Browse the corpus

Reproduction and metabolism are interconnected, but the mediators of the interaction between these physiological processes are poorly understood (1). Up to 40% of men with obesity and/or type 2 diabetes have co-existing hypogonadism, which is associated with an adverse metabolic phenotype (1). Testosterone treatment improves hypogonadism, but there are ongoing safety concerns limiting its use (2). As the prevalence of obesity/diabetes rises, alternative strategies for managing co-existing obesity/diabetes and hypogonadism are required. Glucagon-like peptide-1 (GLP-1), produced by intestinal L-cells postprandially (3), potently reduces food intake, induces weight-loss, and augments insulin secretion (4). Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are currently used to treat obesity/type 2 diabetes. Glucagon-like peptide-1 receptor agonism may also have regulatory effects on the reproductive system (5–8). Glucagon-like peptide-1 administration increases hypothalamic kisspeptin (a key regulator of hypothalamic gonadotropin-releasing hormone [GnRH] secretion) expression (5) and stimulates GnRH secretion from rodent hypothalamic explants (6). Furthermore, liraglutide, a GLP-1RA, increases kisspeptin neuronal firing (7). In female rats, GLP-1 increases the preovulatory luteinizing hormone (LH) surge (8), but exendin-4 (another GLP-1RA) reduces LH levels (8).

[GnRH] secretion) expression (5) and stimulates GnRH secretion from rodent hypothalamic explants (6). Furthermore, liraglutide, a GLP-1RA, increases kisspeptin neuronal firing (7). In female rats, GLP-1 increases the preovulatory luteinizing hormone (LH) surge (8), but exendin-4 (another GLP-1RA) reduces LH levels (8). Acute infusion of GLP-1 to healthy men during a euglycemic clamp (maintaining circulating glucose levels around 5 mmol/L) does not alter mean reproductive hormone levels; but it reduces the number of testosterone pulses with a trend towards longer testosterone pulse duration (9). However, chronic administration of liraglutide to obese, hypogonadal men with type 2 diabetes, increases testosterone to a greater extent than testosterone treatment and metformin alone (10). In obese hypogonadal men, 16 weeks of liraglutide treatment increases reproductive hormone levels (11). It is unclear if GLP-1 receptor agonism has a beneficial effect on the reproductive axis in the absence of weight loss, as these studies reported significant weight loss in the groups that received liraglutide (10, 11). Additionally, in the study reporting no effect of acute GLP-1 infusion on reproductive hormone levels (9), a subanorectic dose of GLP-1 was used (12). Consequently, this dose may have been too low to affect reproductive hormone levels. Therefore, we performed a single-blind, randomized, placebo-controlled crossover study of administration of a biologically active dose of GLP-1 to healthy men to test the hypothesis that GLP-1 has direct effects on the reproductive axis.

Acute infusion of GLP-1 to healthy men during a euglycemic clamp (maintaining circulating glucose levels around 5 mmol/L) does not alter mean reproductive hormone levels; but it reduces the number of testosterone pulses with a trend towards longer testosterone pulse duration (9). However, chronic administration of liraglutide to obese, hypogonadal men with type 2 diabetes, increases testosterone to a greater extent than testosterone treatment and metformin alone (10). In obese hypogonadal men, 16 weeks of liraglutide treatment increases reproductive hormone levels (11). It is unclear if GLP-1 receptor agonism has a beneficial effect on the reproductive axis in the absence of weight loss, as these studies reported significant weight loss in the groups that received liraglutide (10, 11). Additionally, in the study reporting no effect of acute GLP-1 infusion on reproductive hormone levels (9), a subanorectic dose of GLP-1 was used (12). Consequently, this dose may have been too low to affect reproductive hormone levels. Therefore, we performed a single-blind, randomized, placebo-controlled crossover study of administration of a biologically active dose of GLP-1 to healthy men to test the hypothesis that GLP-1 has direct effects on the reproductive axis. Materials and Methods Study participants This study was performed in accordance with the Declaration of Helsinki and received approval from the West London Research Ethics Committee (16/LO/0391). Recruitment via advertisements took place between November 2017 and August 2018. Eighteen healthy men (age 24.7 ± 1years, BMI 22.1 ± 0.4 kg/m2, baseline testosterone 21.9 ± 1.5 nmol/L and calculated free testosterone (13) 0.52 ± 0.03 nmol/L) were enrolled in the study after confirmation of eligibility (ie, absence of active medical or psychiatric conditions and no use of prescription drugs, recreational drugs, and nicotine-containing products within the preceding 3 months) and with the provision of written informed consent.

sterone (13) 0.52 ± 0.03 nmol/L) were enrolled in the study after confirmation of eligibility (ie, absence of active medical or psychiatric conditions and no use of prescription drugs, recreational drugs, and nicotine-containing products within the preceding 3 months) and with the provision of written informed consent. Study visits All participants attended 2 study visits, 1 for GLP-1 administration and 1 for vehicle administration. Infusion order was randomized and participants were blinded as to the identity of the infusions. GLP-17–36 was infused at a rate of 0.8 pmol/kg/min, a dose established to reduce food intake in humans (4). Rate-matched vehicle infusions were comprised of Gelofusine (Braun, Sheffield, UK) only. On each study visit, following an overnight fast, each participant ate a standardised 200 kcal breakfast at 6:00 am and arrived at the clinical research facility at 8.15 am. Two intravenous cannulae were inserted (1 in each arm; 1 cannula was used to administer the infusion and the other cannula was used to obtain blood samples). After baseline sampling, GLP-1/vehicle infusion was started at T = 0 minutes and continued until T = 500 minutes. Visual analogue scales (VAS, 0–10 cm) were used to measure participants’ self-reported nausea at T = -15 minutes, T = 240 minutes, and T = 470 minutes. Participants were given an ad libitum meal at T = 480 minutes. Blood samples were taken every 10 minutes (Fig. 1A). Figure 1. Study protocol and GLP-1 levels.

