| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Endocrinology (F.S., P.V., G.C., L.M., V.R., C.M., E.M., A.P.), Pharmaceutical Sciences (M.M.), and Plastic Surgery (G.G.), University of Pisa, 56124 Pisa, Italy; Endocrinology Unit, University of Pavia, Fondazione Salvatore Maugeri IRCCS (L.C.), Pavia, Italy; Department of Endocrinology, University of California (I.J.C.), Los Angeles, California 90095; and Section of Endocrinology, Boston University School of Medicine (J.D.S., L.E.B.), Boston, Massachusetts 02118
Address all correspondence and requests for reprints to: Dr. Ferruccio Santini, Department of Endocrinology, University of Pisa, Via Paradisa, 2, 56124 Pisa, Italy.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
| Subjects and Methods |
|---|
|
|
|---|
Nine volunteer healthy female subjects (age range, 3239 yr) entered the study. Thyroid disease was excluded by ultrasound examination, hormone assays, and tests for antithyroid antibodies. Following instructions from the manufacturer, 20 ml cream containing 20 mg L-T4 (Somatoline, Manetti & Roberts, Florence, Italy) were applied on d 1, followed by 10 ml/d for 2 wk. At 0800 h each day, subjects applied the thyroid cream to abdominal and thigh skin. The cream requires 20 min for complete absorption. Blood samples were obtained at 0800 h the day before applying the cream; 6, 12, and 24 h after the first application; and on d 15, 24 h after the last application. Serum was assayed for total T4, free T4, total T3, free T3, TSH, and rT3. Urinary iodide was measured in a morning sample obtained the day before starting the treatment and 24 h after the last cream application. Informed consent was obtained from all volunteers.
Experiment 2
Two months after the first experiment, six healthy female volunteers from the original group self-administered 30 ml L-T4 cream. Blood was drawn before and 3, 6, 12, and 24 h after the application. Serum samples were processed as described for experiment 1.
To measure physiological variations in serum TSH and rT3 concentrations, serum samples were obtained from six additional volunteers who did not use the L-T4 cream. Samples were drawn at 0800, 1100, 1400, and 2000 h for measurements of serum TSH and rT3.
Chromatographic analysis of somatoline
To confirm that the cream contained T4, but not T3 or rT3, a sample of the cream was analyzed by reverse phase HPLC. Ten grams of cream were dissolved at 60 C in 100 ml 0.1 N methanolic hydrochloric acid. The resulting mixture was cooled to 0 C and then filtered using first filter paper (no. 1, Whatman, Clifton, NJ) and then an Acrodisc LC (polyvinylidene difluoride, 0.45 µm) filter. Analytical HPLC analysis was performed on a Beckman System Gold apparatus (Fullerton, CA) under the following conditions: Beckman Ultrasphere Ods 5 column (4.6 mm x 25 cm); eluant A, methanol/phosphoric acid (100:0.2, vol/vol); eluant B, water/phosphoric acid (100:0.2, vol/vol); gradient from 5065% A over 22 min; flow, 2.0 ml/min; UV detection, 230 nm (6).
Hormone assays
Serum TSH was measured by an ultrasensitive chemiluminescent assay (Immulite 2000, Diagnostic Products, Los Angeles, CA). Serum total T3 and T4 were measured using indirect methods (ICN Biomedicals, Inc., Milan, Italy). rT3 was measured by specific RIA (BioChem ImmunoSystems Spa, Bologna, Italy). Free T3 and free T4 concentrations were determined by competitive RIA technique (Amerlex-MAB kit, Johnson & Johnson Clinical Diagnostic, Milan, Italy). Normal values in our laboratory are as follows: TSH, 0.43.7 µU/ml; total T4, 54154 nmol/liter; total T3, 1.53.2 nmol/liter; rT3, 0.140.54 nmol/liter; free T3, 3.88.4 pmol/liter; free T4, 8.421 pmol/liter.
Urinary iodide was measured using an autoanalyzer apparatus (Technicon, Rome, Italy) (7). Results were expressed as micrograms of iodide per milligram of creatinine.
Test transdermal cream to demonstrate transcutaneous diffusion
[125I]T4 and unlabeled T4 were independently introduced into a liposome cream (Novasome A, IGI, Inc., Buena, NJ) previously demonstrated to facilitate transcutaneous peptide transfer (8, 9).
