Adrian J. Dunn*
Department of Pharmacology and Therapeutics, Louisiana State University Health Sciences Center. P.O. Box 33932. Shreveport. LA 71103-3932, USA
Abstract: The discovery that peripheral administration of interleukin-1 (IL-1) to rats potently activated the hypothalamo-pituitary-adrenocortical (HPA) axis, initiated a new understanding of the interactions between the immune system and the brain. It was proposed, and widely accepted that the increase in circulating concentrations of glucocorticoids provided a negative feedback to limit immune activity, and to prevent autoimmunity. Administration of certain other cytokines (interleukin-6, IL-6, tumor necrosis factor a, and certain interferons) also activates the HPA axis, but none is as potent or as effective as IL-1. This chapter reviews the evidence that cytokines activate the HPA axis, and the mechanisms by which they do so. Many of the cytokines appear to act via the brain, resulting in the activation of the corticotropin-releasing factor-containing neurons in the hypothalamic paraventricular nucleus. However, direct actions of cytokines on the anterior pituitary and on the adrenal cortex may also occur. Interestingly, some cytokines (most notably IL-1 and IL-6) appear to activate the HPA axis at multiple levels, and by multiple mechanisms. This suggests that the ability of IL-1 and IL-6 (and possibly other cytokines) to elevate circulating glucocorticoids may be critical to the survival of the organism.
The seminal discovery by Besedovsky (1986) that peripheral administration of a purified recombinant preparation of human interleukin-1 (IL-1) potently activated the hypothalamo -pituitary-adrenocortical (HPA) axis triggered a revolution in our understanding of the relationships between the nervous and immune systems. Besedovsky argued that because IL-1 was produced by various cell types early in the immune response, and adrenal glucocorticoids are known to inhibit immune system activity, this action of IL-1 could provide negative feedback to limit immune system activation, thereby limiting immune cell damage of tissues and autoimmunity. Figure 1 depicts these relationships. Activation of the HPA axis has long been associated with stress, and is
""Corresponding author. Tel.: + 1(318)675 7850; Fax: + 1(318)675 7857; E-mail: adunni®lsuhsc.edu considered by many physiologists to be the defining indicator of stress. Thus the effect of IL-1 suggests that activation of the HPA axis associated with immune system activation signals stress from the presence of tissue damage or pathogens. A similar concept had earlier led Blalock (1984) to suggest that the immune system can be regarded as a sixth sensory system, informing the central nervous system (CNS) of the presence in the body of unknown antigens, likely to be pathogens.
The year following Besedovsky's report, a trio of publications appeared in the journal Science, addressing the mechanism of action of IL-1 on the HPA axis. In one of these, Bernton et al. (1987) argued that IL-1 acted directly on the pituitary to stimulate ACTH release. However, Sapolsky et al. (1987) and Berkenbosch et al. (1987) presented compelling evidence that the mechanism of the effects of IL-1 involved the activation of corticotropin-releasing factor (CRF)-containing cells in the hypothalamus,
Fig. 1. Diagram of the relationship between the brain, the HPA axis, and immune cells. Interleukin-1 (IL-1) produced peripherally during immune responses activates the hypothalami pituitary-adrenocortical (HPA) axis. Release of corticotropin-releasing factor (CRF) occurs in the median eminence region of the hypothalamus and is secreted into the portal blood system. CRF then stimulates the secretion of ACTH from the anterior lobe of the pituitary. The ACTH is carried in peripheral blood to the adrenal cortex where it activates the synthesis and secretion of glucocorticoid hormones. The glucocorticoids provide a negative feedback on cytokine production by lymphocytes. IL-1 may act on circumventricular organs (CVOs) such as the median eminence or the OVLT, or may activate vagal afferents which in turn stimulate noradrenergic neurons in the brainstem which innervate the hypothalamus, specifically paraventricular neurons containing CRF. Interleukin-6 (IL-6) and tumor necrosis factor a (TNFa) can also activate the HPA axis, but the mechanism(s) are not established. (Reproduced from Dunn and Wang, 1999.)
in agreement with another report by Uehara et al. (1987b). Thus, the mechanism by which IL-1 activates the HPA axis was immediately controversial and has remained so. The probable reason for this is that there are multiple mechanisms by which IL-1 can activate the HPA axis. The relative importance of each mechanism depends upon the route of injection (or the site of IL-1 production) and also the dose. The existence of redundant mechanisms suggests that the phenomenon of IL-1-induced elevation of circulating glucocorticoids is important for the survival of the organism.