On each study visit, following an overnight fast, each participant ate a standardised 200 kcal breakfast at 6:00 am and arrived at the clinical research facility at 8.15 am. Two intravenous cannulae were inserted (1 in each arm; 1 cannula was used to administer the infusion and the other cannula was used to obtain blood samples). After baseline sampling, GLP-1/vehicle infusion was started at T = 0 minutes and continued until T = 500 minutes. Visual analogue scales (VAS, 0–10 cm) were used to measure participants’ self-reported nausea at T = -15 minutes, T = 240 minutes, and T = 470 minutes. Participants were given an ad libitum meal at T = 480 minutes. Blood samples were taken every 10 minutes (Fig. 1A). Figure 1. Study protocol and GLP-1 levels. A: After an overnight fast and a standardized breakfast, 18 healthy men attended 2 study visits, one with 0.8 pmol/kg/min glucagon-like peptide-1 (GLP-1) infusion and one with (rate-matched) vehicle infusion for 500 minutes. The order of the infusions was randomly determined. Blood samples were taken at 10 minute intervals throughout each study visit (apart from during the ad libitum meal). Visual analogue scales (VAS) were completed by participants to assess subjective nausea preinfusion (at T = -15 minutes), midinfusion (at T = 240 minutes), and premeal (at T = 470mins). An ad libitum meal was given to the participants at T = 480 minutes.

oughout each study visit (apart from during the ad libitum meal). Visual analogue scales (VAS) were completed by participants to assess subjective nausea preinfusion (at T = -15 minutes), midinfusion (at T = 240 minutes), and premeal (at T = 470mins). An ad libitum meal was given to the participants at T = 480 minutes. B: Plasma GLP-1 levels were higher during GLP-1 infusion compared to vehicle infusion. Two-way repeated measures analysis of variance (RM-ANOVA) detected a significant interaction of treatment (ie, vehicle vs. GLP-1) and time (P < 0.0001). Asterisks indicate significant differences at specific timepoints (****P < 0.0001). Biochemical analyses Plasma glucose, serum insulin, LH, follicle-stimulating hormone (FSH), and testosterone were measured (in single sample aliquots) by NorthWest London Pathology on the automated Abbott Architect® platform. Chemiluminescent immunoassays were used to measure serum insulin (intra-assay and interassay coefficient of variation [CV]: ≤7%), serum LH (intra-assay and interassay CV: ≤5%), serum FSH (intra-assay and interassay CV: ≤10%), and serum testosterone (intra-assay and interassay CV: ≤8%). Plasma glucose was measured using a colorimetric hexokinase assay (intra-assay and interassay CV: ≤2%). Total plasma GLP-1 was measured (in duplicate) using an in-house radioimmunoassay (intra-assay and interassay CV: ≤10%) utilizing an antibody that detects GLP-17-36amide and GLP-19-36amide but not glycine-extended forms of GLP-1 (3).

s measured using a colorimetric hexokinase assay (intra-assay and interassay CV: ≤2%). Total plasma GLP-1 was measured (in duplicate) using an in-house radioimmunoassay (intra-assay and interassay CV: ≤10%) utilizing an antibody that detects GLP-17-36amide and GLP-19-36amide but not glycine-extended forms of GLP-1 (3). Statistical methods Based on existing literature (14), a sample size of 18 men provides 90% power to detect a difference in LH (between vehicle and GLP-1 infusion) of 2 IU/L (SD 2.3 IU/L) at a significance level of 0.05. Data from all 18 participants were included in the analyses. Luteinizing hormone pulsatility was determined using a validated blinded deconvolution analysis (15). The differences in hormone levels and nausea during vehicle infusion compared with GLP-1 infusion were compared using a 2-way repeated measures analysis of variance (RM-ANOVA) with Bonferroni’s post hoc multiple correction tests for individual timepoint comparisons. Luteinizing hormone, FSH, and testosterone areas under the curve (AUCs) were calculated using the trapezoidal rule (16). Hormone AUCs and food intake data were compared using paired t-tests. Statistical analyses were performed using STATA 14.1 (STATACorp, College Station, TX, USA) and Prism 8.0.2 (GraphPad, San Diego, CA, USA) software. P-values <0.05 were considered statistically significant. Data are presented as mean ± SEM.

