| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas 77030
Address all correspondence and requests for reprints to: Marvin L. Meistrich, Ph.D., Department of Experimental Radiation Oncology, Box 66, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030. E-mail: meistrich{at}mdanderson.org.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Treatment with agents that suppress ITT, such as GnRH antagonists, can restimulate spermatogonial differentiation in irradiated rats (4), which starts at 4 wk of treatment. The addition of exogenous testosterone to GnRH-antagonist-treated rats blocks recovery of spermatogenesis, further indicating that increased ITT is responsible for spermatogonial arrest (5). Because ITT levels drop by the first week of treatment (6), the reasons for this 4-wk delay in recovery of spermatogonial differentiation are not known.
We have also observed that testicular edema in irradiated rat testes appears to decrease after GnRH antagonist treatment, which suggested that the edema was related to the differentiation blockade. Testicular fluids reside in two compartments, the seminiferous tubules and the interstitium (7). Seminiferous tubule fluid is produced by the Sertoli cells, and its volume is regulated by reabsorption in the efferent ductules (8). Interstitial fluid originates primarily from the testicular vasculature, and its volume is influenced by changes in blood flow, vascular permeability, physiological osmotic pressure differences, and lymphatic drainage (9). Various studies have indicated that the levels of interstitial fluid can be increased by a variety of factors. These include testosterone (10), human chorionic gonadotropin (hCG) or endogenous LH (11), direct effects of GnRH agonists (12), Leydig cell factors other than testosterone (13), and factors produced by Sertoli cells (14).
Exposure to various toxicants also can produce testicular edema. Testicular interstitial fluid production has been shown to increase after radiation (15) and increase in volume after busulfan (16), procarbazine (17), dibromochloropropane (DBCP) (18), and hexanedione (Boekelheide, K., personal communication) treatment. We have shown that three of these, radiation (4), procarbazine (19), and DBCP (20), increase ITT concentrations. Furthermore, it has been shown that hexanedione (21), procarbazine (22), and DBCP (20) also induce blockades in spermatogonial differentiation. These correlations indicate that increased ITT, testicular edema, and the block in spermatogonial differentiation may be interrelated.
In this study, we investigated the role of testicular edema in producing the spermatogonial blockade. Although the spermatogonial blockade is related to ITT levels (3, 4), the delay in development of the blockade after irradiation and its reversal after GnRH treatment did not fit with our preliminary data on the rapid time course of changes in ITT concentrations (2, 6). We propose that the edema, which is modulated by changes in ITT, may be more directly involved in the spermatogonial blockade. To test this hypothesis, we measured the time course of changes in testosterone concentration, interstitial fluid volume (IFV), and the tubule differentiation index after irradiation, and after initiation of GnRH antagonist treatment, with or without concomitant steroid hormone treatment.
| Materials and Methods |
|---|
|
|
|---|
Rats were anesthetized with a ketamine (65 mg/kg) and acepromazine (0.65 mg/kg) mixture im and irradiated by a 60Co
-ray unit (Eldorado 8; Atomic Energy Canada Ltd., Ottawa, Ontario, Canada). Single doses of 3.5 and 6.0 Gy were given at a dose rate of 0.96 Gy/min (5). The field extended distally from a line about 6 cm above the base of the scrotum.
Hormone treatment
The GnRH antagonist, acyline, was obtained from the National Institutes of Health National Institute of Child Health and Human Development (Bethesda, MD) and freshly prepared in sterile water for injection before each use. To determine the dosing regimen required for continuous suppression of testosterone, two preliminary experiments were performed. First, the rats (five rats per group) were given a single dose of acyline at 0.5 or 2.0 mg/kg sc, and blood was collected by saphenous vein nicking under ketamine/acepromazine anesthesia twice weekly for 2 wk. The serum was assayed for testosterone concentration. Based on the results of this experiment, a second experiment was conducted in which rats (five rats per group) were given either acyline as a single dose or as two weekly doses of 1.5 mg/kg, and blood was collected as above, and the serum was assayed for testosterone concentration. The goal was to maintain greater than 90% suppression of serum testosterone levels.
