the physiological range by administration of cortisone acetate, 25mg at 21.00, markedly suppressed IFN-g, TNF-a, IL-1 and IL-12 production, but not the later ...
DIURNAL RHYTHMS OF PRO-INFLAMMATORY CYTOKINES: REGULATION BY PLASMA CORTISOL AND THERAPEUTIC IMPLICATIONS Nikolai Petrovsky,
Peter McNair,
Leonard C. Harrison
Clinical features of certain immuno-inflammatory disorders such as rheumatoid arthritis and asthma exhibit diurnal fluctuation, which could be related to diurnal rhythmicity of proinflammatory cytokine production. To investigate the latter, the authors performed measurements of lipopolysaccharide (LPS)-stimulated whole blood, interferon g (IFN-g), tumour necrosis factor a (TNF-a), interleukin 1 (IL-1) and IL-12 production in 13 healthy volunteers over 24 h. These cytokines exhibited distinct diurnal rhythms that peaked in the early morning and were inversely related to the rhythm of plasma cortisol. Elevation of plasma cortisol within the physiological range by administration of cortisone acetate, 25 mg at 21.00, markedly suppressed IFN-g, TNF-a, IL-1 and IL-12 production, but not the later early morning rise of endogenous plasma cortisol. Suppression of cytokine production was temporally dissociated from changes in numbers of circulating mononuclear cells. Regulation of pro-inflammatory cytokine production by plasma cortisol has potential therapeutic implications. In contrast to standard schedules, a small, late evening, dose of glucocorticoid to suppress the diurnal increase in pro-inflammatory cytokine production could alleviate early morning inflammatory symptoms and minimize side-effects. 7 1998 Academic Press Limited
Cortisol, the major circulating human glucocorticoid, is a powerful natural immuno-suppressant. Plasma cortisol exhibits a prominent diurnal rhythm1 which could impose diurnal variation on immune function. Periods of heightened immune reactivity would then coincide with or follow the early morning nadir in plasma cortisol and may explain why symptoms in some chronic inflammatory diseases, e.g. rheumatoid arthritis or asthma, are exacerbated at night or early morning.2,3 In humans, delayed-type hypersensitivity responses4 and natural killer (NK) cell function5 exhibit diurnal variation that is maximal during night and early morning. Other immune parameters that are diurnal include circulating CD4+ T cell numbers and CD4/CD8 T cell ratio,6 the autologous mixed lymphocyte response,7 serum interleukin (IL)-1b and IL-6 levels,8 the phagocytic index9 and urinary neopterin secretion.10 Some of these rhythms are temporally associated with variations in
plasma cortisol.6,7 Diurnal variation of immune function is not restricted to humans but is also present in a wide range of species including mice,11 rats,12 birds13 and fish.14 Lipopolysaccharide (LPS)- and tetanus-stimulated interferon g (IFN-g) production in human whole blood exhibit a diurnal rhythm in reverse phase to that of plasma cortisol.15 It is well recognized that supra-physiological glucocorticoids inhibit cytokine production,16–18 but whether variation of plasma cortisol within the physiological range, as occurs diurnally, could modulate cytokine production is unknown. We have therefore investigated whether production of the pro-inflammatory cytokines IFN-g, tumour necrosis factor a (TNF-a), IL-1 and IL-12 in whole blood exhibits diurnal variation entrained by plasma cortisol.
