Download Thyrotropin releasing hormone interactions with growth hormone

Thyrotropin releasing hormone interactions
with growth hormone secretion in horses1
H. E. Pruett*, D. L. Thompson, Jr.*2, J. A. Cartmill*, C. C. Williams†, and L. R. Gentry*
*Department of Animal Sciences and †Department of Dairy Science,
Louisiana Agricultural Experiment Station,
Louisiana State University Agricultural Center, Baton Rouge 70803-4210
ABSTRACT: Light horse mares, stallions, and geldings were used to 1) extend our observations on the
thyrotropin releasing hormone (TRH) inhibition of GH
secretion in response to physiologic stimuli and 2) test
the hypothesis that stimulation of endogenous TRH
would decrease the normal rate of GH secretion. In Exp.
1 and 2, pretreatment of mares with TRH (10 ␮g/kg
BW) decreased (P < 0.001) the GH response to exercise
and aspartate infusion. Time analysis in Exp. 3 indicated that the TRH inhibition lasted at least 60 min
but was absent by 120 min. Administration of a single
injection of TRH to stallions in Exp. 4 increased (P <
0.001) prolactin concentrations as expected but had no
effect (P > 0.10) on GH concentrations. Similarly, 11
hourly injections of TRH administered to geldings in
Exp. 5 did not alter (P > 0.10) GH concentrations either
during the injections or for the next 14 h. In Exp. 5, it
was noted that the prolactin and thyroid-stimulating
hormone responses to TRH were great (P < 0.001) for
the first injection, but subsequent injections had little
to no stimulatory effect. Thus, Exp. 6 was designed to
determine whether the inhibitory effect of TRH also
waned after multiple injections. Geldings pretreated
with five hourly injections of TRH had an exercise-
induced GH response identical to that of control geldings, indicating that the inhibitory effect was absent
after five TRH injections. Retrospective analysis of
pooled, selected data from Exp. 4, 5, and 6 indicated
that endogenous GH concentrations were in fact lower
(P < 0.01) from 45 to 75 min after TRH injection but
not thereafter. In Exp. 7, 6-n-propyl-2-thiouracil was
fed to stallions to reduce thyroid activity and hence
thyroid hormone feedback, potentially increasing endogenous TRH secretion. Treated stallions had decreased (P < 0.01) concentrations of thyroxine and elevated (P < 0.01) concentrations of thyroid-stimulating
hormone by d 52 of feeding, but plasma concentrations
of GH and prolactin were unaffected (P > 0.10). In contrast, the GH response to aspartate and the prolactin
response to sulpiride were greater (P < 0.05) in treated
stallions than in controls. In summary, TRH inhibited
exercise- and aspartate-induced GH secretion. The duration of the inhibition was at least 1 h but less than
2 h, and it waned with multiple injections. There is
likely a TRH inhibition of endogenous GH episodes as
well. Reduced thyroid feedback on the hypothalamicpituitary axis did not alter basal GH and prolactin secretion.
Key Words: Horses, Prolactin, Somatotropin, Thyrotropin Releasing Hormone
2003 American Society of Animal Science. All rights reserved.
Introduction
Thyrotropin releasing hormone (TRH), the hypothalamic releasing hormone for thyroid-stimulating hormone
(TSH; Guyton and Hall, 1996), is also an effective secre-
1
Approved for publication by the Director of the Louisiana Agric.
Exp. Sta. as manuscript no. 03-18-1177. We thank A. F. Parlow and
the NIDDKD, National Hormone and Pituitary Program, HarborUCLA Medical Center, Torrance, CA, for reagents.
2
Correspondence—phone: 225-578-3445; fax: 225-578-3279; Email: [email protected].
Received April 1, 2003.
May 29, 2003.
J. Anim. Sci. 2003. 81:2343–2351
tagogue for prolactin in many species (Jacobs et al., 1971;
Convey et al., 1973), including the horse (Thompson et
al., 1986; Johnson, 1987). In some species, TRH stimulates GH secretion as well (Smith et al., 1977; Scanes
and Harvey, 1989; Cabell and Esbenshade, 1990). In
contrast, TRH administration at the doses used to stimulate TSH and prolactin secretion does not stimulate GH
secretion in horses (Thompson et al., 1992), whereas simultaneous administration of TRH with GH secretagogues abolishes the expected increase in GH secretion
(Kennedy et al., 2002). The present series of experiments
was designed 1) to extend our observations on the TRH
inhibition of GH secretion in horses, specifically its effects
on exercise- and aspartate-induced GH secretion, the
2343
2344
Pruett et al.
duration of the inhibition, and the effects on endogenous
GH secretion, and 2) to test the hypothesis that stimulation of endogenous TRH in the horse (by diminishing
thyroid function) would reduce the normal rate of GH secretion.
Materials and Methods
Animals and Experimental Conditions. Mature, light
horse mares (Exp. 1 through 3), geldings (Exp. 5 and 6),
and stallions (Exp. 4 and 7) were used. Mares and geldings were maintained on pasture and were in good body
condition (6 to 8; Henneke et al., 1983). Stallions were
kept in individual outdoor lots with limited access to
native grasses and were fed sufficient pelleted commercial feed and grass hay daily to maintain average to good
body condition (between 5 and 7). All horses were on a
long-term herd health program that included routine
deworming and yearly vaccinations for infectious diseases.
