Introduction

Top Abstract Introduction Methods Results Discussion References

A large number of physiological rhythmic variables are demonstrated in the CNS, in hormone secretion and so on (Kafka et al., 1981; Naber et al., 1981; Thomson et al., 1980). Also, many drugs vary in potency and/or toxicity according to the time in the circadian cycle when they are administered (Ohdo et al., 1988, 1990, 1995a, 1995b, 1996; Frederickson et al., 1977; Walker and Owasoyo, 1974).

Interferons, which belong to a group of cytokines, have been widely used as antiviral and antitumor agents in the human. However, interferons cause unavoidable adverse effects such as fever, fatigue, headache, rigors and myalgias. In particular, fever is an indispensable side effect in nearly all patients during the early phase of interferon treatment. Administration of IFN- in cancer patients is better tolerated in evening than in morning (Abrams et al., 1985). There are also significant dosing time-dependent differences in the antitumor and myelosuppressive activity of IFN- in mice (Koren et al., 1993; Koren and Fleischmann, 1993). However, the rhythmic changes of interferon-induced fever and antiviral activity have not yet been examined.

Rectal temperature and immune functions show significant circadian rhythms in mammals under both nondrugged and drugged conditions (Ohdo et al., 1995a; Refinetti et al., 1990; Haus et al., 1983; Batalla et al., 1994). Therefore, there may be a chronobiologic effect on the fever and antiviral activity induced by IFN-. The increase in body temperature induced by interferons may be one aspect of its antiviral activity as an immunoadjuvant effect, but an excessive febrile reaction may be more detrimental than beneficial.

The purpose of this study was to examine the diurnal change of IFN- induced fever and antiviral activity in mice. The mechanisms underlying these phenomena were also investigated from the perspective of IFN- pharmacokinetics.

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals and treatments. Male ICR mice (5 weeks old) were purchased from Charles River Japan Inc. (Kanagawa, Japan). Mice were housed 6 or 10 per cage in a light-controlled room (light on from 07:00 to 19:00) at a room temperature of 24°C ± 1°C and a humidity of 60% ± 10% with food and water ad libitum. All mice were adapted to their light-dark cycle for 2 weeks before the experiments. In order to study the fever induced by IFN- (Sumiferon, Sumitomo Seiyaku Co., Osaka, Japan), groups of six mice injected i.v. with 1.0, 5.0 or 10.0 MIU/kg IFN- or sterilized saline at the same circadian phase (09:00). IFN- was diluted by sterilized saline to adjust the concentration to 0.2, 1.0 and 2.0 MIU/ml. The volume of injection was 0.05 ml/10.0 g b.wt. The drug solutions were used within 30 min after preparation in order not to decrease their biologic activity. Rectal temperature was continuously determined before, and at 0.5, 1.0, 2.0 and 4.0 hr after, IFN- or saline injection. In the study of the circadian rhythms of IFN--induced fever and plasma IFN- concentrations, groups of 8 to 10 mice were injected i.v. with 10.0 MIU/kg IFN- or saline at one of six times: 09:00, 13:00, 17:00, 21:00, 01:00 or 05:00. Rectal temperature was determined before, and at 0.5, 1.0, 1.5 and 2.0 hr after, IFN- or saline injection. Percent change of rectal temperature (%) from basal level was calculated as follows: % = ([rectal temperature after IFN- injection  rectal IFN- before IFN- injection] / [rectal temperature before IFN- injection]) × 100. Blood samples were drawn by cardiac puncture at 2.5 hr after IFN- injection and placed into polypropylene tubes containing 10 µl of EDTA (4%) solution. To observe the PGE₂ production induced by IFN-, groups of six mice were injected i.v. with 10.0 MIU/kg IFN- on one of two occasions: in the latter half of the light phase (17:00) or in the latter half of the dark phase (05:00). Blood samples were drawn by cardiac puncture at 0.5 hr after IFN- injection and placed into polypropylene tubes containing 10 µl of indomethacin (40 mM) / EDTA (4%) solution. Immediately after blood sample collection, thalamus was removed and placed into ice-cold tubes. To examine 2-5OAS activities induced by IFN-, groups of 8 to 10 mice were injected i.v. with 10.0 MIU/kg IFN- on one of two occasions as described above. Blood samples were collected by cardiac puncture at 24 hr after IFN- injection and placed into polypropylene tubes containing 10 µl of EDTA (4%) solution. Immediately after blood collection, liver was perfused with 0.01 M PBS. The liver was quickly removed, rinsed with saline and placed into ice-cold tubes. To study the time course of plasma IFN- concentrations, groups of six mice were injected i.v. with 10.0 MIU/kg IFN- on one of two occasions as described above. Blood samples were drawn by orbital sinus collection at 0.167, 0.5, 1.0, 2.0, 3.0 and 4.0 hr after IFN- injection.

