.
106 Incubation medium reservoir and individual perifusion chambers were submerged in a constant-temperature water bath at 37°C. Effluent fraction collector tubes were chilled on ice and subsequently stored at -20°C until the contents were assayed. The APs were perifused with Medium-199 (Grand Island Biological) that was modified to contain 0.15% bovine serum albumin (radioimmunoassay [RIA] grade), 250mg/l ampicillin, 0.01 mM bacitracin, and 0.1 mM ascorbate. The pH was adjusted to 7.35 after filtration through a 0.45-/zm Millipore filter. Medium was stored in sterile containers prior to and during continuous perifusion. Bovine serum albumin prevents significant nonspecific adsorption of hormone to polyethylene tubing [19]. Filtration through a Millipore filter, plus the addition of ampicillin to the infusion medium, reduced bacterial contamination below histologically detectable levels during prolonged perifusion periods. Bacitracin was added to inhibit protease degradation of peptide hormone [20]. Finally, 0.1 mM ascorbate was added to the medium to make it similar to that which we are using in concurrent studies that involve the in-series perifusion of hypothalamic and pituitary tissues; ascorbate inhibits dopamine degradation [21]. The medium was prewarmed to 37°C prior to initiation of perifusion and saturated with 02 and CO2 by constant bubbling with a 95% 02-5% CO2 gas mixture. Measurements with a blood gas/pH monitor verified that the 02 content of the perifusion medium before and after passage through the incubation chambers remained at saturation levels and that CO2 production did not increase measurably. The pH of the perifusion medium remained stable after passage through the incubation chambers. The flow rate for perifusion was 0.1 ml/min, and effluent fractions were collected at 10-min intervals. The total clearance time at this flow rate, i.e. the time for a known volume of 12SI-labeled LH to be withdrawn from a side-channel reservoir and passed through the perifusion system into collection tubes, was approximately 40min; more than 37% of the total amount of radiolabeled hormone appeared in the effluent within 20 min, 86% within 30 min, and 95% within 40 min.
Luteinizing hormone release from perifused young and old anterior pituitaries in response to L H R H The APs obtained from each of the treatment groups were used to evaluate the effects of aging on baseline release of LH and LH release in response to varying doses of LHRH. After a 1-h equilibration poriod, i.e. the amount of time needed for LH levels in effluent fractions to reach a near-constant rate of release, the APs were challenged with pulses of increasing L H R H concentrations. These pulses were given in 1.0 ml for 10 rain at concentrations that ranged from 10 -9 M to 10-s M, at hourly intervals. Preliminary studies (unpublished data) indicated that APs from young males respond to L H R H of varying concentrations by releasing LH above baseline levels, and the responses are related to the concentration of
107 L H R H administered, regardless of the sequence in which different concentrations of L H R H are given. The concentration of LH in each 1 ml of effluent fraction was determined with RIA kits obtained from Dr. A. Parlow through the NIADDK. Assays were performed on four dilutions of either unextracted blood plasma or perifusion effluents. The LH concentrations in unknown samples were expressed in terms of rat LH-RP-1. The sensitivity of the rat LH RIA was 1.9 ng/ml, and the intra- and interassay coefficients of variation were 5% and 4.5%, respectively. Baseline release of LH for each age group was defined as the line which best fit the ten lowest mean LH values during the perifusion period. Amounts of LH released in response to L H R H were determined by integration of the areas beneath each peak from the time of administration of L H R H until the time that LH levels returned to baseline. The T RIA was performed on duplicate aliquots of rat plasma in a single assay by Dr. David Hess of the Oregon Regional Primate Research Center as described elsewhere [22]. The sensitivity of this RIA was 0.4ng/ml and the intraassay coefficient of variation was 4%.
