J. Anim. Sci. 2003. 81:2311-2321
© 2003 American Society of Animal Science
Endocrine responses in mares and geldings with high body condition scores grouped by high vs. low resting leptin concentrations1
J. A. Cartmill,
D. L. Thompson, Jr.2,
W. A. Storer,
L. R. Gentry and
N. K. Huff
Department of Animal Sciences, Louisiana Agricultural Experiment Station, Louisiana State University Agricultural Center, Baton Rouge 70803
 |
Abstract
|
|---|
Previous observations from this laboratory indicated that horses with high BCS could have resting plasma leptin concentrations ranging from low (1 to 5 ng/mL) to very high (10 to 50 ng/mL). To study the possible interactions of leptin secretion with other endocrine systems, BCS and plasma leptin concentrations were measured on 36 mares and 18 geldings. From mares and geldings that had a mean BCS of at least 7.5, five with the lowest (low leptin) and five with the highest (high leptin) leptin concentrations were selected. Jugular blood samples were collected twice daily for 3 d from the 20 selected horses to determine average resting hormone concentrations. Over the next 12 d, glucose infusion, injection of thyrotropin-releasing hormone (TRH), exercise, and dexamethasone treatment were used to perturb various hormonal systems. By design, horses selected for high leptin had greater (P < 0.0001) leptin concentrations than horses selected for low leptin (14.1 vs. 2.8 ± 0.92 ng/mL, respectively). In addition, mares had greater (P = 0.008) leptin concentrations than geldings. Horses selected for high leptin had lower (P = 0.027) concentrations of GH but higher (P = 0.0005) concentrations of insulin and thriiodothyronine (T3) than those selected for low leptin. Mares had greater (P = 0.0006) concentrations of cortisol than geldings. There was no difference (P > 0.10) in concentrations of IGF-1, prolactin, or thyroid-stimulating hormone (TSH). Horses selected for high leptin had a greater (P = 0.0365) insulin response to i.v. glucose infusion than horses selected for low leptin. Mares had a greater (P = 0.0006) TSH response and tended (P = 0.088) to have a greater prolactin response to TRH than geldings; the T3 response was greater (P = 0.047) in horses selected for high leptin. The leptin (P = 0.0057), insulin (P < 0.0001), and glucose (P = 0.0063) responses to dexamethasone were greater in horses selected for high leptin than in those selected for low leptin. In addition, mares had a greater (P < 0.0001) glucose response to dexamethasone than geldings. Cortisol concentrations were decreased (P = 0.029) by dexamethasone equally in all groups. In conclusion, differences in insulin, T3, and GH associated with high vs. low leptin concentrations indicate a likely interaction of these systems with leptin secretion in horses and serve as a starting point for future study of the cause-and-effect nature of the interactions.
Key Words: Horses • Insulin • Leptin • Somatotropin • Triiodothyronine
|
Introduction
|
|---|
Gentry et al. (2002a)
reported that leptin concentrations in adult mares with high BCS in September tended to vary widely, and mares tended to fit into two distinct groups based on leptin concentrations: low (<5 ng/mL) or high (7 to 20 ng/mL). Subsequent observations of 18 of those same mares indicated that the high vs. low distinction was consistently observed 2 yr later (our unpublished results), indicating that the underlying cause was relatively permanent. Leptin secretion by adipocytes has been reported to be affected by various other hormones, including insulin (Sivitz et al., 1998; Ramsay and White, 2000), GH (Isozaki et al., 1999), glucocorticoids (Wang et al., 2002; Cartmill et al., 2003), epinephrine (Cammisotto and Bukowiecki, 2002), prolactin (Mastronardi et al., 2000), and thyroid hormones (Ghizzoni et al., 2001; Nowak et al., 2002; Cartmill et al., 2003). The purpose of the current experiment was to test the hypothesis that the high vs. low leptin concentrations observed in horses of good body condition were due to some interaction of leptin secretion with other endocrine systems in the horse. Specific hormonal systems studied were insulin, GH/IGF-I, prolactin, adrenal glucocorticoids, and thyroid hormones.
|
Materials and Methods
|
|---|
Selection of Horses
All horses were of light horse breeds from the resident herd at the Louisiana Agricultural Experiment Station–Central Stations Horse Farm. They were routinely maintained on native grass pastures during the spring, summer, and fall and on ryegrass pasture in winter. On September 25, a total of 36 mares and 18 geldings was assigned a BCS, as described by Henneke et al. (1983; 1 = extremely emaciated through 9 = extremely fat), by two independent, experienced technicians; the mean of the two estimates were used. Previous studies have documented that there are high correlations between direct estimates of empty body fat and backfat thickness determined by ultrasonography (Kane et al., 1987) and between BCS and backfat thickness determined by ultrasonography (Gentry, 2001). Jugular blood samples were collected from each horse into heparinized tubes on the day of BCS estimation for determination of initial leptin concentration. Of the mares and geldings that had a mean BCS of at least 7.5, five each were selected that had the lowest (low leptin) and highest (high leptin) leptin concentrations. Preliminary ANOVA indicated that BCS were approximately equivalent for the low- and high-leptin groups, whereas geldings in the high-leptin group had slightly higher BCS on average (8.6 vs. 7.8 to 7.9 ± 0.3 for the other groups; P = 0.052). A similar analysis of age indicated that horses selected for low vs. high leptin had similar mean ages (10.2 vs. 12.2 yr ± 1.8 yr, respectively), whereas geldings were older (13.7 yr) than mares (8.7 yr; P = 0.013).
Daily Monitoring and Challenges
To establish baseline daily concentrations of the various hormones of interest, samples of jugular blood were collected at 12-h intervals beginning at 0700 on October 4 and continuing through 1900 on October 6. The horses were brought in from pasture on each occasion and were sampled within 1 h or less. These samples were drawn into evacuated, heparinized tubes, and the resulting plasma was stored at -15°C.