For the short-term experiments (Exp. 1 through 6),
mares and geldings were brought in from pasture at
approximately 0700 and were tethered inside a barn and/
or shed; stallions in Exp. 4 were tethered inside their
individual shelters. After insertion of a 14-gauge catheter
in the left jugular vein, the horses were allowed to stand
quietly for a minimum of 1 h before initiation of blood
sampling. Blood samples were collected from, and treatment injections were administered through, the jugular
catheters. In all experiments, blood samples were placed
into disposable glass tubes containing heparin (20 IU/
mL) and were kept at 5°C until centrifugation at 1,200 ×
g for 15 min. Plasma was harvested and stored at −15°C.
Experiment 1. Twelve mares were used in a split-plot
design to characterize the effects of TRH on the GH
response to exercise, a known stimulus for GH in horses
(Thompson et al., 1992; 1994). Stage of the estrous cycle
was not taken into consideration. Mares were randomly
allotted to one of two treatment groups (six per group).
The treated mares received an i.v. injection of TRH (Catalog no. P2161; Sigma Chem. Co., St. Louis, MO) at 10
␮g/kg BW (Kennedy et al., 2002). Control mares received
an equivalent amount of saline i.v. Treatments were immediately followed by 5 min of exercise in a round pen
(moderate to fast trot; Thompson et al., 1992). Jugular
blood samples (5 mL) were collected at −30, −15, 0, 10,
20, 30, 40, 50, 60, 75, 90, 105, and 120 min relative to
the onset of exercise for the measurement of GH and prolactin.
Experiment 2. Twelve mares were used in a split-plot
design to characterize the effect of TRH on the GH response to aspartate infusion (Sticker et al., 2001). Stage
of the estrous cycle was not taken into consideration.
Mares were randomly allotted to one of two treatment
groups (six per group). The treated mares received an
i.v. injection of TRH as described in Exp. 1; control mares
received saline. Immediately after treatment, all mares
were infused with sodium aspartate (2.855 mmol aspartate/kg BW; Catalog no. A9256; Sigma Chem. Co.) in
water (Sticker et al., 2001). The aspartate solutions were
made to 1.427 M, pH of 7.3 to 7.4, and were infused at
2 mL/kg BW.
Samples of jugular blood (5 mL) were collected at −30,
−15, and 0 min relative to saline or TRH treatment.
Aspartate infusion was started immediately following
the 0-min sample and required about 5 min to complete.
Subsequent blood samples were drawn at 10, 20, 30, 40,
50, 60, 75, 90, 105, and 120 min after the onset of aspartate infusion. Prolactin and GH were measured in all
plasma samples.
Experiment 3. Six mares were used in a 6 × 6 Latin
square to characterize the duration of the TRH inhibition
of the GH response to exercise. Stage of the estrous cycle
was not taken into consideration. The six treatments
were 1) control (saline) injection immediately before exercise, 2) TRH immediately before exercise (as described
for Exp. 1), 3) TRH 0.5 h before exercise, 4) TRH 1 h
before exercise, 5) TRH 2 h before exercise, and 6) TRH
3 h before exercise. Dose of TRH in each case was 10 ␮g/
kg BW. Jugular blood samples were drawn at 15-min
intervals starting 60 min before exercise and continuing
through 90 min after onset of exercise. Treatment days
were separated by at least 2 d of inactivity. Concentrations of GH were measured in all plasma samples.
Experiment 4. Six stallions were used in a single
switchback design (three replicates of a 2 × 2 Latin
square) to determine the effect of a single TRH injection
on endogenous GH secretion. The experiment was performed on 2 d, 1 wk apart, in order to minimize any
carryover effects of treatment. On the first treatment
day, three stallions received TRH (10 ␮g/kg BW) in saline
and the remaining three received an equivalent amount
of saline. On the second treatment day, the treatments
were reversed. Blood samples were collected at 15-min
intervals for 4 h to characterize basal GH secretion, and
then the treatments were administered. Subsequent
samples were drawn at 15-min intervals for 6 h. Stallions
were loosely tethered inside sheds within their individual
paddocks during the experiment. They had ad libitum
access to hay and water except when blood samples were
being drawn. Prolactin and GH were measured in all
plasma samples.
Experiment 5. Ten geldings were used in a split-plot
design to determine the effects of multiple hourly TRH
injections on endogenous GH secretion. Jugular blood
samples were drawn from all geldings at 15-min intervals for a total of 25 h beginning at 0800 on d 1. Immediately after the fifth sample (0900), five geldings received
TRH (10 ␮g/kg BW) in saline i.v. through the jugular
catheter; the other five received an equivalent amount
of saline. Injections were repeated hourly through 1900
(total of 11 injections). During the 25-h period, the geldings were loosely tethered in stalls in a barn and were
given access to grass hay and water on a regular basis.
Prolactin and GH were measured in all plasma samples.
Experiment 6. Because the stimulatory effects of TRH
on prolactin and TSH secretion seemed to wane after
the first injection in Exp. 5, 10 geldings were used in a
2345
TRH and GH in Horses
split-plot design to test the hypothesis that the inhibitory
effects of TRH on GH secretion after exercise would also
wane after multiple injections. Five geldings were administered five hourly TRH injections (10 ␮g/kg BW in
saline i.v.) and five geldings were administered saline.