Statistical methods In order to assess differences in baseline LH release from the perifused APs, the differences in elevations of multiple regression lines were determined b y analysis of covariance. If significant differences were found between treatment groups, an analysis of pairwise differences of means was performed with the Student-Newman-Keuls test [23]. The independent or interactive effects of age and L H R H dose on amounts of LH released during perifusion were determined by two-tailed, two-factor analysis of variance with repeated measures [24]. Probabilities less than 0.05 were considered significant. EXPERIMENTALRESULTS Table I summarizes body and AP weights, hormonal data (plasma T and LH) obtained in vivo, and pituitary concentrations of LH for the young and old rats in the several experimental groups. Because the data for the control groups (intact, sham-T-implanted, .and sham-castrated) were similar, they were pooled. As expected, old males weighed more (P <0.0001) than young males and their pituitaries were slightly, but not significantly, heavier; neither T implants nor castration affected AP weight differences between age groups. Plasma T levels under nonstress conditions were higher (P < 0.001) in the young than in the old control males. Plasma T levels in the T-implanted rats were much higher (P < 0.0001) than in the controls and, although attempts were made to correct for differences in body weight and blood volume, the levels were higher (P < 0.025) in the old than in the young T-implanted rats. Conversely, castration produced a marked (P < 0.0001) depression in plasma T concentrations, and the levels found
108
0 0
z
+I
+1
+1 +1
+1 +1
.= +1
e.i -I-I
+1 +i
+1 +1
+1
+1
+1 +1
+1 +1
z M < < ,-1 Z ©
e
t-
t-
< < M
÷I
+1
+1 +1
÷1 +1 ¢~
('4
÷1
÷1
+1 +1
41 +1
¢xl
~
~
ee
+1
+1
"H +l
+1 +1
~
c,,I
V
~V
zI
0
<
ze_
,-1 <
"H
+1
+1 t-I
+1 +1
~.-~ v~ ~ v
109
in the young and old castrated groups were similar. Circulating nonstress levels of LH tended to be higher in the young than in the old intact males, although this difference was not significant. The young intact males showed an approximate 100% increase (P < 0.005) in plasma LH concentrations in response to stress; the old intact males also showed a response to stress (P < 0.025), but it was smaller ( P < 0 . 0 2 5 ) than that in the young rats. Nonstress plasma LH concentrations were similar in the young and old T-implanted and castrated groups, and significant L H responses to stress were not obtained and either the young or the old rats after these manipulations. Pituitary concentrations of LH were similar in glands from young and old intact rats. As expected, castration elevated (P < 0.05) pituitary concentrations of LH in both young and old rats, but T implants did not significantly suppress LH concentrations in either age group. These results were the same when L H was expressed as/zg per AP. Data on the release of L H from the APs of young and old intact males during perifusion are presented in Fig. 1. Comparison of the slopes and elevations of the regression lines indicated that baseline LH release was higher (P < 0.0008) from APs of young rats than that from APs of old rats. Anterior pituitaries from both young and old intact males showed dose-related integrated LH responses (P < 0.025) to increasing concentrations of LHRH, and there was no effect of age on these responses. In addition to the effects noted in vivo (Table I), castration caused elevations (P <0.0008) in baseline levels of LH released during perifusion in both age groups, and there was no effect of age on the amplitude of these elevations (Fig. 2). Although APs of both young and old castrated males released large amounts of LH in response to L H R H , the responses were not dose-dependent. Unlike the LH .
.
.
.
.
.
.
Aged Intact Controls Young Intact Controls
.
--
2'01 m
"~ 1'Ok
t ~ m
~ ~
~
L.=:b~..~_-L~J ,~_r '-I j
-J
/
0
f~L_
-I_1
10-1M
I~ 120
10-8M
I~ 180
f~
I0-7M
I~ 240
10-6M LHRH
I~l 300
I 360
Time (rain) Fig. 1. Mean luteinizing hormone (LH) concentrations (/zg/mi) released from anterior pituitaries (APs) of young and aged intact males during continuous perifusion. Baseline LH concentrations per group were defined as the line that best fit the ten lowest mean LH values during the perifusion period. Amounts of LH released in response to 10-min pulses of increasing concentrations of LH-releasing hormone (LHRH) were determined by integration of the areas beneath each peak. Baseline release of LH was higher (P < 0.0008) from APs of young rats than from those of old rats.
