| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Biochemistry (R.W.) and Medicine (J.C.N.), School of Medicine, Loma Linda University, Loma Linda, California 92354; and Quest Diagnostics Nichols Institute (R.B.W.), San Juan Capistrano, California 92690
Address all correspondence and requests for reprints to: R. Bruce Wilcox, Ph.D., Department of Biochemistry, Mortensen Hall, Room 209, Loma Linda University, Loma Linda, California 92354.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Salsalate (Disalcid), a commonly used nonsteroidal antiinflammatory drug, is a known inhibitor of T4 and T3 binding to serum transport proteins (7). In humans, salsalate is absorbed completely through the gastrointestinal tract and then partially hydrolyzed to form salicylate, also an inhibitor of T4 and T3 binding to transport proteins (7, 8). Salsalate has been reported to reach peak serum levels approximately 1.5 h after a single oral dose, then disappear with a half-life of approximately 1 h (9, 10). Doses of salsalate that reduce serum total T4 (TT4) and total T3 (TT3) concentrations cause minor, if any, side effects. Consequently, salsalate administration should provide an acceptable agent for inducing short-term inhibition in thyroid hormone binding to transport proteins in experimental studies. Chronic administration of salsalate is reported to lead to reductions in serum TT4 and TT3 and to transient reduction in TSH (11, 12, 13, 14). However, the acute thyroid and pituitary hormone changes after a single dose of salsalate have not been reported.
Our hypothesis is that acute inhibition of thyroid hormone binding to transport proteins by salsalate will be followed by a redistribution of thyroid hormones from the circulation into target tissues. The objective of this study was to investigate the thyroid hormone changes in healthy individuals after short-term perturbation by salsalate of thyroid hormone transport by serum proteins and their relationship to serum salsalate concentrations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study protocol was approved by the Loma Linda University Institutional Review Board. Five healthy, lean, nonpregnant, unmedicated adults (3 women and 2 men; age, 23–25 yr), selected for normal serum TSH, free T4 (FT4) and T4-binding globulin (TBG) concentrations, volunteered for this study. Blood samples were obtained on days -7, 0, 1, 2, 3, 4, and 6 of the study. Blood samples were taken at 0800 h by venipuncture on each scheduled day, after an overnight fast. Basal samples were obtained on days -7 and 0, and zero-time values were the means of the two basal samples. Salsalate was administered at 0800 h on day 0 immediately after the second basal sample. On this day (day 0), fasting was extended to 1200 h, and additional blood samples were taken at 2 h, 4 h, and 8 h after salsalate ingestion. Days 1–6 were washout days. Each person took salsalate orally, 60–65 mg/kg BW [in our pilot study, the subject with the smallest BW (48 kg) took the recommended dosage of salsalate (3 g), which was calculated to be 62.5 mg/kg BW; to standardize the dosage of salsalate in our study for body weight, we use 60–65 mg/kg]. The individual dosages ranged between 3.4 g and 5.3 g and were above the usual recommended daily dosage of 3.0 g (which is given independent of body weight). Salsalate (Disalcid) was obtained from Goldline Laboratories, Ft. Lauderdale, FL (product no. 16790; lot no. L26801; 750 mg/tablet). Side effects were mild and transient and typical of high-dose salicylate administration. The most common were mild nausea and tinnitus. Sera were separated within 2 h after phlebotomy and frozen at -20 C until analysis. Sera were assayed for TT4, TT3, FT4, free T3 (FT3), total reverse(r)T3 (TrT3), TSH, TBG, transthyretin (prealbumin), albumin, salsalate, and salicylic acid concentrations.
Analytical methods
Sera were assayed for TT4 and TT3 by RIA (normal ranges: TT4, 5.3–10.5 µg/dL; TT3, 70–204 ng/dL), for FT4 by direct equilibrium dialysis (normal range, 0.8–2.7 ng/dL), for FT3 by equilibrium tracer dialysis (normal range, 260–480 pg/dL), for TrT3 by double-antibody RIA (normal range, 15–40 ng/dL), for TSH by third-generation immunochemiluminometric assay (normal range, 0.4–4.2 mU/L, calibrated to the World Health Organization TSH reference standard 80/558), for thyroid hormone binding proteins by immunoassay methods (normal ranges: TBG, 1.2–3.0 mg/dL; transthyretin, 14–42 mg/dL; albumin, 3200–5500 mg/dL). Except for TrT3, assays were performed at Quest Diagnostics Nichols Institute. TrT3 was measured at the University of Southern California Endocrine Services Laboratory, Los Angeles, CA, using a specific rT3 antiserum provided by Dr. T. J. Visser (15).
