DIURNAL VARIATION OF TNFa IN THE RAT BRAIN

 Rachael A. Floyd and James M. Krueger

 Department of Physiology and Biophysics, University of Tennessee, Memphis, TN 38163  USA

KEY WORDS:  TNFa, circadian rhythm, brain, sleep, cytokine

 ABSTRACT

TNFa is thought to play a physiological role in the brain. Current studies were

performed to determine if a diurnal rhythm of TNFa exist in the rat brain. Brain samples

[hippocampus (HC), hypothalamus (HT), cerebral cortex (CT), cerebellum, pons and

midbrain] were collected at light onset and at 6-hour intervals thereafter over a day. A

TNFa bioassay was used to measure TNFa in each area. TNFa was highest at light onset

in the HT, HC and CT. Values at light onset were about 10-fold greater than minimal

nighttime values. Changes in TNFa activity in other brain areas were also evident but

lesser in magnitude. Current results support the hypothesis that TNFa has physiological

roles in the brain.

INTRODUCTION

Tumor necrosis factor-a (TNFa) is a pleiotropic molecule possessing

immunological, endocrine and central nervous system activities. There is much evidence

regarding the involvement of TNFa in host defense mechanismsreviewed in 1. In contrast, the

physiological roles of TNFa are not as well known although TNFa is implicated in normal

brain developmentreviewed 2 and neuroendocrine regulationreviewed 3. Substantial evidence also

suggests that TNFa is involved in sleep regulation. Thus, administration of exogenous

TNFa into rats4, rabbits5, 6, or mice7 induces excess sleep. In contrast, inhibition of

endogenous TNFa inhibits spontaneous sleep and the sleep rebound that occurs after

sleep deprivationreviewed 8. Further, mice lacking the TNF 55 kD receptor sleep less than

strain controls7. Plasma levels of TNFa vary in phase with electroencephalographic slow-

wave amplitude9 and increase during sleep deprivation10.

TNFa is constitutively expressed in normal brainreviewed 8. Further, in rats there is a

diurnal rhythm of TNFa mRNA levels in the hypothalamus and hippocampus11. Both of

these areas are involved in sleep regulation and TNFa mRNA levels were highest during

daylight hours; this is the period of maximal sleep for rats. Nevertheless, the day/night

differences in TNFa mRNA levels in the brain were only about 2-fold and there is

evidence from other tissues that TNFa is regulated, in part, post-transcriptionally.

Therefore, it was of interest to determine whether TNFa bioactivity in brain varied across

the day. We report herein that there is about a 10-fold greater amount of TNFa

bioactivity in the hippocampus and cortex just after lights are turned on than during the

dark period.

MATERIALS AND METHODS

Animals: Sprague Dawley rats weighting 250-300 g were acclimated to their

experimental cages for ten days. Animals were housed in plastic rat shoeboxes and fed

standard rat chow and water ad libitum. The room was maintained on a 12-hour

light/dark cycle with the lights on at 06:00 hours at a temperature of 21 C. Rats were

sacrificed at 06:00, 12:00, 18:00 and 24:00 hours. Care of the animals was in accordance

with NIH standards published in the "Guide for Care and Use of Laboratory Animals. All

protocols received prior approval by the Animal Care and Use Committee at the

University of Tennessee, Memphis.

Brain Sample preparation for TNFa Bioassay: Brains were removed quickly from

decapitated rats and placed in separate sterile petri dishes on ice containing RPMI-1640

media with 2% fetal calf serum and protease inhibitors (aprotinin 1mg/ml, leupeptin 10

mg/ml, pepstatin A 10 mg/ml, and benzaminide 1mM) (PIB). Each brain was dissected

into the hypothalamus, hippocampus, cortex, pons, midbrain and cerebellum. Each area

was placed in an Eppendorf tube and frozen in liquid nitrogen. The samples were stored

in a -70 C freezer until processed further. Upon removal from the -70 C freezer, the

samples were weighed, combined with PIB added at a ratio of 1 g/ml, sonicated for 10-20

seconds on ice and then centrifuged at 100,000 x g for 1 hour. The supernatants were

collected in sterile tubes and assayed immediately or frozen at -70 C until assayed.

