| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Obstetrics, Gynecology and Reproductive Sciences (A.J., J.N., K.D.D., J.M., M.C.F., L.J.K., K.P.C.), University of Pittsburgh School of Medicine and Magee-Women’s Research Institute; and Department of Cell Biology and Physiology (K.P.C.), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213
Address all correspondence and requests for reprints to: Kirk P. Conrad, M.D., Department of Physiology and Functional Genomics, University of Florida College of Medicine, 1600 Southwest Archer Road, M552, P.O. Box 100274, Gainesville, Florida 32610-0274. E-mail kpconrad{at}ufl.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have recently shown that relaxin up-regulates vascular gelatinase activity during pregnancy, thereby mediating renal vasodilation, hyperfiltration, and loss of myogenic reactivity in small renal arteries through processing of big ET to ET1–32, which activates the endothelial ETB receptor-nitric oxide pathway (12). During pregnancy and after chronic (5 d) relaxin administration to nonpregnant rats, vascular matrix metalloproteinase (MMP)-2 is specifically implicated in these renal circulatory changes (12). Vascular MMP-2 activity is increased in these small renal and mesenteric arteries as a consequence of increased MMP-2 mRNA and protein expression (13).
To mimic the renal hemodynamic changes of pregnancy, we use chronic sc administration of rhRLX to nonpregnant rats at 4 µg/h, which produced serum levels comparable to those at midgestation when the renal circulatory changes are maximal in the gravid rat (e.g.7). In recent dose-response and time-course experiments, the renal vasodilatory effects of relaxin were noted to occur within several hours after rhRLX administration to nonpregnant rats (14). We hypothesized that, similar to pregnancy and chronic relaxin treatment, vascular MMP-2 is up-regulated and mediates the vasodilatory changes after short-term (hours) administration of rhRLX to nonpregnant rats.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Isolation of small renal and mesenteric arteries
After anesthetizing the rats with pentobarbital (60 mg/kg ip) and decapitation, a laparotomy was performed through the linea alba. The mesenteric arcade and small renal arteries were gently isolated as previously described (6, 12, 15). The inner diameter of small arteries ranged from 50–400 µm. Small arteries were snap-frozen, and stored at –80 C. Protein homogenates were prepared as previously described in detail (12, 15). Protein concentrations were determined by the Bio-Rad (Hercules, CA) protein assay, each in triplicate and then averaged.
Evaluation of MMP-2 and -9 activity by gelatin zymography
For gelatin zymography (Ref. 16 ; and refer to detailed descriptions in Refs. 12 and 15), the homogenates were prediluted in Laemmli’s buffer as needed, combined with an equal volume of Novex Tris-Glycine SDS Sample Buffer (Invitrogen Co., Carlsbad, CA) and then allowed to stand at room temperature for 10 min. For both small renal and mesenteric arteries, 7.5–15 µg of protein was loaded in each lane and then electrophoresed on Novex precast 10% Tris-glycine gels containing 0.1% gelatin for approximately 2 h at 100 V. After renaturing and equilibration, buffer was replaced with fresh Novex Developing Buffer and incubated at 37 C for various lengths of time depending on the experiment (18–24 h for MMP-2 and 6–15 d for MMP-9). The incubation time was optimized to obtain distinct bands, but avoiding overdevelopment and consequent saturation of band densities. Staining with Coomassie Blue G250 and destaining were performed as previously described (12, 15). Gels were stored in distilled H2O and then scanned using a Hewlett-Packard Scan Jet 5370C scanner and HP PrecisionScan Pro version 1.4 computer program (Palo Alto, CA). The images were digitized for analysis by UN-SCAN-IT gel automated digitizing system version 4.3 (Silk Scientific Corp., Orem, UT). Bands of interest were delineated and relative densitometries were calculated based on the number of pixels. To combine the results from several zymograms and avoid inter-gel assay variability, the densitometric ratios rhRLX/vehicle of matched pairs were calculated and averaged for presentation.