Immediately after receiving the fifth injection, each gelding was exercised for 5 min as described in Exp. 1. Jugular blood samples were drawn at 15-min intervals starting 120 min before the first injection and were continued
through 120 min after the onset of exercise (8 h total).
Prolactin and GH were measured in all plasma samples.
Experiment 7. Eight stallions were used in a splitplot design to test the hypothesis that reduced thyroid
function, resulting in increased endogenous TRH secretion, would reduce endogenous GH secretion. To reduce
thyroid activity, four randomly selected stallions were
fed 6-n-propyl-2-thiouracil (PTU; Catalog no. P3755;
Sigma Chem. Co.) at 6 mg/kg BW (B. Breuhaus, personal
communication) in 250 g of a molasses-containing balanced grain ration on top of their regular pelleted feed;
four control stallions were given the top dressing with
no added PTU. Stallions were fed their diets daily at
0800 and were allowed to eat throughout the day; hay
and water were available ad libitum throughout the experiment. Blood samples were collected twice weekly for
the measurement of thyroxine and TSH to monitor thyroid function.
Once PTU-fed stallions were determined to have reduced thyroid function (d 52 of treatment), all stallions
were brought from their paddocks and placed in stalls for
an extensive period of blood sampling and secretagogue
challenges (PTU feeding was continued throughout this
period). They were allowed 3 d to equilibrate to their
surroundings in the barn.
At 1200 on d 3, each stallion was fitted with a 14gauge indwelling jugular catheter, which was secured to
the neck. The stallions were tethered lightly within their
stalls and were allowed ad libitum access to hay and
water. After an hour of inactivity, 15-min blood sampling
was started. The first 22 h of sampling was used to assess
endogenous GH secretion. Immediately following the 22h blood sample (1100 on the second day), each stallion
received an i.v. injection of the dopamine antagonist sulpiride (0.10 mg/kg BW; Catalog no. S8010; Sigma Chem.
Co.; Johnson and Becker, 1987), followed immediately
by an infusion of 2.855 mmol sodium aspartate/kg BW in
water (Sticker et al., 2001). Blood samples were collected
thereafter at 15-min intervals for an additional 3 h for
measurement of prolactin and GH. Immediately after
the last sample was drawn, each stallion was given an
i.v. injection of TRH (10 ␮g/kg BW), and blood samples
were collected for another 2 h for the measurement of
prolactin and TSH.
Sample Analyses. Concentrations of GH (Thompson et
al., 1992), prolactin (Colborn et al., 1991), TSH (Sticker
et al., 2001), and thyroxine (ICN Pharmaceuticals, Inc.,
Costa Mesa, CA) were analyzed with RIA previously validated for horse samples. Intra- and interassay coefficients of variation and assay sensitivities were 8%, 11%,
and 0.5 ng/mL for GH; 7%, 12%, and 0.2 ng/mL for prolactin; 5%, 10%, and 0.02 ng/mL for TSH; and 5%, 9%, and
2 ng/mL for thyroxine.
Statistical Analyses. For each experiment, data were
analyzed by univariate ANOVA for repeated measures
(split plot; Gill and Hafs, 1971) within a completely random design (Exp. 1, 2, 5, 6, and 7), 6 × 6 Latin square
design (Exp. 3; Steel and Torrie, 1980), or replicated 2
× 2 Latin square design (Exp. 4; Steel and Torrie, 1980)
using the GLM procedure of SAS (SAS Inst. Inc., Cary,
NC). For Exp. 1, 2, 5, 6, and 7, factors in the analyses were
treatment, horses within treatment (error term used to
test treatment), time, and treatment × time (tested with
residual error). For Exp. 3, the factors were mare, day,
treatment, mare × day × treatment (error term for the
first three factors), time, and treatment × time (tested
with residual error). For Exp. 4, the factors were squares,
stallions within squares, days within squares, treatment,
square × stallion × day × treatment (error term for the
first four factors), time, and treatment × time (tested
with residual error). Differences between treatment
groups were tested within each time period by the LSDtest (Steel and Torrie, 1980).
For Exp. 3, the maximum GH concentration achieved
within the first 30 min after exercise was selected from
the overall data set, and the natural log of these data
were analyzed in a 6 × 6 Latin square ANOVA due to
heterogeneity of variances (Steel and Torrie, 1980). Differences among treatment means were assessed by the
LSD-test (Steel and Torrie, 1980).
Retrospective Analysis. Based on the estimated duration of inhibition from Exp. 3 and the possible waning
of the inhibition over multiple TRH injections (Exp. 6),
it was decided to do a pooled, retrospective analysis on
all horses for which endogenous (no external stimuli) GH
concentrations were available after TRH or saline. The
data for blood samples collected at 0 to 60 min relative
to treatment from horses in Exp. 4, 5 (first TRH injection
only), and 6 (first TRH injection only) were combined
and analyzed by split-plot ANOVA. Factors in the ANOVA were experiment, treatment, experiment × treatment, horse within experiment × treatment (error term
to test the first three factors), time, experiment × time,
treatment × time, and the three-way interaction (the
latter four terms being tested with residual error). Because Exp. 4 was performed as a replicated Latin square,
the appropriate degrees of freedom were subtracted from
the error degrees of freedom in the pooled analysis. The
first 60 min was chosen for this analysis based on the
results of Exp. 3, which indicated a duration of inhibition
of at least 60 min but less than 120 min. Subsequent
pooled analyses were performed using GH data for the
first 75, 90, and 120 min to estimate at what time point
the treatment effect was no longer significant (P > 0.10).
