Abstract
Free full text
Older men exhibit reduced efficacy of and heightened potency downregulation by intravenous pulses of recombinant human LH: a study in 92 healthy men
1 Abstract
Direct sampling of the human spermatic veins has disclosed concomitant LH and testosterone (T) pulses, suggesting pulsatile LH concentration-dependent stimulation of T secretion. However, studies to date have examined this hypothesis using only pharmacological stimulation with hCG. The present study tests the hypothesis that age is marked by decreased T secretory responses to repeated near-physiological iv pulses of recombinant human LH administered in a Clinical Translational Science Center. Participants included 92 healthy men aged 18–75 yr with BMI 18–34 kg/m2. The contribution of endogenous LH pulses was minimized by combined injection of a selective GnRH receptor antagonist sc and successive pulses of biosynthetic LH iv. A new analytical dose response model was applied to estimate the properties of exogenous LH's drive of T secretion. Regression of LH-T dose response potency estimates on age showed that the efficacy of pulses of biosynthetic LH progressively decreased with age (P = 0.014, r = 0.26). Testis sensitivity to exogenous LH pulses also declined with age (P = 0.011, r = 0.27). Moreover, estimated Leydig cell downregulation by LH pulses rose significantly with age (P = 0.039, r = 0.22). These outcomes were selective, since the recovery potency of infused LH was not affected by age but was reduced by increasing BMI (P = 0.011, r = 0.27). Assuming stable bioactivity of infused recombinant human LH, these novel data indicate that factors associated with age and BMI attenuate LH efficacy and testis sensitivity and augment Leydig cell downregulation in healthy men.
testosterone (t) is a primary anabolic hormone in men that exerts important protein-sparing effects. In particular, T and its metabolites augment muscle and bone mass, maintain sexual function, and stimulate visceral lipolysis (3, 14, 28). Cognitive and mood-restoring effects of T are less consistent (1). T secretion is under the control of LH drive (feedforward) and intratesticular modulators such as IGF-I, inflammatory cytokines, and neurogenic signals (24).
Direct sampling of the human spermatic vein has demonstrated that pulses of LH coincide with pulses of T secretion (10, 40). Since testicular venous sampling has limited practicability, noninvasive methods using peripheral LH and T concentrations have been developed to estimate pulsatile LH-T stimulation without catheterization (15, 17). The results of one recent analytical methodology indicated that endogenous LH pulses induce less T secretion in older than in young men (15). This outcome would be analogous to earlier inferences made using chorionic gonadotropin (hCG) stimulation tests, inasmuch as hCG stimulation often, but not exclusively, evokes lower peak T concentrations in older than in young male individuals (7, 13, 22, 26, 30). However, the hCG stimulus is demonstrably nonphysiological compared with endogenous LH pulses. Major differences include slower dissociation kinetics of hCG than LH from the LH/hCG receptor, more prolonged plasma half-life of hCG, and a marked capacity of hCG to downregulate testicular T synthesis (4, 12, 23, 29, 31). LH or hCG-induced downregulation may be particularly confounding if downregulation per se increased with aging, as was suggested recently for endogenous LH pulses in 13 older men (15).
Quantifiable features of pulsatile LH secretion change with age, with pulses becoming more frequent, smaller in amplitude, and more irregular in pattern (17). In addition, the in vitro bioactivity of secreted LH may be decreased in older men (25). Thus, alterations in endogenous LH with age might contribute to decreased Leydig cell steroidogenesis. To eliminate confounding by hCG stimulation and any age-related alterations in secreted LH, the present study quantifies exogenous LH-T dose responsiveness in 92 healthy men using a novel pulsatile LH clamp during acute overnight GnRH receptor blockade.
METHODS
Human subjects.
Ninety-two healthy men were studied after they provided voluntary, witnessed, informed consent that was approved by the Mayo Clinic Institutional Review Board. The protocol was reviewed by the FDA under an investigator-initiated off-label use of ganirelix (Organon, West Orange, NJ) and recombinant human LH (Serono, Rockland, MA).
Protocol.
