| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
Department of Medicine (E.O.), Channing Laboratories (D.L.K.), and the Endocrine-Hypertension Division, Department of Medicine (R.E.G., G.K.A.), Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115
Address all correspondence and requests for reprints to: Gail K. Adler, M.D., Ph.D., Endocrine-Hypertension Division, Brigham and Women’s Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In contrast to the well characterized effect of endotoxin on the HPA axis in humans and animals, there is relatively little information on the response of the HPA axis to other antigens. Studies in animals suggest that antigens have acute (hours) and delayed (days) stimulatory effects on glucocorticoid secretion (5, 6, 7, 8, 9). Furthermore, some animal studies suggest that changes in glucocorticoid levels may alter the humoral response to antigen administration (7, 10, 11, 12).
In this study we investigated the effects on HPA axis activity of administering a protein antigen, tetanus toxoid, vs. placebo to 10 healthy premenopausal women. Plasma ACTH, serum cortisol, and urinary free cortisol (UFC) levels were measured as indicators of HPA axis activity. In addition, we evaluated the relationship between cortisol levels and the increase in tetanus toxoid antibodies.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ten healthy premenopausal female volunteers, aged 19–46 yr (mean ± SEM, 28.6 ± 2.61), were studied. Two were Asian-American, two were African-American, and the remainder were Caucasian. Subjects were recruited by means of advertisements in local newspapers. Signed informed consent was obtained from each volunteer before entry into the study. The study protocol was reviewed and approved by the human research committee as well as by the Clinical Research Center scientific advisory committee of Brigham and Women’s Hospital (Boston, MA).
None of the subjects was pregnant, and all were in good health according to thorough clinical history, physical examination, electrocardiography, pregnancy test, thyroid function tests, and blood and urine analyses. Subjects had received no medications for at least 2 weeks and were excluded if they reported use of corticosteroids within the previous year, use of estrogen/progesterone within the previous 3 months, or a personal or family history of depressive illness. Upon enrollment, subjects were administered the Beck Depression Inventory (13); all subjects scored 3 or less of a possible 63, which is well within the normal range of 0–23 (14). All subjects had received their most recent tetanus shot 5 or more yr previously, and none had a history of adverse reaction to tetanus immunization.
As immune function and HPA axis activity may vary with gender and menstrual phase (15, 16, 17), the study was conducted during the follicular phase of the subjects’ menstrual cycles, as determined by calendar and confirmed by hormone measurement upon admission.
Study protocol
Each subject was admitted to the Clinical Research Center of Brigham and Women’s Hospital for a 2-day stay. On each day the subject was provided lunch at noon and kept without food or water until after the 5-h blood sampling at 2100 h. The in-hospital diet was restricted to approximately 100–125 mEq sodium/day.
An iv catheter was placed into a peripheral arm vein and was kept
patent with 5% dextrose in water infused at a slow rate. The subject
was instructed to remain supine and awake from 1400–2200 h. At
1600 h (time zero), the subject was given an im injection of 0.5
mL placebo (sterile diluent for allergenic extract, ALK Laboratories,
Wallingford, CT) into the right deltoid on day 0 and an injection of
0.5 mL tetanus toxoid (tetanus toxoid aluminum phosphate adsorbed, 5
LFU (level of flocculation unit)/0.5 mL; Wyeth-Ayerst Laboratories,
Philadelphia, PA) into the left deltoid on day 1. The sterile diluent
for allergenic extract contained 0.03% human serum albumin, 0.4%
phenol, and 0.9% sodium chloride. Tetanus toxoid contained tetanus
toxoid adsorbed, aluminum phosphate adsorbed, ultrafine, 5 LFU/0.5 mL
(aluminum content, 
0.85 mg/0.5 mL), 0.02% free formaldehyde, 0.01%
thimersal (preservative), and hydrochloric acid and sodium hydroxide to
adjust pH. The subject was blinded as to the identity of the
injection.