Salsalate and salicylic acid concentrations were measured by the
high-performance liquid chromatography method described by Harrison
et al. (10). Sera were separated as quickly as possible,
then to 0.5-mL aliquots of serum were added 0.9 mL of 0.27 mol/L
HCl, 10 µg of 
-phenylcinnamic acid as an internal standard
(100 µL of a 0.1 mg/mL solution in methanol; Aldrich Chemical
Company, Inc., Milwaukee, WI, Catalog No. P2,200–1; Lot No. 03627JN),
and 10 mL methylene chloride. The tubes were shaken for 15 min on a
mechanical shaker at 125 cycles/min and centrifuged for 5 min at
750 x g. The methylene chloride phases were separated
and evaporated to dryness. The residues were dissolved in 0.5 mL
methanol. This process was completed within 3 h after phlebotomy.
Then, 20-µL samples were injected into the chromatograph. The mobile
phase of methanol-1% acetic acid (60:40 vol/vol) was pumped at 2.0
mL/min (about 2000 psi) using a solvent delivery system and pressure
monitor (Rainin Instrument Co, Inc., Emeryville, CA, model: Rabbit-HP)
through a stainless steel column (30 cm x 4 mm id) packed with a
high-efficiency, reversed-phase packing (µBondapak C18,
P/N 27324, Waters Associates, Milford, MA). Ultraviolet absorbance was
measured at 300 nm by using a V4 absorbance detector (Isco
Instrumentation Special Co., Lincoln, NE). Salsalate or salicylate
concentrations were determined from the ratios of the salsalate or
salicylate peaks to the -phenylcinnamic acid peaks.
Data analysis and statistics
All sera from a single individual were assayed in a single assay run for each analyte. Individual values are the mean of three determinations (except for TrT3, which was the mean of duplicates). Group means are the means of individual values. Repeated-measures ANOVA was used to test the statistical significance of the differences between each time point after salsalate ingestion, as compared with basal (zero-time).
Considering that the potency of salsalate for the inhibition of
T4 binding to serum proteins is 100-fold greater than that
of salicylate (16), salicylate concentrations were divided by 100 to
provide salicylate relative inhibitory potential, then added to
salsalate concentrations to provide a measure of combined inhibitory
potential (Table 1
).
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The combined inhibitory
potential reached a maximum at 2 h, of which salsalate accounted
for 97%. The combined inhibitory potential at 4 h was 80% of
maximum (salsalate accounting for 95%), then fell to 5% and 3%
of maximum at 8 h and 24 h, respectively (Table 1). It is
reasonable to conclude, therefore, that salsalate per se was
the agent primarily responsible for inhibiting T4 and
T3 binding and that the salicylate (produced by
biotransformation of salsalate) contributed little. Serum TBG,
transthyretin, and albumin concentrations were unchanged throughout the
study. Mean TBG concentrations throughout were 2.23 mg/dL, mean
transthyretin concentrations were 25.27 mg/dL, and mean albumin
concentrations were 4.59 g/dL.
Serum TT3 and TT4 concentrations declined
after salsalate ingestion, as expected. Both reached nadir values at
4 h, and both returned to basal levels by 96 h (Fig. 1A). However, their patterns of decline
and recovery were different. TT3 showed the greater
decline, falling to 45% of basal, whereas TT4 fell to 60%
of basal. At 8 h, TT3 had risen to 51% of basal and
TT4 to 77% of basal (Fig. 1A). Serum TrT3
concentration declined to 23.2 ng/dL (81% of basal) at 2 h after
salsalate ingestion, compared with the basal value of 28.7 ng/dL, then
it rose above the basal value to 33.6 ng/dL (117% of basal) at 8
h (P < 0.001) (Fig. 1A). It returned to basal level by
24 h, remaining at basal concentrations thereafter. The ratios of
TT3 to TT4 and TrT3 to
TT4 were compared in Fig. 1B. The ratio of TT3
to TT4 declined after salsalate ingestion and reached its
nadir at 8 h, whereas the ratio of TrT3 to
TT4 increased and reached its peak at 8 h. The former
returned to basal by 72 h, and the latter returned to basal by
24 h.
|
) are likely underestimates of actual free hormone
concentrations (17). For this reason, the FT4 and
FT3 data during the first 24 h after salsalate
administration were deleted from analysis (the deleted FT4
measurements were 1.29 ng/dL at 2 h and 1.05 ng/dL at 4 h,
1.32 ng/dL at 8 h, and 1.44 ng/dL at 24 h, compared with the
basal value of 1.52 ng/dL; the corresponding FT3
measurements were 204 pg/dL at 2 h and 172 pg/dL at 4 h, 187
pg/dL at 8 h, and 247 pg/dL at 24 h, compared with the basal
value of 358 pg/dL). The ratios of FT3 to FT4
were similar to those of TT3 to TT4.