For the main study, starting at 15 wk after 6-Gy irradiation, acyline was given weekly at 1.5 mg/kg for up to 8 wk. Groups of irradiated, GnRH-antagonist-treated rats were also given sc SILASTIC brand implants (2 mm id, 3.2 mm od) (Dow Corning Corp., Midland, MI) containing either 17ß-estradiol (E2) (0.5-cm length) or testosterone (24-cm length given as three 8-cm capsules) (Sigma-Aldrich, St. Louis, MO).
Experimental design
In experiment 1, rats irradiated with 3.5 or 6 Gy were killed by anesthesia overdose (pentobarbital at 100 mg/kg ip) at intervals up to 30 or 60 wk after irradiation, respectively. Age-matched, unirradiated control rats were killed at several corresponding time points. In experiment 2, 6-Gy-irradiated rats were given GnRH antagonist with or without E2 or testosterone implants starting at 15 wk after irradiation. GnRH-antagonist-treated rats were killed at intervals of 18 wk of treatment, corresponding to 1623 wk after irradiation. Different intervals were chosen for different GnRH antagonist and hormone treatments based on the known kinetics of response to different hormone combinations (6). Age-matched unirradiated and irradiated, untreated rats were killed at one or more corresponding time points. Initially, four rats were used per group, but additional rats were added during repeats of the experiments to increase the precision of data points with high variability.
Intratesticular interstitial fluid and seminiferous tubule fluid collection
Rats were killed by anesthesia overdose with 100 mg/kg pentobarbital ip. Each testis was weighed with the tunica albuginea intact, and the right testis was fixed in Bouins fluid for histological determination of tubule differentiation index (TDI), which is the percentage of seminiferous tubules containing at least three differentiated germ cells. Interstitial and seminiferous tubule fluid was collected from the left testis and measured using a modification of the method of Rhenberg (24). Briefly, the testis was pulled from the inguinal canal through a scrotal incision, the testicular artery was ligated with surgical suture, and the testis was carefully removed and trimmed of fat. An approximately 30-cm silk suture was attached to the caput end of the testis with a wound clip. Four small 1-mm incisions that did not intersect were cut into the caudal end of the testis using a fresh no. 11 scalpel blade. The testis was suspended inside a 10-ml syringe barrel by the attached suture, which was taped to the outside of the syringe. A preweighed silicone-coated microcentrifuge tube was attached to the luer-lock tip of the syringe. The syringe assembly was placed inside a 50-ml centrifuge tube and was spun for 30 min at 60 x g at 4 C. At the end of centrifugation, the microcentrifuge tube, now containing the interstitial fluid, was weighed, and 200 µl cold PBS was added, the fluid briefly spun, and then the supernatant stored at 20 C until analysis. IFV was calculated by subtracting the weight of the microcentrifuge tube before fluid collection from the weight of the tube after fluid collection.
For seminiferous tubule fluid collection, the testis was taken from the syringe and the wound clip was removed. The tunica was cut and peeled back to the caput end, exposing the seminiferous tubules. Using the tunica as a handle, the seminiferous tubules were rinsed four times in separate beakers of M199 media (Invitrogen, Carlsbad, CA; catalog no.12350-039, 0.3% BSA, pH 7.4) and blotted on gauze after each rinse. After the rinses, the tunica was removed, and the seminiferous tubules were pushed through a 10-ml syringe into a 15-ml centrifuge tube. The tubes were centrifuged for 30 min at 2500 x g at 4 C. At the end of centrifugation, the supernatant was removed, put into a preweighed microcentrifuge tube, and weighed; 200 µl cold PBS was added, and then the fluid was stored at 20 C until analysis.
RIA
Serum, testis homogenate, interstitial fluid, and seminiferous fluid testosterone levels were measured using an antitestosterone antibody-coated tube RIA kit (catalog no. DSL 4000; Diagnostic Systems Laboratories Inc., Webster, TX) with modifications from the manufacturers instructions (5). The testosterone standards were prepared in PBS with 0.1% gelatin. The intraassay coefficient of variation (CV) was 6.8%, and the interassay CV was 9.0%. The assay for rat interstitial fluid was validated for by assessing linearity (R2 = 0.995; slope = 0.80) and performing added-mass studies (average percent recovery = 112 ± 22%) using fluid collected from the above experiments.