Immunohistochemistry
Kidney sections and mesenteric arteries from rats were prepared for immunohistochemistry, to examine renal arteries in situ (see Ref. 13 for details). Frozen sections were cut 7 µm thick by cryostat, mounted on slides, fixed in acetone for 10 min and washed twice in PBS for 3 min. Slides were allowed to dry at room temperature and stored at –40 C until use.
After investigating a dose response (0.1–10 µg/ml), an optimal dose of 3 µg/ml was used for MMP-9 mouse monoclonal antibody (EMD Biosciences Inc., San Diego, CA). According to the manufacturer, this antibody detects both the active and pro forms of the enzyme. For a negative control, the same concentration of mouse IgG1k (Sigma-Aldrich Inc., St. Louis, MO) was substituted for the primary antibody. The primary and the negative control antibodies were preabsorbed overnight with 10% rat serum to increase specificity. The endothelium was labeled with sheep antirat von Willebrand factor at a dose of 1 µg/ml (Accurate Chemical, Westbury, NY). For negative control, the same concentration of sheep IgG (Sigma-Aldrich Inc., St. Louis, MO) was substituted for the primary antibody.
After permeabilization with 0.3% Triton X-100, quenching of endogenous peroxidase with 1% hydrogen peroxide in methanol and blocking with normal horse serum, the sections were incubated with the primary antibody or its negative control overnight at 4 C. After washing the slides in PBS, sections for MMP-9 and von Willebrand factor were incubated for 30 min with a biotinylated, rat-adsorbed secondary antibody against mouse IgG (Vector Laboratories, Burlingame, CA) and a biotinylated, secondary antibody against sheep IgG (Jackson ImmunoResearch Laboratories, West Grove, PA), respectively. The Vectastain Elite ABC kit was used to detect the immunoreactivity of MMP-9 with diaminobenzidine as the chromogen substrate. Some sections were counterstained with hematoxylin for 15 sec. After dehydration in ethanol and xylene solutions, a coverslip was applied using Cytoseal XYL (Stephens Scientific, Riverdale, NJ).
Myogenic reactivity
Renal artery dissection was performed as described by Gandley et al. (6). Briefly, the kidney was bisected longitudinally and the renal medulla was removed. Typically, the main renal artery divided into three branches in each half of the kidney. These further branched into two small interlobar arteries that were gently harvested.
For the study of myogenic reactivity, two arterial segments from each rat were transferred to a dual chamber isobaric arteriograph (Living Systems, Burlington, VT), and each artery was mounted between two glass microcannulae suspended in the chamber containing 3 ml of HEPES-PSS maintained at pH 7.4 and 37 C (6). The unpressurized inner diameter of these small arteries was typically between 100 and 200 µm. The following MMP inhibitors and their respective vehicles were diluted in the HEPES-PSS buffer to the desired concentration: the specific inhibitor of gelatinase, cyclic CTT (CTTHWGFTLC) and the control peptide STT (STTHWGFTLS) both at 30 µM final concentration (12, 17, 18); MMP-2 neutralizing antibody (catalog no. MS-819-P1ABX, Lab Vision Corp., Fremont, CA) 3 µg/ml, and control antibody, IgG1
; MMP-9 neutralizing antibody (catalog no. MAB13405; Chemicon International, Temecula, CA) 3 or 10 µg/ml and control antibody, IgG1. The proximal cannula from one chamber was filled with a MMP antibody, whereas the proximal cannula in the other was filled with IgG1. Then, an artery was attached to the proximal cannulae in each bath and the lumen was filled with MMP antibody or control antibody as the residual blood was washed from the lumen of the vessel. The MMP neutralizing and control antibodies were only applied intraluminally, thereby requiring less of these expensive reagents. On the other hand, cyclic CTT and STT were introduced inside the vessel and added into the bath. The distal cannula was next occluded to prevent flow. The proximal cannula was attached to a pressure transducer, a pressure servo-controller, and a peristaltic pump. The servo-controller maintained a selected intraluminal pressure, which could be rapidly changed in a stepwise manner. The arteriograph was placed on the stage of an inverted microscope with a video camera to provide an image of the vessel. Arterial diameter was obtained by an electronic dimension analyzing system (Living Systems, Burlington, VT) or electronic filar (Lasico, Los Angeles, CA). After mounting between the two glass microcannulae, the arteries were allowed to equilibrate at 60 mm Hg for 0.5 h in the presence of the various inhibitors or their controls. A conditioning stretch from 60–100 mm Hg was then applied over 1 min, and the vessel was allowed to equilibrate for an additional 15 min. The artery was next slightly constricted by 20% of its initial diameter at 60 mm Hg with phenylephrine. This small amount of constriction was used to induce equivalent tone in the arteries and has been shown to optimize myogenic reactivity (6, 9). Myogenic reactivity of small renal arteries isolated from nonpregnant rats treated for 5 d with either rhRLX or vehicle was assessed in a similar manner using MMP-2 and -9 antibodies (10 µg/ml), as well as control antibody. Likewise, the acute in vitro effects of ETB receptor blockade using RES-701–1 (10 µM) and nitric oxide synthase inhibition, NG-methyl-L-arginine (L-NMA) (0.1 mM) on myogenic reactivity of small renal arteries isolated from rats administered rhRLX or vehicle for 4–6 h was studied.