The involvement of hypothalamic CRF in the IL-1-induced activation of the HPA axis indicates that IL-1 uses the same mechanism (hypothalamic
CRF) believed to be involved in all HPA response in stress. This reinforces the validity of our current concepts of stress, and suggests that IL-1 is a mediator of "immune stress."
Subsequently, it was discovered that certain other cytokines, specifically interleukin-6 (IL-6), tumor necrosis factor-a (TNFa), and interferon-^ (IFNa), also had the ability to activate the HPA axis, although none of these cytokines was as potent or as effective as IL-1. This chapter will review the evidence for the effects of various cytokines on the HPA axis and the mechanisms involved. There are several classic reviews of cytokine activation of the HPA axis that should be consulted to provide comprehensive coverage of this area (Tilders et al, 1994; Besedovsky and del Rey, 1996; Turnbull and Rivier, 1999; Rivest, 2001), and a recent review by Silverman et al. (2003), as well as earlier publications of the present author (Dunn, 1990, 1993a, 2000; Dunn et al, 1999).
The hypothalamo-pituitary-adrenocortical (HPA) axis
The HPA axis is normally thought of as comprised of three hormones: CRF in the hypothalamus, adreno-corticotropin (ACTH) in the anterior pituitary (adenohypophysis), and the glucocorticoids (corti-costerone or Cortisol) in the adrenal cortex. These relationships are indicated in Fig. 2 of Chapter 1.3. An Introduction to the HPA Axis by Allison J. Fulford and Michael S. Harbuz. However, there are more than three hormones associated with the axis. At the hypothalamic level, it is well established that vasopressin synthesized in the hypothalamus and secreted from the neurohypophysis is an important releasing factor for ACTH, but the best evidence indicates that vasopressin is only effective in the presence of CRF (Antoni, 1993). The latter is supported by the lack of a response to vasopressin administration in CRF-knockout mice (Muglia et al, 2000). There are now four members of the CRF family, CRF itself, and the structurally related urocortins (urocortin 1 (Donaldson et al, 1996), urocortin 2 (Reyes et al, 2001) and urocortin 3 (Li et al, 2002). However, whereas there is good evidence that each of the urocortins binds to CRF receptors, there is little evidence that any of them is involved in the secretion of pituitary ACTH.
At the level of the pituitary, ACTH is synthesized by cleavage of the prohormone, proopiomelanocortin (POMC), along with a- and y-melanocyte-stimulating hormones (a-MSH and y-MSH), P-lipotropin and P-endorphin. Although the methionine enkephalin sequence is present in P-endorphin, POMC is not considered to be the natural precursor of met-enkephalin. Interestingly, high proportions of the P-endorphin and related peptides released from the pituitary are acetylated on the N-terminal (Akil et al, 1985). This is biologically significant, because N-acetylated P-endorphin (and related peptides) has very little affinity for opiate receptors, and thus lacks opioid function. However, receptors specific for A^-acetyl-P-endorphin have been identified on immune cells (Sharp and Linner, 1993), although their biological function has not been established. The physiological functions of the POMC-derived pituitary hormones other than ACTH are poorly understood. The glucocorticoid hormones, cortico-sterone, and Cortisol are both produced in the adrenal cortex of most species, but generally one glucocorticoid is predominant, e.g., corticosterone in rodents, and Cortisol in man, dogs, cats, and most large mammals.