TNFa Bioassay: Brain TNFa levels were measured using the WEHI 164 subclone 13 cell

line (ATCC, Rockville, Maryland). This system is highly sensitive and specific for the

measurement of TNF12. This procedure is a modification of the assay used by Denizot and

Lang13. Twenty-five microliters of cells (2.5 x 104 cells/well) in RPMI- 1640

supplemented with 1 mM sodium pyruvate, 50 mg/ml penicillin and 25 mg/ml

streptomycin, 0.5 mg/ml actinomycin D, protease inhibitors and 2% fetal calf serum was

added to 96-well flat bottomed microtitre plates (Costar, Cambridge, MA) and incubated

for 20 hours at 37 C with 25 ml of serial dilutions of the samples. Next, the culture media

was removed, and 25 ml of a 1 mg/ml solution of 3-(4,5-Dimethylthiazol-2-yl)-2,5-

diphenyltetrazolium bromide (MTT) (Sigma, St. Louis MO) in RPMI-1640 media without

phenol was added to each well. The plates were incubated for 3 hours at 37 C. At the

end of the incubation period, plates were centrifuged for 5 minutes at 1,000 x g. After

careful removal of the MTT solution, 50 ml of ethanol was added to each well and the

plates were shaken vigorously on an orbital shaker to dissolve the purple formazan

crystals. The absorbance at 550 nm of each well was read by a BioRad ELISA reader

(Richmond, CA). The rat recombinant TNFa (Biosource International, Camarillo, CA)

was used as the standard for this assay in a range from 0-125 pg/ml. All samples were run

in triplicate.

As an additional control the recombinant rat TNFa (50 pg/ml TNFa bioactivity)

and rat brain samples (50 pg/ml TNFa) were incubated with various concentrations of

rabbit anti-mouse TNFa antibody (0-160 units TNFa antibody) (Biosource International,

Camarillo, CA) for 1 hour prior to adding them to the WEHI tumor cells to determine if

the TNF activity present in the samples was blocked. This anti-mouse TNFa antibody

cross reacts with rat TNFa (Biosource International, Camarillo, CA).

Statistics: Results were analyzed by one-way analysis of variance (ANOVA), Krustal-

Wallis one-way ANOVA on ranks or the Student-Newman-Keuls method.

RESULTS

Diurnal Variation of TNFa Protein in the Brain: The levels of TNFa in the

hippocampus, cortex and hypothalamus had significant (p < 0.05) diurnal variations

(Figure 1). Peak values were obtained in the samples taken at light onset (06:00 hours;

Figure 1). Levels declined in the 12:00 and 18:00 hour samples in the hippocampus,

cortex and hypothalamus but remained above TNFa bioactivity at 24:00 hours. Daytime

levels of TNFa were about 10-fold greater in the hippocampus and cortex than nighttime

values. Although TNFa levels in the cerebellum at 06:00 hours were significantly

different from the 18:00 hour time point (p < 0.05) these differences were only about 2-

fold. No statistically significant differences in TNFa levels were found in the midbrain and

pons.

Inhibition of TNFa Bioactivity using Anti-rat TNFa Antibody: About 83% of the

measured TNFa bioactivity of recombinant rat TNFa was blocked if samples were

pretreated with the rabbit anti-mouse TNFa antibody. The TNFa antibody also inhibited

about 75% of TNFa bioactivity in the hippocampus, hypothalamus and cortex.

DISCUSSION

Present results confirm previous findings showing the presence of TNFa in the

brain and extend those previous results by showing a significant diurnal variation in the

expression of TNFa in hippocampus, cortex and hypothalamus (Figure 1). This finding

supports the hypothesis that TNFa has a role in normal brain physiology.