110
A
B
4.0 --
4.0 -
......
Young C a s t r a t e Young Sham C a s t r a t e
.....
3.0 -
Old C a s t r a t e Sham C a s t r a t e
3.0
E ,,,~ 2.0
2.0-
'
"1" .-I
,.o
,o
[
r-
L r~ !i YT '-, ~: L~. r ~ :__~r ~-'l. -
0
~
iO-IM I~ 120
_
_
I
.
tO-OM ~ 180
~
~
tO-TM 1~ 240
10-$M LHRH ~ I 300 360
I~ 120
~ 180
~ 240
I~ 300
I 360
Time (rain)
Fig. 2. Mean hteinizinghormone (LH) concentrations(/zgJml)released from anterior pituitaries of youngand aged castratedmalesduringcontinuousperifusion.Mean LH concentrationsreleased from APs of young and aged sham-castrated controls are shown for comparison. Castration caused elevations (P < 0.0008) in baselineLH levels released in both age groups.
release rates observed in intact controls (see Fig. 1), baseline rates were similar from APs of young and old sham-castrated rats. Figure 3 shows the effects of implanting T-filled Silastic tubing 2 weeks prior to removal of APs for perifusion. Baseline LH release from glands of young rats was higher (P < 0.001) than that from old rats after implantation with the control, unfilled Silastic tubing. The high circulating T levels that occurred in vivo after T implantation suppressed (P < 0.0008) the in vitro release of LH from APs of both young and old T-implanted males, and the baseline LH levels produced by the young and old APs during perifusion were similar. Comparison of the data from T- and sham-implanted rats indicated that high circulating T levels in vivo reduced (P < 0.025) young and old AP responses to the several doses of L H R H in vitro. DISCUSSION
Despite the accumulated evidence that the decline in circulating T concentration in aging male rodents is in part attributable to extratesticular events [5,7,8,14,25-27], surprisingly few investigations have been carried out to determine whether this age-related change is reflected in AP function in vitro. In one such study, AP cell cultures derived from aging male mice released more LH into
111
A
B
3"5 I [ ] 1"1 I I t I
3.0
--Young T-Implant ........ Young Sham Implant
"TTT
1.~
t.5,
'!i i=
il i L~= I'
E 1.0
ri ' i
'
~L
II I:
',
r-J
,
r-~
LJ
i
~L I 17.
L~
2; ..J
L_
r~ I F l
/ I
- Old T-Implant ........ Old Sham Implant
] (
!,
1.0
L.-.
~J ..-----~
Fi =,
.~ ~ ' J ] " ~ ' ~
L,I
..
r'l
1 I
I •
i,
, L~
,
= l ~
LZ'~
0.5 t___a
10-I)M
120
10"8M
~ 180
tO'?M
~ 240
tO-|M LHflH
r~ 300
I 360
0
L.,_J
I@-tM
tO-SM
io-TM
10-11MLHRH
I~ 120
I~ 180
I~ 240
I~1 300
I 360
Time (rain)
Fig. 3. Mean luteinizing hormone (LH) concentrations ~g/ml) released from anterior pituitaries of young and aged testosterone-implanted males during continuous perifusion. Mean LH concentrations released from APs of young and aged sham-implanted controls are shown for comparison. Testosterone implants suppressed the release of LH (P < 0.001) and the responses to the several doses of LH-releasing hormone (LHRH) (P < 0.025) in both age groups. the culture medium than did cells obtained from young males [25], and the investigators concluded that L H secretion is not compromised in aging male mice. In contrast, Riegle et al. [15] observed that less L H was released from aged than from young rat hemipituitaries during static in vitro incubation. In our study, a perifusion approach also demonstrated that APs from aged male rats release less L H per unit of time under baseline conditions than do pituitaries from young males. The amounts released by young APs during our perifusion study are consistent with those previously reported by others [19,28,29]. Our results are consistent with those reported previously [14] in demonstrating that plasma L H levels are similar in intact nonstressed aged and young male rats. Likewise, the attenuation of the stress-induced elevation of plasma L H that we observed in intact aging male Sprague-Dawley rats is similar to the findings in Long-Evans rats [11]. Although the basis and significance of these findings are uncertain, they could reflect an age-associated decline in the amount of L H R H released in response to stress. Despite the absence of an in vivo age-related difference in circulating L H concentrations under nonstress conditions, a difference in baseline L H release was conspicuous during perifusion; APs from intact aged males released only 48% (P < 0.025) of that released by the glands from young intact rats. The absence of an age-related difference in circulating L H