After the phenylephrine constriction, the diameter at 60 mm Hg was recorded and the intraluminal pressure rapidly increased in a stepwise manner to 80 mm Hg. The pressure was maintained at 80 mm Hg until the artery stabilized at a new diameter (
4 min). This new diameter was then recorded and the pressure returned to 60 mm Hg. After another stabilization period, the entire process was repeated three or four times. The data were expressed as a percent change in diameter at 80 mm Hg to the diameter at 60 mm Hg: % Change = [(D80 – D60)/D60] x 100. The several replicate responses from each vessel were averaged.
Statistical analysis
Data are presented as mean ± SEM. For MMP-2 and -9, the densitometric ratio for each matched pair of rhRLX- and vehicle-treated rats was calculated. Averages of the ratios are presented, and the Student’s t test was applied comparing the ratios to 1.0. A ratio of 1.0 would indicate no difference between the groups. Because increases in gelatinase activity were expected in small arteries from rhRLX-treated nonpregnant rats (12, 13), P values for one tail are presented. For the data on myogenic reactivity, two-factor ANOVA was first applied followed by post hoc comparison of group means using the Fisher’s least squared difference test. P values of less than 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Figure 1A depicts a representative zymogram demonstrating no difference in the activity of either the pro or active MMP-2 forms in small renal arteries isolated from nonpregnant rats administered rhRLX vs. vehicle. In contrast, increased gelatinolytic activity of pro MMP-9 was noted in the same arteries (Fig. 1B). To combine the results from several zymograms, the densitometric ratios of rhRLX/vehicle were calculated for each matched pair and then averaged for presentation. There was no significant change in the activity of either the pro or active forms of MMP-2 (ratios of 1.02 ± 0.14 and 1.28 ± 0.21, both P > 0.05 vs. 1.0, n = 10 pairs; Fig. 1C). In the same small renal arteries, however, pro MMP-9 was approximately 70% higher in those isolated from rats administered rhRLX vs. vehicle (ratio of 1.69 ± 0.12, P < 0.0005 compared with 1.0, n = 10 pairs; Fig. 1D). [The active form of MMP-9 (82 kDa) was inconsistently observed and weak; and therefore, it was not quantified.] Incubation times for MMP-2 were considerably shorter than for MMP-9 indicating an overall higher level of expression of MMP-2 in small arteries.
|
) with an approximately 100% up-regulation of pro MMP-9 activity by rhRLX (ratio of 1.96 ± 0.25, P < 0.005 compared with 1.0, n = 10 pairs; Fig. 2D) and no significant change in MMP-2 activities (ratios of 0.91 ± 0.12 for pro MMP-2 and 0.94 ± 0.20 for active MMP-2, both P > 0.05 vs. 1.0, n = 10 pairs; Fig. 2C).
