| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Division of Neurobiology (M.K., E.F.), German Primate Center, Göttingen D-37077, Germany; Harlow Primate Laboratory (C.L.C.), University of Madison, Madison, Wisconsin 53706; Department of Psychology (C.K.), University of Düsseldorf, 40225 Düsseldorf, Germany; and Department of Psychology (M.K., P.N.), University of Giessen, 35384 Giessen, Germany
Address all correspondence and requests for reprints to: Eberhard Fuchs, Division of Neurobiology, German Primate Center, Kellnerweg 4, D-37077 Göttingen, Germany. E-mail: efuchs{at}gwdg.de
Abstract
This study investigated whether exposing the fetal primate to repeated episodes of maternal stress would have long-lasting effects on the endotoxin-induced cytokine response and corticosteroid sensitivity of peripheral blood cells in juvenile animals. Pregnant rhesus monkeys were acutely aroused on a daily basis for 6 wk using an acoustical startle protocol, either early or late in the 24-wk pregnancy. To quantify cytokine responses and corticosteroid sensitivity in their offspring at 2 yr of age, whole blood cultures were stimulated with lipopolysaccharide and incubated with dexamethasone (DEX). TNF
and IL-6 levels were determined in the culture supernatants. The blood samples were collected from undisturbed monkeys under baseline conditions, as well as in an aroused state induced by a 2 h social separation. Juvenile monkeys from stressed pregnancies had significantly lower cellular cytokine responses compared with the undisturbed controls. When DEX was added to the cell cultures, it systematically inhibited TNF and IL-6 production, bringing the values for control animals down into the range of the prenatally stressed animals. Lipopolysaccharide-induced cytokine production was also markedly suppressed by the experience of acute stress, reducing cytokine responses of controls to the levels found for prenatally disturbed monkeys under baseline conditions. Therefore, this study has demonstrated that prenatal disturbance can induce a lasting change in cytokine biology, which persists well beyond the fetal and infant stage. Further, these effects may be due to elevated hypothalamic-pituitary-adrenal activity in the prenatally stressed animals, because both DEX and acute arousal made the cells from control monkeys appear more similar to those from disturbed pregnancies.
THE PRENATAL ENVIRONMENT is known to influence the development of the nervous, endocrine, and immune systems, with long-lasting effects on offspring postpartum (1). Maternal nutrition, exposure to environmental toxicants, and stressful disturbances of the pregnant female are among the many variables that can affect in utero conditions and impair the maturational trajectory of the fetus (2, 3). It is believed that these effects are often mediated by 1) increased placental transfer of maternal hormones to the fetal compartment; 2) alterations in placental physiology including blood flow; and 3) changes in fetal metabolism impacting oxygen and glucose usage (4).
With regard to prenatal stress, most attention has been directed toward understanding the role of glucocorticoids in mediating the potentially adverse effects on fetal development (5, 6). Maternal cortisol readily transfers across the placenta and thus can exert an influence on the developing organism. In the monkey, cortisol of maternal origin accounts for much of the hormone present in the fetal compartment through mid-gestation (7). At high levels, especially if administered to the mother at pharmacological doses, it exceeds the capacity of the fetus to convert the hormone to less active forms, such as cortisone, or to shunt it back across the placenta (8). Then there can be deleterious effects on organ development, especially for hormone- sensitive glands like the thymus (9). Moreover, the development of the hypothalamic-pituitary-adrenal (HPA) axis in the fetus can be affected, including changes in critical regulatory sites within the brain, e.g. the hippocampus where the activity of the HPA axis is modulated. Administration of dexamethasone (DEX) for 2 d to pregnant rhesus monkeys was found to cause a deterioration of pyramidal neurons in the fetal hippocampus (10). Further, even at 2 yr of age, magnetic resonance imaging revealed that offspring from these DEX-treated pregnancies still had smaller hippocampal volumes, and they secreted higher levels of cortisol when stressed (11). Exposure of pregnant rats to unpredictable light and noise stress similarly increased their corticosterone levels, with cascading effects on their offspring, including a smaller hippocampal size, down-regulated corticoid receptors in the hippocampus, and the maintenance of elevated corticosterone levels in circulation (12).