Results
Experiment 1. There were treatment effects (P < 0.001)
as well as treatment × time interactions (P < 0.0001)
2346
Pruett et al.
GH, ng/mL
10
12
a
8
6
4
Control
TRH
8
6
4
2
2
0
a
10
Control
TRH
GH, ng/mL
12
0
-30
0
30
60
90
120
-30
0
30
6
b
12
8
4
0
0
30
60
120
90
120
4
3
2
1
0
-30
90
b
5
Prolactin, ng/mL
Prolactin, ng/mL
16
60
Minutes
Minutes
90
120
Minutes
Figure 1. Responses in plasma concentrations of GH
(a) and prolactin (b) to exercise at time 0 in control mares
(n = 6) and in mares pretreated with thyrotropin releasing
hormone (TRH; 10 ␮g/kg BW; n = 6). The vertical line in
each panel indicates the LSD-value (P < 0.05) for betweengroup comparisons at each time period. Pooled SEM were
0.84 and 1.7 ng/mL for GH and prolactin, respectively.
for both GH and prolactin concentrations in mares in
response to i.v. administration of TRH and/or saline followed by 5 min of acute exercise (Figure 1). Injection of
TRH stimulated prolactin concentrations, as expected,
and saline did not. Exercise stimulated GH concentrations, as expected, in mares receiving saline, whereas
prior injection of TRH abolished the GH response.
Experiment 2. There were treatment effects (P < 0.001)
as well as treatment × time interactions (P < 0.0001)
for both GH and prolactin concentrations in mares in
response to i.v. administration of TRH and/or saline followed by infusion of sodium aspartate (Figure 2). Injection of TRH stimulated prolactin concentrations, as expected, and saline did not. Aspartate infusion stimulated
GH concentrations, as expected, in mares receiving saline, whereas previous injection of TRH greatly diminished the GH response. In addition to the hormonal responses, the infusion of aspartic acid resulted in sweating
and panting in all mares beginning within 5 min of infusion and lasting from 10 to 15 min.
Experiment 3. There was a treatment effect (P = 0.0002)
as well as a treatment × time interaction (P < 0.0001)
for GH concentrations in response to exercise in mares
treated with saline or TRH at increasing times before
exercise (Figure 3a). There was the expected GH re-
-30
0
30
60
Minutes
Figure 2. Responses in plasma concentrations of GH
(a) and prolactin (b) to aspartate infusion at time 0 in
control mares (n = 6) and in mares pretreated with thyrotropin releasing hormone (TRH; 10 ␮g/kg BW; n = 6).
The vertical line in each panel indicates the LSD-value
(P < 0.05) for between-group comparisons at each time
period. Pooled SEM were 0.71 and 0.37 ng/mL for GH
and prolactin, respectively.
sponse in mares pretreated with saline before exercise,
whereas TRH treatment immediately before exercise
abolished the GH response. There was no GH response
to exercise in mares treated with TRH at 30 min and
1 h before exercise; however, at 2 and 3 h after TRH
treatment, the GH response to exercise was present.
Based on maximum GH concentration achieved in the
first 30 min (Figure 3b), the responses at 2 and 3 h
were greater (P < 0.05) than the response after saline
treatment, and the responses at 0, 0.5, and 1 h were less
(P < 0.05) than after saline treatment.
Experiment 4. Endogenous GH concentrations in stallions were not affected (P > 0.10) by a single injection of
TRH administered during the 10-h period of frequent
sampling (Figure 4a). As expected, there was a treatment
effect (P < 0.0001) and a treatment × time interaction (P
< 0.0001) for prolactin concentrations, which increased
rapidly when the stallions were administered TRH (Figure 4b).
Experiment 5. Endogenous GH concentrations in geldings were not affected (P > 0.10) by 11 hourly injections
of TRH during the 25-h period of blood sampling (Figure
5a). As expected, there was an effect of TRH treatment
(P = 0.035) and a treatment × time interaction (P <
2347
TRH and GH in Horses
2.5
a
7
Saline
TR H 0 h
TR H 0.5 h
TR H 1 h
TR H 2 h
TR H 3 h
GH, ng/mL
6
5
4
a
Control
TRH
2.0
GH, ng/mL
8
3
2
1.5
1.0
0.5
1
0.0
0
-60
-45
-30
-15
0
15
30
45
60
75
-240 -180 -120
90
-60
0
1. 0
e
e
0. 6
0. 4
c
d
0. 2
d
50
Prolactin, ng/mL
Ln maximum GH
60
b
0. 8
60
120 180
240
300
360
Minutes
Minutes
d
b
40
30
20
10
0. 0
0
-0.2
Saline
0
0. 5
1
2
3
Tr eatment
Figure 3. Time analysis of the thyrotropin releasing
hormone (TRH) inhibition of exercise-induced GH secretion in mares. Mares were treated with saline at time 0
or with TRH (10 ␮g/kg BW) at 0, 0.5, 1, 2, or 3 h before
exercise at time 0 in a 6 × 6 Latin square experiment.
(a) Plasma concentrations of GH; (b) natural log of the
maximum GH response in the first 30 min after exercise.