rule (16). Hormone AUCs and food intake data were compared using paired t-tests. Statistical analyses were performed using STATA 14.1 (STATACorp, College Station, TX, USA) and Prism 8.0.2 (GraphPad, San Diego, CA, USA) software. P-values <0.05 were considered statistically significant. Data are presented as mean ± SEM. Results Effects of GLP-1 on LH, FSH, and testosterone Glucagon-like peptide-1 administration resulted in elevated GLP-1 levels (Fig. 1B). However, there were no significant differences between serum LH levels and LH area under the curve (AUC) during vehicle and GLP-1 administration (Fig. 2A and 2B). Furthermore, GLP-1 administration did not alter the number of LH pulses (number of LH pulses/500 min: vehicle 4.2 ± 0.4, GLP-1 4.5 ± 0.3, P = 0.46) nor the mean LH pulse mass (vehicle 5.7 ± 0.7 IU/L, GLP-1 4.6 ± 0.6 IU/L, P = 0.26). Figure 2. Effects of GLP-1 infusion on reproductive hormone levels, food intake, and nausea. A: Mean serum luteinizing hormone (LH) levels were similar during GLP-1 and vehicle infusions. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.16). MEAL = ad libitum meal. B: There was no significant difference between LH area under the curve (AUC) during GLP-1 infusion compared with vehicle infusion (P = 0.95 using paired t-test). C: Mean serum follicle stimulating hormone (FSH) levels were similar during GLP-1 and vehicle infusions. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.29). MEAL = ad libitum meal.

B: There was no significant difference between LH area under the curve (AUC) during GLP-1 infusion compared with vehicle infusion (P = 0.95 using paired t-test). C: Mean serum follicle stimulating hormone (FSH) levels were similar during GLP-1 and vehicle infusions. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.29). MEAL = ad libitum meal. D: There was no significant difference between FSH area under the curve (AUC) during GLP-1 infusion compared with vehicle infusion (P = 0.86 using paired t-test). E: Mean serum testosterone levels were similar during GLP-1 and vehicle infusions. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.71). MEAL = ad libitum meal. F: There was no significant difference between testosterone area under the curve (AUC) during GLP-1 infusion compared with vehicle infusion (P = 0.77 using paired t-test). G: Food intake was lower during GLP-1 infusion compared to rate-matched vehicle infusion (body weight-adjusted food intake at T = 480 mins: vehicle 15.7 ± 1.3 kcal/kg vs. GLP-1 13.4 ± 1.3 kcal/kg, *P = 0.01 using paired t-test). H: Using visual analogue scales (VAS, 0–10 cm), there was no significant difference in participants’ self-reported (change from baseline) nausea during GLP-1 infusion compared with vehicle infusion. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.26).

G: Food intake was lower during GLP-1 infusion compared to rate-matched vehicle infusion (body weight-adjusted food intake at T = 480 mins: vehicle 15.7 ± 1.3 kcal/kg vs. GLP-1 13.4 ± 1.3 kcal/kg, *P = 0.01 using paired t-test). H: Using visual analogue scales (VAS, 0–10 cm), there was no significant difference in participants’ self-reported (change from baseline) nausea during GLP-1 infusion compared with vehicle infusion. Two-way RM-ANOVA did not detect a significant interaction of treatment (vehicle vs. GLP-1) and time (P = 0.26). There were no significant differences between serum FSH levels and FSH AUC during vehicle and GLP-1 administration (Fig. 2C and 2D). Intravenous GLP-1 administration did not alter serum testosterone levels, did not affect the diurnal variation in testosterone levels, and did not affect testosterone AUC (Fig. 2E and 2F). Additionally, GLP-1 administration did not affect testosterone pulsatility (number of testosterone pulses/500 mins: vehicle 4.3 ± 0.6 vs. GLP-1 4.6 ± 0.4, P = 0.76).

id not alter serum testosterone levels, did not affect the diurnal variation in testosterone levels, and did not affect testosterone AUC (Fig. 2E and 2F). Additionally, GLP-1 administration did not affect testosterone pulsatility (number of testosterone pulses/500 mins: vehicle 4.3 ± 0.6 vs. GLP-1 4.6 ± 0.4, P = 0.76). Effects of GLP-1 on glucose, food intake, and nausea Compared to vehicle administration, plasma glucose levels were lower during GLP-1 administration (vehicle 4.99 ± 0.05 mmol/L vs. GLP-1 4.66 ± 0.06 mmol/L, P < 0.0001). Participants were given an ad libitum meal at T = 480 minutes and intravenous administration of GLP-1 resulted in 15% reduction in food intake (Fig. 2G). Therefore the dose of GLP-1 administered was biologically active. Glucagon-like peptide-1 receptor agonists cause dose-dependent nausea (17). Nausea (similar to other types of stress (18)) may have an inhibitory effect on reproductive hormone release. Therefore, we assessed the participants’ self-reported nausea using a 0 to10 cm visual analogue scale (VAS) preinfusion, midinfusion, and just prior to the ad libitum meal (Fig. 1A). Glucagon-like peptide-1 infusion resulted in similar nausea VAS ratings to vehicle (Fig. 2H).

ory effect on reproductive hormone release. Therefore, we assessed the participants’ self-reported nausea using a 0 to10 cm visual analogue scale (VAS) preinfusion, midinfusion, and just prior to the ad libitum meal (Fig. 1A). Glucagon-like peptide-1 infusion resulted in similar nausea VAS ratings to vehicle (Fig. 2H). Discussion This is the first study investigating the effect of high-dose GLP-1 infusion on reproductive hormone secretion in humans. Our data demonstrate that intravenous infusion of GLP-1, administered at a rate of 0.8 pmol/kg/min for 500 minutes, reduces food intake but does not alter serum levels of reproductive hormones in young healthy men. Fertility is dependent on absolute reproductive hormone levels as well as LH pulsatility (19). Therefore, we also assessed the effect of GLP-1 on LH and testosterone pulsatility and demonstrate that GLP-1 administration does not affect LH and testosterone pulsatility. Our results are in agreement with some published human data but not with the animal literature.