RESULTS Whole blood cytokine production
From the Walter and Eliza Hall Institute, Post Office, Royal Melbourne Hospital, Parkville Victoria 3050, Australia Correspondence to: L.C. Harrison, Autoimmunity and Transplantation Division, The Walter and Eliza Hall Institute of Medical Research, Post Office, Royal Melbourne Hospital, Parkville 3050, Australia; E-mail: harrison.wehi.edu.au Received 21 February 1997; accepted for publication 23 May 1997 7 1998 Academic Press Limited 1043–4666/98/040307 + 06 $25.00/0/ck970289 KEY WORDS: blood/cortisol/cytokine/diurnal/inflammation CYTOKINE, Vol. 10, No. 4 (April), 1998: pp 307–312
IFN-g production in response to LPS (Fig. 1B) exhibited diurnal rhythmicity, as confirmed by Cosinor analysis (P Q 0.001). Similarly, the inflammatory cytokines TNF-a and IL-1a also exhibited diurnal rhythmicity (Fig. 1C and D). For all subjects, peak levels for IFN-g, TNF-a and IL-1a occurred at mean times of 00.00, 21.30 and 21.00, respectively. IL-12 is predominantly secreted by macrophages and enhances 307
D.R.
847374—CYTO 10/4 (Issue)————MS 0189
308 / Petrovsky et al.
CYTOKINE, Vol. 10, No. 4 (April, 1998: 307–312)
Plasma cortisol (µM)
600
A
500
500
400
400
300
300
200
200
100
100
0
0
IFN-γ (% 24-h mean)
300
B
200
100
100
0
0
TNF-α (% 24-h mean)
C
100
0
0
D
100
0 17.00
F
200
100
200
E
300
200
200
IL-1α (% 24-h mean)
600
G
200
H
100
21.00
01.00
05.00
09.00
13.00
17.00
0 17.00
21.00
01.00
05.00
09.00
13.00
17.00
Figure 1. Plasma cortisol and LPS-stimulated whole blood IFN-g, TNF-a and IL-1a production. Under normal conditions (left hand panels) and after ingestion of cortisone acetate 25 mg at 21.00 (right hand panels). Seven male and six female subjects were studied and results shown are for one representative subject.
IFN-g production. LPS-stimulated IL-12 production exhibited a diurnal rhythm (peak 23.00) almost synchronous with that of IFN-g (Fig. 2). There was a significant inverse correlation between plasma cortisol and IFN-g (r = −0.9), TNF-a (r = −0.6) and IL-12 (r = −0.75) production.
Effect of exogenous cortisone on cytokine profiles Plasma cortisol rose sharply after oral cortisone acetate 25 mg at 21.00 (Fig. 1E). The single dose increased plasma cortisol to a mean of 526 nmol/l
(range 450–676 nmol/l) and had a mean duration of effect of 8.3 h (range 8–9 h). Comparison with standardized 24-h basal plasma cortisol profiles in the same subjects showed that this dose of cortisone acetate at 21.00 did not significantly suppress the later early morning rise of endogenous plasma cortisol (Fig. 1A and E). Following ingestion of cortisone acetate, whole blood production of pro-inflammatory cytokines fell sharply (Fig. 1F, G, H). After 1 hour, production of IFN-g, TNF-a and IL-1, and IL-12 (not shown), was suppressed by 84%, 95%, 62% and 90%, respectively,
D.R.
847374—CYTO 10/4 (Issue)————MS 0189 Diurnal rhythms of human pro-inflammatory cytokines / 309
plasma cortisol nadir and artificially created a 12-h periodicity in plasma cortisol levels (Fig. 1E). Consistent with a causal relationship to plasma cortisol, pro-inflammatory cytokine production then also demonstrated 12-h periodicity (Cosinor analysis, P Q 0.001) (Fig. 1F, G, H).
16.00
13.00
09.00
05.00
01.00
0
21.00
100
17.00
Cytokine production (% 24-h mean)
200
Figure 2. Diurnal relationship between LPS-stimulated whole blood IFN-g (q) and IL-12 (W) production. Results are the means of individual subject data (n = 10) standardized as a percentage of each subject’s 24-h mean.