It is likely that the brain is the source of TNF responsible for the diurnal changes

observed. Thus, previously we demonstrated a diurnal variation of TNF mRNA in the

hypothalamus and hippocampus although this difference was only about 2-fold. In

contrast, the difference in TNFa protein levels reported were about 10-fold. On the other

hand, there are changes in circulating TNFa associated with the sleep/wake cycle; TNFa

levels in plasma correlate with electroencephalographic slow-wave activity9. Further,

sleep deprivation primes monocytes for lipopolysaccharide-10, 14 or streptococcal15-induced

TNF production. Finally, there is a suggestion that the higher levels of plasma TNFa

found in patients with obstructive sleep apnea syndrome contribute to their daytime

sleepiness16. Banks and colleagues17 have described a TNF active transport system for

TNF uptake by brain. Nevertheless, because the magnitudes of the diurnal variations of

TNF in different parts of the brain were very different it seems likely that brain TNF rather

than circulating TNF is the source of the differences.

The literature indicating that bacterial products induce TNF mRNA production in

brain18, 19 and the diurnal variation of TNF mRNA11 suggest at least some degree of

transcriptional control for TNFa gene expression20, 21. Nevertheless, much other data

suggests that the primary regulatory steps for TNFa production are post-transcriptional22,

23. Thus, if endotoxin is given directly into the brain cerebrospinal fluid TNFa levels

increase within 15 minutes24. In addition, the presence of protein kinase inhibitors

decrease the half-life of TNFa transcripts25. Current results are also consistent with the

idea that post-transcriptional regulation of TNF is dominant. Thus, in the cortex day/night

differences in TNF bioactivity were about 10-fold (Figure 1) whereas previously we were

unable to show a diurnal variation in cortical mRNA11.

The hypothesis that cytokines such as TNF are involved in physiological processes

has received little attention except within the sleep literature. It is posited that under

normal conditions, low basal levels of TNF production/effects vary in subtle ways with

physiological processes. A network of interactions between several cytokines and

hormones could give rise to a specificity of response for any one physiological function

such as sleep26, 27. Regardless of such speculation, current data clearly indicate a diurnal

rhythm in brain of TNF bioactivity; such changes are likely related to the physiological

functions affected by TNF.

CONCLUSION

The studies showing TNFa regulates such functions as sleep, appetite, and release

of neuropeptides strongly suggest a physiological role for TNFa in the brain. Our study

was performed to determine whether there is a diurnal rhythm of TNFa in the rat brain in

the areas of the hypothalamus (HT), hippocampus (HC), cerebral cortex (CT), cerebellum,

pons and midbrain of the rat brain. Brain samples were collected at 6-hour intervals

starting at light onset. TNFa was measured using the TNFa bioassay. TNFa was highest

at light onset in the HT, HC and CT being about 10-fold higher compared to other times

during the 24 hour period. TNFa bioactivity in brain tissues was neutralized by an anti-

TNFa antibody. While it is exciting to speculate on the meaning of these results it is clear

that there is a diurnal variation of TNFa in the brain. These data are consistent with

TNFa having physiological functions in the brain.

GENERAL SUMMARY

TNFa plays an important physiological role in sleep regulation. It is hypothesized

that TNFa levels should be high during the time that rats sleep and lower during the time

they are most active. To determine if a diurnal variation exists for TNFa in the brain we

collected various areas of the rat brain at 6 hour intervals over a 24 hour period. Using

the TNFa bioassay to analyze TNFa activity, we determined that TNFa peaked at light

onset in the hypothalamus, hippocampus, and cerebral cortex then decreased during the

light phase and remained low during the dark phase. TNFa values at light onset were

significantly different from minimal nighttime values. These results are consistent with

diurnal changes observed in TNFa mRNA levels and with the hypothesis that TNFa might

have a physiological role in the brain.