ive hormone levels as well as LH pulsatility (19). Therefore, we also assessed the effect of GLP-1 on LH and testosterone pulsatility and demonstrate that GLP-1 administration does not affect LH and testosterone pulsatility. Our results are in agreement with some published human data but not with the animal literature. In rodents, both stimulatory and inhibitory effects of GLP-1 receptor agonism on reproductive hormones have been reported (5, 7, 8). This might be due to LH-dependent and/or LH-independent mechanisms. For instance, in rodents, testosterone production is suppressed by central administration of norepineprine (20); intravenous administration of a GLP-1RA induces cFos immunoreactivity in catecholamine neurons and increases expression of tyrosine hydroxylase (the enzyme that catalyzes the rate-limiting step of the catecholamine synthesis pathway) (21). However, chronic central administration of a GLP-1RA to rodents reduces urinary norepinephrine secretion (ie, a surrogate marker of norepinephrine production) (22). Therefore, there may be divergent effects of acute versus chronic GLP-1 receptor agonism on catecholamines and testosterone. In the current study, acute administration of GLP-1 to healthy men did not affect testosterone secretion and pulsatility; other groups have reported that chronic administration of GLP-1 receptor agonists increases testosterone levels in obese hypogonadal men (10, 11). Thus, unlike in rodents, acute and chronic GLP-1 receptor agonism does not appear to have adverse effects on testosterone secretion in humans. The reason for this species difference requires further mechanistic study.

nistration of GLP-1 receptor agonists increases testosterone levels in obese hypogonadal men (10, 11). Thus, unlike in rodents, acute and chronic GLP-1 receptor agonism does not appear to have adverse effects on testosterone secretion in humans. The reason for this species difference requires further mechanistic study. In humans, a 6-hour continuous intravenous infusion of 0.4 pmol/kg/min GLP-1 to 9 healthy men during a euglycemic clamp did not affect LH pulsatility or mean serum LH, FSH, and testosterone levels, but GLP-1 administration reduced the number of testosterone pulses with a trend towards increased testosterone pulse duration (9). Volunteers remained euglycemic in both the clamp study (mean plasma glucose was 5 mmol/L during GLP-1 administration and 5.2 mmol/L during saline administration) (9) and in our study (mean plasma glucose was 4.66 mmol/L during GLP-1 administration and 4.99 mmol/L during vehicle administration). As testosterone pulses are not affected by glucose concentrations between 4 and 5.5mmol/L (23, 24), the use of euglycemic clamp methodology would not be expected to influence testosterone pulsatility. Additionally, the differing glucose levels between the two studies do not account for the difference in the effect of GLP-1 administration on testosterone pulsatility reported (ie, reduced number of pulses vs no effect on pulse number respectively). Administration of 0.4 pmol/kg/min GLP-1 does not reduce food intake (12); therefore, the lack of a significant effect on reproductive hormone levels reported may have been due to the low dose of GLP-1 used (9). However, we used a higher dose of GLP-1 (ie, 0.8 pmol/kg/min) that reduced food intake (without causing nausea) but did not alter LH pulsatility, testosterone pulsatility, and circulating reproductive hormone levels.

ant effect on reproductive hormone levels reported may have been due to the low dose of GLP-1 used (9). However, we used a higher dose of GLP-1 (ie, 0.8 pmol/kg/min) that reduced food intake (without causing nausea) but did not alter LH pulsatility, testosterone pulsatility, and circulating reproductive hormone levels. Increases in LH and testosterone levels in hypogonadal obese men with/without type 2 diabetes have been reported following long-term administration of GLP-1RAs (10, 11). In these studies, the men who received GLP-1RA lost more weight than people in the comparator groups. Weight loss is associated with improvement in reproductive hormone levels in obese men with hypogonadism (1). Consequently, the improvement in reproductive hormones reported in these studies may be due to weight loss produced by the GLP-1RAs and not due to a direct effect of GLP-1 receptor agonism.

e comparator groups. Weight loss is associated with improvement in reproductive hormone levels in obese men with hypogonadism (1). Consequently, the improvement in reproductive hormones reported in these studies may be due to weight loss produced by the GLP-1RAs and not due to a direct effect of GLP-1 receptor agonism. Measurement of serum testosterone in this study was performed using a chemiluminescent immunoassay. While this is a well-established assay for the measurement of serum testosterone, using mass spectrometry would have provided greater accuracy. Additionally, in this study the participants consisted solely of healthy eugonadal men who received an acute GLP-1 infusion. Thus, the main limitation of this study is the lack of an obese control group and/or a lack of data on chronic use in healthy men (in contrast to the positive data with chronic liraglutide in obese men). Therefore, we cannot conclude with certainty that the beneficial reproductive effects with chronic GLP1-RA in obese hypogonadal men are due to concomitant weight loss or not. As the prevalence of obesity and type 2 diabetes is increasing in both men and women, and both men and women are receiving GLP-1RAs, further studies are required to determine the direct effects (if any) of GLP-1 and GLP-1RAs on reproductive hormone secretion in women and hypogonadal patients with obesity/diabetes.

