点击显示 收起
Divisions of 1Endocrinology and 2Infectious Diseases, Department of Internal Medicine, School of Medicine of Ribeirão Preto, University of São Paulo, São Paulo, Brazil
Received 18 June 2002; revised 6 September 2002; electronically published 13 December 2002.
The functional status of adrenocortical hormones and their relationship to the pattern of inflammatory cytokines in paracoccidioidomycosis were investigated in a prospective study. Patients were evaluated before treatment and 1 and 6 months after receiving antifungal therapy. Interleukin (IL)1, IL-6, and tumor necrosis factor plasma levels, C-reactive protein (CRP) concentrations, and erythrocyte sedimentation rate (ESR) were significantly higher in untreated patients than in control subjects. After 6 months of treatment, levels of the 3 cytokines, CRP concentrations, and the ESR decreased significantly. Both baseline and stimulated adrenocorticotropic hormone and cortisol plasma levels were not different between patients and control subjects. In contrast, adrenal androgen dehydroepiandrosterone sulfate (DHEA-S) plasma levels were significantly lower in patients than in sex- and age-matched control subjects. There was a significant inverse correlation between DHEA-S and IL-6 plasma levels. This finding may be of pathogenetic significance in this disease and in other inflammatory states.
Informed consent was obtained from patients and control subjects. The ethics committee of the University Hospital of the School of Medicine of Ribeirão Preto approved the study.
Reprints or correspondence: Dr. Angela M. O. Leal, Div. of Endocrinology, Dept. of Internal Medicine, School of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, SP, Brazil 14049-900 .
Paracoccidioidomycosis (PCM) is a chronic and granulomatous infective disease caused by the dimorphic fungus Paracoccidioides brasiliensis. It is the most common systemic mycosis in South America. PCM clinical forms vary from a benign self-limited infection to a severe, progressive, and sometimes fatal disease involving pulmonary and extrapulmonary tissues [1]. Previous studies (which used different methods) found impairment of adrenal function in a high percentage of patients with PCM [26]. In another study, the adrenal glands were involved in 44%80% of cases in an autopsy series [7].
The relationship between the pathogenetic factors of PCM and adrenocortical hormones has not been explored. In recent years, many studies have documented that numerous cytokine family members influence the secretory activity of the hypothalamic-pituitary-adrenal (HPA) axis. During inflammation, cytokines stimulate the HPA axis through direct and indirect actions on the central nervous system and pituitary and adrenal glands, which, in turn, generally inhibit or modulate inflammation through the immunossuppressive effects of the glucocorticoids [8]. More recently, the adrenal androgen dehydroepiandrosterone sulfate (DHEA-S) was implicated as an immunomodulator hormone [9, 10]. Since the pathogenetic influence of the inflammatory cytokines may be involved in the modulation of host defense against the pathogen in PCM [11], this disease offers an attractive model in which to serially investigate the pathogenic correlation between cytokine pattern and adrenocortical hormones. In this study, we prospectively examined the functional status of adrenocortical hormones and their relationship with inflammatory cytokine patterns in patients with active PCM before and during antifungal therapy.
PATIENTS AND METHODS
Patients and control subjects. We evaluated 14 men who exhibited active PCM (mean age, 40.7 ± 2.7 years) and 9 healthy male control subjects (mean age, 39.3 ± 2.7 years). All cases were drawn from the infectious disease outpatient unit of University Hospital, School of Medicine of Ribeirão Preto, Ribeirão Preto, Brazil. The clinical diagnosis of active PCM was reaffirmed by serologic tests and microscopic observation and/or culture isolation of P. brasiliensis from sputum, mucocutaneous ulcer scrapings, or biopsy specimens. The patients were consecutively evaluated during the active phase of the disease, regardless of the presence of any symptoms suggestive of hypocortisolism. Three patients had the acute form, and 11 had the chronic form of PCM, according to PCM classification criteria proposed elsewhere [1].
Protocol. Every patient was studied before treatment (BT) and 1 and 6 months after receiving therapeutic drug (sulfadiazine). None of the healthy subjects had a history of PCM or adrenal disease or had received antifungal or glucocorticoid treatment. All tests were started at 9:00 A . M ., after an overnight fast. An indwelling catheter was placed into a forearm vein 60 min before each test. The study maneuvers were carried out in the order given below.
Baseline blood sample and ovine corticotropinreleasing hormone (oCRH) test. On day 1, we obtained a baseline blood sample for adrenocorticotropic hormone (ACTH), cortisol, DHEA-S, tumor necrosis factor (TNF), interleukin (IL)1, IL-6, C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) determination. oCRH (1 g/kg; Bachem) was then given as an intravenous (iv) bolus. Blood samples for ACTH and cortisol assays were collected 15, 30, 45, 60, and 120 min after the injection.
ACTH test. On day 2, a baseline blood sample was taken for cortisol assay, and ACTH 1-24 (0.25 mg of tetracosactrin; Organon) was given as an iv bolus. Another blood sample for cortisol measurement was obtained 60 min after the injection.
