Luteinizing hormone and growth hormone secretion in ewes infused intracerebroventricularly with neuropeptide Y

2.1. Animal care 

McShane et al. [6] have previously described animal care procedures for lambs subjected to single i.c.v. injections of NPY in experiment 1. In experiments 2 and 3, use of animals for continuous infusions of NPY was approved by the University of Missouri-Columbia, Animal Care and Use Committee (protocol #3273). Prior to the experiments, postpubertal yearling ewes were ovariectomized (OVX) and fitted with two lateral cerebroventricular cannulae [6] for the infusion of saline or porcine NPY (Penninsula Laboratories, Belmont, CA). Ewes were housed indoors under a 12h light/12h dark photoperiod and were maintained in individual 2.5m2 pens when not on study. During treatments, ewes were maintained in stainless steel metabolic pens, and prior to the beginning of each experiment were fitted with indwelling jugular cannulae for the collection of repeated blood samples.

2.2. Experiment 1: effect of a single i.c.v. injection of NPY on serum GH 

Four OVX ewes and four OVX ewes treated with subcutaneous estradiol-17β (E) implants [22] were randomly assigned to receive varying doses of NPY i.c.v. in a 4×4 Latin Square arrangement of treatments as described by McShane et al. [6]. Ewes received i.c.v. injections of 0, 0.5, 5, or 50μg of porcine NPY. Treatments were replicated over four periods, each separated by 1 week, such that each ewe received each dose of NPY. On the morning of each treatment period, ewes were fed between 07:00 and 09:00h, after which feed was removed. At 10:00h, blood was collected every 20min for 4h prior to and for 4h following an i.c.v. injection of NPY. Samples were stored at 4°C until centrifugation, and serum was stored at −20°C until the determination of serum GH via radioimmunoassay.

2.3. Experiment 2: effect of a 20-h continuous, linearly increasing i.c.v. infusion of NPY 

Ten ewes, averaging 32±0.9kg were used to determine the effects of continuous infusion of NPY on pulsatile secretion of LH and GH. Ewes were randomly assigned to receive either NPY or saline for a 20-h infusion period. Delivery of the infusate was achieved by a computer-controlled syringe pump that administered incremental increases in volume every 10min, thus producing a constant, linearly increasing delivery of the infusate. A maximum dose of 50μg/h of NPY (at 1ml/h) was reached at hour 20, with half the maximum dose being delivered at hour 10. Control ewes received a volume of saline equal to that of the NPY-treated ewes. Blood samples were collected via jugular cannulae every 10min during hour −4–0 (interval 1, pre-treatment), hour 6–10 (interval 2) and hour 16–20 (interval 3) relative to the initiation of infusion (0h). Blood samples were allowed to clot overnight at 4°C, and serum LH and GH were determined via radioimmunoassay. Ewes were fed immediately before the infusion, and were allowed access to feed during the 20-h infusion. Feed remaining after 20h was weighed as a measure of cumulative feed intake.

2.4. Experiment 3: effect of a 10-h continuous, linearly increasing i.c.v. infusion of NPY 

Four OVX ewes were assigned to receive an infusion of NPY in a continuous, linearly increasing dose as in experiment 2, except that: (a) ewes were infused so as to achieve a maximum dose of 50μg/h NPY at hour 10; rather than at hour 20 as is experiment 2 and (b) blood samples were collected from ewes every 10min from −4 to 14h relative to initiation (0h) of the 10-h infusion; rather than only prior to and during infusions as in experiment 2. The specific objective of experiment 3, therefore, was to more accurately characterize the temporal patterns of secretion of LH and GH within each animal prior to, during, and following infusion of NPY, rather than as in experiment 2, which was only prior to and at defined intervals during infusion of NPY.

2.5. Radioimmunoassays 

Blood samples were allowed to clot overnight at 4°C and serum was harvested and stored at −20°C until assayed. Serum concentrations of LH and GH were determined via double antibody radioimmunoassay procedures previously validated in our laboratory [23], [24]. Briefly, serum concentrations of both LH and GH were separately assayed in duplicate 200μl serum samples. Luteinizing hormone was analyzed using RAoLH TEA #35 heterologous ovine antisera [25] and reagents provided by the National Institute of Diabetes and Digestive and Kidney Diseases National Hormone and Pituitary Program (NIDDK-NHPP; courtesy Dr. A.F. Parlow). Growth hormone was analyzed using AFPC0123080 anti-ovine growth hormone primary antisera and other reagents also provided by the NIDDK-NHPP (courtesy Dr. A.F. Parlow). The GH and LH intra- and inter-assay coefficients of variation were: GH 4 and 11%; LH 5 and 10%, respectively.

2.6. Statistical analysis 

In experiment 1, samples were classified as pre- and post-i.c.v. injection of treatment, and the effects of treatment (NPY versus saline), estradiol-17β (present versus absent), time (pre- versus post-injection), dose (0, 0.5, 5, and 50μg) and all interactive effects on mean GH were tested. In experiment 2, GH and LH values within animal within each 4-h sampling interval were characterized using the CLUSTER pulse analysis program [26], thereby generating mean, pulse frequency, and pulse amplitude values for each animal within each interval. The effects tested were treatment, sampling interval (or time of infusion), and the two-way interaction. The main effect of treatment was tested using ewe within treatment as the error term, and the residual error was used to test all other effects. In experiment 3, the length of the sampling intervals were inappropriate to test for effects on pulse frequency, and consequently hourly hormone means were computed, and the effects of time relative to the onset of infusion within each animal were tested.