T4 diffusion assessment
Epidermal samples taken from human surgical discards were drawn across Franz Diffusion Cell (TheraTech, Inc., Salt Lake City, UT) lumens according to the manufacturers instructions. Epidermal samples so placed separated tested substances from small reservoirs of 0.9% saline. The substances tested were [125I]T4 in liposome cream, [125I]T4 in saline, unlabeled T4 in liposome cream, unlabeled T4 in saline, liposome cream alone, and saline alone. A 50-µl aliquot of each test substance was placed on the exposed side of an epidermal sample. The saline was withdrawn from the reservoir and replaced with fresh saline after 2, 4, and 24 h.
The withdrawn samples from the reservoirs below radiolabeled substances and controls were counted in a
-counter. The withdrawn samples from the reservoirs below cold T4 samples and controls were subjected to measurement of total T4 and total T3 using standard RIA kits (ICN Biomedicals, Inc., Orangeburg, NY). Customized standards were designed with known concentrations of each hormone in saline. Each condition was repeated with four randomly selected skin samples.
Measurement of D3 activity
A skin specimen was obtained from a 13-yr-old girl who underwent plastic surgery for an ectopic breast, and D3 activity was determined by measuring the conversion of T3 to 3,3'-T2 as previously described (10). T3 is the favorite physiological substrate for D3, and using T3 as a substrate for the 5-monodeiodinase assay makes the assay more sensitive and precise than when T4 is used. For this reason we preferred to use T3 instead of T4 for measurement of D3 activity in skin. The sample was homogenized in 50 mmol/liter phosphate buffer (pH 7.4) containing 10 mmol/liter EDTA and 0.4 mmol/liter phenylmethylsulfonylfluoride (Sigma-Aldrich Corp., St. Louis, MO). Microsomes were prepared by ultracentrifugation and suspended in the same buffer by sonication. The protein concentration was determined using a microassay reagent (Bio-Rad Laboratories, Inc., Richmond, CA). Microsomes (540 µg protein) were incubated with [125I]T3 (0.62.4 nmol/liter; Amersham International, Milan, Italy) in the presence of dithiothreitol (10 mmol/liter; Sigma-Aldrich Corp.), in 0.1 mol/liter Tris buffer, pH 7.4, at 37 C for 30 to 240 min (final volume, 0.25 ml). The reaction was stopped by adding 0.1 ml 5% BSA, followed by 2 vol ethanol. The mixture was centrifuged and [125I]3,3'-T2 was quantified in an aliquot (75 µl) of the supernatant by binding to a highly specific rabbit anti-3,3'-T2 antibody in a 16-h incubation at 4 C. The antibody bound [125I]3,3'-T2 was precipitated by adding a previously determined excess of goat antirabbit
-globulin.
Statistical analysis
Data for D3 activity are reported as the mean of duplicate determinations that differed from each other by less than 10%. Other data are reported as the mean ± SD. Variables were compared with controls using t test and ANOVA.
| Results |
|---|
|
|
|---|
After 2 wk of treatment with L-T4-containing creams, urinary iodide did not change significantly relative to pretreatment values (posttreatment, 77.8 ± 16 µg/g creatinine; pretreatment, 48.1 ± 10.3 µg/g creatinine; P > 0.05). T4 and T3 concentrations did not change significantly during the treatment period. rT3 concentrations increased significantly 6 and 12 h after cream application (Fig. 1
). Serum TSH concentrations exhibited a circadian rhythm in subjects receiving L-T4 cream, which was similar to that observed in controls not receiving the L-T4-containing cream (Fig. 2
). Further, serum TSH measured after 2 wk of cream application did not differ from pretreatment values. No significant intraday variation of rT3 concentration was observed in controls not applying the L-T4 cream (data not shown).
|
|
After a single cutaneous application of 30 mg L-T4, no changes in serum T4 or T3 concentrations were observed. A significant increase in serum rT3 was seen 6 and 12 h after L-T4 application (Fig. 3
). No changes in the TSH rhythm were observed relative to controls (Fig. 4
).