For the subsequent challenges (manipulations) that were used to perturb the hormonal systems of interest, the general procedure was that horses were fitted with jugular catheters, loosely tethered in a shed, and allowed to rest at least 1 h before blood sampling was initiated. Frequent blood samples from the challenges were placed into tubes containing heparin and were centrifuged within 15 min; plasma was stored at -15°C. There was a minimum of 2 d between consecutive challenges, administered in the following order: i.v. glucose infusion, i.v. thyrotropin-releasing hormone (TRH) injection, exercise, and i.m. dexamethasone injection. The doses of glucose (Sticker et al., 1995a), TRH (Thompson and Nett, 1984), and dexamethasone (Cartmill et al., 2003) were the same as those used in previous experiments; all three were purchased from Sigma Chemical Co. (St. Louis, MO).
For the glucose infusion administered on October 7, all horses were deprived of feed overnight and were administered glucose i.v. in the morning. Glucose (0.2 g/kg of BW as a 0.5-g/mL solution in 0.155 M saline) was administered through the jugular catheter. Blood samples drawn at -10, 0, 5, 10, 15, 20, 25, 30, 45, 60, 90, 120, 150, and 180 min relative to onset of infusion were used to assess insulin response to glucose.
An i.v. TRH challenge (4 µg/kg of BW) was administered to each horse at 1400 on October 10. Blood samples collected at -10, 0, 10, 20, 30, 45, 60, 90, 120, 150, and 180 min relative to injection of TRH were used to assess the responses of prolactin, thyroid-stimulating hormone (TSH), and thyroid hormones.
A brief exercise bout on October 13 was used to assess the responses of GH, prolactin, cortisol, and leptin. Beginning in the morning, each horse was lunged in a circular pen for 5 min, usually at a trot, but with occasional cantering. This regimen has been used previously as a secretory stimulus for GH and prolactin in horses (Thompson et al., 1994). Blood samples were collected at -10, 0, 10, 20, 30, 45, 60, 90, 120, 150, and 180 min relative to onset of exercise.
Lastly, on October 16 at 0700, each horse was administered dexamethasone at 125 µg/kg of BW as an i.m. injection in corn oil. Blood samples were collected immediately before injection (time 0) and then at 12-h intervals through 0700 on October 20; daily samples were collected at 0700 on October 22, 24, and 26. Plasma from these samples was used for the measurement of leptin, insulin, glucose, IGF-I, and cortisol.
Leptin Antiserum Preparation and Assay Development
A 16-amino acid fragment of porcine leptin (KQRVTGLDFIPGLHPV) encompassing the connecting loop between the A and B 
-helices (Richards et al., 2000) was purchased from Sigma-Genosys (The Woodlands, TX) and subsequently conjugated to succinylated keyhole limpit hemocyanin via the water-soluble carbodiimide reaction enhanced with N-hydroxysulfosuccinimide (Staros et al., 1986; 7.2 mg of peptide, 20 mg of hemocyanin, 62 mg of N-hydroxysulfosuccinimide, and 55 mg of 1-ethyl-3[3-dimethylamino-propyl] carbodiimide-HCl in 4.7 mL of 0.06 M phosphate buffer at pH 6). The reaction was allowed to proceed overnight at room temperature, after which the conjugate was sequentially dialyzed against deionized water and 0.01 M phosphate-buffered (pH = 7.4) 0.155 M saline containing 1.0 g/L of sodium azide. The resulting conjugate was diluted to a final concentration of 1.7 mg/mL. Two rabbits each were immunized with 1.0 mg of the conjugate emulsified in 1.0 mL of Freund’s complete adjuvant, injected into 12 to 14 s.c. sites along the back, haunches, and shoulders. Booster injections of 1.0 mg of conjugate emulsified in Freund’s incomplete adjuvant were similarly injected at 23, 51, 67, and 89 d after the primary immunization. Serum was harvested from each rabbit 10 d after the final booster injection. Of the two rabbits, one produced antiserum that proved useful for radioimmunoassay (KLH-red). This antiserum bound approximately 25% of radiolabeled human recombinant leptin at a final tube dilution of 1:32,000 (800 µL = total volume).
For the assay of leptin in horse plasma, duplicate aliquots of 100 µL of plasma were mixed with 100 µL of assay buffer (PBS containing 1.0 g/L of gelatin and 0.05% Triton-X100) in 12- x 75-mm disposable glass tubes. Eight levels of porcine leptin standards in saline were run in the same manner (0.66 to 84 ng/mL), as well as six buffer controls (zero standard) and six nonspecific binding (0.95% normal rabbit serum) tubes. After the addition of 200 µL of leptin antiserum (diluted 1:8,000 in PBS containing 0.05 M EDTA and 0.95% normal rabbit serum) to all tubes except the nonspecific binding tubes, all tubes were vortexed briefly and incubated at 5°C for 48 h. A fixed amount of radioiodinated human recombinant leptin (approximately 40 nCi) was then added to all tubes in 200 µL of PBS containing 1.0 g/L of gelatin; the tubes were vortexed briefly and incubated for 24 h at 5°C. Precipitation of the rabbit gamma globulin was achieved by the addition of 200 µL of anti-rabbit gamma globulin serum (diluted 1:4.5) generated in a sheep. After the tubes were vortexed, they were incubated for at least 24 h at 5°C. Final separation of antibody-bound leptin from free leptin was achieved by centrifugation of the tubes at 1,200 x g for 30 min at 5°C and decanting of the supernantant; the remaining pellets were washed once with 1.0 mL of PBS at 5°C followed by a second centrifugation. Radioactivity in the pellets was assessed by solid scintillation counting for 1 min. The resulting data were used in a logit-log transformation analysis, and leptin concentrations in unknowns were determined from the predictive equation of the standard curve regression analysis.