not. As the prevalence of obesity and type 2 diabetes is increasing in both men and women, and both men and women are receiving GLP-1RAs, further studies are required to determine the direct effects (if any) of GLP-1 and GLP-1RAs on reproductive hormone secretion in women and hypogonadal patients with obesity/diabetes. Conclusions Our data demonstrate that in healthy eugonadal men, administration of a biologically active dose of GLP-1 has no effect on LH pulsatility and does not alter serum levels of LH, FSH, or testosterone. This important data contributes to our understanding of the interaction between metabolic and reproductive systems in humans. Acknowledgments

Conclusions Our data demonstrate that in healthy eugonadal men, administration of a biologically active dose of GLP-1 has no effect on LH pulsatility and does not alter serum levels of LH, FSH, or testosterone. This important data contributes to our understanding of the interaction between metabolic and reproductive systems in humans. Acknowledgments Financial Support: This article presents independent research funded by the National Institute for Health Research (NIHR) and supported by the NIHR Imperial Clinical Research Facility and NIHR Imperial Biomedical Research Centre at Imperial College Healthcare NHS Trust. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health. The Section of Endocrinology and Investigative Medicine is funded by grants from the UK Medical Research Council (MRC), Biotechnology and Biological Sciences Research Council, and is supported by the NIHR Biomedical Research Centre Funding Scheme. C.I. (MR/M004171/1), R.R. (MR/N020472/1), and L..Y (MR/R000484/1) are funded by MRC Clinical Research Training Fellowships. D.P. is supported by NIHR Clinical Research Network funding. T.T. is supported by grants from the MRC and Wellcome Trust. A.A. is funded by an NIHR Clinician Scientist Award (CS-2018-18-ST2-002). A.N.C. is funded by the NHS. W.S.D. is funded by an NIHR Professorship (RP-2014-05-001). Additional Information: Disclosure Summary: The authors have nothing to disclose.

Obesity is a major global health problem that increases morbidity (1), mortality (2), and has a significant detrimental impact on healthcare budgets (3). In 2016, 39% of adults and 13% of children worldwide were obese, and more people died as a consequence of obesity than undernutrition (4). Bariatric surgery is the most effective treatment for obesity, but it is not universally available or acceptable to patients, and postprocedure complications limit its use (5). Lifestyle modifications are difficult to maintain outside research settings with extended care required for weight loss maintenance (6). Therefore, medications remain the mainstay of obesity management. An emerging class of antiobesity medication is peptide YY (PYY) analogs, with over 20 patents for this new class of medication listed on the European Patent Register (7). Physiologically, PYY is predominantly produced by intestinal L cells in response to nutrient ingestion. Resulting rapid elevations in circulating postprandial PYY levels are directly proportional to the size of the ingested meal (8, 9). PYY has a powerful anorectic effect via activation of central Y2 receptors (10) with central and peripheral administration of PYY dose-dependently reducing food intake by up to 25% in rodents (10). In normal weight, overweight, and obese people, administration of PYY potently suppresses appetite and reduces food intake (by ~30%) (10, 11).

norectic effect via activation of central Y2 receptors (10) with central and peripheral administration of PYY dose-dependently reducing food intake by up to 25% in rodents (10). In normal weight, overweight, and obese people, administration of PYY potently suppresses appetite and reduces food intake (by ~30%) (10, 11). Furthermore, there is evidence that PYY may have other effects in addition to weight loss, which are imperative to decipher given the ongoing development of PYY analogs as antiobesity therapies. PYY stimulates the release of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) from isolated prepubertal rat pituitaries (12). Additionally, PYY administration increases LH and FSH levels in adult male rats, an effect that is potentiated by fasting (13). Secondary hypogonadism occurs in up to 40% of obese men, which is associated with higher body weight and increased insulin resistance (14). Therefore, the anorectic effects of PYY, and the potential to stimulate reproductive hormone release, would be advantageous in the treatment of obesity with coexisting hypogonadism.

econdary hypogonadism occurs in up to 40% of obese men, which is associated with higher body weight and increased insulin resistance (14). Therefore, the anorectic effects of PYY, and the potential to stimulate reproductive hormone release, would be advantageous in the treatment of obesity with coexisting hypogonadism. As there are no reports of the effects of PYY on the reproductive system in humans, we undertook a randomized single-blinded placebo-controlled crossover study to determine the effects of PYY administration on reproductive hormone release in healthy men. Fertility is dependent on absolute reproductive hormone levels as well as LH pulsatility (15). Therefore, we sought to determine the effect of PYY administration on several key parameters of LH secretion, as well as circulating FSH and testosterone levels. Materials and Methods Study participants This study was approved by the West London Research Ethics Committee (16/LO/0391) and performed in accordance with the Declaration of Helsinki. Healthy eugonadal men (aged 18–40 years) were recruited via online and print advertisements. Written informed consent was obtained from each participant prior to study enrolment. Exclusion criteria included: body mass index (BMI) <18.5 or >25 kg/m2, history of medical and psychological conditions, use of prescription, recreational or investigational drugs within the preceding 2 months, blood donation within 3 months of study participation, ingestion or inhalation of nicotine-containing substances within 3 months, alcoholism, and history of cancer.