ACKNOWLEDGMENTS

This research was supported by National Institutes of Health Grants NS-01727

and NS-31453. We thank Ms. Maria Swayze for her secretarial assistance.

REFERENCES

1. Beutler B and Cerami A. Annu Rev Immunol 7, 625-655 (1989).

2. Merrill JE. Dev Neurosci 14, 1-10 (1992).

3. Rivier C. Front Neuroendocrinol 16, 151-182 (1995).

4. Nistico G, DeSarro G and Rotiroti D. Behavioural and electrocortical spectrum power changes of interleukins and tumor necrosis factor after their microinfusion into different areas of the rat brain. In: Smirne, S, Franceschi M, Ferini-Strambi L, et al., eds. Sleep, Hormones and Immunological System, Masson: Milan, 1992; 11-22.

5. Kapás L, Hong L, Cady AB, et al. Am J Physiol 263, R708-R715 (1992).

6. Shoham S, Davenne D, Cady AB, et al. Am J Physiol 253, R142-R149 (1987).

7. Fang J, Kapás L, Wang Y, et al. Soc for Neurosci Absts 22, 147 (1996).

8. Krueger JM and Majde JA. Crit Rev in Immunol 14, 3555-3579 (1994).

9. Darko DF, Mitler MM and Henriksen SJ. Adv Neuroimmunol 5, 57-77 (1995).

10. Yamasu K, Shimada Y, Sakaizumi M, et al. Eur Cytokine Netwk 3, 391-398 (1992).

11. Bredow S, Obál F Jr, Guha-Thakurta N, et al. Soc. for Neurosci. Absts. 22: 146 (1996).

12. Eskandari MK, Nguyen DT, Kunkel SL, et al. Immunol Invest 19, 69-79 (1990).

13. Denizot F and Lang R. J Immunol Methodol 89, 271-277 (1986).

14. Hohagen F, Timmer J, Weyerbrock A, et al. Neuropsychobiology 28, 9-16 (1993).

15. Uthgenannt D, Schoolmann D, Pietrowsky R, et al. Psychosom Med 57, 97-104 (1995).

16. Entzian P, Linnemann K, Schlaak M, et al. Am J Respir Crit Care Med 153, 1080-1086 (1996).

17. Gutierrez EG, Banks WA and Kastin AJ. J Neuroimmunology 47, 169-176 (1993).

18. Gatti S and Bartfai T. Brain Res 624, 291-294 (1993).

19. Breder CD, Hazuka C, Ghayur T, et al. Proc Natl Acad Sci USA 91, 11393-11397 (1994).

20. de Kossodo S, Critico B and Grau GE. Eur J Immunol 24, 769-772 (1994).

21. Zhang F, zur Hausen A, Hoffman R, et al. Biol Chem Hoppe-Seyer 375, 249-254 (1994).

22. Han J, Brown T and Beutler B. J Exp Med 171, 465-475 (1990).

23. Probert L, Akassoglou K, Pasparakis M, et al. Proc Natl Acad Sci USA 92, 11294-11298 (1995).

24. Sanna PP, Weiss F, Samson ME, et al. Proc Natl Acad Sci USA 92, 272-275 (1995).

25. Lieberman AP, Pitha PM and Shin ML. J Exp Med 172, 989-992 (1990).

26. Krueger JM and Obál F Jr. FASEB J 7, 645-652 (1993).

27. Krueger JM, Obál F Jr, Opp M, et al. Yale J Biol Med 63, 157-172 (1990).

Figure Legend

Figure 1. TNFa bioactivity in various areas of the rat brain at different times of the

day. Each point represents an average value of 8 rats. The data are double

plotted. The direction of the results, high TNFa at the onset of sleep in the

hippocampus, cortex and hypothalamus (p < 0.05) are what one would

predict if TNFa is involved in sleep regulation. The black bars indicate the

dark period. Hippocampus (l), hypothalamus (n), cerebellum (!),

cortex (!), midbrain (u), and pons ( ).