Leptin Assay Validation
Inhibition curves generated by serial dilution of a pool of horse plasma, previously estimated to be 20 ng/mL of leptin (multispecies leptin kit, Linco Research Inc., St. Charles, MO; McManus and Fitzgerald, 2000; Cartmill et al., 2003), and the porcine leptin standard were parallel (Figure 1). Cross-reactivity assessments using commercially available preparations and a pituitary extract of known hormonal content indicated no practical interference in the assay by these hormones (Table 1). Fractionation of a 1-mL aliquot of equine plasma (20 ng/mL) on a 0.8- x 21-cm column of Sephadex G-200 followed by assay of the fractions indicated one major peak (93%) of immunoreactivity (Figure 2) coincident with the elution of radiolabeled human leptin plus a minor amount (7%) of activity in the first few fractions off the column. Recovery of graded amounts porcine leptin added to a pool of equine plasma selected for low leptin concentration resulted in quantitative recovery (105%) with high correlation between leptin added and that measured (r = 0.99). An assay of five plasma pools that had been previously measured in the Linco multispecies assay resulted in very good agreement between this assay and those previous values (r = 0.98; ng/mL measured = 0.96 x ng/mL in Linco assay + 2.3). Estimates of the intra- and interassay CV averaged 6 and 4%, respectively. Sensitivity of the assay, based on a 100-µL sample size, was 0.2 ng/mL.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 1. Inhibition curves generated in the equine leptin assay with recombinant porcine leptin standard and a pool of equine plasma.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Figure 2. Leptin immunoreactivity detected by the equine leptin assay in sequential fractions off a column of Sephadex G-200 eluted with phosphate-buffered saline. One milliliter of a pool of horse plasma selected for high leptin immunoreactivity was added to the column.
|
|
Other Sample Analyses
Concentrations of glucose were determined spectrophotometrically (Pointe Scientific, Inc., Lincoln Park, MI). Plasma concentrations of insulin (DePew et al., 1994), IGF-I (Sticker et al., 1995b), GH (Thompson et al., 1992), prolactin (Colborn et al., 1991), and TSH (Sticker et al., 2001) were assessed by RIA previously validated for horse samples. Plasma concentrations of cortisol (Diagnostic Systems Laboratories, Webster, TX), triiodothyronine (T3), and thyroxine (ICN Pharmaceuticals, Costa Mesa, CA; kits 07-290102, and 07-292102, respectively) were assessed with commercially available RIA reagents. Intra- and interassay CV and assay sensitivities were 5%, 8%, and 0.5 µIU/mL for insulin; 5%, 12%, and 2 ng/mL for IGF-I; 8%, 11%, and 0.5 ng/mL for GH; 7%, 12%, and 0.2 ng/mL for prolactin; 5%, 8%, and 0.02 ng/mL for TSH; 6%, 8%, and 0.11 µg/dL for cortisol; 5%, 8%, and 3.7 ng/dL for T3; and 5%, 9%, and 1.2 µg/dL for thyroxine.
Statistical Analyses
Data were analyzed using the PROC GLM procedure of SAS (SAS Inst., Inc., Cary, NC) in a 2 x 2 factorial ANOVA (gender and group fixed; Steel and Torrie, 1980); repetitive sampling was taken into account when appropriate (split-plot; Gill and Hafs, 1971). Gender, group (low leptin or high leptin), and their interaction were tested with the horses within gender-group term; time and its interactions with gender and group were tested with residual error. Differences among the four groups of horses within each time period were compared with the LSD test (Steel and Torrie, 1980) when a significant effect of group or an interaction involving group was detected; because four means were being compared within each time period, only differences at the P < 0.01 level were considered to be meaningful in these comparisons.
|
Results
|
|---|
Daily Hormone Concentrations
Plasma leptin concentrations in horses selected for high leptin averaged 14.1 ng/mL over the 3-d period of daily sampling (Figure 3a), which was greater (P < 0.0001) than the 2.8 ng/mL in horses selected for low leptin. In addition, mares had greater (P = 0.008) concentrations of leptin than geldings (10.4 vs. 6.6 ng/mL). Horses selected for high leptin had lower (P = 0.027) daily GH concentrations over the 3-d period than horses selected for low leptin (Figure 3b). In contrast, horses selected for high leptin had greater (P = 0.0005) daily insulin concentrations than horses selected for low leptin (Figure 3c).
View larger version (14K):
[in this window]
[in a new window]
|
Figure 3. Resting concentrations of leptin (a), GH (b), and insulin (c) in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar in a panel indicates the LSD value (P < 0.01) for comparison among groups within each time period. Pooled SEM were 0.95 ng/mL for leptin, 0.86 ng/mL for GH, and 0.83 µIU/mL for insulin.
|
|
Cortisol concentrations in daily samples were greater (P = 0.0006) in mares than in geldings, and there was an interaction (P = 0.03) between gender and group (Figure 4a); mares selected for high leptin had greater cortisol concentrations than mares selected for low leptin, whereas the opposite was true for geldings. There was also a time trend (P = 0.084) in cortisol concentrations, with the highest concentrations observed in the mornings.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 4. Resting concentrations of cortisol (a), triiodothyronine (T3; b), and thyroxine (c) in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar in a panel indicates the LSD value (P < 0.01) for comparison among groups within each time period. Pooled SEM were 0.65 µg/dL for cortisol, 2.3 ng/dL for T3, and 0.21 µg/dL for thyroxine. Thyroxine concentrations were not affected (P = 0.60) by group.
|
|
Horses selected for high leptin had greater (P = 0.0005) daily concentrations of T3 than horses selected for low leptin (Figure 4b
). Moreover, there was an interaction (P = 0.058) of group x time, with T3 concentrations decreasing more over the 3-d period in the high- vs. low-leptin group. In contrast to T3 concentrations, there was no effect of group (P = 0.6) on concentrations of thyroxine. However, there was an interaction (P = 0.015) of gender, group, and time for concentrations of thyroxine (Figure 4c). Mares selected for low leptin tended to have greater thyroxine concentrations than those selected for high leptin for most but not all time periods, whereas in geldings the opposite was true. There was no effect (P > 0.10) of group or gender for daily TSH, IGF-1, or prolactin concentrations (Figure 5).