Statistical analysis
Results were presented as either mean ± SEM calculated from untransformed data or as the mean ± SEM calculated from log-transformed data obtained from individual rats. The statistical significance of differences was determined using SPSS version 11.5 software (Lead Technologies, Chicago, IL) using Students t test. If more than two groups were being compared, ANOVA was used first, with P < 0.05 being considered significant.
| Results |
|---|
|
|
|---|
|
Interstitial fluid testosterone (IFT) averaged 83 ng/ml in unirradiated rats across all time points, and there were no significant differences with age of the rats. The CV for IFT between rats was 37% for unirradiated rats and 18% for 6-Gy irradiated rats at 19 wk after irradiation. Irradiation doubled IFT by 2 wk after irradiation for 3.5-Gy and by 6 wk for 6-Gy rats (Fig. 1C
). IFT levels then plateaued through the remainder of the experiment. Between 6 and 30 wk after irradiation there were no significant differences between 3.5- and 6-Gy irradiated rats, indicating that IFT levels may not be radiation dose dependent within this range of doses.
The TDI in 6-Gy irradiated rats fell to 10% at 4 wk, but it increased to 25% at 6 wk, dropped to 1% at 8 wk, and remained at 1% or less through 60 wk after irradiation (Fig. 1D
). Similarly, in the 3.5-Gy irradiated rats, the TDI was 26% at 4 wk, increased to 60% at 6 wk, dropped to 33% at 10 wk, and remained at less than 35% through 30 wk after irradiation.
Effects of GnRH antagonist treatment on testes of irradiated rats
We first determined the effective dose of acyline for suppression of testosterone production. Whereas acyline given as a 0.5-mg/kg dose failed to maintain 90% suppression of serum testosterone levels in all rats for 7 d, 2.0 mg/kg sustained this suppression for 11 d after injection (Fig. 2A
). To identify a dose and frequency that would achieve continuous suppression of testosterone, rats were given acyline at 1.5 mg/kg in either a single dose or two weekly doses. The single dose of acyline maintained 99% suppression of testosterone for the first 10 d after injection, whereas all rats given weekly does of acyline could maintain greater than 99% suppression of testosterone (Fig. 2B
). Our results, using acyline at 1.5 mg/kg weekly, were similar to others delivered acyline with osmotic minipumps, but we were able to save the expense of such pumps (25).
|
|
|
IFT concentrations in irradiated rats dropped precipitously, from 170 to 8.7 ng/ml at 1 wk of GnRH antagonist treatment and plateaued at about 2 ng/ml through 8 wk of treatment (Fig. 3C
). The TDI appeared to rise marginally to 0.6% at 3 wk of treatment with GnRH antagonist but rose more appreciably to 5.5% at 4 wk and reached 69% by 8 wk (Fig. 3D
).
Effects of exogenous steroid hormone treatment on testes of GnRH-antagonist-treated irradiated rats
The addition of 0.5-cm E2 decreased testicular weight of irradiated rats to a greater extent than did GnRH antagonist alone. This difference was already significant at 2 wk of treatment, reaching 49% of untreated irradiated rat testes at 4 wk of treatment (Fig. 3A
). Addition of 24-cm testosterone significantly blunted the effect of GnRH antagonist treatment by decreasing testis weight to only 78% of untreated irradiated rats. There was no significant different in body weights between GnRH-antagonist-treated rats and rats treated with GnRH antagonist and 0.5-cm E2 implants (P > 0.1).
The addition of 0.5-cm E2 capsules to the GnRH antagonist treatment induced an additional decline in IFV (Fig. 3B
), so that it was significantly less than the value for rats given GnRH antagonist alone at 2 wk of treatment and reached a minimum of 58 µl at 2 wk of treatment. Starting from 2 wk of treatment, the IFV values for GnRH- plus E2-treated rats were not significantly different (P > 0.56) from those for unirradiated rats. In contrast, the addition of 24-cm testosterone capsules to the GnRH antagonist treatment largely blocked GnRH antagonist-induced reduction of IFV, as levels averaged 133 µl from wk 28, which was still significantly below the irradiated-only values (P < 0.01).