but recovered to pre-cortisone levels over 6–8 h, concomitant with the decay in plasma cortisol. Administration of cortisone at 21.00 induced a peak of plasma cortisol at the time of the normal
Effect of exogenous cortisone on circulating white blood cells The reduction in IFN-g, TNF-a, IL-1 and IL-12 production after cortisone ingestion could have been due to a decrease in circulating white blood cell numbers or specific T cell subsets, rather than suppression of T cell function per se. However, 1 hour after cortisone, when cytokine production was maximally suppressed, we detected no reduction in the total white blood cell count or lymphocyte count, or numbers of CD3+, CD4+, CD8+ and CD56+ cells (data not shown). The lymphocyte fraction decreased by 30–40%, but with a lag of
Plasma cortisol (% 24-h mean)
300
200
100
0
Lymphocytes (% 24-h mean)
130
110
90
70
Non-lymphocytes (% 24-h mean)
120
100
80 17.00
21.00
01.00
05.00
09.00
13.00
Figure 3. Effect of oral cortisone acetate 25 mg at 21.00 on the lymphocyte and non-lymphocyte white blood cell (WBC) counts. Results shown are the means of individual subject data standardized as a percentage of each subject’s 24-h mean.
D.R.
847374—CYTO 10/4 (Issue)————MS 0189
310 / Petrovsky et al.
2–3 h after the plasma cortisol peak, whereas the non-lymphocyte fraction (principally monocytes and polymorphonuclear cells) increased by 20–30% (Fig. 3).
DISCUSSION The production of the pro-inflammatory cytokines IFN-g, TNF-a, IL-1a and IL-12 exhibited diurnal rhythmicity that correlated inversely with plasma cortisol. Pro-inflammatory cytokine levels were highest in blood taken in the late evening or early morning, the time of the nadir of plasma cortisol. Elevation of plasma cortisol within the physiological range by the oral administration of cortisone acetate significantly reduced pro-inflammatory cytokine production. The dose of cortisone that attenuated the early morning rise in IFN-g, TNF-a, IL-1a and IL-12 production had a reasonably short duration of action and did not markedly suppress the normal, early morning increase in endogenous plasma cortisol. Thus, it is conceivable that this mode of glucocorticoid administration could be used to suppress diurnal inflammatory symptoms without suppressing the hypothalamic–pituitary–adrenal (HPA) axis. The numbers of circulating T cells, in particular CD4+ T cells, peak during the early morning hours,6,19–22 when plasma cortisol is low. The peaks in circulating lymphocytes and whole blood cytokine production were synchronous, raising the possibility that diurnal variation in cytokine production is the consequence of variations in circulating cell numbers. However, elevation of plasma cortisol within the physiological range which suppressed cytokine production had no immediate effect on the total white cell count, lymphocyte count or numbers of circulating CD4+ or CD8+ T cells or CD56+ NK cells. Two to three hours after the rise of plasma cortisol there was a modest decrease in circulating lymphocytes and an increase in the non-lymphocyte white blood cell fraction, as previously reported.23,24 These findings indicate that cortisol does not suppress cytokine production simply by reducing circulating lymphocyte numbers. Glucocorticoids downregulate cytokine expression by binding to and activating negative regulatory elements in the promoters of cytokine genes and by inducing IkBa, a protein which binds and neutralizes the cytokine transcription factor NF-kB.25 Pro-inflammatory cytokines whose transcription is downregulated by glucocorticoids include IL-2 and IFN-g,16 IL-3, GM-CSF and TNF-a,26 IL-618 and IL-8.17 The symptoms of chronic immuno-inflammatory disorders, for example rheumatoid arthritis or asthma, exhibit diurnal rhythmicity. Joint symptoms in
CYTOKINE, Vol. 10, No. 4 (April, 1998: 307–312)
rheumatoid arthritis are most severe in the early morning,2 whereas asthma exacerbations commonly occur at night.3 Dysregulated HPA function may contribute to rheumatoid arthritis27 and nocturnal exacerbations of asthma have been related to increased immune reactivity associated with the early morning nadir in plasma cortisol.3,28 Recently, there has been a resurgence of interest in glucocorticoid treatment for rheumatoid arthritis because of evidence that low dose therapy reduces joint destruction, in addition to acutely relieving inflammatory symptoms.29 Glucocorticoids are commonly administered as a single or major morning dose, because administration in phase with the normal diurnal cortisol rhythm may cause less adrenal suppression than multiple daily doses.30 However, in the treatment of inflammatory conditions with early morning exacerbations, it would be advantageous if the peak anti-inflammatory effect occurred during the night. If glucocorticoids are given in the morning, effective nocturnal levels can only be achieved with high doses or long-acting preparations. The present study demonstrates the potential to use short-acting glucocorticoids as a single, late evening dose to suppress the night-time increase in proinflammatory cytokine production. This could maximize therapeutic efficacy whilst minimizing dosedependent side-effects.