View larger version (14K):
[in this window]
[in a new window]
|
Figure 5. Resting concentrations of thyroid-stimulating hormone (TSH; a), IGF-I (b), and prolactin (c) in mares (M) and geldings (G) selected for high vs. low leptin concentrations. Pooled SEM were 0.01 ng/mL for TSH, 3.7 ng/mL for IGF-I, and 0.87 ng/mL for prolactin. There was no effect (P > 0.10) of group for any hormone.
|
|
Responses to Challenges
Horses selected for high leptin had a greater (P = 0.036) insulin response to glucose infusion than did horses selected for low leptin (Figure 6).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 6. Concentrations of insulin following i.v. infusion of glucose in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar indicates the least significant difference (LSD) value (P < 0.01) for comparison among groups within each time period. Pooled SEM was 1.7 µIU/mL.
|
|
Mares had a greater (P = 0.0006) TSH response to the i.v. injection of TRH than geldings (Figure 7a), however there was no effect (P > 0.1) of group. In addition, mares tended (P = 0.088) to have a greater prolactin response to TRH injection (Figure 7b). Average concentrations of T3 during the TRH challenge were greater (P = 0.047) in horses selected for high leptin compared to horses selected for low leptin (Figure 7c), and there was an interaction between group and time (P = 0.022) for concentrations of thyroxine (Figure 7d).
View larger version (30K):
[in this window]
[in a new window]
|
Figure 7. Concentrations of thyroid stimulating hormone (TSH; a), prolactin (b), triiodothyronine (T3; c), and thyroxine (d) following i.v. administration of thyrotropin-releasing hormone (TRH) in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar in each panel indicates the LSD value (P < 0.01) for comparison among groups within each time period. Pooled SEM were 0.05 ng/mL for TSH, 2.1 ng/mL for prolactin, 10.9 ng/dL for T3, and 0.23 µg/dL for thyroxine.
|
|
Plasma concentrations of GH increased (P = 0.0004) in response to 5 min of exercise (Figure 8a), as did cortisol concentrations (Figure 8b; P < 0.0001). However, there was no effect (P > 0.1) of gender or group on the GH or cortisol responses. Leptin concentrations in the four groups around the exercise bout (Figure 8c) were consistent with the daily concentrations, but were not affected (P > 0.1) by exercise.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 8. Concentrations of GH (a), cortisol (b), and leptin (c) following exercise in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar for leptin indicates the LSD value (P < 0.01) for comparison among groups within each time period. Pooled SEM were 0.69 ng/mL for GH, 0.55 µg/dL for cortisol, and 0.81 ng/mL for leptin. Growth hormone and cortisol concentrations were affected by time (P < 0.05) but not by group (P > 0.10).
|
|
A single injection of dexamethasone increased (P < 0.0001) plasma concentrations of leptin in all groups (Figure 9a), and the response was greater (P = 0.006) in horses selected for high leptin than in horses selected for low leptin. Similarly, the insulin response to dexamethasone was also greater (P < 0.0001) in horses selected for high leptin compared to horses selected for low leptin (Figure 9b). Concentrations of glucose in response to dexamethasone were greater (P = 0.0063) in horses selected for high leptin (Figure 9c). In addition, the glucose response to dexamethasone was greater (P < 0.0001) in mares compared to geldings.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 9. Concentrations of leptin (a), insulin (b), and glucose (c) following administration of dexamethasone in mares (M) and geldings (G) selected for high vs. low leptin concentrations. The vertical bar in each panel indicates the LSD value (P < 0.01) for comparison among groups within each time period. Pooled SEM were 2.4 ng/mL for leptin, 7.2 µIU/mL for insulin, and 0.40 mmol/L for glucose.
|
|
There was no difference (P > 0.1) in IGF-I concentrations due to group or gender (Figure 10a), although there was an effect of time (P < 0.0001); IGF-I concentrations increased in all groups starting approximately 3 d after dexamethasone injection. Cortisol concentrations were equally suppressed (P = 0.029) by dexamethasone in all groups (Figure 10b), and remained so through 6 d after injection, after which they gradually began to recover.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 10. Concentrations of IGF-I (a) and cortisol (b) following administration of dexamethasone on d 0 in mares (M) and geldings (G) selected for high vs. low leptin concentrations. Pooled SEM were 6.4 ng/mL for IGF-I and 0.45 µg/dL for cortisol. There was no effect (P > 0.10) of group on either hormone.
|
|
|
Discussion
|
|---|
In other species, leptin has been reported to be primarily derived from adipose tissue, thus circulating concentrations are generally correlated to body fat mass (Prolo et al., 1998; Chilliard et al., 2000). Similar results have been reported for the horse, in which body fat mass was estimated by ultrasonography of backfat (Fitzgerald and McManus, 2000) or visually by BCS (Buff et al., 2002; Gentry et al., 2002a). In the study of Gentry et al. (2002a), 24 mares of BCS between 6.5 and 8.0 in September had widely variable plasma leptin concentrations and tended to fit into two distinct groups of either high (7 to 20 ng/mL) or low (<5 ng/mL) leptin concentrations. Subsequent nutrient restriction of half of those mares resulted in a decrease in leptin concentrations, regardless of their initial starting point. The question arose as to what factors might contribute to the large variability in plasma leptin concentrations in a group of mares with similar body fat content. Approximately 2 yr later, we sampled 18 mares that were available from that experiment, all of which were still in good body condition, and found that their original high vs. low classification was repeatable (r = 0.91). This indicated that the underlying cause was relatively permanent, thus leading to the present experiment.