In addition to effects on the endocrine and nervous systems, it is known that prenatal stress can have a lasting influence on the development of certain immune responses. For example, lymphocyte proliferation and cytolytic responses were found to be reduced in offspring of rats stressed during pregnancy (13, 14, 15, 16). In addition, prior research on monkeys exposed to pregnancy manipulations similar to the ones employed in the current study showed that proliferation responses in mixed lymphocyte cultures were abnormal in the neonates that had been stressed prenatally (17). Further, when ACTH was administered to the pregnant monkey for 2 wk to stimulate endocrine activity in the mother, the offspring exhibited many behavioral alterations, as well as reduced cytokine responses (18). Specifically, when tested at 2 yr of age, they produced lower amounts of IL-6 both in blood and cerebrospinal fluid following an iv challenge with IL-1ß (19).
Many of these effects involving both the HPA axis and immunity appear to reflect alterations in the hormone regulation of lymphoid tissues, especially in the regulatory feedback on proinflammatory cytokines via the glucocorticoid receptor (20, 21). We tested this hypothesis directly by evaluating the corticosteroid sensitivity of leukocytes. An ex vivo assay procedure was used, which involved incubating whole blood cultures with lipopolysaccharide (LPS) to stimulate cytokine release, and then adding DEX, a potent corticoid type 2 receptor agonist, to the cultures (22, 23, 24). This approach enabled us to determine if prenatal stress exposure caused alterations in cytokine production, as well as changed the leukocytes’ sensitivity to corticosteroids. Both IL-6 and TNF were quantified in the culture supernatants because previous studies in humans and monkeys had found that IL-6 tends to be more resistant to glucocorticoid inhibition compared with other proinflammatory cytokines (22, 25). Infants were generated from two types of stressed pregnancies, compromising development either early or later in gestation. This permitted us to investigate whether the changing pattern of maternal hormone transfer across pregnancy, as well as the maturational state of the fetus, would influence the magnitude of the postnatal effects. Several reproductive and developmental traits make the monkey an important species for bridging the findings between rodents and humans. Monkeys give birth to one infant at a time, whereas studies of prenatal stress in litter-bearing animals introduce the additional variable of parallel effects on multiple fetuses. The fetal development of the HPA axis and hippocampus is also more similar in rhesus monkeys and humans, when compared with rodents that continue to have considerable maturation of these structures after birth (26, 27).
Materials and Methods
Subjects
Rhesus monkey infants (Macaca mulatta) were generated from three pregnancy conditions in a breeding colony at the University of Wisconsin. Twenty healthy male and female offspring were selected, and at the time of these postnatal assessments, the juveniles were 24–28 months of age and prepubertal. All procedures were approved by the Institutional Animal Care and Use Committee; the husbandry conditions were in accord with NIH guidelines for the proper care and treatment of laboratory animals.
Housing
The mothers were multiparous females, born and raised at the Harlow Primate Laboratory from defined pedigrees with known histories. These breeders had similar rearing histories: they had been maternally reared through 6 months of age and then housed in juvenile peer groups until menarche, when they were individually housed to facilitate timed-mating protocols. To verify the date of conception, they were time-mated with one adult male during the 4 d around ovulation. After breeding, the gravid female was then transferred to another room in the animal facility, where she remained undisturbed until the appropriate prenatal condition was initiated (see below). Following the experimental manipulations, the female’s pregnancy continued in an undisturbed manner until parturition. Housing conditions were standardized. Pregnant females and the mother-infant pairs were housed in stainless steel cages (0.9 x 0.9 x 0.9 m). The infants were reared by the mother through 6–7 months of age, after which they were initially transferred into small social groups with other weaned monkeys (1.2 x 1.2 x 1.9 m). These groups were composed of 4–8 animals from both control and disturbed pregnancies. At the time of testing, the monkeys were housed as pairs in standardized cages (0.9 x 1.8 x 0.9 m), which enabled samples to be collected quickly. Commercial monkey chow (Ralston Purina, St. Louis, MO) and water were available ad libitum. The light:dark schedule was 16:8, with lights on at 0600. Ambient room temperature was maintained at 21 C.