The vertical line in (a) indicates the LSD-value (P < 0.05)
for between-group comparisons at each time period.
Means with different superscripts in (b) differ (P < 0.05).
Pooled SEM were 0.58 ng/mL and 0.072 for GH concentrations and maximum GH response, respectively.
0.0001) for prolactin concentrations (Figure 5b), which
increased soon after first TRH injection and then gradually declined throughout the rest of the sampling period.
There was also a treatment × time interaction (P <
0.0001) for TSH concentrations (Figure 5c), which increased soon after the first TRH injection and then declined gradually over the next 8 h. For both prolactin
and TSH concentrations, injections of TRH after the first
one or two produced little response in the already elevated concentrations.
Experiment 6. Plasma GH concentrations were affected
by time (P < 0.0001) in the geldings in Exp. 6 (Figure
6a), but there was no effect of multiple TRH treatment
nor any interaction between treatment and time (P >
0.10); the GH response to exercise at 360 min was virtually identical in the two groups. There was the expected
prolactin response to TRH injection (P < 0.0001; Figure
6b), with concentrations in TRH-treated geldings being
-240 -180 -120 -60
0
60
120
180
240
300
360
Minutes
Figure 4. Plasma concentrations of GH (a) and prolactin
(b) in stallions relative to a single injection of thyrotropin
releasing hormone (TRH; 10 ␮g/kg BW) at time 0. The
experiment was performed as a single switch back (three
simultaneous replicates of a 2 × 2 Latin square) with six
stallions. The vertical line in (b) indicates the LSD-value
(P < 0.05) for between-group comparisons at each time
period. Pooled SEM were 0.73 and 9.4 ng/mL for GH
and prolactin, respectively.
higher than in control geldings from 15 min after the
first injection through the onset of exercise. Exercise resulted in a sharp rise in prolactin concentrations in control geldings, but little to no change in TRH-treated
geldings.
Experiment 7. There was a treatment effect (P < 0.0001)
as well as a treatment × day interaction (P < 0.0001) for
thyroxine concentrations (Figure 7a) when stallions were
fed PTU. Although variable and apparently affected by
environmental conditions, thyroxine concentrations in
stallions fed PTU decreased over time and were approximately half of those in control stallions by d 52. In addition, there was an effect of treatment (P < 0.0001) as
well as a treatment × day interaction (P < 0.0001) for TSH
concentrations (Figure 7b), which increased gradually in
PTU-fed stallions beginning approximately d 20 and
were 7 to 8 times higher than in control stallions by d 52.
During the 22 h of frequent sampling, there was no
effect (P > 0.10) of treatment (PTU feeding), nor was
there any interaction of treatment with time on endogenous GH or prolactin concentrations (data not shown).
After infusion of sulpiride and aspartate at 22 h, concen-
2348
Pruett et al.
' ' ' ' ' ' ' ' ' ' '
16
Control
TRH
a
a
Control
TRH
12
10
GH, ng/mL
GH, ng/mL
15
5
8
4
0
0
4
8
12
16
20
24
0
Hours
0
' ' ' ' ' ' ' ' ' ' '
60
120
180
240
300
360
420
480
300
360
420
480
Minutes
b
16
b
15
Prolactin, ng/mL
Prolactin, ng/mL
20
10
5
12
8
4
0
0
4
8
12
16
20
24
Hours
1.2
0
' ' ' ' ' ' ' ' ' ' '
TSH, ng/mL
0
c
1.0
60
120
180
240
Minutes
0.8
0.6
0.4
0.2
0.0
0
4
8
12
16
20
24
Hours
Figure 5. Plasma concentrations of GH (a), prolactin
(b), and thyroid stimulating hormone (TSH; c) in geldings
administered 11 hourly injections of thyrotropin releasing
hormone (TRH; 10 ␮g/kg BW; indicated by the arrows;
n = 5) or saline (control; n = 5) beginning at time 0. The
vertical lines in (b) and (c) indicate the LSD-values (P <
0.05) for between-group comparisons at each time period.
Pooled SEM were 2.0, 1.2, and 0.06 ng/mL for GH, prolactin, and TSH, respectively.
trations of GH and prolactin increased (P < 0.001) as
expected in both groups of stallions, and the response
was greater (P < 0.05) in PTU-fed stallions relative to
controls for both hormones (Figure 8). Immediately before TRH was injected at 24.5 h, concentrations of both
TSH and prolactin (Figure 9) were already higher (P <
0.05) in PTU-fed stallions relative to controls. Injection
of TRH increased (P < 0.001) concentrations of TSH and
prolactin in both groups of stallions, and the TSH response was greater (P < 0.05) in PTU-fed stallions than
in controls.
Retrospective Analysis. There was an effect of treatment (P = 0.002) and a treatment × time interaction (P
Figure 6. Plasma concentrations of GH (a) and prolactin
(b) in geldings administered five hourly injections of thyrotropin releasing hormone (TRH; 10 ␮g/kg BW; n = 5)
or saline (control; n = 5) beginning at 120 min and then
exercised at 360 min. The vertical line in (b) indicates the
LSD-value (P < 0.05) for between-group comparisons at
each time period. Pooled SEM were 2.5 and 2.9 ng/mL
for GH and prolactin, respectively.