All data were analyzed as repeated measures using the mixed model procedures (PROC MIXED) of SAS [27]. Multiple covariance structures were tested and the model fitting statistics were compared to determine the best fitting model. Mean separation was performed using the LSMEANS statement with the PDIFF option, and therefore represent least significant differences tests for pre-planned comparisons. Data are presented as least-squares means±S.E.M.

3.1. Experiment 1: effect of a single i.c.v. injection of NPY on serum GH 

Treatment of OVX lambs with E implants resulted in a significant increase in serum E concentrations (OVX 2.4±0.31pg/ml versus OVX+E 7.26±0.33pg/ml; P<0.01). Peripheral concentrations of GH were affected by both NPY treatment (pre- versus post-injection, P=0.035) and dose of NPY injected (P=0.049). More specifically, a single i.c.v. injection of either 0, 0.5, or 5μg of NPY did not significantly affect mean GH as compared to pre-injection levels; however, ewes receiving a single 50μg injection had greater mean concentrations of GH during the 4-h post-injection period as compared to the 4-h pre-injection period (P=0.003; Fig. 1).

View Large Image Download to PowerPoint

Fig. 1. Mean serum concentrations of GH following a single i.c.v. injection of NPY. NPY increased mean GH at the highest dose (50μg), regardless of estrogen treatment (P=0.003). Means consist of averaged concentrations of GH determined at 20-min intervals during the 4-h interval prior to and the 4-h interval following an i.c.v. injection of NPY.

3.2. Experiment 2: effect of a 20-h continuous, linearly increasing i.c.v. infusion of NPY 

Treatments (NPY versus saline) were delivered as a 20-h linearly increasing i.c.v. infusion to examine the effects of both chronic NPY exposure and dose of NPY on secretion of LH and GH. In contrast to experiment 1, neither mean GH nor GH pulse frequency differed in saline versus NPY-infused ewes (P>0.23; Fig. 2). Alternatively, chronic infusion of NPY significantly reduced serum concentrations of LH. Mean LH and LH pulse frequencies (Fig. 3) were similar between NPY versus saline-treated ewes prior to delivery of treatments (interval 1; P>0.25), but then were dramatically reduced in NPY versus saline-treated ewes during interval 2 and 3 (P<0.02). Individual LH profiles in Fig. 4 illustrate the robust inhibitory effect of NPY on LH. Also of interest, during the 20-h i.c.v. infusion, NPY-treated ewes consumed 1216±241.5g of feed and saline-treated ewes consumed 880±241.5g of feed (P=0.36).

View Large Image Download to PowerPoint

Fig. 2. Mean serum concentrations of GH and GH pulse frequency during a 20-h infusion of NPY. Blood samples were collected at 10-min intervals during three 4-h sampling intervals: hour −4–0 (interval 1, pre-treatment), hour 6–10 (interval 2), and hour 16–20 (interval 3). No influence of NPY on serum GH was detected (P>0.23).

View Large Image Download to PowerPoint

Fig. 3. Mean serum concentrations of LH and LH pulse frequency during a 20-h infusion of NPY. Samples were collected at 10-min intervals during three 4-h sampling intervals: hour −4–0 (interval 1, pre-treatment), hour 6–10 (interval 2), and hour 16–20 (interval 3). NPY significantly suppressed both mean LH and LH pulse frequency during sampling intervals 2 and 3 (P<0.02).

View Large Image Download to PowerPoint

Fig. 4. Individual LH profiles during the three sampling intervals for lambs receiving either NPY (left column; profiles for four lambs) or saline (right column; profiles for four lambs). Samples were collected at 10-min intervals during three 4-h sampling intervals: hour −4–0 (interval 1, pre-treatment), hour 6–10 (interval 2), and hour 16–20 (interval 3) with each box representing a single animal. NPY markedly decreased the secretion of LH during sampling intervals 2 and 3.

3.3. Experiment 3: effect of a 10-h continuous, linearly increasing i.c.v. infusion of NPY 

In experiment 3, serum LH decreased significantly over time (P=0.0001; Fig. 5) as in experiment 2, with mean LH beginning to decrease within 2h following the start of NPY infusion. Mean LH was significantly lower by hour 3 (versus hour −4 to −2; P<0.01), and remained suppressed throughout the duration of the NPY infusion. LH subsequently increased approximately 2h after the cessation of infusion of NPY to levels not different from pre-treatment values (P>0.15). Serum GH also changed significantly over time (P=0.001; Fig. 5). Mean GH tended to increase after 2h of NPY infusion (hour 2 versus hour −4, −3, and −1; P<0.08), however this tendency was transient as mean GH returned to levels not different from pre-treatment values for the duration of the infusion. Unexpectedly, peripheral concentrations of GH again increased following the end of the infusion of NPY, such that GH levels were significantly greater 2h after the end of the infusion compared with the final 2h of infusion (hour 9 and 10 versus hour 12 and 13; P<0.01).

View Large Image Download to PowerPoint

Fig. 5. Effect of a 10-h infusion of NPY on mean hourly LH or GH (open squares). Pattern of NPY delivery is also depicted (dark line). Mean LH levels significantly decreased by infusion hour 3 (P<0.01; vs. hour −4 to −2), and then increased approximately 2h after the end of infusion such that levels were no longer different from pre-treatment (P>0.15). Mean GH transiently increased by infusion hour 2 (P<0.08; hour 2 vs. hour −4, −3, and −1). Mean GH increased again in response the end of NPY infusion (P<0.01; hour 9 and 10 vs. hour 12 and 13).