Prenatal manipulations
Juvenile monkeys from undisturbed, normal pregnancies (control, n = 9) were compared with those that had been disturbed for 6 wk during the 24-wk pregnancy, in either early gestation (early stress, n = 5) or late gestation (late stress, n = 6). The early stress period began on d 50 and lasted until d 92, whereas late stress spanned d 105–147 post conception. The pregnant female was disturbed daily by being relocated to a darkened test room between 1430 and 1600. While located there in a small transport cage, the female was aroused by an acoustical startle protocol for 10 min (three 1-sec broadcasts of a car horn at 1–5 min intervals, 110 decibels). Both before and after the 6-wk manipulations, the females remained undisturbed in their home cages until the natural birth of their infants. Prior research using the same stress paradigm had confirmed that it significantly elevates cortisol above the normal levels of pregnant monkeys (18).
Postnatal assessment: blood-sampling
To assess the cytokine responses of the offspring and their cellular response to DEX, blood samples (7 ml) were collected via femoral venipuncture at 1000–1030, when they were 2 yr of age. Ketamine was administered (10 mg/kg) to facilitate rapid collection of the blood samples and to minimize the stress of handling. One sample was obtained under nonaroused, baseline conditions. Another sample was purposefully obtained in a stressed state, which was induced by a 2 h removal from the home cage and separation from the social partner. To control for an effect of repeat testing, the two conditions were spaced at 3-wk intervals, and the order of the two conditions was counter-balanced across the monkeys. At each blood sampling, 5 ml of blood were collected in a heparinized vacutainer for determination of cytokine responses, and 2 ml were collected in an EDTA tube for a complete blood count.
LPS stimulation of blood leukocytes
Blood samples were processed within 1 h after collection for the cytokine-suppression assay, employing a slight modification of previously published ex vivo whole blood culture techniques (22, 23, 24). Briefly, 3.6 ml blood were combined with 0.4 ml of 10% RPMI 1640 (Life Technologies, Inc., Gaithersburg, MD), and then mixed with LPS and dexamethasone-21-phosphate, both dissolved in sterile phosphate-buffered saline (LPS from Escherichia coli 055: B5, L6529, Sigma, St. Louis, MO). To standardize the volume incubated in every well, 400 µl of blood were always added to 50 µl of LPS and 50 µl DEX or saline solution in the 48-well plate (Costar, Cambridge, MA). Two different concentrations of LPS (10 ng/ml and 100 ng/ml), and four concentrations of DEX (10-8, 10-7, 10-6, and 10-5 M) were tested, resulting in 10 wells per animal. For DEX treatments, the blood was incubated for 1 h with DEX before adding the LPS. All procedures were performed under sterile conditions in a biosafety cabinet. As a further control for possible bacterial contamination, 2 wells per plate were processed in an analogous manner but were left unstimulated, by just adding 100 µl of phosphate-buffered saline to the diluted blood. After 24 h incubation in a humidified atmosphere at 37 C and 5% CO2, the blood cultures were transferred to 0.5 ml microcentrifuge tubes and centrifuged for 7 min at 10,000 x g. Supernatants were collected and stored at -60 C until assayed for cytokines.
Cytokine assays
TNF and IL-6 were determined using commercial ELISA kits (PharMingen, San Diego, CA). This sandwich ELISA uses antihuman monoclonal antibody as the capture antibody and then a second biotinylated antihuman monoclonal antibody as the detection antibody. Briefly, the 96-well plate was coated with the capture antibody and incubated overnight. Then, the plate was blocked with 200 µl of assay diluent and washed. One hundred microliters of a known standard or the experimental sample (diluted 1:600 for IL-6, and 1:50 for TNF) were added to each well and incubated for 2 h. After another wash step, 100 µl of detection antibody were added and the plates incubated for 1 h and washed again. Finally, 100 µl of substrate solution (tetramethylbenzidine and hydrogen peroxide) were added for 30 min, followed by the addition of 50 µl stop solution (2N H2SO4). The plate was read in an ELISA reader (Dynatech Corp., Denkendorf, Germany) at 450 nm and 630 nm.
Complete blood count
A 2 ml blood sample was collected in an EDTA tube, and total leukocyte count and cell differential counts determined by a clinical laboratory familiar with monkey hematology (General Medical Laboratories, Madison, WI).