Each subject underwent an overnight study in the Mayo Clinical Translational Science Center. The protocol comprised 2200 sc injection of 1 mg/m2 ganirelix, a selective GnRH receptor antagonist (2). Age does not affect ganirelix absorption or blood levels (33). Intravenous (6- or 8-min square-wave) pulses of 6.25–50 IU Serono LH were given every 1–3 h for 8 h to 2 days, beginning 2 h after the ganirelix injection. Blood was sampled every 10 min concurrently for LH and T measurements.
Hormone assays.
LH was measured by robotics-automated immunochemiluminesce assay in duplicate in each subject exactly as described (33). No samples were undetectable (all >0.1 U/l, Second International Reference Preparation). In this assay, 75 IU Serono LH represents 30 IU Second International Reference Preparation LH by direct assay. T was measured by the same platform, which correlates at r2 = 0.98, slope = 1.07 (P < 0.001), with tandem mass spectrometry (20).
Analytical methods.
The goal was to relate time-varying LH concentrations (input, effector) to time-varying T secretion rates (output, response) via a new hysteresis-based dose response model in healthy men. The relationship was illustrated in Figs. 1 and and44 of Ref. 15. This methodology represents an extension of the classical four-parameter logistic concept of dose-dependent efficacy, sensitivity, potency, and basal (unstimulated) secretion. Efficacy denotes maximal (asymptotic) T secretion. Sensitivity is a slope term. The revised model comprises a nonlinear (logistic) dose response function with allowance for two potencies of stimulation, one during the ascending and the other during the descending phase of the pulsatile stimulus. The potency term is rendered as an exponent or as an estimated LH concentration stimulating one-half maximal T secretion (EC50) during the onset (rising phase) or recovery/offset (falling phase) of the LH pulse. The idea is to estimate possible testis downregulation within an LH pulse after an apparent time lag. The model was developed for corticotropin's feedforward onto cortisol secretion in Ref. 16 and applied to endogenous pulsatile LH's drive of T secretion (15).
Body mass index (BMI) elevates the EC50 of infused recombinant human LH in 92 healthy men. Both onset (initial; top) and recovery (delayed; bottom) LH EC50 values are affected. Potency was evaluated as the EC50 [estimated LH concentration stimulating one-half maximal testosterone (T) secretion rates]. Higher EC50 values denote lower potency, and vice versa for low EC50 values. Pearson's correlation coefficients and matching P values are given.
Negative association between the natural logarithm of the efficacy of pulsatile intravenous rhLH infusions (y-axis) and age (x-axis) (top) as well as age × BMI (bottom) in 92 healthy men.
Dose response estimation involves three steps. First, LH concentration profiles (10-min time series) are analyzed by a validated deconvolution method that represents LH concentration profiles as a train of variable-amplitude secretory bursts superimposed upon basal (time-invariant) nonpulsatile secretion (19). LH elimination was embodied in a biexponential function. Second, T concentration profiles are deconvolved analogously to yield sample T secretion rates. In the third automated step, reconvolved pulsatile LH concentrations are aligned with time-lagged deconvolved T secretory bursts to allow estimation of nonlinear LH→T feedforward (logistic-like) dose response parameters (basal, onset and offset potency, sensitivity, and efficacy) simultaneously with a time delay before downregulation within T secretory bursts (15). The result is that pulsatile LH concentrations (reconvolved) are related to T secretion rates (deconvolved) in paired LH-T profiles. Random effects on T secretory burst amplitude and the standard deviation (SD) of residual model error are calculated concurrently.
Statistical contrasts.
Baseline-screening hormone concentrations were regressed on age. Analytical estimates of LH potency, LH efficacy, and testis sensitivity to LH pulses were regressed linearly on age or body mass index (BMI) after natural logarithmic transformation. Two-tailed P < 0.05 was construed as significant. Data are given as the geometric means ± SE and/or median plus range.
RESULTS
Age, BMI, and recombinant human LH-infusion schedules are given in Table 1. All subjects completed the full sampling protocol. Mild injection site tenderness was noted after ganirelix administration, not requiring treatment. There were no other adverse events. Age ranged from 18 to 75 yr and BMI from 18 to 34 kg/m2. By linear regression, age was associated with increases in BMI (P < 0.001), sex hormone-binding globulin (P = 0.004), and FSH (P < 0.001) and decreases in both prolactin (P < 0.001) and bioavailable T concentrations (P < 0.001) at screening. LH, E2, and total T did not differ with age (P > 0.07).