Determination of IFN- induced fever. IFN- induced fever was determined by measuring the rectal temperature after IFN- or saline injection. Rectal temperature was measured on a digital thermometer (digital thermometer TD-300, Shibaura Electronics, Tokyo, Japan). A lubricated thermocouple was inserted 1.5 cm into the rectum of mice. Rectal temperature was measured at least every 30 min to avoid hyperthermia occasioned by continuous handling stress (Briese et al., 1991).

Determination of PGE₂ concentration in plasma and thalamus. Plasma samples were obtained after centrifugation at 3000 rpm for 3 min. The plasma samples (300 µl) were added to ethanol solution to give a final concentration of 10% ethanol. The ice-cold thalamus was weighed and homogenized with the cold absolute ethanol (200 µl), distilled water being added to give final concentration of 10% ethanol. The supernatant, after centrifugation at 1500 × g × for 20 min, was used as the thalamus homogenate sample. PGE₂ was extracted from plasma and thalamus samples according to the method of Powell (Powell, 1982) and Shono (Shono et al., 1988). The plasma and thalamus homogenate samples were acidified to pH 3.0 by acetic acid and applied to a SEP-PAK C₁₈ column (Waters, Massachusetts). The solvent of crude extract was evaporated and dissolved in the HPLC mobile-phase buffer (methanol / H₂O/acetic acid, 60 / 40 / 0.01, v/v/v). Further purification was performed by HPLC using ODS-80Ts column (4.6 mm I.D. × 150 mm) connected to a pump (655A-11 Liquid Chromatograph, Hitachi, Tokyo, Japan). The flow rate was 0.8 ml/min. The PGE₂ fractions for assay were collected from 21 to 26 min. The solvent was evaporated, and the residue was dissolved in assay buffer (0.1% bovine serum albumin/0.1 M phosphate-buffered saline pH 7.4). PGE₂ concentrations were determined by enzyme immunoassay (PGE₂ immunoassay system, Amersham, Bukinghamshire, U.K.).

Determination of 2-5OAS activities in plasma and liver. Plasma samples were obtained after centrifugation at 3000 rpm for 3 min and then stored at 20°C until assayed. The ice-cold liver was immediately homogenized with modified lysis buffer (10 mM HEPES-KOH/50 mM KCl/3 mM Mg(OAc)₂/0.3 mM EDTA/10% glycerol/0.01% NaN₃/0.5% Triton-100/100 µM PMSF/7 mM 2-mercaptoethanol, pH 7.5) (Sokawa et al., 1994). The supernatants, after centrifugation at 9000 × g for 20 min at 4°C, were used as the liver sample. The protein concentrations in the liver homogenate sample were determined by Lowry's method. The plasma and liver 2-5OAS activities were determined by radioimmunoassay (2-5A kit, Eiken, Tokyo, Japan). The 2-5OAS activities in liver were expressed as 2-5oligoadenylate fmol per liver protein concentration.

Determination of IFN- concentration in plasma. Plasma samples were obtained after centrifugation at 3000 rpm for 3 min and stored at 20°C until assayed. Plasma IFN- concentrations were determined by enzyme-linked immunosorbent assay (ELISA) (IFN- immunoassay kit, BioSource International Inc, California). The titer was expressed in international units (IU) per milliliter, and the detection limit in the sample was 10 IU/ml. There was no cross-reactivity with endogenous mouse interferons.

Statistical analysis. Pharmacokinetic parameters were calculated by the nonlinear least-squares method, following the two-compartment model: CL, V_c, K₁₂ and K₂₁. Analysis of variance (ANOVA) and Tukey's test were applied for the multiple comparison. Student's t test was used for independent comparison between groups. The 5% level of probability was considered to be significant.

	Results

Top Abstract Introduction Methods Results Discussion References

Influence of IFN- on body temperature. The effects of three dosages (1.0, 5.0 and 10.0 MIU/kg) of IFN- on rectal temperature in mice injected with the drug at the same circadian phase (09:00) are shown in figure 1. The rectal temperature increased from basal level at all dosages of IFN-. The rectal temperature at 0.5 hr after IFN- 10.0 MIU/kg injection was significantly different from that after saline injection (P < .01). However, the rectal temperature of mice injected with IFN- 1.0 or 5.0 MIU/kg was not significantly different from that of mice injected with saline.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   The time course of rectal temperature after IFN- injection. Each point represents the mean ± S.E. of six mice. ** P < .01 when compared with the saline group using Tukey's test. , saline; , IFN- 10.0 MIU/kg; , 5.0 MIU/kg; , 1.0 MIU/kg.