In irradiated rats treated with GnRH antagonist and exogenous E2, IFT dramatically decreased within 1 wk, reached 0.9 ng/ml after 2 wk of treatment, and plateaued through 4 wk of treatment (Fig. 3C
). Treatment with GnRH antagonist plus exogenous testosterone reduced IFT levels only to between 49 ng/ml at 2 wk of treatment and 27 ng/ml at 8 wk of treatment. These levels were intermediate between those of irradiated-only rats and irradiated rats treated with GnRH antagonist alone.
The additional testosterone treatment did not appear to inhibit the initial GnRH-antagonist-induced rise in TDI at about 4 wk, but it did inhibit any further increase, which remained at an average of 2.6% through 8 wk. In contrast, addition of E2 to the GnRH antagonist treatment accelerated the GnRH-induced recovery of spermatogenesis, with the TDI beginning to rise significantly at 3 wk of treatment (P < 0.001), about 1 wk earlier than with GnRH antagonist alone. The TDI had already reached 30% at 4 wk, which was the longest period of E2 treatment used in this study, compared with 34% at 6 wk with GnRH antagonist alone.
Seminiferous tubule fluid volumes (STFV) and testosterone levels
To obtain a more complete description of the fluid levels, we collected seminiferous tubule fluid and measured its volume (STFV) from unirradiated rats, irradiated rats, and irradiated rats treated for 4 wk with GnRH antagonist with or without E2 or testosterone implants. In unirradiated rats, STFV averaged 219 µl, compared with 50 µl for IFV, indicating that the majority of testicular fluids resided in the seminiferous tubules (Fig. 4A
). In irradiated rats receiving no hormone treatment, STFV decreased to 42 µl, whereas the IFV increased to 241 µl. GnRH antagonist treatment, alone or with added E2, further decreased STFV to less than 10 µl, and addition of testosterone to GnRH antagonist treatment raised STFV to 23 µl as expected from previous reports of effects of testosterone on STFV (27). In contrast to the relationship between increases in IFV and the inhibition of spermatogonial differentiation, changes in STFV were unrelated to this blockade, because GnRH antagonist treatment further reduced STFV.
We also compared ITT, IFT, and seminiferous tubule fluid testosterone (STFT) in the same rats. In general, the same trends in changes in testosterone levels were observed in whole-testis homogenates (ITT), interstitial fluid (IFT), and seminiferous tubule fluid (STFT) (Fig. 4B
). That is, the testosterone concentrations increased after irradiation, decreased after GnRH antagonist treatment, decreased even further upon addition of E2 during the GnRH antagonist treatment, and after addition of testosterone to GnRH antagonist treatment, increased above the levels of GnRH antagonist alone. There were no significant differences between IFT and STFT levels seen in any treatment. There were no significant differences between ITT, IFT, and STFT levels in any of the treatments except for the rats treated with GnRH antagonist and testosterone; ITT levels (41 ± 2.5) in that group were greater than IFT levels (30 ± 3.8) (P = 0.04). Hence the current measurements using IFT can be compared with previous results measuring ITT in rats not treated with exogenous testosterone (28).
| Discussion |
|---|
|
|
|---|
IFT
Although in this study we measured testosterone in testicular interstitial fluid, whereas we previously measured it in testicular homogenates, we have shown with all treatment conditions that IFT levels are similar to ITT levels (5, 6, 29). The causes of the increases in IFT concentrations after irradiation are complex and include increases in LH (30) and decreases in blood flow as a result of decreased testicular mass caused by germ cell loss (31). These could be counteracted by dilution effects of increasing IFV and direct radiation effects on Leydig cells (32). GnRH antagonists reduce IFT by reducing levels of LH, which is responsible for inducing testosterone production by Leydig cells (33, 34).
The addition of E2 to GnRH-antagonist-treated rats further reduced IFT below that of GnRH antagonist treatment alone, possibly by direct action on Leydig cells (35). However, it is not possible to determine whether it induced further suppression of LH, because LH levels were already below the limits of detection (5). Exogenous 24-cm testosterone was used to increase IFT levels in GnRH-antagonist-treated rats. However, despite supraphysiological serum testosterone levels that are produced by 24-cm testosterone implants (5), IFT levels were only 23% of those in untreated irradiated rats. It appears that exogenous testosterone treatment increased IFT in GnRH-antagonist-treated irradiated rats by equilibration of the high testosterone levels in the serum with the interstitial fluid.