Similar to what was previously observed by Gentry and coworkers (2002a), the mares and geldings with BCS of 6.5 or greater originally sampled for the present experiment (n = 51 of the 54) had widely variable leptin concentrations in September; however, there was less of a dichotomy per se than that described by Gentry et al. (2002a). That is, there was more of a continuum of plasma leptin concentrations from 0.6 to 50 ng/mL, with 25 horses <5 ng/mL, 18 horses 10 ng/mL or greater, and eight horses between those limits. The scattergram of leptin concentrations vs. BCS was similar to that presented by Buff et al. (2002), with horses in the lower leptin range found up through BCS of 8, whereas no horses with leptin concentrations above 10 ng/mL were observed with BCS <7. In contrast to the report of Buff et al. (2002), which indicated that serum leptin concentrations were greater in stallions and geldings compared to mares, mares in the present experiment had greater average leptin concentrations than geldings, albeit both genders had an equivalent number of animals fall into the high leptin and low leptin classifications. Moreover, the geldings in the present experiment had an average BCS slightly higher than the mares, indicating even a lesser leptin output per unit of fat mass.
Of the hormonal systems studied in these horses selected for high vs. low leptin, three seemed to be consistently different between the high- vs. low-leptin groups: GH, insulin, and T3. The greater daily concentrations of GH in horses with low leptin are consistent with reports in other species indicating a negative relationship between leptin and GH concentrations (Spurlock et al., 1998; Isozaki et al., 1999; Elimam et al., 2001), perhaps via a direct inhibitory effect of GH on adipocytes. Long-term treatment of growing foals with GH (Capshaw et al., 2001) decreased plasma leptin concentrations by about 75% at 16 mo of age (Gentry, 2001); however, that effect was likely due to decreased fat stores in the GH-treated horses (Kulinski et al., 2002).
Insulin is one hormone that consistently increases leptin secretion and/or is associated with elevated leptin concentrations. Hyperinsulinemia has been shown to increase leptin concentrations within 3 to 5 h in both rats and humans (Cusin et al., 1995; Saladin et al., 1995), plasma leptin concentrations are markedly reduced under conditions of insulin deficiency and are rapidly increased by insulin treatment (Sivitz et al., 1998), and insulin directly stimulates leptin secretion from in vitro-cultured adipocytes (Ramsay and White, 2000; Cammisotto and Bukowiecki, 2002). Horses selected for high leptin in the present experiment had greater resting concentrations of insulin, as well as a greater insulin response to glucose infusion, all of which are characteristic of insulin insensitivity in other species (Boden et al., 1997; Haffner et al., 1997; Appleton et al., 2002). Therefore, it is possible that the elevated insulin observed in the horses selected for high leptin may have directly increased the adipocyte production and secretion of leptin. Alternatively, given that leptin administration enhances systemic insulin sensitivity and whole-body glucose utilization in rats (Ogawa et al., 1999; Wang et al., 1999), the high insulin in horses selected for high leptin may be a result of them being leptin insensitive.
The lower daily concentrations of T3 in horses selected for low leptin in the present experiment did not seem consistent at first with our previous findings (Cartmill et al., 2003) in which hypothyroidism (reduced T3 and thyroxine) induced by propylthiouracil feeding in stallions increased leptin concentrations. However, the results from both experiments are consistent with the model proposed by Ghizzoni et al. (2001), which states that under baseline physiological conditions, the hypothalamic-pituitary-thyroid axis in humans has a prevailing inhibitory effect on leptin secretion, whereas leptin has a prevailing positive effect on the hypothalamic-pituitary-thyroid axis. Based on that model, the high leptin in the current experiment in horses selected for high leptin would be the causitive agent for the higher T3 concentrations. Similarly, in rats, prolonged treatment with leptin increased thyroid hormones while decreasing TSH (Nowak et al., 2002). Unlike T3, thyroxine and TSH concentrations in these horses did not vary in any significant manner.
In some species, IGF-1 and prolactin can directly affect concentrations of leptin (Gualillo et al., 1999; Mastronardi et al., 2000). In horses, IGF-I concentrations are very responsive to alterations in feed intake (Sticker et al., 1995b), and prolactin concentrations rise consistently after a meal (DePew et al., 1994). However, no differences were observed for IGF-I or prolactin concentrations between the horses selected for high leptin and low leptin in the current experiment, likely because their nutrient intakes, like their BCS, were similar.
Following administration of TRH, mares had greater concentrations of prolactin and TSH when compared to geldings. We previously reported (Thompson et al., 1994) that mares and stallions had greater resting prolactin concentrations than geldings in summer, although the exercise-induced rise in prolactin did not differ among genders. In rats, testosterone increased basal TSH and response to TRH; however, estrogen had no effect (Christianson et al., 1981), whereas Donda et al. (1990) reported that androgens inhibited and estrogen stimulated TRH receptor content in the pituitary. This latter report is consistent with the possibility that mares may have greater TRH receptor numbers and therefore a greater TSH response to TRH compared with geldings. Similar to the daily concentrations of T3, horses selected for high leptin had a greater T3 response to TRH administration.
