TNF-␣ suppresses the expression of clock genes by interfering with E-box-mediated transcription Gionata Cavadini*, Saskia Petrzilka*, Philipp Kohler*, Corinne Jud†, Irene Tobler‡, Thomas Birchler*§, and Adriano Fontana*§ *Division of Clinical Immunology, University Hospital Zurich, Haeldeliweg 4, CH-8044 Zurich, Switzerland; †Institute of Biochemistry, University of Fribourg, Rue du Muse´e 5, CH-1700 Fribourg, Switzerland; and ‡Institute of Pharmacology and Toxicology, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland Edited by Charles A. Dinarello, University of Colorado Health Sciences Center, Denver, CO, and approved June 15, 2007 (received for review February 22, 2007) Production of TNF-␣ and IL-1 in infectious and autoimmune dis- by the suprachiasmatic nuclei (SCN) of the hypothalamus being eases is associated with fever, fatigue, and sleep disturbances, entrained by light stimuli to the environment. This self- which are collectively referred to as sickness behavior syndrome. In sustaining circadian pacemaker uses a molecular mechanism mice TNF-␣ and IL-1 increase nonrapid eye movement sleep. Be- similar to the one used in subsidiary oscillators present in any cause clock genes regulate the circadian rhythm and thereby type of cell in the organism. The molecular clockwork involves locomotor activity and may alter sleep architecture we assessed the transcriptional repressor genes Per1, Per2, Cry1, and Cry2,as the influence of TNF-␣ on the circadian timing system. TNF-␣ is well as the transcriptional activators Bmal1 and Clock. The shown here to suppress the expression of the PAR bZip clock- heterodimerized transcription factor BMAL1:CLOCK activates controlled genes Dbp, Tef, and Hlf and of the period genes Per1, Per and Cry gene transcription by binding to E-box mo- Per2, and Per3 in fibroblasts in vitro and in vivo in the liver of mice tives in their promoters. PER and CRY proteins inhibit infused with the cytokine. The effect of TNF-␣ on clock genes is BMAL1:CLOCK complexes, thereby inhibiting their own gene shared by IL-1␤, but not by IFN-␣, and IL-6. Furthermore, TNF-␣ expression. This feedback-loop mechanism generates circadian interferes with the expression of Dbp in the suprachiasmatic oscillations of Per and Cry expression. The same positive and nucleus and causes prolonged rest periods in the dark when mice negative regulatory components also govern the rhythmic ex- show spontaneous locomotor activity. Using clock reporter genes pression of the nuclear orphan receptor Rev-Erb␣, which in turn TNF-␣ is found here to inhibit CLOCK-BMAL1-induced activation of represses the transcription of Bmal1 through direct binding to a E-box regulatory elements-dependent clock gene promoters. We REV-ERB␣ response element in the Bmal1 promoter. Thereby, suggest that the increase of TNF-␣ and IL-1␤, as seen in infectious REV-ERB␣ interconnects the cyclic expression of positive- and and autoimmune diseases, impairs clock gene functions and causes negative-loop members (for reviews, see refs. 15 and 16). The fatigue. targeted inactivation of Bmal1 showed that this gene is indis- pensable for the maintenance of circadian functions (17). Clock- behavior ͉ circadian rhythms ͉ cytokines ͉ innate immunity deficient mice show robust circadian patterns of locomotor activity with period lengths shortened by only 20 min; Dbp mRNA rhythm, however, was severely blunted in both the SCN n microbial infections, host defense mechanisms activate the and the liver (18). The deletion of the clock-controlled genes innate and adaptive arms of the immune response. Microbial I (CCG) PAR bZip transcription factors Dbp, Tef, and Hlf only recognition by Toll-like receptors (TLR) expressed by macrophages moderately affects the circadian clock but leads to pronounced and dendritic cells leads to the activation of signal transduction disturbances of locomotor activity. In the present study, our pathways with induction of various genes including IL-1, TNF-␣, objective was to examine (i) whether and how TNF-␣ influences IL-6, and IFN-␣/␤ (1–3). These cytokines mediate the acute-phase the circadian timing system and modulates the expression of response, which is a systemic generalized reaction characterized by clock genes and CCGs, and (ii) whether TNF-␣ affects locomo- fever, fatigue, and weight loss, an increase in the number of tor activity, rest time, and periodicity in mice. neutrophils, and the induction of synthesis of acute-phase proteins in the liver with increased haptoglobin, antiproteases, complement Results components, fibrinogen, ceruloplasmin, and ferritin in the blood Suppressed Expression of Clock Genes in TNF-␣-Treated Synchronized (4). An acute