.5 or >25 kg/m2, history of medical and psychological conditions, use of prescription, recreational or investigational drugs within the preceding 2 months, blood donation within 3 months of study participation, ingestion or inhalation of nicotine-containing substances within 3 months, alcoholism, and history of cancer. Study visits Each participant attended 2 study visits, 1 for PYY administration and 1 for vehicle administration. Infusion order was randomized and participants were blinded to the infusion identity. PYY infusions were prepared by dissolving PYY3-36 (Bachem, UK) in 1 ml of 0.9% NaCl (Braun, UK) and adding the PYY solution to 49 ml of Gelofusine (Braun, UK). PYY was infused at a rate of 0.4 pmol/kg/min, a dose previously established to be biologically active in humans (9, 16). Vehicle infusions consisted of Gelofusine (Braun, UK), administered at the equivalent rate to the PYY infusion for each participant.

K) and adding the PYY solution to 49 ml of Gelofusine (Braun, UK). PYY was infused at a rate of 0.4 pmol/kg/min, a dose previously established to be biologically active in humans (9, 16). Vehicle infusions consisted of Gelofusine (Braun, UK), administered at the equivalent rate to the PYY infusion for each participant. After an overnight fast starting at 10 pm on the night before each study visit, each participant ate a standardized 200 kcal breakfast (1 pot of Oat So Simple® porridge, Quaker Food Products, UK) at 6 am on the morning of each study visit. The participants arrived at the Clinical Research Facility at 8.15 am on the morning of each study visit. After a period of acclimatization, 2 intravenous cannulae (1 in each arm) were inserted (1 for blood samples and 1 to administer the infusion). Following baseline sampling, PYY or vehicle infusion was started at T = 0 minutes (9 am) and infused until T = 500 minutes. Visual analog scales (VASs, 0–10 cm), used to measure participants’ self-reported nausea, were performed at T = –15 minutes, T = 240 minutes, and T = 470 minutes. Blood samples were taken at 10-minute intervals throughout the study (Fig. 1). Participants were not allowed to eat during the infusion until after T = 480 minutes. Figure 1. Study visit protocol.

After an overnight fast starting at 10 pm on the night before each study visit, each participant ate a standardized 200 kcal breakfast (1 pot of Oat So Simple® porridge, Quaker Food Products, UK) at 6 am on the morning of each study visit. The participants arrived at the Clinical Research Facility at 8.15 am on the morning of each study visit. After a period of acclimatization, 2 intravenous cannulae (1 in each arm) were inserted (1 for blood samples and 1 to administer the infusion). Following baseline sampling, PYY or vehicle infusion was started at T = 0 minutes (9 am) and infused until T = 500 minutes. Visual analog scales (VASs, 0–10 cm), used to measure participants’ self-reported nausea, were performed at T = –15 minutes, T = 240 minutes, and T = 470 minutes. Blood samples were taken at 10-minute intervals throughout the study (Fig. 1). Participants were not allowed to eat during the infusion until after T = 480 minutes. Figure 1. Study visit protocol. After a period of acclimatization following their arrival at the Clinical Research Facility, 18 healthy men (mean age 24.1 ± 0.9 years, mean body mass index 22.2 ± 0.4 kg/m2), received an 8-hour infusion of 0.4 pmol/kg/min Peptide YY3-36 (PYY) during one study visit and rate-matched vehicle infusion at a second study visit, in random order. Blood samples were taken at ten-minute intervals starting from 30 minutes before the infusion was started (ie, T = –30 minutes). Visual analog scales (VASs) were completed by the participants pre-infusion (T = –15 minutes), mid-infusion (T = 240 minutes) and prior to the end of the infusion (T = 470 minutes).

om order. Blood samples were taken at ten-minute intervals starting from 30 minutes before the infusion was started (ie, T = –30 minutes). Visual analog scales (VASs) were completed by the participants pre-infusion (T = –15 minutes), mid-infusion (T = 240 minutes) and prior to the end of the infusion (T = 470 minutes). Biochemical analyses PYY was measured using an established in-house radioimmunoassay (17). Serum insulin, plasma glucose, serum LH, serum FSH, and serum testosterone were measured in the Clinical Chemistry Laboratory of Imperial College Healthcare NHS Trust on the automated Abbott Architect® platform. Chemiluminescent immunoassays were used to measure serum insulin (intra-assay and interassay coefficient of variation (CV): ≤7%), serum LH (intra-assay and interassay CV: ≤5%), serum FSH (intra-assay and interassay CV: ≤10%), and serum testosterone (intra-assay and interassay CV: ≤8%). Plasma glucose was measured using a colorimetric hexokinase assay (intra-assay and interassay CV: ≤2%). Statistical analysis LH pulsatility was determined using blinded deconvolution analysis (18). Longitudinal nonindependent data were analyzed with generalized estimating equations (GEEs). Paired t-tests were performed on parametric data and Wilcoxon matched pairs sign rank tests were performed on paired nonparametric data. STATA 14.1 (STATACorp, USA) and Prism 8.0.2 (GraphPad, USA) software were used to perform statistical analyses. P-values <.05 were considered statistically significant. Data are presented as mean ± standard error of the mean.

c data and Wilcoxon matched pairs sign rank tests were performed on paired nonparametric data. STATA 14.1 (STATACorp, USA) and Prism 8.0.2 (GraphPad, USA) software were used to perform statistical analyses. P-values <.05 were considered statistically significant. Data are presented as mean ± standard error of the mean. Results Participants Twenty-four healthy men were recruited, with 18 men (mean age 24.1 ± 0.9 years, mean BMI 22.2 ± 0.4 kg/m2) completing the study. Four men did not complete the study due to their other commitments and 2 were unable to tolerate the PYY infusion due to nausea. Data from the 18 men who completed the study are included in the analyses below. Effects of PYY on LH levels There was no significant difference between serum LH levels during PYY infusion and during vehicle infusion (Fig. 2A). Intravenous PYY infusion did not alter LH pulsatility (mean number of LH pulses/8 hours: PYY 4.4 ± 0.3 vs vehicle 4.4 ± 0.4, P > .99). Furthermore, mean LH (PYY 2.8 ± 0.2 IU/L vs vehicle 3.0 ± 0.2 IU/L, P = .31) and LH area under the curve (AUC) (PYY 1503 ± 79 IU.min/L vs vehicle 1574 ± 86 IU.min/L, P = .36) were not significantly altered by PYY infusion.