MATERIALS AND METHODS Subjects Seven male and six female healthy, young adult volunteers had blood taken each hour for 24 h for measurement of LPS-stimulated cytokine production. The subjects maintained their normal sleep/wake cycle and activity patterns. Blood samples were taken hourly via an indwelling venous cannula inserted in a forearm cubital fossa vein. In a second 24-hour study four weeks later, whole blood assays as described above were performed on nine of the original subjects after each had received an oral dose of 25 mg cortisone acetate at 21.00.
Whole blood assay Heparinized venous blood was aliquoted at 280 ml/well into 96-well tissue culture plates (Falcon) preloaded in quadruplicate with LPS (E.coli serotype 0127:B8) in 20 ml of human tonicity (HT)-RPMI-1640 medium. Negative control wells were preloaded with 20 ml HT-RPMI-1640 alone. The plates were incubated at 37°C in 5% CO2 atmosphere for 48 h and the plasma supernatants from quadruplicate wells were then pooled and stored at −20°C until assayed for cytokines. We have previously demonstrated15 that LPSor tetanus-stimulated cytokine levels in whole blood are maximal after incubation for 36–48 h. The intra-assay coefficient of variation for IFN-g production in the whole blood assay is 15–20%.31
D.R.
847374—CYTO 10/4 (Issue)————MS 0189 Diurnal rhythms of human pro-inflammatory cytokines / 311
Cytokines Cytokines were measured by enzyme-linked immunosorbent assay (ELISA). The IFN-g, IL-1a and IL-12 kits were from Commonwealth Serum Laboratories (CSL) Australia, Immunotech S.A. and R&D Systems, respectively. The IL-10 ELISA used rat monoclonal antibodies (mAb) 9D7 and 12G8-NIP and anti-rat detection mAb J4 conjugated to horseradish peroxidase (J4-HRP), provided by Dr Robert Coffman of DNAX. The TNF-a ELISA used the PharMingen mouse mAb pair, 18631D and biotinylated 18642D, for cytokine capture and detection, respectively. For both the IL-10 and TNF-a ELISAs, capture mAb was coated overnight at 4°C onto Nunc Maxisorb plates at a concentration of 5 mg/ml. The plates were then washed and blocked with 10% bovine serum albumin (BSA) for 1 h at room temperature (RT). Samples (50 ml) were added to wells and incubated overnight at 4°C, followed by washing and incubation with a secondary detection mAb at 1 mg/ml for 1 h at RT. In the IL-10 ELISA this was followed by incubation with J4-HRP mAb (1:5000) for 1 h, and in the TNF-a ELISA by incubation with 100 ml streptavidin– peroxidase (1:500) for 1 h. Colour was developed with 100 ml TMB peroxidase substrate.
Plasma cortisol Plasma cortisol was measured with the Orion Diagnostica cortisol (125I) radioimmunoassay.
Peripheral blood counts To assess the effect of cortisone administration on circulating T cell subsets, venous blood were taken from healthy subjects immediately before and after oral cortisone acetate 25 mg. The total white cell count and lymphocyte count were measured by a Coulter Counter and the number and proportion of CD3+, CD4+, CD8+ and CD56+ cells enumerated by flow cytometry with commercially available antibodies (Becton-Dickenson).