The endocrine response to a single injection of dexamethasone in the current experiment was virtually identical to our previous observations (Gentry et al., 2002b; Cartmill et al., 2003), even though horses in the previous experiments had been administered the same dose of dexamethasone (per injection) over four consecutive days. Concentrations of both leptin and insulin began increasing at 12 h after injection, peaked at 36 to 48 h, and returned to pre-dexamethasone values by d 10. Horses selected for high leptin had much greater leptin and insulin responses to dexamethasone compared to horses selected for low leptin. Gentry et al. (2002b) reported a similar difference between mares full fed and those feed-restricted to produce BCS of about 3. Full fed mares had an average pre-dexamethasone leptin concentration of about 10 ng/mL, and increased to 40 to 50 ng/mL after dexamethasone, whereas mares with low BCS had an average pre-dexamethasone leptin concentration of 0.5 ng/mL and rose only to about 2.5 ng/mL after dexamethasone. A major difference between the present experiment and that of Gentry et al. (2002b) is that the horses in the present experiment all have high BCS, thus, the differential effects on leptin and insulin response to dexamethasone were not due to large differences in body fat, but likely were due to the long-term differences in insulin sensitivity described herein.
The gradual rise and fall in plasma IGF-I concentrations observed 2 to 10 d after dexamethasone administration in the present experiment have been consistently observed in previous experiments (Cartmill et al., 2003) in which untreated (no dexamethasone) mares and geldings were included. Although the rise in IGF-I concentrations does not coincide in time with the increases in leptin and insulin concentrations, both leptin (Miyakawa et al., 1998; Houseknecht et al., 2000) and insulin (Bereket et al., 1999; Thrailkill et al., 2000) have been implicated in stimulating IGF-I production and secretion in other species.
In conclusion, mares and geldings with similar high BCS selected for high vs. low resting leptin concentrations had distinctly different insulin characteristics and daily GH and T3 concentrations. Based on reports in other species, the apparent insulin insensitivity/Type-II diabetic-like condition of the horses with high leptin likely contributes to their high resting leptin concentrations. The lower daily GH concentrations in horses selected for high leptin are consistent with models in other species in which a negative relationship between leptin and GH concentrations exists; however, the nature of the relationship (cause and effect) needs to be identified in future experiments. In a similar manner, there is evidence from other species that high leptin stimulates thyroid function, which may explain the higher T3 concentrations in horses selected for high leptin, but this too needs to be studied further in the horse.
|
Implications
|
|---|
Mares and geldings of similar high body condition (body fat) can have widely variable plasma concentrations of leptin, a hormone produced by fat cells. Horses selected for high leptin concentrations also have higher concentrations of insulin and triiodothyronine and lower concentrations of growth hormone. Taken together, these results are consistent with a model in which a percentage of horses with high body fat display endocrine characteristics similar to Type-II diabetes in humans, which is associated with high insulin and leptin concentrations. Given the similarity to the human disease, these observations may be useful for improving the nutritional management of the horse.
|
Footnotes
|
|---|
1 Approved for publication by the Director of the Louisiana Agric. Exp. Stn. as manuscript No. 03-18-0932. We thank A. F. Parlow and the Natl. Inst. of Diabetes and Digestive and Kidney Diseases, National Hormone and Pituitary Program, Harbor-University of California-Los Angeles Medical Center, Torrance, CA, for reagents; and T. G. Ramsay, Growth Biology Laboratory, Agricultural Research Service, USDA, Beltsville, MD, for recombinant porcine leptin. 
2 Correspondence— phone: 225-578-3445; fax: 225-578-3279; E-mail: dthompson{at}agctr.lsu.edu.
Received for publication January 21, 2003.
Accepted for publication May 22, 2003.
|
Literature Cited
|
|---|
Appleton, D. J., J. S. Rand, and G. D. Sunvold. 2002. Plasma leptin concentrations are independently associated with insulin sensitivity in lean and overweight cats. J. Feline Med. Surg. 4:83–93.[Medline]
Bereket, A., C. H. Lang, and T. A. Wilson. 1999. Alterations in the growth hormone-insulin-like growth factor axis in insulin dependent diabetes mellitus. Horm. Metab. Res. 31:172–181.[Medline]
Boden, G., X. Chen, J. W. Kolaczynski, and M. Polansky. 1997. Effects of prolonged hyperinsulinemia on serum leptin in normal human subjects. J. Clin. Invest. 100:1107–1113.[Medline]
Buff, P. R., A. C. Dodds, C. D. Morrison, N. C. Whitley, E. L. McFadin, J. A. Daniel, J. Djiane, and D. H. Keisler. 2002. Leptin in horses: Tissue localization and relationship between peripheral concentrations of leptin and body condition. J. Anim. Sci. 80:2942–2948.[Abstract/Free Full Text]
Cammisotto, P. G., and L. J. Bukowiecki. 2002. Mechanisms of leptin secretion from white adipocytes. Am. J. Physiol. Cell. Physiol. 283:C244–250.[Abstract/Free Full Text]
Capshaw, E. L., D. L. Thompson, Jr., K. M. Kulinski, C. A. Johnson, and D. D. French. 2001. Daily treatment of horses with equine somatotropin from 4 to 16 months of age.J. Anim. Sci. 79:3137–3147.[Abstract/Free Full Text]
Cartmill, J. A., D. L. Thompson, Jr., L. R. Gentry, H. E. Pruett, and C. A. Johnson. 2003. Effects of dexamethasone, glucose infusion, adrenocorticotropin, and propylthiouracil on plasma leptin concentrations in horses. Domest. Anim. Endocrinol. 24:1–14.[Medline]
Chilliard, Y., A. Ferlay, Y. Faulconnier, M. Bonnet, J. Rouel and F. Bocquier. 2000. Adipose tissue metabolism and its role in adaptations to undernutrition in ruminants. Proc. Nutr. Soc. 59:127–134.[Medline]
Christianson, D., E. Roti, A. G. Vagenakis, and L. E. Braverman. 1981. The sex-related difference in serum thyrotropin concentration is androgen mediated. Endocrinology 108:529–535.[Abstract/Free Full Text]
Colborn, D. R., D. L. Thompson, Jr., T. L. Roth, J. S. Capehart, and K. L. White. 1991. Responses of cortisol and prolactin to sexual excitement and stress in stallions and geldings. J. Anim. Sci. 69:2556–2562.[Abstract]
Cusin, I., A. Sainsbury, P. Doyle, R. Rohner-Jeanrenaud, and B. Jeanrenaud. 1995. The ob gene and insulin. A relationship leading to clues to the understanding of obesity. Diabetes 44:1467–1470.[Abstract]
DePew, C. L., D. L. Thompson, Jr., J. M. Fernandez, L. S. Sticker, and D. W. Burleigh. 1994. Changes in concentrations of hormones, metabolites, and amino acids in plasma of adult horses relative to overnight feed deprivation followed by a pellet-hay meal fed at noon. J. Anim. Sci. 72:1530–1539.[Abstract]
Donda, A., F. Reymond, F. Rey, and T. Lemarchand-Beraud. 1990. Sex steroids modulate the pituitary parameters involved in the regulation of TSH secretion in the rat. Acta Endocrinol. 122:577–584.