112 in vivo in the presence of a marked age difference in the amount of LH released in vitro is unexplained but may reflect the effect of variations in metabolic clearance rates in vivo or the effects of complete removal of hypothalamic influences on pituitary function during perifusion. The finding that the in vitro
secretion rate of LH under baseline conditions is reduced in pituitaries from old rats supports the concept that the decline in androgen status experienced by aging male rats is the result of long-term understimulation of relatively normal testes by LH [5,14,261. The possibility exists that APs of aged male rats release less LH because of a reduced sensitivity to LHRH. The finding of Riegle et al. [15] that incubated hemipituitaries from aged male Long-Evans rats release less LH in response to L H R H than d/0 those from young rats is consistent with this possibility. However, our study, involving perifusion methods that obviated feedback effects on the AP, demonstrated that old and young rat APs release equivalent amounts of LH in response to L H R H stimulation. This discrepancy in results may be due to one or more methodological differences. For example, Riegle et ai. [15] used a static in vitro system, with the potential for complications due to the effects of feedback, hemipituitaries and a 4-h exposure to LHRH. The latter could have produced AP desensitization to L H R H [30]. In contrast, our perifusion approach probably precluded feedback effects, utilized smaller fragments of pituitary tissue and involved a series of 10-rain exposures to LHRH. Our results suggest that the age-related decline in pituitary release of LH is not due to reduced sensitivity of the AP to LHRH. The LH content in APs of the young intact males in our study is in agreement with that found in Long-Evans rats by Riegle et al. [15]. Unlike these investigators, however, we observed only a slight and nonsignificant loss of LH content in APs of old intact Sprague-Dawley males. The basis for this discrepancy in results is unknown, but it may reflect a strain difference. That APs from old intact males released 48% less LH per unit of time during perifusion with no detectable change in AP LH content implies a reduction in the rate of pituitary synthesis of LH. Such a reduction might reflect an intrinsic decrease in pituitary biosynthetic capacity or a decrease in stimulation by the hypothalamus. The ability of APs from aged castrated males to release high baseline levels of LH in vitro and the similarity of the responses to L H R H by the APs from the two age groups favor the latter alternative. Although castration produced changes in LH release from old and young APs that were similar, the baseline LH levels declined during perifusion of the old APs. This was the only decline noted in this series of experiments, and the basis for it is unknown. Because AP LH content of aged castrated males was less (P < 0.025) than that of young castrated males, a deficit in physiological reserve cannot be discounted. However, if this were the case, the potential deficit did not limit the LH responses to LHRH. Testosterone and its metabolites exert negative feedback effects on circulating levels of LH, and recent findings suggest that old men are more sensitive than