Circadian rhythm of IFN- induced fever. The rectal temperature in mice injected with saline showed significant circadian rhythm with a lower level during the light phase and a higher level during the dark phase (P < .01; fig. 2). The rectal temperature after IFN- 10.0 MIU/kg injection was significantly higher during the 24-hr cycle when compared with that after saline injection (P < .01). The rhythmic pattern of IFN--induced fever resembled overall the rhythm that occurred after saline injection. However, fever was not induced by IFN- injection in the latter half of the dark phase (05:00). The time course of rectal temperature was expressed as percent change from basal level, the level before IFN- injection (fig. 3). The percent changes in rectal temperature at 0.5 hr after IFN- injection were significantly higher in the light phase than in the dark phase (P < .01). However, the rectal temperature after IFN- injection in the latter half of dark phase (05:00) showed no significant difference from that after saline injection.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Circadian rhythm of rectal temperature at 0.5 hr after IFN- (10.0 MIU/kg i.v.) injection () or saline injection (). Each point represents the mean ± S.E. of 8 to 10 mice. ** P < .01 when compared with the corresponding saline group using Tukey's test.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   The time course of percent change of rectal temperature after IFN- (10.0 MIU/kg i.v.) injection at 09:00 (), 13:00 (), 17:00 (), 21:00 (), 01:00 () and 05:00 (). Value of rectal temperature is expressed as percent change from basal level. Each point represents the mean ± S.E. of 8 to 10 mice.

Influence of dosing time on PGE₂ levels in plasma and thalamus. The effect of dosing time of IFN- on PGE₂ production in plasma and thalamus is shown in figure 4. There was no significant difference in PGE₂ levels in plasma and thalamus between mice injected with saline at 17:00 and 05:00. The PGE₂ levels in plasma after IFN- injection at 17:00 tended to be higher than those after saline injection at 17:00 (P < .10). The PGE₂ levels in thalamus at 0.5 hr after IFN- injection were significantly higher in mice injected with the drug at 17:00 than in those injected at 05:00 (P < .01). The PGE₂ levels in thalamus after IFN- injection at 17:00 also increased significantly when compared with those after saline injection at 17:00 (P < .01).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Influence of dosing time on the plasma and thalamus PGE2 levels after IFN- (10.0 MIU/kg i.v.) injection at 17:00 or 05:00. Each column represents the mean ± S.E. of six mice. ** P < .01 when compared with the saline group or between the two dosing times using Tukey's test. : saline; : IFN-.

Influence of dosing time on 2-5OAS activity in plasma and liver. The effect of time dosing with IFN- on 2-5OAS activity in plasma and liver is shown in figure 5. 2-5OAS activities in plasma and liver showed no significant difference between mice injected with saline at 17:00 and 05:00. 2-5OAS activity in plasma at 24 hr after IFN- injection was significantly higher for injection at 05:00 than for injection at 17:00 (P < .05), but 2-5OAS activity in liver at 24 hr after IFN- injection showed no dosing time-dependent difference. 2-5OAS activities in plasma and liver after IFN- injection at 05:00 increased significantly when compared with those after saline injection (P < .01, P < .05 respectively).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Influence of dosing time on the plasma and liver 2-5OAS activities after IFN- (10.0 MIU/kg i.v.) injection at 17:00 or 05:00. Each column represents the mean ± S.E. of 8 to 10 mice. * P < .05; ** P < .01 when compared with the saline group or between the two dosing times using Tukey's test. : saline; : IFN-.

Circadian rhythm of IFN- concentrations in plasma. The plasma IFN- concentrations at 2.5 hr after IFN- injection showed a significant circadian rhythm with higher levels from late dark phase to early light phase and lower levels from late light phase to early dark phase (P < .01; fig. 6). The time course of plasma IFN- concentrations after IFN- injection decayed biphasically (fig. 7). IFN- concentrations at 2.0, 3.0 and 4.0 hr after IFN- injection were significantly higher for injection at 05:00 than for injection at 17:00 (P < .05). CL was significantly higher in mice injected with IFN- at 17:00 than in those injected at 05:00 (P < .05, table 1). There was no significant difference in any other pharmacokinetic parameters between mice injected with the drug at 17:00 and those injected at 05:00.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Circadian rhythm of plasma IFN- concentrations at 2.5 hr after IFN- (10.0 MIU/kg i.v.) injection. Each point represents the mean ± S.E. of 8 to 10 mice.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   The time course of plasma IFN- concentrations after IFN- (10.0 MIU/kg i.v.) injection at 17:00 () or 05:00 (). Each point represents the mean ± S.E. of six mice. * P < .05 when compared between the two groups using Tukey's test.