r LH pulsatility (mean number of LH pulses/8 hours: PYY 4.4 ± 0.3 vs vehicle 4.4 ± 0.4, P > .99). Furthermore, mean LH (PYY 2.8 ± 0.2 IU/L vs vehicle 3.0 ± 0.2 IU/L, P = .31) and LH area under the curve (AUC) (PYY 1503 ± 79 IU.min/L vs vehicle 1574 ± 86 IU.min/L, P = .36) were not significantly altered by PYY infusion. Figure 2. Effects of intravenous peptide YY (PYY) administration (0.4 pmol/kg/min PYY3-36 for 8 hours) on reproductive hormone secretion in 18 healthy men. (A) Serum luteinizing hormone (LH) levels were similar throughout PYY and rate-matched vehicle infusions (generalized estimating equation (GEE), P = .50). (B) Serum follicle-stimulating hormone (FSH) levels were similar throughout PYY and rate-matched vehicle infusions (GEE, P = .79). (C) Serum testosterone levels were similar throughout PYY and rate-matched vehicle infusions (GEE, P = .53). Effects of PYY on FSH levels Similar to LH, there was no significant difference between serum FSH levels during PYY infusion and during vehicle infusion (Fig. 2B). Additionally, mean FSH (PYY 2.2 ± 0.2 IU/L vs vehicle 2.3 ± 0.2 IU/L, P = .48) and FSH AUC (PYY 1158 ± 513 IU.min/L vs vehicle 1199 ± 476 IU.min/L, P = .49) did not significantly differ between PYY and vehicle infusion.

icant difference between serum FSH levels during PYY infusion and during vehicle infusion (Fig. 2B). Additionally, mean FSH (PYY 2.2 ± 0.2 IU/L vs vehicle 2.3 ± 0.2 IU/L, P = .48) and FSH AUC (PYY 1158 ± 513 IU.min/L vs vehicle 1199 ± 476 IU.min/L, P = .49) did not significantly differ between PYY and vehicle infusion. Effects of PYY on testosterone levels Consistent with the absence of an effect on LH and FSH secretion, intravenous PYY infusion did not alter serum testosterone levels during the 8-hour infusion, and diurnal variation in testosterone levels was unchanged by PYY (Fig. 2C). Similarly, mean testosterone (PYY 19.8 ± 1.3 IU/L vs vehicle 21.1 ± 1.5 IU/L, P = .24) and testosterone AUC (PYY 10 485 ± 684 IU.min/L vs vehicle 11 133 ± 803 IU.min/L, P = .24) were unaffected by PYY infusion. Effects of PYY on nausea and fullness PYY infusion resulted in significantly higher circulating PYY levels than vehicle infusion (Fig. 3A). PYY is known to cause nausea at biologically active doses (9). Therefore, we assessed the effect of the PYY infusion on nausea. PYY infusion resulted in significantly higher self-reported nausea than vehicle infusion over the 8-hour study period peaking midinfusion (Fig. 3B). However, the absolute nausea levels were low (ie, ~2/10 cm). Additionally, PYY infusion resulted in a smaller reduction in the feeling of fullness at T = 480 minutes compared with vehicle infusion (Fig. 3C).

ly higher self-reported nausea than vehicle infusion over the 8-hour study period peaking midinfusion (Fig. 3B). However, the absolute nausea levels were low (ie, ~2/10 cm). Additionally, PYY infusion resulted in a smaller reduction in the feeling of fullness at T = 480 minutes compared with vehicle infusion (Fig. 3C). Figure 3. Effects of intravenous peptide YY (PYY) administration (0.4 pmol/kg/min PYY3-36 for 8 hours) on plasma PYY levels, nausea and fullness in 18 healthy men. (A) PYY infusion resulted in significantly higher plasma PYY levels compared to vehicle infusion (generalized estimating equation (GEE), ****P < .0001). (B) Self-reported nausea, measured using a 0–10 cm visual analog scale (VAS), was increased by PYY infusion compared to vehicle infusion (generalized estimating equation (GEE), ***P < .001). C: Self-reported fullness, measured using a 0–10 cm visual analog scale (VAS), reduced by a lesser extent at T = 480 minutes during PYY infusion compared with vehicle infusion (Wilcoxon matched-pairs signed rank test, **P = .01).