Table 1.
Table of rhLH subjects according to study design
| Means ± SE (Ranges) | ||
|---|---|---|
| Study (Infusion Paradigm) | Age, yr | BMI, kg/m2 |
| 6.25, 12.5, 25, and 37.5 IU rhLH pulses, each twice iv in random order every 3 h for 20 h (n = 15) | 48 ± 4.6 (21–75) | 27 ± 0.82 (21–32) |
| 12.5 (every hour) or 25 IU (every 2 h) rhLH over 22 h (n = 15) | 40 ± 5.4 (19–73) | 26 ± 1.1 (19–34) |
| 12.5 IU rhLH every 2 h over 22 h (n = 23) | 41 ± 3.0 (19–72) | 26 ± 0.71 (20–32) |
| 37.5 IU rhLH boluses administered 2 h apart over 8 h (n = 20) | 35 ± 2.8 (18–70) | 26 ± 0.73 (18–31) |
| 50 IU rhLH iv every 2 h for 2 days (n = 19) | 41 ± 4.9 (19–73) | 28 ± 0.75 (22–32) |
| 6.25–50 IU rhLH iv every 1–3 h for 8 h to 2 days (all 5 studies; n = 92) | 40 ± 1.8 (18–75) | 27 ± 0.37 (18–34) |
Values are means ± SE; n = 92 men.
rhLH, recombinant human LH; iv, intravenous; BMI, body mass index.
Pulses were infused as 6- or 8-min square-wave boluses.
Estimated potency of initial (onset) LH stimulation averaged 2.5 ± 0.47 IU/l (median 2.3) in the 92 volunteers (Table 2). Downregulated (offset) LH potency averaged 6.6 ± 0.79 IU/l (median 5.2). The latter values denote significant pulse-by-pulse downregulation [P < 0.001 by paired 2-tailed t-test compared with initial (onset) EC50]. Regression analysis revealed no significant effect of age on the initial (onset) LH EC50 (P = 0.64) or on the delayed (recovery) LH EC50 (P = 0.16). In contrast with age, BMI predicted onset (initial) LH EC50 (P = 0.054, r = 0.20; Fig. 1, top) and offset (downregulated) LH EC50 (P = 0.011, r = 0.27; Fig. 1, bottom). The positive slopes of BMI on EC50 define loss of LH potency. The hypothesis of heightened downregulation of LH potency with age was affirmed by regressing the untransformed LH potency (exponent) difference (delayed minus initial) on age; P = 0.039, r = 0.22 (Fig. 2). Strikingly, the potency difference also rose markedly (signifying greater downregulation) as a function of an age × BMI interaction (P = 0.008, r = 0.28; Table 3). The interaction reflects the joint effect of the two variables age and BMI such that an increase in either variable augments the potency difference (thereby defining greater downregulation), whereas an increase in both variables amplies the potency difference even more (as their product).
Table 2.
LH-T 2-potency dose response data in men: rhLH
| rhLH | ||
|---|---|---|
| Means ± SE | Median (range) | |
| EC50 onset, IU/l | 2.5 ± 0.47 | 2.3 (0.003–19) |
| EC50 recovery, IU/l | 6.6 ± 0.79 | 5.2 (0.81–38) |
| Potency onset (exponent) | −3.5 ± 0.76 | −4.9 (−40 to −0.003) |
| Potency recovery (exponent) | −9.3 ± 1.5 | −9.4 (−40 to −0.18) |
| Sensitivity (slope) | 1.4 ± 0.21 | 1.7 (0.001–8.7) |
| Efficacy, ng·dl−1·min−1 | 9.5 ± 0.58 | 9.9 (1.7–29) |
| Basal, ng·dl−1·min−1 | 4.1 ± 0.76 | 5.4 (0.001–26) |
| Downregulation delay, min* | 30 ± 1.5 | 40 (10–50) |
| Model SD, ng·dl−1·min−1 | 3.1 ± 0.13 | 3.0 (0.68–6.7) |
| Random effects, ng·dl−1·min−1 | 0.77 ± 0.21 | 2.1 (0.001–9.1) |
Data are the geometric means ± SE and median (range); n = 92.