The status of T4 and T3 binding
to carrier proteins is shown in Fig. 1, D and E. Fig. 1D compares the
time courses of TT4 concentrations with FT4
concentrations, represented as percentages of basal values.
TT4 showed the greater reduction and slower recovery, not
returning to basal levels until 96 h. FT4 did return
to basal levels at 48 h. TT4 was reduced, relative to
FT4, at 48 h, suggesting reduced binding to transport
proteins at a time when salsalate and salicylate were undetectable. The
time courses of TT3 and FT3 are compared in
Fig. 1E. The decline and recovery patterns of the two were similar,
providing no evidence of decreased T3 binding during
recovery. As a measure of T4 to T3 conversion,
FT3 was compared with FT4 in Fig. 1C. The
pattern of FT3 recovery was different from that of
FT4 recovery. FT4 recovered within 24 h,
whereas FT3 did not recover until 96 h.
FT3 was reduced, compared with FT4, between
48 h and 72 h.
Serum TSH concentrations fell to a nadir at 8 h, which was 41% of
basal value (Fig. 1F). This was 4 h after the nadirs of
TT4 and TT3 and coincident with the nadir of
the TT3/TT4 ratio and the peak of the
TrT3/TT4 ratio (Fig. 1B). TSH returned to basal
level by 48 h, at a time when TT4, TT3,
and FT3 (but not FT4) were still reduced. TSH
then rebounded to levels above basal levels at 96 h and continued
to rise for the duration of the study (6 days), whereas all thyroid
hormones remained at basal levels (Fig. 1F).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) to
data previously reported by Harrison et al. (9). The
salsalate peak of 82 µg/mL (0.32 mmol/L) was below its therapeutic
range of 100–300 µg/mL (0.39–1.16 mmol/L; see Ref. 18) and the
salicylate peak of 211 µg/mL (1.32 mmol/L) was within its therapeutic
range of 120–350 µg/mL (0.75–2.19 mmol/L; see Ref. 19). In a
separate study, the potency of salsalate in displacing T4
from serum transport proteins was approximately 100-fold greater than
that of salicylate (16). Similar data have been reported by Bishnoi
et al. (7).
The hormone changes, after binding inhibition, were complex (Figs. 1, A–F). TT4 and TT3 concentrations both
declined, as expected, with TT3 demonstrating a
proportionately greater decline (Fig. 1A). The simplest explanation for
the observed fall in TT4 and TT3 concentrations
is an acute release of T4 and T3 from their
serum transport proteins, resulting in rapid, transient elevations of
circulating free hormone levels associated with redistribution of these
hormones from the circulation into tissue compartments. The decline in
total thyroid hormone concentrations could not be attributed to
reductions in carrier protein concentrations, because these protein
concentrations did not change. Neither can it be attributed to
drug-induced primary hypothyroidism, because TSH did not rise (Fig. 1F).
FT4 returned to basal levels at 48 h, whereas
TT4 did not return to basal levels until 96 h (Fig. 1D). The reduction of TT4, relative to FT4, at
48 h suggests a decrease of T4 binding to carrier
proteins at a time when circulating salsalate and salicylate levels
were negligible (Table 1, Fig. 1D). The reason(s) for this finding is
unclear. Because TT3 and FT3 recovered in
parallel (Fig. 1E), the low TT3 concentration during
recovery can be attributed to the low FT3 concentration
(the T3 available for binding), rather than reduced protein
binding. In theory, the low FT3 might be caused by reduced
peripheral conversion of T4 to T3 or to a slow
reentry of T3 into the circulation from tissue, as compared
with the reentry of T4. The ratios of TT3 to
TT4 and TrT3 to TT4 may represent
the peripheral conversion of T4 to T3 and
rT3 to 3,3'-diiodothyronine (T2), respectively.
The decreased ratio of TT3 to TT4 plus the
increased ratio of TrT3 to TT4 suggest the
conclusion that there was a transient reduction in peripheral tissue
5'-monodeiodination (Fig. 1B).