Statistics For group analyses, cytokine levels were first expressed as percentages of the 24-h mean, as previously described.15 Diurnal rhythmicity was analysed with Chronolab, a software package for chronobiological time series analysis32 kindly provided by Dr Fernida, Bioengineering and Chronobiology Lab., ETSI Telecommunie, University of Vigo, Spain. Correlation coefficients were calculated using the Pearson correlation matrix.
Acknowledgements We thank Sue Horton for assistance with the analysis of T cell subsets and Margaret Thompson and Sharen Petrovsky for secretarial assistance. NP was a Post-graduate Scholar and LCH a Senior Principal Research Fellow of the National Health and Medical Research Council of Australia.
REFERENCES 1. Pincus G (1943) Diurnal rhythm in the excretion of urinary ketosteroids by young men. J Clin Endocrinol Metab 3:195–199. 2. Harkness JA, Richter MB, Panayi GS, Van de Pette K, Unger A, Pownall R, Geddawi M (1982) Diurnal variation in disease activity in rheumatoid arthritis. Br Med J Clin Res 284:551–554. 3. Martin JR, Cicutto LC, Smith HR, Ballard RD, Szefler SJ (1991) Airways inflammation in nocturnal asthma. Am Rev Resp Dis 143:351–357. 4. McGovern J, Smolensky M, Reinberg A (1977) Chronobiology in Allergy and Immunology. In McGovern J, Smolensky M, Reinberg A (eds). Charles C Thomas, pp. 79–116. 5. Angeli A, Gatti G, Sartori ML, Masera RG (1992) Chronobiological aspects of the neuroendocrine-immune network. Regulation of human natural killer (NK) cell activity as a model. Chronobiologia 19:93–110. 6. Miyawaki T, Taga K, Nagaoki T, Seki H, Suzuki Y, Taniguchi N (1984) Diurnal changes of T lymphocyte subsets in human peripheral blood. Clin Exp Immunol 55:618–622. 7. Indiveri F, Pierri I, Rogna S, Poggi A, Montaldo P, Romano R, Pende A, Morgano A, Barabino A, Ferrone S (1985) Diurnal variations of autologous mixed lymphocyte reactions and endogenous cortisol. J Immunol Methods 82:17–24. 8. Gudewill S, Pollmacher T, Vedder H, Schreiber W, Fassbender K, Holsboer F (1992) Nocturnal plasma levels of cytokines in healthy men. Eur Arch Psychiatry Clin Neurosci 242:53–56. 9. Melchart D, Martin P, Hallek M, Holzmann M, Jurcic X, Wagner H (1992) Diurnal variation of the phagocytic activity of polymorphonuclear leukocytes and of various other parameters in 13 healthy male adults. Chronobiol Int 19:35–45. 10. Auzeby A, Bogdan A, Krosi Z, Touitou Y (1988) Time-dependence of urinary neopterin, a marker of cellular immune activity. Clin Chem 34:1866–1867. 11. Fernandes G, Halberg F, Yunis EJ, Good RA (1976) Diurnal rhythmic plaque-forming cell response of spleens from mice immunized with SRBC. J Immunol 117:962–966. 12. Loubaris N, Cros G, Serrano JJ, Boucard M (1983) Diurnal and circannual variation of the carrageenin inflammatory effect in rat. Life Sci 32:1349–1354. 13. Stinson R, McCorkle F, Mashaly M, Taylor R, Martin D, Glick B (1980) The effects of diurnal rhythms on immune parameters in New Hampshire chickens. Int Arch Allergy Appl Immunol 61:220–226. 14. Nevid NJ, Meier AH (1993) A day-night rhythm of immune activity during scale allograft rejection in the gulf killifish, Fundulus grandis. Dev Comp Immunol 17:221–228. 15. Petrovsky N, McNair P, Harrison LC (1994) Diurnal rhythmicity of interferon-gamma production in antigen-stimulated whole blood. Chronobiologia 21:293–300. 16. Kelso A, Munck A (1984) Glucocorticoid inhibition of lymphokine secretion by alloreactive T lymphocyte clones. J Immunol 133:784–791. 17. Mukaida N, Gussella GL, Kasahara T, Ko Y, Zachariae CO, Kawai T, Matsushima K (1992) Molecular analysis of the inhibition of interleukin-8 production by dexamethasone in a human fibrosarcoma cell line. Immunology 75:674–679. 18. Ray A, LaForge KS, Sehgal PB (1990) On the mechanism for efficient repression of the interleukin-6 promoter by glucocorticoids: enhancer, TATA box, and RNA start site (Inr motif) occlusion. Mol Cell Biol 10:5736–5746. 19. Kawate T, Abo T, Hinuma S, Kumagai K (1981) Studies of the bioperiodicity of the immune response. II. Co-variations of murine T and B cells and a role of corticosteroid. J Immunol 126:1364–1367.