|
|
Samples taken from Franz cells treated with [125I]T4 in saline, saline alone, and liposome vehicle alone did not have significantly different quantities of radioactivity measured with the
-counter. In contrast, significantly more radioactivity was detected in samples taken from Franz cells treated with [125I]T4 in the liposome vehicle (Fig. 5
). At 2 h, radioactivity equal to 4.3 ± 0.4% of the original sample (P < 0.005) was observed. The activity equaled 5.4 ± 1.4% of the original sample at 4 h (P < 0.01) and 10.2 ± 3.4% of the original sample at 24 h (P < 0.05). Total T3 was not detectable at any time point under any tested condition. The detection limit of the T3 assay was 1% of the compound transferred or 0.1% of the original compound applied. Over the course of the experiment, total T4 concentrations represented 10.5 ± 4% (P < 0.05) of the total compound transferred (or
1% of the original compound).
|
The increase in the serum rT3 concentration, without a concomitant rise in serum T4 concentration, led us to investigate and characterize the presence of inner ring deiodinating activity in human skin. A specimen from breast skin of a young woman was tested for D3 activity immediately after surgical resection. D3 activity increased with enzyme content up to 40 µg microsomal protein and with incubation time up to 240 min (Fig. 6
). Figure 7
shows the kinetic parameters of 5-monodeiodination of T3 by Lineweaver-Burk analysis. The Km was 1.74 nmol/liter, and the maximum velocity was 23.5 fmol/µg microsomal protein·h.
|
|
Figure 8a
shows a typical HPLC chromatogram of an extract of the cream, and Figure 8b
demonstrates a chromatogram obtained from HPLC analysis of a reference solution of T4, T3, and rT3. The chromatogram of the extract of the cream (Fig. 8a
) contained a well resolved peak corresponding to T4, and did not show any appreciable peaks at the retention times corresponding to T3 and rT3.
|
| Discussion |
|---|
|
|
|---|
Our study was conducted by topically applying amounts of T4 that greatly exceed the physiological production rate in humans. Indeed, when such high doses are taken orally they produce overt thyrotoxicosis. In contrast, the measurement of serial serum samples after skin application of T4 to our volunteers did not show any significant change in concentrations of T4 or T3, suggesting that little T4, if any, reached the bloodstream. Further confirmation came from the failure of serum TSH concentrations to fall despite 2 wk of topical T4 administration.
Skin provides a barrier between the body and the environment, arresting the penetration of microorganisms or destructive chemicals, absorbing radiation from the sun, and preventing the loss of fluids (12, 13). The primary compartment that limits the percutaneous absorption of compounds is the stratum corneum, a highly differentiated structure made of "bricks" of bundled, water-insoluble proteins (the corneocytes) embedded within a continuous lipid-rich matrix. It is mainly through intercellular lipid that skin allows some permeation of almost every substance, although rates of penetration of different materials may vary by several thousand fold. Skin contains a wide range of enzymatic activity, including oxidative, reductive, hydrolytic, and conjugative reactions, as well as a full complement of drug-metabolizing enzymes (14). Eventually, the extent of blood flow regulates resorption of compounds by the cutaneous microvasculature (15). Thus, to undergo percutaneous absorption, a compound must penetrate the stratum corneum, diffuse through the viable epidermis into the dermis, and gain access to the systemic compartment through the vascular system.
We investigated the mechanisms making the human skin a barrier to T4, as the hormone could be either blocked on the skin surface or metabolized before entering the bloodstream. Ten milligrams of T4 contain approximately 6.5 mg iodide that would be mostly removed from the molecule by T4 metabolizing enzymes once T4 had entered the bloodstream. Although the metabolic clearance rate of T4 is slow, its serum half-life approximating 7 d, the iodide contained in a daily dose of T4 administered for 2 wk would saturate the thyroid and appear in the urine. In our volunteers, urinary iodide concentrations did not change significantly relative to measurements preceding T4 administration. These data suggest that the majority of T4 applied on the skin surface did not reach the deepest layers where it could be either metabolized or enter the circulation. We did demonstrate a clear increase in serum levels of rT3 612 h after T4 application. Circulating rT3 is cleared about 120 times faster than T4. Therefore, the rise in serum rT3 without a concomitant rise in serum T4 suggests that some T4 did cross the external layers of the skin. but was inner ring deiodinated before reaching the bloodstream.