nfusion compared to vehicle infusion (generalized estimating equation (GEE), ***P < .001). C: Self-reported fullness, measured using a 0–10 cm visual analog scale (VAS), reduced by a lesser extent at T = 480 minutes during PYY infusion compared with vehicle infusion (Wilcoxon matched-pairs signed rank test, **P = .01). Discussion This is the first study investigating the effects of PYY on the reproductive system in humans. Our study demonstrates that, an 8-hour infusion of 0.4 pmol/kg/min PYY3-36 does not alter LH pulsatility in healthy men and does not change circulating levels of LH, FSH, and testosterone. This is in contrast to rodent studies, where PYY administration (centrally or peripherally) had specific effects on reproductive hormone levels depending on pubertal status and route of administration as follows. Incubation of pituitaries from prepubertal rats with PYY results in increased LH and FSH secretion within 60 minutes (12). However, incubation of PYY with hypothalamic fragments from prepubertal and adult rats results in reduced gonadotrophin-releasing hormone (GnRH) secretion (12, 13). Central administration of PYY to prepubertal rats inhibited LH secretion but did not affect FSH secretion (12). In contrast, intracerebroventricular administration of PYY increases LH and FSH in fed and fasted adult male rats, and co-administration of a GnRH antagonist abolishes these effects (13). However, peripheral administration of PYY (via intraperitoneal injection) has no effect on LH secretion but increases FSH secretion in prepubertal male rats (12).

icular administration of PYY increases LH and FSH in fed and fasted adult male rats, and co-administration of a GnRH antagonist abolishes these effects (13). However, peripheral administration of PYY (via intraperitoneal injection) has no effect on LH secretion but increases FSH secretion in prepubertal male rats (12). It is possible that no effect on reproductive hormone secretion was detected in response to PYY administration in this study due to the peripheral route of administration. As outlined in the rodent studies above, central administration of PYY stimulates gonadotrophin secretion in adult male rats (13), whereas peripheral administration has been shown to increase FSH (but not LH) in prepubertal male rats (12). The anorectic effects of PYY are thought to be mediated by agonism of the Y2 receptor (10, 19). However, the effect of PYY on reproductive hormone secretion may not occur via the Y2 receptor. Central administration of a selective Y2 receptor agonist reduces LH and FSH secretion in adult male rats, while administration of a Y2 receptor antagonist increases LH and FSH secretion in adult male rats (20). In contrast, central administration of PYY to adult male rats stimulates LH and FSH secretion (13). In the present study, peripheral PYY administration increased nausea but had no effect on reproductive hormone levels. Therefore, Y2 agonism in healthy men does not modulate reproductive hormone secretion.

ale rats (20). In contrast, central administration of PYY to adult male rats stimulates LH and FSH secretion (13). In the present study, peripheral PYY administration increased nausea but had no effect on reproductive hormone levels. Therefore, Y2 agonism in healthy men does not modulate reproductive hormone secretion. Although participants were instructed to fast overnight and eat a standardized breakfast (which was provided to them in advance), the overnight fast and consumption of the standardized breakfast were not monitored. However, this limitation is unlikely to have significantly influenced the results as all participants were closely monitored as fasting throughout the 8-hour infusion and no reproductive hormone changes were evident at any point. In this present study, PYY administration did not alter LH, FSH, and testosterone levels in healthy (non-obese) men. Further studies are required to determine the effects of PYY on reproductive hormone secretion in obese people, ie, the target population for PYY-based therapeutic agents. Conclusions Although animal data suggest that PYY affects reproductive hormone secretion, our data demonstrate that in humans, acute administration of a biologically active dose of PYY does not have an adverse effect on LH pulsatility and does not alter levels of LH, FSH, and testosterone. This has important clinical and safety implications for the continuing development of PYY analogs for the treatment of obesity. Abbreviations BMIbody mass index GEEgeneralized estimating equation GnRHgonadotrophin-releasing hormone FSHfollicle-stimulating hormone

Conclusions Although animal data suggest that PYY affects reproductive hormone secretion, our data demonstrate that in humans, acute administration of a biologically active dose of PYY does not have an adverse effect on LH pulsatility and does not alter levels of LH, FSH, and testosterone. This has important clinical and safety implications for the continuing development of PYY analogs for the treatment of obesity. Abbreviations BMIbody mass index GEEgeneralized estimating equation GnRHgonadotrophin-releasing hormone FSHfollicle-stimulating hormone LHluteinizing hormone PYYpeptide YY VASvisual analog scale Acknowledgments This article presents independent research funded by the NIHR and supported by the NIHR CRF and BRC at Imperial College Healthcare NHS Trust. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

LHluteinizing hormone PYYpeptide YY VASvisual analog scale Acknowledgments This article presents independent research funded by the NIHR and supported by the NIHR CRF and BRC at Imperial College Healthcare NHS Trust. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Financial Support: The Section of Endocrinology and Investigative Medicine is funded by grants from the Medical Research Council (MRC), Biotechnology and Biological Sciences Research Council (BBSRC), National Institute for Health Research (NIHR), an Integrative Mammalian Biology (IMB) Capacity Building Award, an FP7- HEALTH- 2009-241592 EuroCHIP grant and is supported by the NIHR Biomedical Research Centre Funding Scheme. C.I. is funded by an MRC Clinical Research Training Fellowship (MR/M004171/1). D.P. is supported by NIHR CLRN funding. R.R. is funded by an MRC Clinical Research Training Fellowship (MR/N020472/1). L.Y. is funded by an MRC Clinical Research Training Fellowship (MR/R000484/1). T.T. is supported by grants from the MRC and Wellcome Trust. A.A. is funded by an NIHR Clinician Scientist Award (CS-2018-18-ST2-002). A.N.C. is funded by the NHS. W.S.D. is funded by an NIHR Professorship (RP-2014-05-001). Additional Information Disclosure Summary: The authors have nothing to disclose.