|
. MMP-9 is not localized to the endothelium, but the brown staining appeared to be more intense just beneath the internal elastic lamina (Fig. 3A, inset). To clearly define the endothelium, we also stained for von Willebrand’s factor (Fig. 3C). Staining for von Willebrand’s factor is evident on the luminal side of the internal elastic lamina. [In our hands, bright-field immunohistochemistry is not quantitative unless the signal is rendered virtually on or off by the experimental intervention, which was clearly not the case here (e.g. Fig. 1
). Therefore, we did not attempt to assess differences using this technique, but relied instead on zymography for quantitation as described above.]
|
). This inhibition of myogenic reactivity persisted despite the neutralization of vascular MMP-2 activity with antibody (3 µg/ml; Fig. 4A). However, incubation of small renal arteries with the same concentration of MMP-9 neutralizing antibody (3 µg/ml) partially restored myogenic reactivity of small renal arteries isolated from nonpregnant rats treated with rhRLX (Fig. 4B). A higher dose of MMP-9 neutralizing antibody (10 µg/ml) completely reversed relaxin-mediated inhibition of myogenic reactivity (Fig. 4C). There was no significant effect of either the low or the high dose of MMP-9 neutralizing antibody (Fig. 4, B and C), or the MMP-2 antibody (Fig. 4A) on the robust myogenic reactivity of small arteries isolated from the vehicle-treated control rats.
|
We studied small renal arteries isolated from nonpregnant rats administered rhRLX for 5 d, to corroborate our earlier work, which showed that vascular MMP-2 specifically mediates the inhibition of myogenic reactivity in this setting (12). Our concern was that the MMP-2 neutralizing antibody used in this study was a newer lot number, and each batch requires dialysis to remove sodium azide; these differences may have in some way compromised antibody function in the current study. As before, however, the MMP-2 neutralizing antibody reversed the loss of myogenic reactivity in small renal arteries from rats chronically treated with rhRLX for 5 d (Fig. 5; average percent change in diameter –0.2% vs. 5.6% after incubation with the IgG control antibody, n = 2 rats). In contrast, the MMP-9 neutralizing antibody did not affect the loss of myogenic reactivity (Fig. 5; average percent change in diameter of 5.7 vs. 5.6% after incubation with IgG control antibody, n = 2 rats). On balance, these results support the concept that vascular MMP-9, but not MMP-2 mediates the inhibition of myogenic reactivity after 4–6 h of rhRLX administration, and that vascular MMP-2, but not MMP-9 activity mediates the inhibition of myogenic reactivity after chronic relaxin administration.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The major results of this study are as follows. First, small renal arteries isolated from nonpregnant rats that were treated with rhRLX for 4–6 h demonstrated increased MMP-9, but not MMP-2 activity by gelatin zymography. Comparable findings were observed in mesenteric arteries from the same relaxin-treated rats. Second, MMP-9 protein was expressed by the vascular smooth muscle, but not by endothelium in the small renal arteries. Third, myogenic reactivity was inhibited in small renal arteries isolated from nonpregnant rats treated with relaxin for 4–6 h, similar to that observed in small arteries isolated from midterm pregnant rats and nonpregnant rats treated with relaxin for 2–5 d (6, 9, 12). However, in sharp contrast to vessels isolated from midterm pregnant rats or from nonpregnant rats chronically treated with rhRLX, neutralization of MMP-9 with specific antibodies, rather than neutralization of MMP-2, restored robust myogenic reactivity in small renal arteries isolated from rats after short-term rhRLX administration. Fourth, comparable with midterm pregnancy and with chronic relaxin administration, the endothelial ETB receptor and nitric oxide mediate the inhibition of myogenic reactivity of small renal arteries isolated from rats after short-term relaxin administration. The role of vascular MMP-9 in mediating the vasodilatory effects of relaxin after short-term treatment is an unexpected and novel finding.