Vital signs were monitored at 15- to 30-min intervals, from 1.5 h before to 6 h after tetanus or placebo injection and at 8, 12, and 16 h after each injection on each of the 2 days of the study. Heart rate and mean arterial pressure were measured with a Dynamap monitor (Critikon, Tampa, FL). Oral temperature was measured with an IVAC thermometer (IVAC Corp., San Diego, CA).
Local reactions to tetanus toxoid and placebo injections, including pain, swelling, and erythema, were monitored by the investigators every hour from -1 to 6 h and at 8, 12, and 16 h after each injection. Subsequently, subjects measured local reactions and oral temperature daily at 1600 h for the 7 days after the study and recorded the extent of pain at each injection site using a scale from 0–10, with 0 representing no pain and 10 representing severe pain. Each subject also completed a symptom survey at the same time points, reporting such subjective sensations as nausea, fatigue, and headache. Pain scores, an indicator of local reaction at the injection sites, were considered positive if a subject reported a pain level of 2 at any one time point or a level of 1 on at least two consecutive reports.
Urine was collected in batches for measurement of UFC from 1500–2400 h and from 2400–0900 h after each injection. Peripheral venous blood was collected at various intervals from -30 min to 16 h. Specimens were processed for measurement of plasma ACTH, serum cortisol, and plasma cytokines. Blood for ACTH and cytokines was collected on ice in chilled, ethylenediamine tetraacetate-containing glass tubes (siliconized glass for ACTH). These samples were spun in the cold, aliquoted immediately, stored at -20 C for 24 h, and then kept at -70 C until processing. Tetanus antibody titers were determined from blood collected before vaccination and 1 month after tetanus toxoid administration.
Assays
The cortisol and ACTH assays were performed in the General Clinical Research Center’s Core Laboratory of Brigham and Women’s Hospital. Samples were assayed with commercially available kits (cortisol: GammaCoat 125I Cortisol RIA Kit, Incstar Corp., Stillwater, MN; ACTH: Nichols Allegro HS-ACTH Kit, Nichols Institute Diagnostics, San Juan Capistrano, CA). For cortisol, the lower limit of detection was 1 µg/dL, with an intraassay coefficient of variation (CV) of 5–8% and an interassay CV of 8–11%. For ACTH, the lower limit of detection was 1 pg/mL, with an intraassay CV of 3% and an interassay CV of 7–8%.
IL-1 receptor antagonist (IL-1ra) and IL-6 assays were performed by enzyme-linked immunosorbent assay with Quantikine human IL-1ra and Quantikine HS human IL-6 kits (R&D Systems, Minneapolis, MN). The reported sensitivity of the IL-1ra assay is 22.0 pg/mL. In our laboratory, the inter- and intraassay CVs were 1% and 6.4%, respectively. The reported sensitivity of the IL-6 kit is less than 0.094 pg/mL.
Tetanus toxoid antibody titers were measured as previously described (18).
Statistical analysis
Results are reported as the mean ± SEM.
Statistical analyses were performed with a statistical software package
(SAS for Windows, version 6.1) using paired t tests,
2 analysis, correlation analysis, and
two-factor repeated measures ANOVA as indicated.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Local and systemic responses to tetanus toxoid
There was no reported pain, swelling, or other local reaction at
the placebo injection site either initially (1–16 h) or during the
following week. Two subjects reported transient pain (<2 h) within
8 h of receiving tetanus toxoid injection that was no longer
present between 8–16 h after injection. In contrast to the lack of
pain in all subjects after placebo injection, six subjects developed
pain at the site of the tetanus toxoid injection at 24–48 h
(P < 0.05, by 2 analysis; Fig. 1
). This pain lasted an average of 4
± 0.9 days. Two of these subjects also reported the concurrent
development of swelling at the tetanus injection site.