T, testosterone.
Increased downregulation (greater rightward shift due to exponential potency loss) of the LH-T dose-response curve with age in 92 healthy men. Downregulation was defined as the onset minus recovery exponential potency of infused recombinant human LH (RhLH).
Table 3.
Significance of age-related regressions
| Parameter | P Value | r Value |
|---|---|---|
| Sensitivity to LH* | 0.011 | −0.27 |
| Potency difference† | 0.039 | 0.22 |
| EC50 difference | 0.023 | 0.24 |
| Efficacy of LH pulses | 0.014 | −0.26 |
| Model residual SD | 0.004 | −0.30 |
| Basal T secretion | 0.634 | −0.051 |
Data are from n = 92 men given repeated iv pulses of rhLH after ganirelix (GnRH receptor blocker) injection.
The sensitivity of pulsatile T secretion to pulsatile exogenous LH drive (slope term) averaged 1.4 ± 0.21 (median 1.7) (ng·dl−1·min−1)·(IU/l) in the 92 men studied (Table 2). There were significantly negative regressions of testis sensitivity on age (P = 0.011, r = −0.27), BMI (P = 0.029, r = −0.23), and age × BMI product (P = 0.005, r = −0.29; Fig. 3).
Inverse relationships between age, BMI, and age × BMI and the natural logarithm of T secretory sensitivity to infused pulses of rhLH in 92 healthy men. The data point in the box is an outlier at P < 0.001 by F ratio testing.
Efficacy of infused LH pulses averaged 9.5 ± 0.58 (median 9.9) in the full cohort (Table 2). Regression analysis showed a strongly negative influence of age on pulsatile LH efficacy (P = 0.014, r = −0.26; Fig. 4, top). This indicates that maximal pulsatile exogenous LH drive of pulsatile T secretion decreases approximately linearly as age increases. In addition, there was a significant interaction between age and BMI in jointly predicting decreased LH efficacy (P = 0.047, r = −0.21; Fig. 4, bottom). There was no effect of BMI alone.
Basal (nonpulsatile) T secretion averaged 4.1 ± 0.76 ng·dl−1·min−1 (median 5.4), with no change with age (Table 2). The time delay to downregulation averaged 30 ± 1.5 min (median 40 min; n = 92 men). Neither age nor BMI affected this value. Model fit, defined statistically by the model residual SD, was 3.1 ± 0.13 (3.0) ng·dl−1·min−1. Model SD declined with age (P = 0.004, r = −0.30). Random effects on T secretory burst mass were invariant of age (P = 0.27) and BMI (P = 0.46).
DISCUSSION
In 92 healthy men aged 18–75 yr, pulsatile intravenous (iv) infusion of biosynthetic human LH at 1- to 3-h intervals for 8–48 h unveils a strong age-associated loss of stimulatory efficacy. Confounding by endogenous LH was minimized by prior subcutaneous administration of a potent selective GnRH receptor antagonist, ganirelix, which lowers LH concentrations by ≥85% (21, 34, 37, 38). Earlier studies showed that age per se does not affect ganirelix blood levels in healthy men (21). Accordingly, assuming validity of the pulsatile recombinant human LH clamp paradigm, the present outcomes establish an age-associated reduction of maximal Leydig cell steroidogenic responses to nearly physiological LH pulses. This inference is strengthened by evidence that GnRH antagonists exert no significant direct gonadal effects (9) and that deconvolution estimates of the mass and half-life of injected LH pulses were well within the range of normal endogenous LH pulses (15). Deconvolution analysis of T concentration series obtained during the exogenous pulsatile recombinant human LH clamp provided an independent verification and a mechanistic explanation for diminished testis responsiveness in older individuals. In particular, the mass (amount of) T secreted per burst declined significantly with age (P = 0.003, r = −0.31). The age-related finding was selective, since nonpulsatile (basal) T secretion was independent of age across the five-decade span evaluated here.