phase response with a dose-dependent state of Fibroblasts. To determine whether TNF-␣ interferes with circa- lethargy and severe fatigue has been described in cancer patients ␣ ␣ dian gene expression, we used NIH 3T3 fibroblast cultures IMMUNOLOGY treated with TNF- (5). A link between production of TNF- and synchronized by a 2-h serum shock; this system elicits a well daytime fatigue has also been suggested in rheumatoid arthritis described circadian expression of central clock genes and CCGs (RA) and in the obstruction sleep apnea syndrome (OSAS). for at least three cycles (19). We analyzed serum shocked Inhibition of TNF-␣ by soluble TNF-receptor p75 improves dis- abling fatigue in patients with RA (6). A TNF-␣-308 (A-G) single-nucleotide polymorphism and elevated TNF-␣ serum levels Author contributions: G.C. and S.P. contributed equally to this work; T.B. and A.F. contrib- have been described in OSAS (7, 8). Moreover, neutralization of uted equally to this work; G.C., S.P., I.T., T.B., and A.F. designed research; G.C., S.P., P.K., C.J., TNF-␣ reduces daytime sleepiness in sleep apneics (9). Direct T.B., and A.F. performed research; C.J. and I.T. contributed new reagents/analytic tools; G.C., S.P., I.T., T.B., and A.F. analyzed data; and T.B. and A.F. wrote the paper. effects of TNF-␣ on spontaneous sleep are also shown in animal studies. i.v., i.p., or intracerebroventricular injections of TNF-␣ or This article is a PNAS Direct Submission. IL-1 enhance nonrapid eye-movement (NREM) sleep (for reviews, Abbreviations: TLR, Toll-like receptor; RA, rheumatoid arthritis; NREM, nonrapid eye movement; SCN, suprachiasmatic nuclei; CCG, clock-controlled gene; ZT, Zeitgeber time; see refs. 10 and 11). This increase in NREM sleep is independent LD, 12-h light/12-h dark. of the fever-inducing capacity of these cytokines (12). Although §To whom correspondence may be addressed. E-mail: [email protected] or there is good evidence for TNF-␣ and IL-1 as mediators of altered [email protected]. sleep–wake behavior, the underlying mechanisms remain elusive. This article contains supporting information online at www.pnas.org/cgi/content/full/ The regulation of sleep depends on a circadian control and a 0701466104/DC1. homeostatic drive (13, 14). The circadian influence is provided © 2007 by The National Academy of Sciences of the USA www.pnas.org͞cgi͞doi͞10.1073͞pnas.0701466104 PNAS ͉ July 31, 2007 ͉ vol. 104 ͉ no. 31 ͉ 12843–12848 Downloaded by guest on October 1, 2021 A B Dbp Per 1 3 Clock 3 Per1 3 Dbp 120 140 rol 2 2 2 Rev-erb α nt 100 * 120 1 1 1 Clock 100 o t=0 o 80 t Bmal1 Per1 80 ** n 3 6 Per2 3 Tef * 60 *** 60 tio 2 4 2 Per2 ** a 40 2 1 *** 40 1 Per3 ** 0 elative to co 20 20 vari Dbp r Per3 Hlf *** ld 8 Rev-erb α 2 3 0 0 % 6 Tef 2 * 4 1 o x-f 1 Hlf 2 *** 20 806040 100 120 140 Per 2 Per 3 0 4 8 12162024283236 0 4 8 12162024283236 0 4 8 12162024283236 Zeitgeber Time (hours) % relative to control 140 160 120 140 100 120 3 Dbp ** 100 0 24h TNF 80 ** C = 80 t D 60 24h TNF/ 24h TNF 60 *** to 2 24h TNF/ 24h mock 40 n n 40 elative to control 20 20 r 0 0 % atio ari Dbp 1 α β α γ v ø TNF-αIL-1β IL-6 IFN-αIFN-γ ø TNF- IL-1 IL-6 IFN- IFN- d Per1 l 24 Per2 x-fo 0 8 16 24 32 40 48 56 Clock Fig. 2. Cytokine effects on Dbp, Per1, Per2, and Per3 in NIH 3T3 fibroblasts. Zeitgeber Time (hours) Confluent cells were synchronized with 50% horse serum for 2 h (ZT 0–2). Dbp * After serum shock, cells were kept in serum-free medium with TNF-␣ (10 TNF-α ␤ ␣ ␥ E Per1 *** ng/ml), IL-1 (10 ng/ml), IL-6 (10 ng/ml), IFN- (10 ng/ml), IFN- (20 ng/ml) or 1.6 (ng/ml): Zeitgeber time: 1.4 without cytokines and analyzed at ZT 24 with quantitative real-time RT-PCR. 0.1 48 Per2 *** 1.2 1 Results are shown as percent of expression to noncytokine-treated fibroblast 1.0 Clock Ϯ 0.8 10 cultures; one representative experiment of three; mean values of triplicates 0.6 20 Յ Յ Յ 0 20406080100120140 SD; independent sample t test; *, P 0,05; **, P 0,005; ***, P 0.0005. 0.4 0.2 % expression compared to mock-treated 0 to mock-treated rel. mRNA expression Dbp Per1 Per2 Clock second circadian cycle and analyzed rhythmic expression up to Fig. 1. TNF-␣ impairs expression of clock genes in synchronized NIH 3T3 60 h. In this, emphasis was placed on Dbp expression, because its ␣ fibroblasts. (A) TNF- attenuates circadian Per1/2/3 and PAR bZIP family (Dbp, amplitude was the most affected by TNF-␣. Rhythmicity and Tef, and Hlf) gene expression. After serum shock (ZT 0–2), cells were kept in period length of Dbp expression were not changed by TNF-␣, but serum-free medium with TNF-␣ (10 ng/ml; open circles) or without the cyto- kine (filled circles) and analyzed every 4 h with quantitative real-time RT-PCR.