|
Effect of tetanus toxoid on ACTH and cortisol
UFC levels were measured in urine collected from 1500–2400 and
2400–0900 h with administration of placebo and tetanus toxoid at
1600 h. UFC (1500–2400) levels were 1.8-fold higher after the
injection of tetanus toxoid than of placebo (31.7 ± 6.2
vs. 17.4 ± 2.1 µg/dL; P < 0.05, by
paired t test; n = 9; Fig. 2). However, UFC (2400–0900) levels did
not differ between tetanus and placebo days (31.2 ± 7.9 and
36.8 ± 7.4 µg/dL, respectively; n = 9), which suggests
that the stimulatory effect of tetanus toxoid on cortisol levels was
not prolonged. The fact that urinary creatinine excretion did not
differ between the 2 days indicates that complete collections were
obtained. Thus, tetanus toxoid administration caused a significant
increase in UFC levels in the succeeding 8 h.
|
). However,
by two-way repeated measures ANOVA, serum cortisol levels differed
significantly (P < 0.05) between the 2 study days.
Mean serum cortisol levels were higher after tetanus toxoid injection
than after placebo injection (Fig. 3a). The maximum difference occurred
4–5 h after the injection, when mean serum cortisol levels were
1.6-fold higher on the tetanus vs. the placebo day. By
0900 h, serum cortisol levels were the same. For each individual,
for the time of intensive sampling (1600–2200 h), we determined the
difference in serum cortisol between the 2 days (
cortisol).
Cortisol correlated with the difference in UFC (1500–2400) between
the 2 days (r = 0.70; P < 0.05).
|
). Thus, it appears that tetanus toxoid induced an
increase in adrenal cortisol secretion, most likely through activation
of the HPA axis. IL-1 and IL-6 are known CRH secretagogues and are involved in endotoxin-induced activation of the HPA axis (19). Therefore, we tested the hypothesis that these factors may be involved in tetanus toxoid activation of the HPA axis. As IL-1 receptor antagonist appears to be a very sensitive indicator of increases in IL-1 (20, 21), we examined plasma IL-1 receptor antagonist levels in nine women 0, 3, 6, 8, and 16 h after tetanus toxoid and placebo injections. Plasma IL-6 levels were measured in six women. No significant differences in levels of IL-1 receptor antagonist or IL-6 were seen after tetanus and placebo injections (data not shown). The lack of a detectable increase in IL-1 receptor antagonist and IL-6 may be due to a dose-response effect or may reflect differences in the response to tetanus toxoid compared with that to endotoxin.
Relationship between cortisol and pain
As cortisol has an antiinflammatory function, we investigated the
relationship between cortisol (serum cortisol on the tetanus day
minus that on the placebo day from 1600–2200 h) and the development of
pain 48 h after tetanus toxoid injection. cortisol was
significantly higher in the four women who did not develop pain than in
the six women who did develop pain at the tetanus toxoid injection site
(49.2 ± 15.6 vs. 2.7 ± 3.4 µg/dL,
respectively; P = 0.007). Likewise, UFC (1500–2400 h)
levels on the tetanus day were 2.7-fold higher in women who did not
develop pain than in women who did develop pain (48.8 ± 5.8
vs. 18.0 ± 3.6 µg/dL, respectively;
P = 0.002). Baseline UFC excretion after placebo
injection did not differ between subjects who did and did not develop
pain 48 h after tetanus toxoid injection.