Elimam, A., S. Norgren, and C. Marcus. 2001. Effects of growth hormone treatment on the leptin system and body composition in obese prepubertal boys. Acta Paediatr. 90:520–525.[Medline]
Fitzgerald, B. P., and C. J. McManus. 2000. Photoperiodic versus metabolic signals as determinants of seasonal anestrus in the mare. Biol. Reprod. 63:335–340.[Abstract/Free Full Text]
Gentry, L. R. 2001. Body condition, leptin, and reproductive characteristics of horses. Ph.D. Diss., Louisiana State Univ., Baton Rouge.
Gentry, L. R., D. L. Thompson, Jr., G. T. Gentry, Jr., K. A. Davis, and R. A. Godke. 2002b. High versus low body condition in mares: Interactions with responses to somatotropin, GnRH analog, and dexamethasone. J. Anim. Sci. 80:3277–3285.[Abstract/Free Full Text]
Gentry, L. R., D. L. Thompson, Jr., G. T. Gentry, Jr., K. A. Davis, R. A. Godke, and J. A. Cartmill. 2002a. The relationship between body condition, leptin, and reproductive and hormonal characteristics of mares during the seasonal anaovulatory period. J. Anim. Sci. 80:2695–2703.[Abstract/Free Full Text]
Ghizzoni, L., G. Mastorakos, M. Ziveri, M. Furlini, A. Solazzi, A. Vottero, and S. Bernasconi. 2001. Interactions of leptin and thyrotropin 24-hour secretory profiles in short normal children. J. Clin. Endocrinol. Metab. 86:2065–2072.[Abstract/Free Full Text]
Gill, J. L., and H. D. Hafs. 1971. Analysis of repeated measurements of animals. J. Anim. Sci. 33:331–336.
Gualillo, O., F. Lago, M. Garcia, C. Menendez, R. Senaris, F. F. Casanueva, and C. Dieguez. 1999. Prolactin stimulates leptin secretion by rat white adipose tissue. Endocrinology 140:5149–5153.[Abstract/Free Full Text]
Haffner, S. M., H. Miettinen, L. Mykkanen, P. Karhapaa, D. L. Rainwater, M. Laakso. 1997. Leptin concentrations and insulin sensitivity in normoglycemic men. Int. J. Obes. Relat. Metab. Disord. 21:393–399.[Medline]
Henneke, D. R., G. D. Potter, J. L. Kreider, and B. F. Yeates. 1983. Relationship between condition score, physical measurements and body fat percentage in mares. Equine Vet. J. 15:371–376.[Medline]
Houseknecht, K. L., C. P. Portocarrero, S. Ji, R. Lemenager, and M. E. Spurlock. 2000. Growth hormone regulates leptin gene expression in bovine adipose tissue: Correlation with adipose IGF-I expression. J. Endocrinol. 164:51–57.[Abstract]
Isozaki, O., T. Tsushima, M. Miyakawa, H. Demura, and H. Seki. 1999. Interaction between leptin and growth hormone (GH)/IGF-1 axis. Endocrine J. 46(Suppl.):17–24.
Kane, R. A., M. Fisher, D. Parrett, and L. M. Lawrence. 1987. Estimating fatness in horses. Pages 127–131 in Proc. 10th Equine Nutr. Physiol. Symp., Colorado State Univ., Ft. Collins.