= 0.007) in the retrospective analysis of endogenous GH
concentrations from horses administered TRH or saline
in Exp. 4, 5, and 6 (Figure 10). Horses administered TRH
had lower GH concentrations at 45 and 60 min after
injection. Similar analyses on data for the first 75, 90,
and 120 min after treatment indicated that the difference
between groups was present through 75 min (P < 0.02)
but was absent thereafter (P > 0.10; data not shown).
Discussion
Administration of TRH immediately before exercise or
aspartate infusion inhibited the normal GH response to
these stimuli just as it inhibited the GHRH- and
EP51389-induced GH responses reported by Kennedy et
al. (2002). Although the mechanism by which exercise
and aspartate infusion stimulate GH release is not
known, reports for other species indicate that many of
these indirect stimuli act via GHRH secretion from the
hypothalamus, which in turn stimulates GH secretion
from the pituitary (Wehrenberg et al., 1985; Magnan et
al., 1994). Alternatively, acute GH secretion may also be
2349
TRH and GH in Horses
35
a
a
Control
PTU
12
25
GH, ng/mL
Thyroxine, ng/mL
30
16
Control
PTU
20
15
10
8
4
5
0
0
10
0
30
40
-30
50
0
30
b
100
Prolactin, nglmL
2.0
1.5
1.0
0.5
0.0
0
10
20
60
90
120
150
90
120
150
Minutes
Days
2.5
TSH, ng/mL
20
30
40
50
Days
Figure 7. Plasma concentrations of thyroxine (a) and
thyroid stimulating hormone (TSH; b) in control stallions
(n = 4) and in stallions fed 6-n-propyl-2-thiouracil (PTU;
n = 4) daily to decrease thyroid activity. The vertical
line in each panel indicates the LSD-value (P < 0.05) for
between-group comparisons at each time period. Pooled
SEM were 1.8 and 0.21 ng/mL for thyroxine and TSH, respectively.
due to an abrupt decrease in somatostatin secretion from
the hypothalamus (Miki et al., 1988; Guyton and Hall,
1996). Because TRH inhibits GH secretion induced by
GHRH itself, it is likely that TRH either affects the somatotropes directly to block GH release or acts via stimulation of somatostatin, which is known to block GH release
via inhibition of adenyl cyclase formation in the somatotrope (Bass et al., 1996). Receptors for TRH have been
reported on somatotropes of the rat (Konaka et al., 1997)
in numbers approximately equal to those on lactotropes.
Receptors for TRH have also been localized in various
areas of the brain and in peripheral tissues (Zabavnik
et al., 1993).
Time analysis of the TRH inhibition of exercise-induced GH secretion indicated that the inhibition lasted
at least 1 h but was gone by 2 h. In fact, a rebound-like
recovery and/or potentiation was evident at 2 and 3 h,
such that the responses to exercise at these times were
greater than the uninhibited response. Gentry et al.
(2002) described a similar rebound-like response in
mares administered the GH secretagogue EP51389 and
TRH simultaneously; mares had no GH response in the
first 90 min after injection but had surges in GH concentrations between 90 and 180 min. Clark et al. (1988)
described a similar rebound secretion of GH after with-
b
80
60
40
20
0
-30
0
30
60
Minutes
Figure 8. Plasma concentrations of GH (a) and prolactin
(b) in response to administration of aspartate and sulpiride at time 0 in control stallions (n = 4) and in stallions
previously fed 6-n-propyl-2-thiouracil (PTU; n = 4) daily
to reduce thyroid activity. The vertical line in each panel
indicates the LSD-value (P < 0.05) for between-group comparisons at each time period. Pooled SEM were 1.1 and
9.1 ng/mL for GH and prolactin, respectively.
drawal of somatostatin infusion in rats; that rebound
was subsequently determined to be mediated by hypothalamic release of GHRH.
At first glance, the results of Exp. 4 and 5 seemed
to indicate that there is no inhibitory effect of TRH on
endogenous GH secretion. However, focusing on only the
first 60 min after injection, the time during which TRH
inhibition is now known to be present, neither TRHtreated nor control horses had much endogenous GH
activity in either experiment. Extending the TRH exposure to several hours in Exp. 5 with multiple injections
seemed a logical way to average over the inconsistent
(infrequent) endogenous surges among horses; however,
the fallacy of that approach was subsequently borne out
by Exp. 6. That is, just as the stimulatory effects of TRH
on prolactin and TSH secretion waned with multiple
injections, so did the inhibitory effect of TRH on exerciseinduced GH release. Thus, given the episodic nature of
GH release in horses (Thompson et al., 1992) and the
infrequent occurrence of episodes, determination of the
effect of a single injection of TRH on endogenous episodic
GH release would require averaging over more replications than presented in any one experiment. For example, endogenous episodes result in elevated GH concen-
2350
Pruett et al.
8
a
Control
PTU
TSH, ng/mL
6
4
2
0
-30
0
30
60
90
120
90
120
Minutes
Prolactin, ng/mL
50
b
40
30
20
10
0
-30
0
30
60
Minutes
Figure 9. Plasma concentrations of thyroid stimulating
hormone (TSH; a) and prolactin (b) in response to administration of thyrotropin releasing hormone (TRH) at time
0 in control stallions (n = 4) and in stallions previously
fed 6-n-propyl-2-thiouracil (PTU; n = 4) daily to decrease
thyroid activity. The vertical line in each panel indicates
the LSD-value (P < 0.05) for between-group comparisons
at each time period. Pooled SEM were 0.31 and 1.8 ng/
mL for TSH and prolactin, respectively.