The regulation of vascular function by relaxin is complex. An emerging concept is that the mechanisms underlying the vasodilatory actions of relaxin vary with the duration of exposure. In earlier work, we elucidated the mechanisms of pregnancy-induced renal vasodilation, hyperfiltration, and reduced myogenic reactivity of small renal arteries in rats: circulating relaxin stimulates the endothelial ETB receptor-nitric oxide vasodilatory pathway (19). More recently, we showed that vascular gelatinase activity, specifically MMP-2, mediates the activation of the endothelial ETB receptor during midterm pregnancy and in nonpregnant rats chronically administered relaxin through processing big ET-1 to ET1–32 (12). In addition to elevations in both pro and active MMP-2 activities, pro MMP-9 activity was also increased in small renal arteries at least during pregnancy, although overall levels of MMP-2 greatly exceeded those of MMP-9. The increase in vascular MMP-2 activity was due to increased protein and mRNA expression (13). MMP-2 protein was immunolocalized to both vascular smooth muscle and endothelium, although in which of these compartment(s) pregnancy or chronic rhRLX administration up-regulated MMP-2 was not determined (13).
In the current study involving short-term administration of rhRLX to nonpregnant rats, we fully expected up-regulation of vascular MMP-2 as we observed after chronic hormone administration. However, much to our surprise, the zymographic activities of both the pro and active forms of MMP-2 were not increased in small renal arteries isolated from rats treated with relaxin for 4–6 h. Rather, only pro MMP-9 activity was increased. We were unable to determine whether active MMP-9 was also up-regulated because generally low expression precluded accurate imaging and quantitation. Nevertheless, by using MMP-9 neutralizing antibodies we showed a functional role for vascular MMP-9 activity (see below).
We also investigated small mesenteric arteries because both pregnancy and relaxin per se vasodilate the mesenteric circulation (6, 9, 20). Once again, in contrast to midterm pregnancy or chronic administration of rhRLX to nonpregnant rats (12, 13), we observed that only MMP-9 activity was increased by short-term hormone administration.
We found MMP-9 immunoreactivity to be associated with the vascular smooth muscle but not the endothelium in small renal arteries. Because immunohistochemistry is not quantitative unless the signal is rendered virtually on or off by the experimental intervention, we did not attempt to assess differences using this technique and instead we relied upon the zymography for quantitation. This exclusive localization to vascular smooth muscle with increased density of MMP-9 in the immediate subendothelial region beneath the internal elastic lamina, raises the possibility that MMP-9 may be secreted toward the endothelium where it processes big ET to ET1–32 at a gly-leu bond, thereby activating the endothelial ETB receptor and endothelial nitric oxide synthase localized in caveolae (21, 22). Indeed, we demonstrated in the present study that the inhibition of myogenic reactivity after short-term rhRLX administration is mediated by the endothelial ETB receptor and nitric oxide by using specific inhibitors. Thus, a common vasodilatory pathway mediates the inhibition of myogenic reactivity after short-term and chronic rhRLX administration distal to the step of vascular gelatinase activity.
Our functional analyses of myogenic reactivity are consistent with the molecular studies. Neutralization of MMP-9 reversed the inhibition of myogenic reactivity of small renal arteries after short-term rhRLX in vivo. In contrast, MMP-2 neutralizing antibodies were ineffectual. The opposite was also found to be true; that is, neutralization of MMP-2 reversed the inhibition of myogenic reactivity of small renal arteries after chronic rhRLX in vivo, thereby corroborating our earlier work (12). In contrast, MMP-9 neutralizing antibodies were ineffectual is this setting. We also employed a specific gelatinase inhibitor, cyclic CTT, a decapeptide reported by Koivunen and co-workers (17, 18) to inhibit gelatinases with greater selectivity for MMP-2 than -9. We used a dose of 30 µM previously shown to reverse the blunted myogenic reactivity of small renal arteries isolated from midterm pregnant rats and nonpregnant rats chronically administered relaxin (12). In small renal arteries isolated from rats administered rhRLX for 4–6 h, cyclic CTT at the same concentration failed to reverse the inhibition of myogenic reactivity. These results suggest that, at least in small renal arteries, cyclic CTT inhibits MMP-2, but not MMP-9.