Statistical analysis
Data are presented in the figures as mean ± SE. For statistical analysis, two-, three-, and four-factor analyses of variance were run. Prenatal conditions (control, early stress, and late stress) were always considered as a between factor. LPS concentration and DEX concentration were analyzed as repeated measures. The analyses are presented sequentially in the Results: 1) baseline cytokine responses without DEX, 2) cytokine responses in the stressed state without DEX, and finally 3) DEX effects on cytokine production. DEX effects were analyzed in two ways: absolute cytokine levels and as a percent change from the control well with just saline. A few missing cytokine values in the latter analysis caused the N to decrease from 20 to 17–19 monkeys, due to the need to have data from all wells in the repeated measures. The effect of the acute social stress on cell numbers was tested in a two-factor ANOVA, with prenatal condition as a between factor and baseline vs. with separation stress as a within subjects factor.
Results
Monkeys from the disturbed pregnancies produced significantly lower amounts of both TNF and IL-6 when their leukocyte cytokine responses were tested at 2 yr of age. Figure 1
portrays the data from the undisturbed baseline samples and shows that LPS stimulation resulted in the production of high levels of both cytokines, although significantly more IL-6 than TNF (F[1,17] = 113.32, P < 0.0001). Further, when a higher concentration of LPS was used, it stimulated more cytokine to be released into the supernatant (F[1,17] = 45.85, P < 0.0001). However, the effect of prenatal treatment was clearly evident across these assay conditions. Leukocytes from the control animals produced more cytokine overall, and the absolute difference between the control and prenatally stressed monkeys was larger for IL-6 than for TNF (F[1,17] = 4.10, P < 0.036). Moreover, a significant interaction term in the ANOVA indicated that the magnitude of the prenatal stress effect was greater at the higher LPS stimulation (F[1,17] = 3.57, P < 0.05).
|
). Concurring with the baseline data, more IL-6 was produced than TNF (F[1,17] = 39.12, P < 0.0001), and higher levels of both cytokines were released after LPS stimulation at 100ng/ml compared with 10 ng/ml (F[1,17] = 22.34, P < 0.0002). However, there was still a main effect for prenatal treatment, with control animals producing the highest levels across all assay conditions (F[1,17] = 4.34, P < 0.03). In this ANOVA, there were no significant interaction terms because the prenatal effect was so consistent, and also because the stress manipulation had induced a systematic reduction in cytokine levels in all monkeys. When the two data sets from the baseline and acute stress conditions were analyzed together in one larger ANOVA, this inhibition in cytokine production after the 2 h removal from the home cage was highly significant (F [1,17] = 19.89, P < 0.0003). Because of this difference in the quantity of TNF and IL-6 produced, the magnitude of the stress-induced reduction was significantly greater for IL-6 than for TNF (F[1,17] = 17.41, P < 0.0006). Finally, it is also important to highlight that the TNF and IL-6 values for control monkeys in the aroused blood sampling condition were now in the range of the cytokine levels found for prenatally disturbed animals in the baseline condition (compare Figs. 1
and 2).
|
, and Table 1 summarizes the complete results for the DEX-treated cultures from both base and stress conditions. High molar concentrations of DEX significantly suppressed TNF production to a greater degree (F[3,39] = 20.28, P < 0.0001) and did so more when a lower level of LPS stimulant was used (F[1,13] = 7.23, P < 0.02). Similarly, the DEX inhibition was much more pronounced when cell cultures were run from blood collected in the Arousal condition (see Table 1). Significant interaction terms in the ANOVA indicated that the influence of pregnancy conditions on TNF were still evident at 10-8 M DEX, but no longer evident at the higher DEX concentrations (Fig. 3) (F[6,48] = 3.19, P < 0.04). Similar results were found for IL-6, but now the effect of pregnancy conditions only approached significance after DEX was added to the cultures (P < 0.08), unless the data from monkeys in both early and late stress pregnancies were combined (control vs. stress, P < 0.03). Considering these data as a percent change from the control saline well did not alter these conclusions, and even more clearly demonstrated that DEX abrogated the influence of prenatal treatment. The quantity of DEX required to induce a 50% inhibition (IC50) is shown in Table 1 and indicates only that IL-6 was more difficult to inhibit than TNF. In addition, samples collected after acute arousal were easier to suppress with less DEX than were cells obtained from undisturbed animals. Most inhibition was observed between 10-8 M and 10-7 M, and cytokine levels were more similar and tended to be maximally suppressed at 10-6 M and 10-5 M. As a consequence, prenatally stressed animals did not reveal a differential gradient of inhibition at the higher DEX concentrations.