D.R.
847374—CYTO 10/4 (Issue)————MS 0189
312 / Petrovsky et al. 20. Ritchie AW, Oswald I, Micklem HS, Boyd JE, Elton RA, Jazwinska E, James K (1983) Diurnal variation of lymphocyte subpopulations: a study with monoclonal antibodies. Br Med J Clin Res 286:1773–1775. 21. Abo T, Kawate T, Itoh K, Kumagai K (1981) Studies on the bioperiodicity of the immune response. I. Diurnal rhythms of human T, B, and K cell traffic in the peripheral blood. J Immunol 126:1360–1363. 22. Levi FA, Canon C, Touitou Y, Sulon J, Mechkouri M, Ponsart ED, Touboul JP, Vannetzel JM, Mowzowicz I, Reinberg A (1988) Diurnal rhythms in circulating T lymphocyte subtypes and plasma testosterone, total and free cortisol in five healthy men. Clin Exp Immunol 71:329–335. 23. Fauci AS, Dale DC (1975) The effect of hydrocortisone on the kinetics of normal human lymphocytes. Blood 46:235–243. 24. Thomson SP, McMahon LJ, Nugent CA (1980) Endogenous cortisol: a regulator of the number of lymphocytes in peripheral blood. Clin Immunol Immunopathol 17:506–514. 25. Scheinman RI, Cogswell PC, Lofquist AK, Baldwin A Jr (1995) Role of transcriptional activation of I kappa B alpha in mediation of immunosuppression by glucocorticoids. Science 270:283–286.
CYTOKINE, Vol. 10, No. 4 (April, 1998: 307–312) 26. Homo-Delarche F, Dardenne M (1993) The neuroendocrine-immune axis. Springer Sem Immunopathol 14:221–238. 27. Chikanza IC, Petrou P, Kingsley G, Chrousos G, Panayi GS (1992) Defective hypothalamic response to immune and inflammatory stimuli in patients with rheumatoid arthritis. Arthr Rheum 35:1281–1288. 28. Reinberg A, Ghata J, Sidi E (1963) Nocturnal asthma attacks: their relationship to the diurnal adrenal cycle. J Allergy 34:323. 29. Kirwan JR (1995) The effect of glucocorticoids on joint destruction in rheumatoid arthritis. The Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group. N Eng J Med 333:142–146. 30. Nichols T, Nugent C, Tyler G (1965). Diurnal variations in suppression of adrenal function by glucocorticoids. J Clin Endocrinol 25:343. 31. Petrovsky N, Harrison LC (1995) Cytokine-based human whole blood assay for the detection of antigen-reactive T cells. J Immunol Methods 186:37–46. 32. Mojon A, Fernandez JR, Hermida RC (1992) Chronolab: an interactive software package for chronobiologic time series analysis written for the Macintosh computer. Chronobiol Int 9:403–412.