View this table: [in this window] [in a new window]   TABLE 1 Influence of dosing time of IFN- (10.0 MIU/kg i.v.) injection on pharmacokinetic parameters

	Discussion

Top Abstract Introduction Methods Results Discussion References

Rectal temperature in mice showed a significant circadian rhythm with higher levels during the dark phase and lower levels during the light phase under nondrugged conditions. Our result confirms previous observations (Ohdo et al., 1995a). Normal body temperature is regulated by the relative balance of catecholamine and serotonin levels in the anterior hypothalamus (Feldberg and Myers., 1964). The changes in locomotor activities, eating, drinking and secretion of several hormones influence the rhythm of rectal temperature (Refinetti and Menaker, 1992). The rectal temperature at 0.5 hr after IFN- injection increased significantly during the 24-hr cycle, except for the latter half of the dark phase (05:00), when compared with that after saline injection. The rhythmic pattern of IFN--induced fever resembled overall the rhythm occurring after saline injection, but the percent changes from basal level of rectal temperature after IFN- injection varied according to the dosing time. IFN- acts on the thermosensitive neurons in the preoptic and anterior hypothalamus and increases body temperature via PGE₂ production and/or opioid receptor (Nakashima et al., 1988, 1995; Dinarello et al., 1984). Certainly, cyclooxygenase inhibitors, by decreasing PGE₂ production, suppress IFN- induced fever. PGE₂ levels in the thalamus at 0.5 hr after IFN- injection were significantly higher in mice injected with the drug at 17:00 than in those injected at 05:00. This seems to coincide with the circadian rhythm of IFN--induced fever. Although plasma IFN- concentrations showed significant circadian rhythm, it was out of phase with the rhythm of IFN--induced fever. Thus the rhythmicity of IFN--induced fever seems to be due to that of the sensitivity of mice to the drug.

The important question still remains whether the antiviral activity of IFN- declines at the dosing time that alleviates IFN--induced fever. The antiviral activity of interferon due, at least in part, to the 2-5oligoadenylate synthetase system (Baglioni, 1979). 2-5OAS is the enzyme directly related to the antiviral action of interferon. Serum 2-5OAS activity is used as an index of the antiviral effect of interferon in patients with hepatitis. There was no significant dosing time-dependent difference in 2-5OAS activity between saline injection at 17:00 and that at 05:00. However, both plasma and liver 2-5OAS activities induced by IFN- were higher in mice injected with drug at 05:00 than in those injected at 17:00. The rhythm corresponded well to the rhythmicity of IFN- concentration. Therefore, the diurnal difference of 2-5OAS activity induced by IFN- can be explained, at least in part, by the rhythm of plasma IFN- concentration. The circadian rhythm of antitumor activity induced by IFN- exhibits higher activity in the early light phase (Koren et al., 1993). In the circadian phase, plasma IFN- concentration was higher in the present study. The rhythm of IFN--induced antitumor activity also seems to be due to that of IFN- pharmacokinetics.

Plasma IFN- concentrations at 2.5 hr after IFN- injection showed a significant circadian rhythm. A significant dosing time-dependent difference was also demonstrated for the pharmacokinetic parameter of IFN-, which showed higher CL for injection at 17:00 than for injection at 05:00. The rhythmicity in CL seems to be closely related to that in plasma IFN- concentration. IFN- concentrations in plasma have been shown to decay biphasically after an i.v. injection of IFN-, and the distribution phase lasted for 1.0 hr after the drug injection (Cantell and Pyhärä, 1973). IFN- is quickly eliminated from the body by several pathways. The main route of excretion of IFN- is the kidneys (Bino et al., 1982). Renal tubular cells take up and break down many plasma proteins (Strober and Waldmann, 1974). IFN- is also internalized and catabolized intracellularly in kidney via receptor-mediated endocytosis (Bocci et al., 1983). Both renal elimination rate and liver metabolism rate increase during the active period in mice (Ohdo et al., 1995a). The rhythmicity can reflect not only the rhythmic activity of the enzyme in kidney and liver but also the rhythmic rate of blood flow (Labrecque et al., 1988). The rhythmicity of CL in our study corresponds well to that of blood flow. Thus the circadian rhythm in IFN- pharmacokinetics may be caused by the diurnal rhythm of renal function. Although the circadian rhythm of receptor-mediated endocytosis has not been investigated yet, this should be clarified in future.

The present findings in this mouse model support the concept that the choice of the most appropriate time of day for administration of interferons may reduce their side effects and increase their antiviral activity in clinical situations.