Relationship between cortisol and the increase in tetanus toxoid antibody titers
Tetanus antibody titers were measured at baseline in all subjects
and at 1 month (mean, 36.9 ± 2.8 days) after vaccination in eight
subjects. Tetanus toxoid administration induced a rise in tetanus
antibody levels from a titer of 10,830 ± 2,843 to 75,850 ±
15,324. There was no correlation between baseline antibody titers and
the cortisol response to vaccination (cortisol). In contrast, there
was a significant inverse correlation between cortisol and the
antibody response to vaccination (r = 0.87; P <
0.005; Fig. 4). Thus, it appeared that
activation of the HPA axis in response to vaccination correlated with a
reduced antibody response to vaccination.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Many studies in man and animals have demonstrated the stimulatory effect of endotoxin, a lipopolysaccharide, on the HPA axis (2, 3, 4). However, the interactions between protein antigens and the HPA axis are not well characterized. A number of studies suggest that in animals, antigens such as horse and sheep red blood cells and trinitrophenylated hemocyanin stimulate the HPA axis (5, 6, 7, 8, 11, 22). In one study, sheep red blood cell administration induced a dose-dependent increase in corticosterone levels, with maximum stimulation 2–4 h after injection and a return to baseline 24–96 h after sheep red blood cell injection (7). Other investigators have shown delayed increases in corticosterone with maximum increases occurring 5–7 days after administration of horse and sheep red blood cells, phosphocholine-keyhole limpet hemocyanin, and trinitrophenylated hemocyanin (5, 9, 11, 22). The reason for these differences in corticosterone responsiveness is unknown, but could be differences in the antigen or in the sex and/or strain or species of animal studied.
In humans, few studies have examined the effect of protein vaccines on the HPA axis. Administration of diphtheria tetanus pertussis vaccine to infants stimulated an increase in UFC levels in the 24 h after vaccination (23). However, this vaccine may contain endotoxin. Thus, it is difficult to determine whether the increase in cortisol was a response to the endotoxin and/or the protein antigens. In a study in adults, serum cortisol levels were significantly higher 2 h after administration of tetanus toxoid than after placebo (24). Although this report suggested that administration of tetanus toxoid activated the HPA axis, the investigators were unable to show a stimulatory effect of tetanus toxoid on plasma ACTH levels. In addition, infrequent sampling and the performance of the study in the morning meant that the stimulatory effect of tetanus toxoid on serum cortisol levels was noted at only one time point and at a time when baseline cortisol levels were falling rapidly. We have performed a detailed, placebo-controlled study on the effect of tetanus toxoid on ACTH and cortisol that includes administration of the tetanus toxoid in the late afternoon when the activity of the HPA axis was relatively quiet, frequent blood sampling, and correlation of blood and urinary glucocorticoid levels. We showed that tetanus toxoid administration significantly increased plasma ACTH, serum cortisol, and UFC levels measured in the 8 h after tetanus toxoid administration. Thus, our study confirms and significantly expands a previous report showing that tetanus toxoid administration increases serum cortisol.
There are several potential limitations to our study. The tetanus toxoid and placebo injections were administered sequentially because of the unknown duration of tetanus toxoid actions on the HPA axis. In addition, because of concerns that in premenopausal women hormonal variations may alter immune and endocrine reactivities (25, 26), we administered the injections in the follicular phase of the menstrual cycle. We did not administer the injections in consecutive menstrual cycles due to concerns for subject compliance and for potential variations in hormones, immune function, life events, and overall health status from 1 month to the next. Although the lack of randomization is a potential limitation, we believe that the similar levels of cortisol in the baseline period on each of the study days suggest that this lack of randomization is not significant. We attributed the activation of the HPA axis with tetanus toxoid vaccination to the tetanus toxoid itself. However, there were other differences in the composition of the placebo and tetanus toxoid injections. The placebo contained phenol and human serum albumin, whereas the tetanus toxoid vaccine contained formaldehyde, aluminum phosphate, thimerosal, and tetanus toxoid. It is possible that components of the vaccine other than tetanus toxoid contributed to the observed activation of the HPA axis. Finally, it is possible that pain contributed to HPA axis activation. The placebo injection was used to control for the injection procedure itself. Although no further pain occurred in any subject after placebo administration, two subjects experienced mild, transient pain in the early (1–6 h) period after tetanus toxoid administration. Exclusion of these two subjects from the data analysis revealed a 1.7-fold increase in UFC (1500–2400 h) after tetanus toxoid injection compared to a 1.8-fold increase if all subjects were included in the analysis. Thus, early pain alone is unlikely to account for the observed increase in cortisol after tetanus toxoid administration.