113 young men to such feedback [31]. Likewise, middle-aged rats show a greater and more prolonged suppression of L H than do young rats when exposed to the same circulating T levels [32]. We did not detect such an age-dependent difference in sensitivity to the negative feedback effects of T, perhaps owing to the supraphysiologic levels of T that were used. Plasma T levels in the T-implanted males were approximately eight times higher than normal; these high levels probably maximally inhibited L H release in both age groups. The administration of T in vivo for 2 weeks before perifusion clearly inhibited the release of L H from the isolated APs of both young and old rats in vitro, an indication that this feedback effect is due, at least in part, to changes in the AP that persist in the absence of continued exposure to circulating T. Treatment with T in vivo also reduced the magnitude of the AP responses to L H R H , and the reductions were similar in the APs of young and old males. The similarity of the responses to L H R H in the young and old T-treated groups is consistent with the similarity of responses obtained with APs of intact and castrated rats in suggesting that the reduced L H release in aged males does not involve pituitary hyporesponsiveness to LHRH. In summary, these results support the concept that the decline in plasma T levels experienced by aging male rats reflects in part the long-term understimulation of relatively normal testes due to a reduction in LH release. Because LH responses to L H R H were consistently similar for APs from young and old rats, it is likely that the onus of declining androgen status in old rats rests with functional deficits at the level of the hypothalamus. This conclusion is consistent with the reported reductions of hypothalamic L H R H activity [15] and immunocytochemically stained L H R H fibers [16] in old male rats. ACKNOWLEDGEMENTS Special thanks are due to Dr. Robert Mioduszweski for assistance in the development of the perifusion system. We thank Jessie Kroning and Paul Ballinger for their excellent technical assistance. The materials used in the protein hormone assays were kindly provided by the N I A D D K through Dr. A. Parlow. The work described in this article, Publication No. 1295 of the Oregon Regional Primate Research Center, was supported in part by the Animal Resources Branch Grant RR-00163 and by a National Research Service Award AG05220 (LWK), the Medical Research Foundation of Oregon Grant 8130--882 (LWK), and the National Institutes of Health Research Grant AM32442 (VC). REFERENCES 1 R. Ghanadian, J.G. Lewis and G.D. Chisholm, Serum testosterone and dihydrotestosterone changes with age in rats. Steroids, 25 (1975)753-762. 2 A.E. Miller and G.D. Riegle, Aging effects on hypothalamic catecholamine and testosterone secretion in the male rat. Fed. Proc., 34 (1975)303. 3 S.W.C. Chan, J.H. Leathernand T. Esashi, Testieularmetabolism and serum testosteronein aging male rats. Endocrinology, 101 (1977) 128-133.
114 4 G.D. Gray, Changes in the levels of luteinizing hormone and testosterone in the circulation of ageing male rats. ,L Endocrinol., 76 (1978) 551-552. 5 S.M. Harman, R.L. Danner and G.S. Roth, Testosterone secretion in the rat in response to chorionic gonadotropir): alterations with age. Endocrinology, 102 (1978) 540-544. 6 A.E. Miller and G.D. Riegle, Serum testosterone and testicular response to HCG in young and aged male rats. J. Gerontol., 33 (1978) 197-203. 7 K.M. Pirke, M. Geiss and H.J. Vogt, In vitro and in vivo studies on Leydig cell function in old rats. Acta Endocrinol. Suppl., 215 (1978) 37 (Abstract). 8 K.M. Pirke, H.J. Vogt and M. Geiss, In vitro and in vivo studies on Leydig cell function in old rats. Acta Endocrinol., 89 (1978) 393--403. 9 C.L. Bethea and R.F. Walker, Age-related changes in reproductive hormones and in Leydig cell respomivity in the male Fischer 344 rat. J. Gerontol., 34 (1979) 21-27. 