Calculated testis sensitivity to recombinant human LH pulses decreased markedly with age, BMI, and their interaction. The largest absolute effect was for the interaction between age and BMI, where r was −0.29 (P = 0.005). Sensitivity corresponds to a slope term defined by the degree of augmentation of T secretion rate induced by each unit rise in LH concentrations. Lower sensitivity to individual LH pulses could be due to reductions in either LH receptor activation or postreceptor signaling. In this regard, data in the aged male rat demonstrate decreases in testicular LH-driven AMP accumulation and in cholesterol's delivery to inner mitochondrial membranes (18, 32). Postulated pathophysiological mechanisms in aging include formation of reactive oxygen species and impairment of antioxidant defenses in aging steroidogenic cells (5). In contrast, we speculate that an absolute reduction in the number of LH receptors or Leydig cells could better explain reduced LH efficacy (lower maximal T secretion) (27). Accumulation of systemic inflammatory cytokines in aging and obesity may also confer a mechanism for decreased Leydig cell function (39). The pulsatile LH clamp model provides a means to test such hypotheses prospectively.
The finding of age-invariant LH potency is consistent with infusion of recombinant human LH of fixed biological potency. An implication of unchanged potency is that age does not engender competitive inhibitors of LH action, since the latter would be expected to reduce even exogenous LH potency. The earlier inferred reduction in endogenous LH potency in aging men (15) might be due to reduction in circulating LH bioactivity (6, 11, 25, 36). Whereas detailed analysis of LH isoforms will be required to clarify this issue, the pulsatile recombinant human LH clamp paradigm used here bypasses endogenous LH due to overnight hypogonadotropism induced by ganirelix pretreatment. Under these experimental circumstances, LH potency estimates were comparable in young and older men whether assessed during the rising (onset) or falling (offset) phases of LH pulses.
Intraindividual EC50 differences (recovery EC50 − initial EC50) provide a measure of intrapulse downregulation of LH-stimulated T secretion. This difference (downregulation) term increased significantly with age in the cohort of 92 men (P = 0.023, r = 0.24). Heightened downregulation of Leydig cell T responses to distinct LH pulses in older men suggests a novel mechanism for reduced testis responsiveness in aging individuals. If verified, this mechanism could also contribute to decreased hCG effects in older men, since pharmacological hCG is itself a strong downregulating signal to T secretion in men and animals (see introduction) (8, 35).
Caveats include the issue of parameter correlation in all models, including dose response models. Here, the concern is mitigated by deconvolving the otherwise strongly autocorrelated T concentration time series into a T secretion profile and by parameter estimation using a maximum-likelihood-based convergence method. The asymptotic (tending to a plateau) nature of the logistic dose response function also provides an implicit upper and lower boundary. Additional caveats include the need to replicate these outcomes in other cohorts of healthy men, the restriction that exogenous LH pulses may not be identical to endogenous LH pulses in shape (viz., the former were ~2 min shorter) or in biochemical isoform distribution, the recognition that the present experimental infusions test mainly short-term testis responses, and the need for longitudinal data to test cause and effect.
In summary, by using a paradigm of pulsatile iv infusion of recombinant human LH along with analytical LH-T dose response estimation, the present studies identify three distinguishable age-related defects in testis responsiveness to LH action: decreased LH efficacy, diminished Leydig cell sensitivity, and increased LH-T downregulation. Together, these mechanisms could provide a powerful basis for reduced effectual LH drive in older men.
GRANTS
This work was supported in part via Center for Translational Science Activities Grant no. 1-UL-1-RR-024150 from the National Center for Research Resources (Rockville, MD) and AG-023133, AG-031763, and DK-050456 (Metabolic Studies Core of the Minnesota Obesity Center) from the National Institutes of Health (Bethesda, MD). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging or the National Institutes of Health. Matlab versions of the deconvolution methodology are available from J. D. Veldhuis at ude.oyam@sennahoj.siuhdlev.