In our analysis we sought to confirm that thyroid hormone could be transported transcutaneously in an appropriate vehicle. To that end, thyroid hormone was introduced into liposome cream (see Subjects and Methods). The experiment confirmed that a simple liposome cream facilitated thyroid hormone transfer. Using 0.9% saline as a vehicle, T4 did not cross the epidermis in significant quantities. When liposome cream was used as a vehicle, T4 was consistently transported in measurable quantities. Substantial inner ring deiodination was suggested by the fact that only 10% of transferred thyroid hormone remained as T4, and T3 was not detected.
To test the hypothesis that the human skin contains iodothyronine inner ring deiodinase activity, we characterized D3 enzyme activity in a skin specimen from a young woman. The results indicate that iodothyronine inner ring deiodinase activity was abundant, with an affinity constant comparable to those of the iodothyronine inner ring deiodinases in placenta and brain (16, 17), thus confirming that conversion of T4 to rT3 occurs in skin tissue.
The specific iodothyronine inner ring deiodinase D3 has been previously detected in human placenta, brain, and liver (10, 17, 18, 19, 20, 21). D3 has also been detected in skin and gut in other species (22, 23). The cDNA for a selenoprotein with the kinetic properties of D3 has been isolated in the human placenta, and the corresponding mRNA has been detected in both placenta and lung (24). D3 causes irreversible degradation of T4 and T3. T3 is the preferred substrate for D3. Recent studies indicate that D3 is responsible for the differential regulation of T3 levels in selected tissues, such as brain, by interplay with the activating deiodinases D1 and D2 (17, 25, 26, 27). In fetal life, D3 keeps serum T3 levels low while maintaining high levels of rT3 (28). Placental D3 also plays a role in limiting the passage of maternal T3 to the fetus (29, 30, 31). D3 activity has been detected in the fetal rat skin (22). Our data suggest that skin D3 may contribute to limiting fetal transdermal passage of maternal thyroid hormone present in the amniotic fluid.
Our experiments used thigh skin for the in vivo studies and surgical breast skin discards for the in vitro studies. Relative thyroid hormone transfer may differ for skin in other regions of the body. Although our study is reassuring regarding general risks of cutaneous T4 administration, the degree of danger is not known in situations compromising the barrier function of the skin for T4. The skin barrier may be altered in pathological conditions that alter the structure of the stratum corneum (13, 14), including environmental conditions and physical trauma. Agents affecting the permeability of skin may also influence its barrier function. Thus, the use of keratolitics or fat solvents may allow increased T4 penetration through the skin by disrupting the epidermal barrier. Tape stripping of the skin has also been shown to disrupt the stratum corneum barrier (32). Further, total blood flow in skin may vary up to 100-fold, a process primarily regulated by vascular shunts as well as by recruitment of new capillary beds (15). Thus, changes in temperature, sun exposure, or vasoactive compounds may influence skin blood flow and percutaneous absorption of exogenous compounds. We therefore suggest caution with the use of large doses of T4 applied to the skin, particularly in patients with concomitant skin disease or traumatic injuries.
In conclusion, our study indicates that the normal human skin serves as a substantial barrier to T4 passage from the outside surface, with D3 playing an important role in this function by inactivating T4 to rT3.
| Footnotes |
|---|
Received September 12, 2002.
Accepted February 21, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Dentice, C. Luongo, S. Huang, R. Ambrosio, A. Elefante, D. Mirebeau-Prunier, A. M. Zavacki, G. Fenzi, M. Grachtchouk, M. Hutchin, et al. Sonic hedgehog-induced type 3 deiodinase blocks thyroid hormone action enhancing proliferation of normal and malignant keratinocytes PNAS, September 4, 2007; 104(36): 14466 - 14471. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. J. Visser The elemental importance of sufficient iodine intake: a trace is not enough. Endocrinology, May 1, 2006; 147(5): 2095 - 2097. [Full Text] [PDF] |
||||
![]() |
J. Kohrle, F. Jakob, B. Contempre, and J. E. Dumont Selenium, the Thyroid, and the Endocrine System Endocr. Rev., December 1, 2005; 26(7): 944 - 984. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Santini, A. Pinchera, A. Marsili, G. Ceccarini, M. G. Castagna, R. Valeriano, M. Giannetti, D. Taddei, R. Centoni, G. Scartabelli, et al. Lean Body Mass Is a Major Determinant of Levothyroxine Dosage in the Treatment of Thyroid Diseases J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 124 - 127. [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 |