The declines in circulating TT4 (40% disappearance in 4 h) and TT3 (55% in 4 h) after salsalate ingestion were many folds more rapid than the reported hormone half-lives in healthy adults with normal transport proteins (6–8 days for TT4 and 1 day for TT3) (20). This rate of decline is best explained by reduced hormone binding and a rapid redistribution of T4 and T3 from the circulation into tissue compartments. In healthy adults, the mass of circulating T4 is approximately 450 µg. The tissue T4 pool is also 450 µg. Consequently, the extrathyroidal T4 pool is 900 µg (21). The mass of circulating T3 is 6 µg, and that of tissue T3 is 34 µg, with a extrathyroidal T3 pool of 40 µg (21). Because TT4 was reduced 40% and TT3 was reduced 55%, it could be calculated that approximately 200 µg T4 and 3.3 µg T3 moved from the circulation into tissues within 4 h. The amount of T4 transferred was large, relative to the amount of T3 transferred, even after correcting for a 4-fold greater biological potency of T3. Based on this, we conclude that T4 was the primary effector of the TSH changes observed in this study. Raising the levels of T4 (and T3) in pituitary thyrotrophs and hypothalamic TRH-secreting neurons would be expected to suppress TSH secretion (22). Raising levels of T4, and perhaps T3, in rapidly equilibrating tissues (liver and kidney) may reduce 5'-monodeiodinase activity.
Spencer et al. reported serum TSH suppression after single or multiple doses of thyroid hormones (23). They showed that the rate of TSH decline was largely independent of, whereas the duration of TSH suppression was directly dependent upon, dosage. In that study, the half-life of disappearance of TSH from the circulation was approximately 6 h, which is similar to the 59% decline of TSH at 8 h in the present study.
The recovery of serum total thyroid hormone concentrations can be attributed, in part, to restored thyroid hormone binding to transport proteins, with a subsequent return of thyroid hormones from peripheral tissue compartments into the circulation. It cannot be attributed to increased TSH stimulation of thyroid hormone secretion because TSH was at or below basal levels during T4 and T3 recovery.
There was an apparent biphasic derangement in the relationship of serum
TSH to thyroid hormone concentrations after salsalate ingestion. At
4 h, TSH was suppressed at a time when total thyroid hormone
levels were reduced (Fig. 1, A and F). From 96 h onward, TSH was
elevated at a time when total and free T4 and
T3 were at basal levels. This TSH elevation may be
attributable to the prior exit of thyroid hormones from tissue
compartments as binding to serum proteins recovered. The progressive
increase in TSH levels from 96 h onward occurred without a
detectable response of total or free T4 or T3.
This may be attributable to one or both of two possible mechanisms: 1)
disturbances in the circadian rhythm of TSH secretion without changes
in the mean 24-h TSH concentrations (24, 25); and 2) reduced
bioactivity of TSH, caused by reduced TSH glycosylation (26, 27).
Faber et al. studied the sequelae of a single oral dose of salicylate (1.5 g) and reported reduced TSH levels and increased FT4 levels, as measured by both ultrafiltration and equilibrium dialysis, with a greater increase as determined by ultrafiltration, compared with dialysis (17). There are three points of comparison with our study: 1) In Faber’s study, salicylate concentrations peaked at 0.7 mmol/L. In our study, salsalate peaked at 0.32 mmol/L. Both peaked at 2 h. 2) Faber reported a rapid increase in serum-free T4 levels, by equilibrium dialysis, which we did not detect. 3) They observe a nadir in serum TSH that was 80% of basal value at 2 h. We observed a TSH nadir of 41% of basal value at 8 h. The greater inhibitory potency of salsalate may be one reason we observed reduced free T4 values and Faber observed increased free T4 values using the same equilibrium dialysis free T4 method (see Results).
Chopra et al. reported that salicylate decreases the activity of rat hepatic 5'-monodeiodinase (28). In unpublished data, Chopra has also observed salsalate inhibition of rat 5'-monodeiodinase (personal communication). This may have contributed to reduced T3 formation and impaired rT3 degradation after salsalate ingestion in the present study.
Figure 2 is a cartoon summarizing the
time course of the complex inhibitor-induced hormone changes observed
in the present study. At different time periods, the various patterns
of abnormal hormone relationships were similar to those reported in the
sick euthyroid syndrome (29, 30, 31), indicating that inhibitor-induced
compartmental shifts in T3 and T4 could be
important mechanisms in the genesis of the complex thyroid hormone
disorders associated with acute nonthyroidal illness, and that
administration of thyroid hormone-binding inhibitors may provide useful
pharmacologic models for studies of these changes.
|
Received September 18, 1997.
Revised April 20, 1998.
Revised June 2, 1998.
Accepted June 10, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. H. Samuels, K. Pillote, D. Asher, and J. C. Nelson Variable Effects of Nonsteroidal Antiinflammatory Agents on Thyroid Test Results J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5710 - 5716. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