Volume of interstitial fluid
The changes in IFV after irradiation could have been a direct consequence of irradiation or an indirect effect resulting from changes that follow depletion of germ cells and loss of testis mass. The fact that IFV initially decreased at 2 wk after irradiation indicates that the increase was not an immediate direct effect of irradiation. However, radiation effects on the vascular system in other tissues occur over longer time frames, from 425 wk, and hence the increase in IFV could have been a direct late effect of irradiation (36, 37). It is unlikely that there is a delayed effect on the blood-testis barrier, because it has been reported to remain intact after irradiation (38). The mechanism causing the decreased STFV in irradiated rats, which have increased IFV, is not known.
Although there is general agreement that LH or hCG increase IFV (7), there has been some disagreement in the literature about whether testosterone, rather than another Leydig cell product or the action of LH/hCG on other cells, is the major regulator of IFV. We believe that the study by Maddocks and Sharpe (10) that involved elimination of Leydig cells with ethane dimethane sulfonate, conclusively showed that testosterone and not LH is the major direct regulator of IFV. Furthermore, rats treated with ethane dimethane sulfonate do not respond to hCG treatment with increased IFV and vascular permeability, supporting the above conclusion (39). The ways increased testosterone production could lead to edema include changes in testicular capillary permeability, lymph flow, and blood flow (40). The rise in IFV, which began at 6 wk after irradiation in the present study, is consistent with the observations that increased testosterone is responsible for the development of edema. However, the edema progressed after the increase in testosterone, which indicates that the fluid accumulation was a gradual process affected by the changes caused by testosterone.
Another factor responsible for the increase in IFV in the irradiated rats could have been the loss of tubular mass. The passage of fluid between the vasculature and the interstitial space is a function of the pressure difference between the two compartments (7). In the irradiated testis, the loss of tubular mass occurs, but the tunica albuginea does not appear to contract, and hence the interstitial pressure must be very low in the flaccid testis, which would favor fluid flow from the vasculature to the interstitium (31). The losses in testis weight correlated with the increase in IFV through wk 6 after irradiation. However, the additional losses in testis parenchymal weights that occurred after GnRH antagonist treatment with or without E2 was associated with decreases rather than increases in IFV, indicating involvement of another mechanism.
The greater reduction in IFV with GnRH antagonist treatment than with combined testosterone and GnRH antagonist is also consistent with the correlation between the edema and IFT levels. Furthermore, the initial rapid decline in the IFT levels after GnRH treatment followed by a reduction in IFV indicated that the testosterone levels were regulating edema, rather than that the two were simultaneously caused by the reduction in LH. Additionally, GnRH antagonist treatment decreased STFV, also by suppressing testosterone, which normally would stimulate Sertoli cell secretions (41).
The greater extent of reduction of IFV with the combination of GnRH antagonist and E2 than with GnRH antagonist alone may have been a result of further reductions in IFT or a direct effect of E2 on interstitial fluid production or reabsorption. E2 is responsible for seminiferous tubule fluid absorption via action on the efferent ductule epithelium (42), but it is not known whether that could have an indirect effect on reduction of extratubular fluid.
Relationship between testicular edema and spermatogonial differentiation
To elucidate the factors related to the 6- to 8-wk lag between irradiation and the onset of spermatogonial arrest, the time courses of changes in IFV, IFT, and TDI were compared at the early time points after irradiation (Fig. 5
, A and B). IFT levels did rise rapidly at 26 wk after 3.5-Gy irradiation, although the results at the earlier points were variable. After 6-Gy irradiation, IFT remained close to control levels for 4 wk but showed a marked rise at wk 6. This time course differs from a previous result from our laboratory, in which ITT concentrations were already elevated within 2 wk after 6 Gy (2). The IFV, however, first began to significantly increase over the values in unirradiated rats at 6 wk after irradiation after either 3.5 or 6 Gy and showed progressive increases at 8 or 10 wk. Thus at 3.5 Gy, the rise in IFV followed the rise in testosterone; at 6 Gy, the rise in IFV seemed to coincide with the rise in IFT, but the previous data on ITT indicated that the IFV followed the increase in testosterone. Overall, the onset of the spermatogonial blockade appeared to be closer to the increase in edema than the increase in testosterone. Thus the 68 wk required for edema to develop may account for the lag between irradiation and spermatogonial arrest.