Kulinski, K. M., D. L. Thompson, Jr., E. L. Capshaw, D. D. French, and J. L. Oliver. 2002. Daily treatment of growing foals with equine somatotropin: Pathologic and endocrinologic assessments at necropsy and residual effects in live animals. J. Anim. Sci. 80:392–400.[Abstract/Free Full Text]
Mastronardi, C. A., A. Walczewska, W. H. Yu, S. Karanth, A. F. Parlow, and S. M. McCann. 2000. The possible role of prolactin in the circadian rhythm of leptin secretion in male rats. Proc. Soc. Exp. Biol. Med. 224:152–158.[Abstract/Free Full Text]
McManus, C. J., and B. P. Fitzgerald. 2000. Effects of a single day of feed restriction on changes in serum leptin, gonadotropins, prolactin and metabolites in aged and young mares. Domest. Anim. Endocrinol. 19:1–13.[Medline]
Miyakawa, M., T. Tsushima, H. Murakami, O. Isozaki, H. Demura, and T. Tanaka. 1998. Effect of growth hormone (GH) on serum concentrations of leptin: Study in patients with acromegaly and GH deficiency. J. Clin. Endocrinol. Metab. 83:3476–3479.[Abstract/Free Full Text]
Nowak, K. W., P. Kaczmarek, P. Mackowiak, A. Ziolkowska, G. Albertin, W. J. Ginda, M. Trejter, G. G. Nussdorfer, and L. K. Malendowicz. 2002. Rat thyroid gland expresses the long form of leptin receptors, and leptin stimulates the function of the gland in euthyroid non-fasted animals. Int. J. Mol. Med. 9:31–34.[Medline]
Ogawa, Y., H. Masuzaki, K. Hosoda, M. Aizawa-Abe, J. Suga, and M. Suda. 1999. Increased glucose metabolism and insulin sensitivity in transgenic skinny mice overexpressing leptin. Diabetes 48:1822–1829.[Abstract]
Prolo, P., M. L. Wong, and J. Licinio. 1998. Leptin. Int. J. Biochem. Cell Biol. 30:1285–1290.[Medline]
Ramsay, T. G., and M. E. White. 2000. Insulin regulation of leptin expression in streptozotocin diabetic pigs. J. Anim. Sci. 78:1497–1503.[Abstract/Free Full Text]
Richards, M. P., T. J. Caperna, T. H. Elsasser, C. M. Ashwell, and J. P. McMurtry. 2000. Design and application of a polyclonal peptide antiserum for the universal detection of leptin protein. J. Biochem. Biophys. Methods 45:147–156.[Medline]
Saladin, R., P. DeVos, M. Guerre-Millo, A. Leturque, J. Girard, B. Staels, and J. Auwerx. 1995. Transient increase in obese gene expression after food intake or insulin administration. Nature 377:527–529.[Medline]
Sivitz, W. I., S. Walsh, D. Morgan, P. Donohoue, W. Haynes, and R. L. Leibel. 1998. Plasma leptin in diabetic and insulin-treated diabetic and normal rats. Metabolism 47:584–591.[Medline]
Spurlock, M. E., M. A. Ranalletta, S. G. Cornelius, G. R. Frank, G. M. Willis, S. Ji, A. L. Grant, and C. A. Bidwell. 1998. Leptin expression in porcine adipose tissue is not increased by endotoxin but is reduced by growth hormone. J. Interferon Cytokine Res. 18:1051–1058.[Medline]
Staros, J. V., R. W. Wright, and D. M. Swingle. 1986. Enhancement by N-hydroxysulfosuccinimide of water-soluble carbodiimide-mediated coupling reactions. Anal. Biochem. 156:220–222.[Medline]
Steel, R. G. D., and J. H. Torrie. 1980. Principles and Procedures of Statistics: A Biometrical Approach. 2nd ed. McGraw-Hill Book Co., New York.
Sticker, L. S., D. L. Thompson, Jr., L. D. Bunting, J. M. Fernandez, and C. L. DePew. 1995a. Dietary protein and/or energy restriction in mares: Plasma glucose, insulin, nonesterified fatty acid, and urea nitrogen responses to feeding, glucose, and epinephrine. J. Anim. Sci. 73:136–144.[Abstract]
Sticker, L. S., D. L. Thompson, Jr., J. M. Fernandez, L. D. Bunting, and C. L. DePew. 1995b. Dietary protein and/or energy restriction in mares: Plasma growth hormone, IGF-I, cortisol, and thyroid hormone responses to feeding, glucose, and epinephrine. J. Anim. Sci. 73:1424–1432.[Abstract]
Sticker, L. S., D. L. Thompson, Jr., and L. R. Gentry. 2001. Pituitary hormone and insulin responses to infusion of amino acids and N-methyl-D,L-aspartate in horses. J. Anim. Sci. 79:735–744.[Abstract/Free Full Text]
Thompson, Jr., D. L., C. L. DePew, A. Ortiz, L. S. Sticker, and M. S. Rahmanian. 1994. Growth hormone and prolactin concentrations in plasma of horses: Sex differences and the effects of acute exercise and administration of growth hormone releasing hormone. J. Anim. Sci. 72:2911–2918.[Abstract]
Thompson, Jr., D. L., and T. M. Nett. 1984. Thyroid stimulating hormone and prolactin secretion after thyrotropin releasing hormone administration to mares: Dose response during anestrus in winter and during estrus in summer. Domest. Anim. Endocrinol. 1:263–268.
Thompson, Jr., D. L., M. S. Rahmanian, C. L. DePew, D. W. Burleigh, C. J. DeSouza, and D. R. Colborn. 1992. Growth hormone in mares and stallions: Pulsatile secretion, response to growth hormone-releasing hormone and effects of exercise, sexual stimulation, and pharmacological agents. J. Anim. Sci. 70:1201–1207.[Abstract]
Thrailkill, K. M. 2000. Insulin-like growth factor-I in diabetes mellitus: Its physiology, metabolic effects, and potential clinical utility. Diabetes Technol. Ther. 2:69–80.[Medline]
Wang, J. L., N. Chinookoswong, S. Scully, M. Qi, and Z. Q. Shi. 1999. Differential effects of leptin in regulation of tissue glucose utilization in vivo. Endocrinology 140:2117–2124.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
W. A. Storer, D. L. Thompson Jr., C. A. Waller, and J. A. Cartmill
Hormonal patterns in normal and hyperleptinemic mares in response to three common feeding-housing regimens
J Anim Sci,
November 1, 2007;
85(11):
2873 - 2881.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. M. Vick, A. A. Adams, B. A. Murphy, D. R. Sessions, D. W. Horohov, R. F. Cook, B. J. Shelton, and B. P. Fitzgerald
Relationships among inflammatory cytokines, obesity, and insulin sensitivity in the horse
J Anim Sci,
May 1, 2007;
85(5):
1144 - 1155.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. R. Buff, B. R. Spader, C. D. Morrison, and D. H. Keisler
Endocrine responses in mares undergoing abrupt changes in nutritional management
J Anim Sci,
October 1, 2006;
84(10):
2700 - 2707.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. Cartmill, D. L. Thompson Jr., W. A. Storer, J. C. Crowley, N. K. Huff, and C. A. Waller
Effect of dexamethasone, feeding time, and insulin infusion on leptin concentrations in stallions
J Anim Sci,
August 1, 2005;
83(8):
1875 - 1881.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