10 C.J. Shaar, J.S. Euker, G.D. Riegle and J. Meites, Effects of castration and gonadal steroids on serum luteinizing hormone and prolactin in old and young rats. Y. Endocrinol., 66 (1975) 45-51. 11 G.D. Riegle and J. Meites, Effects of aging on LH and prolactin after LHRH, L-dopa, methyldopa, and stress in male rats. Proc. Soc. Exp. Biol. Med., 151 (1976) 507-511. 12 J.F. Bruni, H. Huang, S. Marshall and J. Meites, Effects of single and multiple injections of synthetic CmRH on serum LH, FSH, and testosterone in young and old male rats. Biol. Reprod., 17 (1977) 309-312. 13 P.D. Tsitouras, M.A. Kowatch and S.M. Harman, Age-related alterations of isolated rat Leydig cell function: gonadotropin receptors, adenosine 3',5'-monophosphate response, and testosterone secretion. Endocrinology, 105 (1979) 1400-1405. 14 L.W. Kaler and W.B. Neaves, The androgen status of aging male rats. Endocrinology, 108 (1981) 712-719. 15 G.D. Riegle, J. Meites, A.E. Miller and S.M. Wood, Effect of aging on hypothalamic LH-releasing and prolactin inhibiting activities and pituitary responsiveness to LHRH in the male laboratory rat. J. Gerontol., 32 (1977) 13-18. 16 G.E. Hoffman and J.R. Sladek, Age-related changes in dopamine, LHRH and somatostatin in the rat hypothalamus. Neurobiol. Aging, I (1980) 27-37. 17 S.J. Legan, G.A. Coon and F.J. Karsch, Role of estrogen as initiator of daily LH surges in the ovariectomized rat. Endocrinology, 96 (1975) 50-56. 18 R. Mioduszewski and V. Critchlow, Marked and selective suppression of prolactin secretion after interruption of neural connections of the medial basal hypothalamus of the rat. Brain Res., 216 (1981) 35-44. 19 A.J. Dowd, A.L. Barofsky, N. Chaudhuri, C.W. Lloyd and J. Weisz, Patterns of LH and FSH release from perifused rat pituitaries in response to infusions of hypothalamic extract. Endocrinology, 96 (1975) 243-252. 20 K. Maeda and L.A. Frohman, Release of somatostatin and thyrotropin-releasing hormone from rat hypothalamic fragments/n vitro. Endocrinology, 106 (1980) 1837-1842. 21 A. Negro-Vilar, S.R. Ojeda and S.M. McCann, Catecholaminergic modulation of luteinizing hormone-releasing hormone release by median eminence terminals in vitro. Endocrinology, 104 (1979) 1749-1757. 22 J.A. Resko, A. Malley, D. Begley and D.L. Hess, Radioimmunoassay of testosterone during fetal development of the rhesus monkey. Endocdno/ogy, 93 (1973) 156-161. 23 J.H. Zar, BiostatJstical Analysis, Prentice-Hall, Englewood Cliffs, New Jersey, 1974. 24 B.J. Winer, Stat~tical Principles in Exper~ental Design, 2nd edn., McGraw-HiE, New York, 1971. 25 C.E. Finch, V. Jones, J.R. Wisher, Y.N. Sinha, J.S. de Vellis and R.S. Swerdloff, Hormone production by the pituitary and testes of male C57BL/6J mice during aging. Endocrinology, I01 (1977) 1310-1317. 26 L.W. Kaler and W.B. Neaves, The steroidogenic capacity of the aging rat testis. J. Gerontol., 36
(1981) 398-404. 27 G. Schafer, A.F. Holstein and H. Hilz, Steroidogenic capacity of isolated adult mouse Leydig cells does not decrease with age. Endocrinology, 110 (1982) 1362-1366. 28 M.J. Kilpatrick, W.P. Collins and J.R. Newton, Studies on the release of gonadotrophins during the superfusion of isolated rat pituitaries in a continuous flow system. Y. Reprod. Fer61., 46 (1976) 25-30.
115 29 M..l. Kilpatrick, J.R. Newton and W.P. Collins, Studies on factors affecting release of gonadotrophins during the superfusion of isolated rat pituitaries. J. Reprod. Fertil., 49 (1977) 347-351. 30 M.A. Smith, M.H. Pert'in and W.W. Vale, Desensitization of cultured pituitary cells to gonadotropin-releasing hormone: evidence for a post-receptor mechanism. Moi. cell. Endocrinol., 30 (1983) 85-96. 31 S.J. Winters and P. Troen, The gonadotropin-suppressive potency of androgen is increased in elderly men. Endocrinology, (1982) 222 (Abstract No. 572). 32 G.D. Gray, E.R. Smith and J.M. Davidson, Oonadotropin regulation in middle-aged male rats. Endocrinology, 107 (1980) 2021-2026.