|
To elucidate the factors related to the 4-wk delay in initiation of spermatogonial differentiation after suppression of testosterone, the time courses of changes in IFV, IFT, and TDI after initiation of GnRH antagonist treatment were plotted together (Fig. 5C
). The GnRH antagonist treatment induced a major decrease in testosterone within 1 wk, and a previous study showed that GnRH antagonist can induce a marked decrease even within 1 d (6). The decrease in fluid volume first became significant at wk 2 of GnRH antagonist treatment, and the decline continued for one more week, reaching a minimum of about 80 µl at wk 3. At wk 3, spermatogonial differentiation was initiated in only 0.6% of tubules, but this increased to 5.5% of tubules by wk 4. Thus, the initiation of TDI recovery clearly followed the decline in the interstitial edema much more closely than it followed the decline in IFT levels.
Exogenous E2 treatment in conjunction with GnRH antagonist treatment enhanced the decline in IFV and also accelerated the recovery of spermatogonial differentiation (Fig. 4D
). The IFV reached a minimum of about 60 µl at wk 2. This correlated with the increase in TDI beginning at 3 wk and reaching high levels with 4 wk of GnRH antagonist plus E2 treatment, whereas the initiation of TDI recovery required 4 wk and did not reach high levels until 6 wk of treatment with GnRH antagonist alone. These results further support the hypothesis that testicular edema is responsible for the spermatogonial block.
Mechanisms and applications
Although the data presented here are all consistent with a relationship between the interstitial fluid and the block in spermatogonial differentiation, additional studies will be necessary to prove whether or not there is a causal relationship and, if so, the mechanism involved. One approach to determining a causal relationship would be to modulate testicular edema without hormonal changes by pharmacological or surgical methods. Another approach would be to determine whether the interstitial fluid inhibits the differentiation of spermatogonia to spermatocytes in tissue culture (43).
The mechanism by which interstitial edema inhibits spermatogonial differentiation might involve direct or indirect effects. Edematous interstitial fluid may contain proteins or other factors or generate reactive oxygen species (23) that directly suppress spermatogonial differentiation or induce apoptosis of differentiating spermatogonia. The edema might render the seminiferous tubules hypoxic because of the increased separation between the vasculature and the tubules, although preliminary results indicate that this is not the case (Porter, K. L., unpublished observations). However, it is not possible that the edema acts by increasing intratesticular pressure, because the edematous testes were always very flaccid.
In conclusion, we have shown that testicular edema, as measured by IFV, is highly correlated with spermatogonial arrest in irradiated rats. We suggest that irradiation causes an increase in IFV by increasing IFT levels, reduced pressure by loss of parenchymal tissue, and/or vascular damage. Hormonal treatment can reduce IFV by reducing IFT and lead to the reversal of spermatogonial differentiation arrest. It will be of interest to determine whether testicular edema also contributes to the prolonged irradiation- or chemotherapy-induced azoospermia in humans. If it does, understanding how it is caused and its modulation can be important in the development of a treatment to prevent or reverse such iatrogenic sterility in humans.
| Acknowledgments |
|---|
| Footnotes |
|---|
Current address for K.L.P.: U.S. Army Center for Environmental Health Research, 568 Doughten Drive, Fort Detrick, Maryland 21702.
First Published Online November 23, 2005
Abbreviations: CV, Coefficient of variation; DBCP, dibromochloropropane; E2, 17ß-estradiol; hCG, human chorionic gonadotropin; IFT, interstitial fluid testosterone; IFV, interstitial fluid volume; ITT, intratesticular testosterone; STFT, seminiferous tubule fluid testosterone; STFV, seminiferous tubule fluid volume; TDI, tubule differentiation index.
Received July 15, 2005.
Accepted for publication November 14, 2005.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |