Circadian Immune Measures in Healthy Volunteers: Relationship to
Hypothalamic-Pituitary-Adrenal Axis Hormones and Sympathetic
Neurotransmitters
ZIAD KRONFOL, MD, MADHAVAN NAIR, PHD, QUN ZHANG, PHD, ELIZABETH E. HILL, PHD,
AND MORTON B. BROWN, PHD
Objective: The purpose of this study is to examine the circadian pattern of specific immunologic measures and
to compare observed circadian rhythms of these measures with the well-established circadian rhythms of
hypothalamic-pituitary-adrenal axis hormones and sympathetic neurotransmitters.
Methods: Blood samples were collected every 2 hours for a total of 24 hours from nine healthy volunteers. The
blood samples were assayed for hormones and immune measures, including adrenocorticotropic hormone
(ACTH), cortisol, norepinephrine, and epinephrine. The immune measures included percentage and absolute
number of neutrophils, lymphocytes, the lymphocyte subsets CD3+ (T cells), CD4+ (T helper/inducer), CD8+
(T suppressor/cytotoxic), CD56+ (natural killer [NK] cells) and NK cell activity (NKCA).
Results: The following immune measures exhibited a significant circadian rhythm: the percentages of
neutrophils, CD4+ cells, and CD56+ cells; the absolute numbers of total lymphocytes, CD3+ cells, CD4+
cells, and CD8+ cells; and NKCA. Cross-correlations between the circadian rhythms of selected hormones and
immune measures indicated a strong inverse association between the circadian rhythms of cortisol and the
different T cell subsets on the one hand, and a strong direct association between the rhythms of cortisol and
the percentage of CD56+ and NKCA on the other. Cross-correlations involving the circadian rhythms of
norepinephrine and the same immune measures were in general much weaker and statistically nonsignificant.
Conclusion: In healthy individuals, both enumerative and functional immune measures exhibit circadian
rhythms that seem to be associated most closely with the circadian rhythm of cortisol.
Key words: circadian, cortisol, norepinephrine, T cells, NK cells, psychoneuroimmunology.
INTRODUCTION
It is now well established that the central nervous
system and the immune system are closely interre-
lated (1). Although mediators of such interactions
have not always been identified with certainty, the
hypothalamic-pituitary-adrenal (HPA) axis (2-4)
and the sympathetic nervous system (SNS) (5-7)
have often been assumed to play key roles. Studies
have shown that glucocorticoids (8) and cat-
echolamines (9), when added in vitro to cultures of
human lymphocytes, were associated with inhibi-
tion of various immunoregulatory activities. When
the same products were assessed in vivo, the results
varied. Corticosteroid administration resulted in an
increase in natural killer (NK) cell activity (10),
From the Departments of Psychiatry (Z.K., Q.Z., E.E.H.) and
Biostatistics (M.B.B.), University of Michigan Medical Center,
Ann Arbor, Michigan; and Department of Medicine, State Univer-
sity of New York at Buffalo, New York (M.N.).
Address reprint requests to: Ziad Kronfol, MD, Department of
Psychiatry, University of Michigan Medical Center, 1500 E. Med-
ical Center Drive, Ann Arbor, MI 48109-0722.
Received for publication March 29, 1995; revision received
February 12, 1996.
whereas epinephrine injections were accompanied
by a decrease in mitogen-induced lymphocyte pro-
liferation (11). Similarly, there were conflicting re-
sults regarding the effects of corticosteroids and
catecholamines on circulating lymphocyte subsets
(12,13). Part of the discrepancy may be related to the
dose of hormone given, the route of administration,
the duration of treatment, and the specific com-
pound being used (ie, natural or synthetic). Al-
though these studies of "pharmacologic" interven-
tion have helped advance our knowledge in this
area, they provided little information on day-to-day
interactions between HPA-axis hormones, cat-
echolamines, and specific immune measures under
"physiologic" conditions.
One research strategy to address this issue has
been the study of circadian rhythmicity. HPA-axis
hormones such as cortisol and adrenocorticotropic
hormone (ACTH) have well-established circadian
rhythms (14, 15). Similarly, although perhaps not as
widely known, plasma catecholamines show circa-
dian fluctuations (16, 17). Several studies have
therefore examined the circadian rhythmicity of spe-
cific immune measures in relation to HPA hor-
mones, especially cortisol. Ritchie et al. (18), for
instance, were able to document a circadian rhythm
42
0033-3174/97/5901 -0042S03.00/0
Copyright © 1997 by the American Psychosomatic Society
Psychosomatic Medicine 59:42-50 (1997)