AUTHOR CONTRIBUTIONS
J.D.V. and P.Y.L. did the conception and design of the research; J.D.V., P.Y.L., and P.Y.T. performed the experiments; J.D.V., D.M.K., and P.Y.T. interpreted the results of the experiments; J.D.V. and D.M.K. prepared the figures; J.D.V. drafted the manuscript; J.D.V., P.Y.L., and P.Y.T. edited and revised the manuscript; J.D.V. approved the final version of the manuscript; D.M.K. analyzed the data.
ACKNOWLEDGMENTS
We thank Jill Smith for support in manuscript preparation, Ashley Bryant for data analysis and graphics, the Mayo Immunochemical Laboratory for assay assistance, and the Mayo research nursing staff for implementing the protocol.
REFERENCES
Articles from American Journal of Physiology - Endocrinology and Metabolism are provided here courtesy of American Physiological Society
Full text links
Read article at publisher's site: https://doi.org/10.1152/ajpendo.00450.2011
Read article for free, from open access legal sources, via Unpaywall:
https://europepmc.org/articles/pmc3328086
Citations & impact
Impact metrics
Citations of article over time
Alternative metrics
Article citations
Male Sex Hormones, Metabolic Syndrome, and Aquaporins: A Triad of Players in Male (in)Fertility.
Int J Mol Sci, 24(3):1960, 19 Jan 2023
Cited by: 0 articles | PMID: 36768282 | PMCID: PMC9915845
Review
Free full text in Europe PMCAging and androgens: Physiology and clinical implications.
Rev Endocr Metab Disord, 23(6):1123-1137, 02 Dec 2022
Cited by: 5 articles | PMID: 36459352 | PMCID: PMC10370404
Review Free full text in Europe PMC
Male sex hormones, aging, and inflammation.
Biogerontology, 24(1):1-25, 04 Jan 2023
Cited by: 7 articles | PMID: 36596999 | PMCID: PMC9810526
Review
Free full text in Europe PMCSleep, testosterone and cortisol balance, and ageing men.
Rev Endocr Metab Disord, 23(6):1323-1339, 24 Sep 2022
Cited by: 11 articles | PMID: 36152143 | PMCID: PMC9510302
Review
Free full text in Europe PMCAnimal models of male reproductive ageing to study testosterone production and spermatogenesis.
Rev Endocr Metab Disord, 23(6):1341-1360, 23 May 2022
Cited by: 4 articles | PMID: 35604584
Review
Go to all (24) article citations
Similar Articles
To arrive at the top five similar articles we use a word-weighted algorithm to compare words from the Title and Abstract of each citation.
Dynamic testosterone responses to near-physiological LH pulses are determined by the time pattern of prior intravenous LH infusion.
Am J Physiol Endocrinol Metab, 303(6):E720-8, 17 Jul 2012
Cited by: 9 articles | PMID: 22811471 | PMCID: PMC3468430
Aging in healthy men impairs recombinant human luteinizing hormone (LH)-stimulated testosterone secretion monitored under a two-day intravenous pulsatile LH clamp.
J Clin Endocrinol Metab, 90(10):5544-5550, 19 Jul 2005
Cited by: 15 articles | PMID: 16030163
Age diminishes the testicular steroidogenic response to repeated intravenous pulses of recombinant human LH during acute GnRH-receptor blockade in healthy men.
Am J Physiol Endocrinol Metab, 288(4):E775-81, 30 Nov 2004
Cited by: 18 articles | PMID: 15572655
Steroidal regulation of biologically active luteinizing hormone secretion in men and women.
Hum Reprod, 8 Suppl 2:84-96, 01 Nov 1993
Cited by: 5 articles | PMID: 8276976
Review
Fetal Leydig cell culture--an in vitro system for the study of trophic hormone and GnRH receptors and actions.
J Steroid Biochem, 23(5b):743-755, 01 Nov 1985
Cited by: 13 articles | PMID: 3001417
Review
Funding
Funders who supported this work.
NCRR NIH HHS (2)
Grant ID: 1UL1-RR-024150
Grant ID: UL1 RR024150
NIA NIH HHS (4)
Grant ID: AG-031763
Grant ID: R01 AG031763
Grant ID: R01 AG023133
Grant ID: AG-023133
NIDDK NIH HHS (2)
Grant ID: P30 DK050456
Grant ID: DK-050456