Control
TRH
GH, ng/mL
2.0
1.5
1.0
0.5
0
10
20
30
40
50
60
70
Minutes
Figure 10. Plasma concentrations of GH (n = 16 per
point) in the first 60 min after thyrotropin releasing hormone (TRH) administration at time 0 in horses from Exp.
4, 5, and 6. In each case, these data were for the first or
only TRH injection given; no other stimuli were administered. The vertical line indicates the LSD-value (P < 0.05)
for between-group comparisons at each time period.
Pooled SEM was 0.17 ng/mL.
trations lasting 60 min or less (Thompson et al., 1992),
and between-episode intervals range from 3 to 8 h
(Thompson et al., 1992; individual horses in Exp. 4 and
5); thus, the chance that a randomly drawn blood sample
will be drawn during an endogenous GH episode is 12
to 33%. The retrospective analysis performed with the
data from Exp. 4, 5, and 6 was an attempt to average
over these infrequent episodes, and although the approach may be less than ideal from a statistical standpoint, the results are consistent with a TRH inhibition
of endogenous GH secretion in the first 60 to 75 min
after injection.
The loss of inhibitory effect of TRH on exercise-induced
GH secretion, as well as the wane in stimulatory effect
on prolactin and TSH secretion, is characteristic of a
down-regulation of receptors and refractoriness commonly associated with high doses of hypothalamic hormones. Although the TRH dose used in the present experiments (10 ␮g/kg of BW) was higher than the dose needed
to produce maximal TSH response in mares (4 ␮g/kg of
BW; Thompson and Nett, 1984), Gentry et al. (2002) used
the lower dose in mares and observed the same inhibition
of GH release in response to EP51389.
Experiment 7 was performed in an attempt to raise
physiologic levels of TRH by reducing thyroid hormone
secretion and feedback. Given the long-term increase
in TSH concentrations in stallions fed PTU and their
exaggerated response to exogenous TRH, it is likely that
TRH secretion from the hypothalamus was increased to
some degree. However, the relative importance of the
hypothalamus vs pituitary gland as sites of negative
feedback by thyroid hormones in the horse has yet to be
determined; thus, we can only assume that TRH secretion was increased. Moreover, the direct effects of reduced thyroid activity must also be taken into consideration. There was no alteration in basal GH or prolactin
secretion associated with PTU feeding and hypersecretion of TSH, indicating that 1) reduction of thyroid hormone feedback does not affect the secretion of these two
pituitary hormones or 2) little or no change in endogenous TRH secretion was induced by the feeding of PTU.
In contrast, the enhanced responses in GH and prolactin
secretion to aspartate and sulpiride infusion in stallions
fed PTU may indicate an increased pituitary content of
these hormones or an increased sensitivity to stimulation
(although such an increase was not evident for prolactin
after TRH injection 3 h later). The increase in prolactin
secretion is consistent with the fact that TRH stimulates
prolactin secretion and maybe production (Laverriere et
al., 1988; Guyton and Hall, 1996), but the increase in
GH secretion after aspartate infusion is contrary to our
working hypothesis that increased endogenous TRH
would reduce GH secretion.
In summary, the TRH inhibition of GHRH- and
EP51389-induced GH release reported by Kennedy et
al. (2002) has been extended to two other GH-releasing
stimuli, exercise and aspartate infusion. The duration of
the inhibition is at least 1 h but less than 2 h, and, like
the stimulatory effect on prolactin and TSH, wanes with
TRH and GH in Horses
multiple injections. There is likely a TRH inhibition of
endogenous GH episodes as well; however, more research
on this question is needed to confirm our limited analysis.
Although PTU feeding reduced thyroid feedback on the
hypothalamic-pituitary axis, as evidenced by increased
TSH concentrations, basal GH and prolactin concentrations were unaffected. The increases in aspartate-induced GH secretion and sulpiride-induced prolactin secretion in stallions fed PTU may indicate an alteration
in pituitary content of these hormones or sensitivity to
stimuli, but again, further research is needed to clarify
these possibilities.
Implications
Exogenous administration in horses of the hypothalamic releasing hormone that normally regulates thyroid
activity prevents the secretion of growth hormone in response to various stimuli, including its own releasing
hormone, growth hormone releasing hormone. It is not
known whether this inhibition occurs in horses naturally. The results of this research, although limited, indicate that physiologic alteration of thyrotropin releasing
hormone secretion likely does not alter normal growth
hormone or prolactin secretion.
Literature Cited
Bass, R. T., B. L. Buckwalter, B. P. Patel, M. H. Pausch, L. A. Price,
J. Strnad, and J. R. Hadcock. 1996. Identification and characterization of novel somatostatin antagonists. Mol. Pharmacol.
50:709–715.
Cabell, S. B., and K. L. Esbenshade. 1990. Effect of feeding thyrotropinreleasing hormone to lactating sows. J. Anim. Sci. 68:4292–4302.
Clark, R. G., L. M. Carlsson, B. Rafferty, and I. C. Robinson. 1988. The
rebound release of growth hormone (GH) following somatostatin
infusion in rats involves hypothalamic GH-releasing factor release. J. Endocrinol. 119:397–404.