Hormone assays. Plasma ACTH, cortisol, and DHEA-S levels were determined by RIA, as described elsewhere [5, 12]. The assay sensitivity and the intra- and interassay coefficients of variation were, respectively, 9 pg/mL and 3.9% and 18% for ACTH; 0.6 g/dL and 4.2% and 18% for cortisol; and 1.9 g/dL and 1.5% and 10% for DHEA-S.
IL determination. Plasma concentrations of TNF-, IL-1, and IL-6 were determined by specific commercial ELISAs (Quantikine [HSTA50, HSLB50, and HS600, respectively]; R&D Systems). The lower limits of detection of these assays were 1, 0.15, and 0.12 pg/mL, respectively.
CRP quantification, ESR determination, and counterimmunoelectrophoresis (CIE). We measured CRP concentration by the turbidimetric method [13] and ESR by the Wintrobe method. CIE was used to determine serum antibodies to P. brasiliensis.
Statistical analysis. We calculated the total integrated ACTH and cortisol responses to oCRH by the trapezoid method and expressed this as the area under the hormone concentration time curve (AUC) from 0 to 120 min. Differences between group means were assessed by the Friedman test, by Dunn multiple comparison test for paired data, and by Wilcoxon Mann-Whitney U test for unpaired data. Correlations were examined by Spearman rank correlation test. All data analyses were done with GraphPad Prism software (version 3.00 for Windows). P < .05 was considered to be statistically significant.
RESULTS
Baseline plasma cortisol and ACTH. Mean (±SEM) baseline plasma levels of cortisol (control subjects, 11 ± 1 g/dL; patients BT, 11 ± 1.2 g/dL; patients at 1 month, 9 ± 1.2 g/dL; patients at 6 months, 9.5 ± 0.8 g/dL) and ACTH (control subjects, 36 ± 5.6 pg/mL; patients BT, 32 ± 6 pg/mL; patients at 1 month, 41 ± 6.9 pg/mL; patients at 6 months, 33 ± 2.6 pg/mL) were not significantly different between control subjects and patients before and after treatment.
oCRH stimulation test. The total integrated AUC CRH-stimulated ACTH and cortisol concentrations did not differ significantly between control subjects and untreated patients with chronic PCM (ACTH: control subjects, 6123 ± 1536 pg/mL × min; patients BT, 8356 ± 2576 pg/mL × min; cortisol: control subjects, 1808 ± 76 g/dL × min; patients BT, 2009 ± 143 g/dL × min). Among patients after 1 and 6 months of treatment, the ACTH (1 month, 8474 ± 1483 pg/mL × min; 6 months, 6659 ± 930 pg/mL × min) and cortisol (1 month, 1813 ± 102 g/dL × min; 6 months, 1840 ± 119 g/dL × min) responses to oCRH were not different from pretreatment values.
ACTH stimulation test. The cutoff point for the lower normal cortisol response to ACTH was obtained from the data generated in the control group (95% confidence limit). The net (control subjects, 12 ± 2 g/dL; patients BT, 11 ± 1 g/dL) and mean total ACTH 1-24stimulated cortisol (control subjects, 22 ± 2 g/dL; patients BT, 22.5 ± 1.5 g/dL) concentrations were similar in patients and control subjects. Among patients after 1 and 6 months of treatment, both the net (1 month, 10 ± 1.3 g/dL; 6 months, 9 ± 0.9 g/dL) and mean total stimulated (1 month, 18 ± 0.8 g/dL; 6 months, 18 ± 0.8 g/dL) cortisol concentrations did not differ from pretreatment values.
DHEA-S concentrations. Plasma DHEA-S levels of patients BT were significantly lower than those of the control group (control subjects, 260 ± 34.2 g/dL; BT, 87 ± 19.2 g/dL; P = .0015). Eight patients (57%) had low DHEA-S levels for their respective ages. Plasma DHEA-S levels of patients BT were not different from the levels after 1 and 6 months of treatment (1 month, 90 ± 13.8 g/dL; 6 months, 92 ± 13.8 g/dL).
Cytokines. Cytokine concentrations are shown in . The 3 cytokinesIL-1, IL-6, and TNF-were significantly elevated in patients with active PCM BT, compared with control subjects (1.5 ± 0.6 vs. 0.4 ± 0.1 pg/mL; 7 ± 1.4 vs. 3 ± 1.4 pg/mL; and 9 ± 1.5 vs. 5 ± 1.1 pg/mL, respectively). After 1 month of treatment, TNF- was still higher in patients than in control subjects. After 6 months of treatment, all 3 cytokine concentrations decreased significantly. A similar pattern of cytokine plasma concentration was observed for IL-1, IL-6, and TNF-.
fig.ommitted
Figure 1. Plasma cytokine concentrations (mean ± SEM) in control subjects (C) and 14 patients with active paracoccidioidomycosis (PCM) before (BT) and after 1 (1MO) and 6 (6MO) months of treatment. *P < .05, vs. control subjects. IL, interleukin; TNF, tumor necrosis factor.