There is a complex relationship between HPA axis activity and antibody production. Studies in rodents suggest that increases in circulating glucocorticoid levels inhibit the antibody response to specific antigenic challenges. In response to sheep red blood cell injection, an inverse relationship was noted between endogenous levels of corticosterone measured 2–4 h after sheep red blood cell injection and the number of splenic plaque-forming colonies (i.e. antibody-forming cells) (7). Other investigators found that the development of specific IgG after inoculation with ovalbumin was reduced in rats receiving glucocorticoids and increased in animals receiving RU486 (a glucocorticoid and progesterone receptor antagonist) (10). In addition, studies of antigenic competition suggest that increases in endogenous corticosterone induced by the first antigen play a role in inhibiting the antibody response to the second antigen (11). Finally, the timing of elevated glucocorticoid levels relative to antigen administration appears to be critical. Administration of a single dose of 2.5 mg hydrocortisone to mice before or at the time of exposure to sheep red blood cell reduced the antibody response to sheep red blood cell, with a maximum effect between days -4 and 0 (12). This inhibitory effect of hydrocortisone was lost if it was administered more than 1 day after antigen exposure. The above studies suggest that in animals, appropriately timed increases in endogenous or exogenous glucocorticoids inhibit the antibody response to antigens. Thus, it has been hypothesized that in animals, the immune response to protein antigens involves activation of the endocrine system and that these changes in the endocrine system, in turn, may alter the immune response (7, 10, 11, 19, 22).
In humans, high dose glucocorticoids are efficacious in treating antibody-mediated autoimmune diseases (27) and, in general, appear to impair Ig production (28). However, in some studies glucocorticoids have had either a neutral or a stimulatory effect on Ig synthesis (29, 30, 31). The reasons for the variable effects of glucocorticoids on Ig production are unknown, but could be related to the type of Ig, the antigen, the timing and dose of glucocorticoid, and/or differences in the study population. For example, studies examining antibody production in individuals with medical illnesses may be complicated by underlying abnormalities in the immune-neuroendocrine system (31). Indeed, compared with the normal white leghorn chicken, obese strain chickens that develop spontaneous autoimmune thyroiditis show a marked reduction in the corticosterone response to sheep red blood cell immunization (9).
In the present study, we found a significant inverse correlation between the tetanus toxoid-induced rise in cortisol and the development of pain at the tetanus toxoid injection site 24–48 h after injection. Thus, activation of the HPA axis around the time of vaccination may have suppressed the later development of injection site pain and inflammation. In addition, the rise in cortisol correlated inversely with the rise in tetanus toxoid antibody levels. That is, women who had higher increases in serum cortisol had lower immune responses to tetanus toxoid. We hypothesize that antigen-mediated activation of the HPA axis tends to limit the humoral immune response. Although our data are consistent with the hypothesis that relatively small, appropriately timed physiological changes in the endogenous activity of the HPA axis have significant effects on immune function, it is also possible that tetanus toxoid-induced activation of the HPA axis persists for a longer period than we monitored and that this prolonged activation influences immune function. If the HPA axis alters the humoral immune response, the efficacy of vaccine administration could be diminished if vaccination occurs when endogenous glucocorticoid levels are high. It is also possible that those individuals in whom antigens have a limited ability to activate the HPA axis may be more prone to exuberant antibody responses and perhaps even to autoimmune diseases. We suggest that tetanus toxoid is a safe stimulus that may be a useful tool for studying in vivo the dynamic relationship between the immune and neuroendocrine systems in healthy and ill individuals.
In summary, we have shown that in healthy women, administration of the protein antigen, tetanus toxoid, acutely activated the HPA axis. This acute increase in cortisol correlated inversely with the subsequent rise in tetanus toxoid antibody titers. Thus, it appears that protein antigens can activate the HPA axis and that this activation may have a role in limiting the immune response to these antigens.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received November 14, 1997.
Revised January 7, 1998.
Revised February 4, 1998.
Accepted February 10, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
infusion in humans. Cytokine. 6:215–219.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