|
|
). However, the significant decrease in both IL-6 and TNF seen after the acute stress manipulation was clearly associated with the marked lymphocytopenia that was evident in all monkeys. Following removal from the home cage and separation from other monkeys for 2 h, there was a significant drop in lymphocyte numbers in the blood samples (P < 0.0001).
|
These findings indicate that maternal conditions during pregnancy can continue to exert a lingering influence on the proinflammatory cytokine response of monkeys up to 2 yr after birth. This conclusion concurs with an earlier study demonstrating that administration of ACTH to gravid female monkeys caused their offspring to have deficient IL-6 responses when challenged with IL-1ß (19). In the current study, differences in the production of TNF and IL-6 by LPS-stimulated leukocytes were found in undisturbed baseline samples and were preserved in an aroused state, even after the acute stress response induced an overall decrease in cytokine levels. Because the addition of DEX to the cultures brought the cytokine levels of control animals down into the range of the monkeys disturbed prenatally, it implicates HPA activity as a likely mediator of the variation in the cellular responses. A parsimonious explanation is that the prenatally disturbed animals maintain higher set points for HPA activity. The prenatally stressed animals may also be more behaviorally and emotionally reactive because prior research has demonstrated that they tend to be more submissive in the dominance hierarchies in the social groups and react more to acute disturbance. Uno (10, 11) proposed a similar explanation to account for the abnormalities in monkeys generated from DEX-treated pregnancies. His animals had larger cortisol responses when they were psychologically aroused as juveniles, an effect that appeared to be associated with structural damage to the hippocampus.
The DEX inhibition assay proved to be a useful tool for evaluating differences in cytokine production. As found in rodents and humans, increasing molar concentrations of DEX led to a progressive suppression of cytokine levels (24), although IL-6 was more difficult to inhibit than TNF (22, 25). While the effect of prenatal conditions was still evident at 10-8 M, higher concentrations of DEX brought values of control animals down into the range of prenatally stressed ones and then the cytokine response of the cells became progressively suppressed. The combined effect of an acute stress induced by social separation and adding DEX to the cell cultures was summative and produced the greatest inhibition. This action of contemporaneous arousal was associated with a marked lymphocytopenia in the blood. Cell margination after the stress of rehousing or disruption of social relationships has been described previously in many reports on monkeys (28, 29). Prior studies in humans have also documented that cytokine responses and the DEX inhibition gradient can be affected by both acute and chronic processes, including strenuous exercise (22, 24).
It is also important to consider these results within the context of the growing literature demonstrating that prenatal conditions can have a pervasive influence on other physiological systems in the developing infant. These observations span many species, including laboratory rodents, farm animals, and primates, and have led to considerable speculation that such effects may occur in humans after deprivation or maternal disturbance (4, 30, 31). In general, previous research in primates has indicated that stressful events experienced earlier in pregnancy have a more deleterious effect than do the same stressors encountered later in pregnancy. However, in the current project, the effects appeared to be comparable after early and late stress. Compared with this psychological disturbance of the mother, a prior project indicated that a 2-wk treatment of the gravid monkey with ACTH had an even larger impact on cytokine responses in the offspring than did the 6-wk protocol of daily stress used in the current project (19).
Our finding of reduced IL-6 release in the disturbed offspring does differ from the growing literature in adult humans indicating that stress and psychopathology can be indexed by a hypersecretion of IL-6 and other proinflammatory cytokines (32, 33, 34). Here it should be noted that these reports are often on circulating levels of cytokine found in unstimulated blood and likely reflect a contribution from many other types of cells and organs, including the liver. It is also possible that the current findings highlight an important developmental distinction. A reduced response in the immature animal could later be manifest as a hyperactive response in adulthood. There is already precedent for such bidirectional shifts: for example, physical abuse in childhood is associated initially with elevated HPA activity, but later evinced by a hypocortisolemia after the emergence of post-traumatic stress disorder in adulthood (35, 36). Moreover, research on cytokines over the last 2 decades has revealed the dynamic nature and remarkable complexity of hormone-cytokine relationships. Cytokines are known to be potent stimulators of hormone release by the HPA axis, and a number of investigators have suggested that this hormone secretion provides a feedback signal to ensure against overproduction of cytokine (37). Proinflammatory cytokines, such as IL-1, TNF, and IL-6, can also act directly on the corticosteroid receptors, changing receptor number, affinity and function (38, 39, 40). Thus, it is not possible at this point to say whether the disturbance of pregnancy first affected hormone set points and then cytokine biology, or if the cytokine differences led to changes in the physiology of the endocrine system.