CRP concentrations and ESR. CRP concentration and ESR of patients BT were higher than in the control group (3.2 ± 1.2 vs. 0.5 ± 0 mg/dL and 33.9 ± 1.2 vs. 6.9 ± 2.3 mm/h 1, respectively). After 6 months of treatment, both CRP and ESR decreased significantly (0.6 ± 0.05 mg/dL and 17.7 ± 3.9 mm/h 1, respectively).
Correlation between cytokines and DHEA-S. There was a significant negative correlation between IL-6 and DHEA-S concentrations of patients before and after 1 and 6 months after treatment (r = -0.31; P < .05; ).
fig.ommitted
Figure 2. Correlation between dehydroepiandrosterone sulfate (DHEA-S) and interleukin (IL)6 plasma concentrations (r = -0.31; P < .05).
Correlation between cytokines and specific antibody titer. There was a significant positive correlation between both IL-6 and TNF- (r = 0.69 and P = .006 and r = 0.85 and P < .0001, respectively) and antibody titers against fungus antigen.
Correlation between cytokines and CRP concentrations and ESR. No correlation was observed between cytokine concentrations (IL-1, IL-6, or TNF-) and CRP concentrations or ESR.
Correlation between DHEA-S and specific antibody titer, CRP concentrations, and ESR. There was a significant inverse correlation between DHEA-S plasma levels and ESR (r = -0.39; P = .01). However, no correlation was observed between DHEA-S and CRP concentrations or antibody titers.
DISCUSSION
This prospective study demonstrates an increase in inflammatory cytokines in untreated patients with PCM and a significant inverse correlation between IL-6 and DHEA-S plasma levels. Clinically manifested adrenocortical hypofunction was not observed before or after treatment, and both baseline and stimulated ACTH and cortisol secretion levels were not different from control values. Diminished adrenal reserve, as shown by individual subnormal cortisol responses to ACTH (60 min after ACTH, <16.7 g/dL, 95% confidence limit), was found in only 2 patients (14.3%) BT and in 3 patients (21.4%) 6 months after treatment.
Several studies have reported adrenal function studies in PCM. These studies usually focused on cortisol responses to synthetic ACTH administration and established graded impairment of adrenal dysfunction in such diseases as Addison disease and limited cortisol production [26]. Because differing methodologies were used to interpret the data, comparisons are difficult. Our results agree with one study that used the same criterion applied in this study [6]. No correlation was observed between cytokine concentrations and baseline and stimulated cortisol secretion. This finding argues against the functional influence of the cytokines studied in corticosteroid secretion in PCM. To our knowledge, our study is the first to prospectively investigate the relationship between adrenal function and inflammatory cytokines pattern in PCM.
We observed lower levels of DHEA-S in patients with PCM than in age- and sex-matched control subjects, even after 6 months of treatment. We previously demonstrated the reduction of DHEA-S levels in 50% of untreated male patients with PCM [5]. The dissociation of adrenal corticosteroid and androgen secretion and the decrease of DHEA-S have been found in chronic diseases [14]. However, the explanation for low DHEA-S serum levels, specifically in chronic inflammatory diseases, is not yet known. A shift in adrenal steroid synthesis away from adrenal androgens and toward the essential cortisol pathway and a blockage of the 17,20-desmolase within the adrenal zona reticularis are presumed but not yet proved [14].
In the present study, DHEA-S plasma levels were inversely correlated with IL-6 plasma concentrations, which may suggest the influence of inflammatory cytokines on the adrenal androgen secretion in inflammatory states. Multiple immunomodulating properties have been attributed to DHEA-S and its precursor (e.g., the modulation of T helper lymphocyte function and cytokine production) [9, 10]; however, a link between low plasma levels of DHEA or DHEA-S and pathogenesis of systemic inflammatory processes in humans has been poorly explored. Nevertheless, the ability of IL-6 to act on steroidogenesis in all 3 adrenocortical zones via autocrine/paracrine mechanisms may contribute to the coordination of adrenal responses to maintain glucocorticoid secretion in situations of prolonged stress [15]. After 6 months of treatment, decreased cytokine plasma levels were observed, despite the maintenance of low levels of DHEA-S concentrations. This may be related to a long-lasting action of immunomodulation at the adrenal level. Finally, we cannot rule out that the significant correlations reported here are just coincidences.
In conclusion, our findings demonstrate a significant inverse correlation between DHEA-S and IL-6 plasma levels in PCM. Because DHEA is an important immunomodulating hormone, this finding may be of pathogenetic significance in this and others inflammatory diseases. Further studies are required.
Acknowledgements
We thank Lucimara Bueno, Adriana Rossi, and Margarida M. P. Nascimento, for excellent technical assistance.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.