Colborn, D. R., D. L. Thompson, Jr., M. S. Rahmanian, and T. L. Roth.
1991. Plasma concentrations of cortisol, prolactin, luteinizing hormone, and follicle-stimulating hormone in stallions after physical
exercise and injection of secretagogue before and after sulpiride
treatment in winter. J. Anim. Sci. 69:3724–3732.
Convey, E. M., H. A. Tucker, V. G. Smith, and J. Zolman. 1973. Bovine
prolactin, growth hormone, thyroxine and corticoid response to
thyrotropin-releasing hormone. Endocrinology 92:471–476.
Gentry, L. R., D. L. Thompson, Jr., and A. M. Stelzer. 2002. Treatment
of seasonally anovulatory mares with thyrotropin releasing hormone and/or gonadotropin releasing hormone analog. J. Anim.
Sci. 80:208–213.
Gill, J. L., and H. D. Hafs. 1971. Analysis of repeated measurements
of animals. J. Anim. Sci. 33:331–336.
Guyton, A. C., and J. E. Hall. 1996. Textbook of Medical Physiology.
9th ed. W. B. Saunders, Philadelphia.
Henneke, D. R., G. D. Potter, J. L. Kreider and B. F. Yeates. 1983.
Relationship between body condition score, physical measurements and body fat percentage in mares. Equine Vet. J.
15:371–372.
Jacobs, L. S., P. J. Snyder, J. F. Wilber, R. D. Utiger, and W. H.
Daughaday. 1971. Increased serum prolactin after administration
of synthetic thyrotropin-releasing hormone (TRH) in man. J. Clin.
Endocrinol. Metab. 33:996–998.
2351
Johnson, A. L. 1987. Seasonal and photoperiod-induced changes in
serum prolactin and pituitary responsiveness to thyrotropin-releasing hormone in the mare. Proc. Soc. Exper. Biol. Med.
184:118–122.
Johnson, A. L., and S. E. Becker. 1987. Effects of physiologic and
pharmacologic agents on serum prolactin concentrations in the
nonpregnant mare. J. Anim. Sci. 65:1292–1297.
Kennedy, S. R., D. L. Thompson, Jr., H. E. Pruett, P. J. Burns, and
R. Deghenghi. 2002. Growth hormone response to a novel GH
releasing tripeptide in horses: Interaction with GnRH, TRH, and
sulpiride. J. Anim. Sci. 80:744–750.
Konaka, S., M. Yamada, T. Satoh, H. Ozawa, E. Watanabe, K. Takata,
and M. Mori. 1997. Expression of thyrotropin-releasing hormone
(TRH) receptor mRNA in somatotrophs in the rat anterior pituitary. Endocrinology 138:827–830.
Laverriere, J. N., A. Tixier-Vidal, N. Buisson, A. Morin, J. A. Martial,
and D. Gourdji. 1988. Preferential role of calcium in the regulation
of prolactin gene transcription by thyrotropin-releasing hormone
in GH3 pituitary cells. Endocrinology 122:333–340.
Magnan, E., M. Cataldi, V. Guillaume, L. Mazzocchi, A. Dutour, H.
Razafindraibe, N. Sauze, M. Renard, and C. Oliver. 1994. Role of
growth hormone (GH)-releasing hormone and somatostatin in the
mediation of clonidine-induced GH release in sheep. Endocrinology 134:562–567.
Miki, N., M. Ono, and K. Shizume. 1988. Withdrawal of endogenous
somatostatin induces secretion of growth hormone-releasing factor in rats. J. Endocrinol. 117:245–252.
Scanes, C. G., and S. Harvey. 1989. Somatostatin inhibition of thyrotropin-releasing hormone- and growth hormone-releasing factor-induced growth hormone secretion in young and adult anesthetized
chickens. Gen. Comp. Endocrinol. 75:256–264.
Smith, V. G., R. R. Jacker, J. H. Burton, and D. M. Veira. 1977. Response
of bovine serum prolactin and GH to duodenal, abomasal, and
oral administration of thyrotropin-releasing hormone. J. Dairy
Sci. 60:1624–1628.
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., and L. R. Gentry. 2001. Pituitary
hormone and insulin responses to infusion of amino acids and Nmethyl-D,L-aspartate in horses. J. Anim. Sci. 79:735–744.
Thompson, D. L., Jr., 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.
Thompson, D. L., Jr., 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, D. L., Jr., 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 hormonereleasing hormone and effects of exercise, sexual stimulation, and
pharmacological agents. J. Anim. Sci. 70:1201–1207.
Thompson, D. L., Jr., J. J. Wiest, and T. M. Nett. 1986. Measurement
of equine prolactin with an equine-canine radioimmunoassay:
Seasonal effects on the prolactin response to thyrotropin releasing
hormone. Domest. Anim. Endocrinol. 3:247–252.
Wehrenberg, W. B., B. Bloch, and N. Ling. 1985. Pituitary secretion
of growth hormone in response to opioid peptides and opiates is
mediated through growth hormone-releasing factor. Neuroendocrinology 41:13–16.
Zabavnik, J., G. Arbuthnott, and K. A. Eidne. 1993. Distribution of
thyrotrophin-releasing hormone receptor messenger RNA in rat
pituitary and brain. Neuroscience 53:877–887.