The continuation of these effects into the second year of life raises the possibility that the shift in physiological set points may persist into adulthood. Longitudinal studies in rats indicate that prenatal stress can cause increased HPA activity throughout the whole life span (5). Lasting effects of prenatal and early rearing conditions on immune responses have also been found in mice and rats (41, 42). Because of the longevity of primates, there have not been comparable studies conducted with regard to immunity, but lifelong alterations in the anatomy and neurochemistry of midbrain and limbic structures have been found in aged monkeys that were disturbed during early rearing (43). These diverse effects implicate other processes beyond just cortisol, but the HPA axis does appear to offer one hormonal pathway that could mediate many of these changes. Administration of corticosteroids to the gravid female induces postnatal effects on the infant that are very similar to those seen after psychological disturbance (6, 44). Thus, the placental transfer of hormones may be a critical process that shapes the developmental trajectory of the baby before and after birth (45). It remains to be determined whether some of these alterations are sometimes adaptive responses with beneficial consequences, or if they primarily reflect pathological reactions that ultimately portend of a deleterious outcome for the developing infant as it matures.
Acknowledgments
We thank H. Crispen, R. Engel, N. Finkler, and G. Lubach for their help with sample collection and assays.
Footnotes
1 M. Kramer died suddenly and unexpectedly after the initial submission of this manuscript, and the paper was revised posthumously by his colleagues, who still mourn the loss of this talented young scientist. He had been the recipient of predoctoral fellowship from the Studienstiftung des Deutschen Volkes and was supported by a grant from G. A. Lienert-Stiftung für die Nachwuchsförderung in Biopsychologischer Methodik. Additional research support was provided by Deutsche Forschungsgemeinschaft Fu 174/16–1, Ki 537/6–1, and a National Institute of Mental Health grant (MH41659) (to C.L.C.). 
Abbreviations: DEX, Dexamethasone; HPA, hypothalamic-pituitary-adrenal; LPS, lipopolysaccharide.
Received April 3, 2001.
Accepted October 23, 2001.
References
-melanocyte-stimulating hormone on fetal and maternal plasma steroids. Endocrinology 104:1805–1813
-B in mediation of immunosuppression by glucocorticoids. Science 270:283–286 B- independent downregulation of NF--B activity by glucocorticoid receptor. EMBO J 16:4698–4707[CrossRef][Medline]
production in humans: high sensitivity of TNF and resistance of IL-6. J Clin Endocrinol Metab 82:2182–2191 decreases and interleukin 10 increases the sensitivity of monocytes to dexamethasone: potential regulation of the glucocorticoid receptor. J Clin Endocrinol Metab 84:2834–2839This article has been cited by other articles:
 |
|
 |
|
  G. N. Neigh, C. F. Gillespie, and C. B. Nemeroff The Neurobiological Toll of Child Abuse and Neglect Trauma Violence Abuse, October 1, 2009; 10(4): 389 - 410. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Buske-Kirschbaum, S. Krieger, C. Wilkes, W. Rauh, S. Weiss, and D. H. Hellhammer Hypothalamic-Pituitary-Adrenal Axis Function and the Cellular Immune Response in Former Preterm Children J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3429 - 3435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Pincus-Knackstedt, R. A. Joachim, S. M. Blois, A. J. Douglas, A. S. Orsal, B. F. Klapp, U. Wahn, E. Hamelmann, and P. C. Arck Prenatal Stress Enhances Susceptibility of Murine Adult Offspring toward Airway Inflammation J. Immunol., December 15, 2006; 177(12): 8484 - 8492. [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 |