MMP-2 is highly and constitutively expressed in a wide range of tissues including the vasculature (13, 23). It should be noted that the even in arteries isolated from the rats administered rhRLX for 4–6 h, basal MMP-2 activity was high as evidenced by the rapid development time of gelatin zymograms (18 h); however, it was not increased by the rhRLX treatment. In contrast, the optimal development time for MMP-9 was 7–15 d indicating overall lower activity. Nevertheless, the functional studies using MMP-2 and -9 neutralizing antibodies confirm that, despite overall lower levels of expression, the increase in vascular MMP-9 by short-term rhRLX administration mediates the reduction in myogenic reactivity. One possible explanation for this apparent paradox relates to anatomical location. That is, after chronic rhRLX administration or during midterm pregnancy, vascular MMP-2 may be up-regulated specifically in caveolae where it would be ideally situated to interact with the endothelial ETB receptor (21) and endothelial nitric oxide synthase (22). Indeed, MMP-2 as well as membrane type 1-MMP and tissue inhibitor of metalloproteinase-2, which facilitate processing of pro to active MMP may localize to caveolae under certain conditions (24, 25). Thus, although overall expression of vascular MMP-2 remains higher than MMP-9 even after short-term rhRLX administration, which increases the latter (supra vide), the MMP-2 may not be located in caveolae, thereby precluding its activation of the endothelial ETB-NO vasodilatory pathway.
The intracellular signaling pathways and transcription factors by which relaxin regulates vascular gelatinases are unknown. Relaxin has been shown to up-regulate MMP-9 in the human acute leukemia cell line, THP-1 (26) and breast cancer cells (27), as well as increase secretion of MMP-9 during uterine and cervical growth and remodeling in the pig (28). One potential mechanism is through activation of nuclear factor B (26). Up-regulation of MMP-2 by relaxin has been demonstrated in lower uterine segment fibroblasts from women (29) as well as rat cardiac fibroblasts (30) and in the left ventricle of spontaneously hypertensive rats treated with rhRLX (31). A possible mechanism is through activation of a tyrosine kinase signaling pathway (29). Interestingly, relaxin up-regulates both MMP-9 and -2 in breast cancer cells (27), during uterine and cervical remodeling in the pig (28), and in human renal fibroblasts (32).
One caveat of the current investigation is that we have not assessed whether inhibition of MMP-9 prevents renal vasodilation and hyperfiltration in the conscious rat model after short-term administration of rhRLX. Unfortunately, the expense of the MMP-9 neutralizing antibodies precludes their application in vivo. Nevertheless, myogenic reactivity of small renal arteries ex vivo has served as a reliable functional bioassay and surrogate of the renal vasodilation and hyperfiltration caused by pregnancy or rhRLX administration to nonpregnant rats (6, 9, 10, 12).
In conclusion, the finding that vascular MMP-9 rather than MMP-2 mediates the reduced myogenic reactivity of small renal arteries isolated from nonpregnant rats administered rhRLX in the short-term is a novel and unexpected finding. It suggests that vasodilatory mechanisms vary according to the time course of rhRLX administration, although all pathways converge on endothelial NOS and nitric oxide production. Further delineation of the differences in vasodilatory mechanisms after the various times of relaxin exposure ranging from minutes to days is needed.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: A.J., J.N., K.D.D., J.M., M.C.F., L.J.K have nothing to disclose. K.P.C. holds two use patents for recombinant human relaxin as primary inventor (with BAS Medical Inc., San Mateo, CA) and is an unpaid consultant for BAS Medical Inc.
First Published Online October 19, 2006
Abbreviations: ET, Endothelin; L-NMA, NG-methyl-L-arginine; MMP, matrix metalloproteinase; rhRLX, recombinant human RLX; RLX, relaxin.
Received July 24, 2006.
Accepted for publication October 6, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B in human THP-1 cells. Ann NY Acad Sci 1041:314–316This article has been cited by other articles:
 |
|
 |
|
  C. A. Reichel, M. Rehberg, P. Bihari, C. M. Moser, S. Linder, A. Khandoga, and F. Krombach Gelatinases mediate neutrophil recruitment in vivo: evidence for stimulus specificity and a critical role in collagen IV remodeling J. Leukoc. Biol., April 1, 2008; 83(4): 864 - 874. [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 |