The regulation of hypothalamic CRF secretion is only partially understood. It is generally accepted that adrenergic mechanisms (both noradrenaline (NA) and adrenaline) are important, and that a major stimulatory effect occurs via a,-adrenergic receptors (Al-Damluji, 1988; Plotsky et al, 1989). However, there is also evidence for the involvement of P-adrenergic receptors, perhaps inhibitory, although this has not been fully resolved (Al-Damluji, 1988; Plotsky et al, 1989; Saphier, 1989). There is also evidence for excitatory effects of cholinergic, GABAergic, and serotonergic (5-HT|A and 5-HTt) agonists (Plotsky et al, 1989; Saphier and Welch, 1994).
Before discussing the effects of specific cytokines, a few general points need to be made. First of all, it is important that the cytokines used for such studies be pure, and that they be free of endotoxin. Endotoxin (also known as lipopolysaccharide, LPS) is a common contaminant of recombinant preparations of cytokines because it is a breakdown product of the cell walls of Gram-negative bacteria, often used to synthesize recombinant cytokines. LPS has long been known to be a potent stimulator of the HPA axis (e.g., Chowers et al, 1966). Its administration induces the synthesis and secretion of cytokines, such as IL-1, IL-6, and TNFa, and possibly others. Nevertheless, it should not be assumed that all the biological effects of LPS are mediated by these cytokines. LPS acts on specific receptors (Toll-like receptor 4, TLR4) many of which are not coupled to the synthesis of cytokines (e.g., those on endothelial cells). Secondly, there is the question of species differences in cytokine structure. There are very substantial differences in the structures of cytokines from different species. Human IL-1 a and IL-1 (3 have only 22% sequence homology, and the sequences of IL-1 differ substantially among species. Nevertheless, IL-1 a and IL-1 p bind with relatively similar affinities to the IL-1 Type I and Type II receptors. Furthermore, the affinities of rat and mouse IL-ls for the human receptors are quite similar, and, in general, most forms of IL-1 are active in most species. However, this is not necessarily the case for other cytokines. In some cases, cytokines from one species are inactive in another. A classic example is TNFa. The amino acid sequence homology between mouse and human TNFa is 79% (Fransen et al, 1985). Mouse TNFa is a glycosylated dimer, whereas hTNFa is not glycosylated (Fransen et al, 1985; Sherry et al, 1990). Human TNFa does not bind to mouse type 2 TNF receptors (mTNF-R2, also known as p75), while mTNFa binds to both mTNF-Rl (p55) and mTNF-R2 (Lewis et al, 1991). Thus, hTNFa virtually lacks some actions of mouse TNFa in mice, such as antitumor (B16BL6) activity (Brouckaert et al, 1986) and lethality (Brouckaert et al, 1992). Another factor is glycosylation and other posttranslational modification of cytokines. As indicated above, TNFa may be glycosylated in one species, but not in another. Another example is IL-6; human IL-6 is glycosylated, whereas mouse IL-6 is not. The significance of these posttranslational modifications for function is poorly understood. Thus it is important to consider the form of the cytokine. For most purposes, it is preferable to study the actions of homologous cytokines.
Yet another factor is the route of injection and the dose. It should be obvious that both these factors will influence the amount or concentration of cytokine reaching a specific site, but this has not always been adequately taken into consideration. As will become clear when the mechanisms of the effects of IL-1 are discussed below, the mechanisms involved in HPA activation depend on the route of injection and the dose.
As indicated above, multiple mechanisms exist for the HPA-activating effect of IL-1, and the details are still incompletely resolved. It is likely that IL-1 acts at all three levels of the HPA axis, the brain, the hypothalamus, the anterior pituitary, and the adrenal cortex. The most important mechanism will vary with the route of administration, or the site of its production, and with the dose, and probably, with the physiological state of the animal. However, the preponderance of the evidence suggests that under normal physiological circumstances, the principal action of IL-1 on the HPA axis involves hypothalamic CRF.
An early study showed that intraperitoneal (ip) administration of a very high dose of IL-1 (70|xg) increased plasma corticosterone in rats (Roh et al., 1987). A direct action was suggested because in this study, 3.5-35 (ig of IL-1 increased corticosterone output by perfused rat adrenals in vitro. Some subsequent studies have shown that IL-1 stimulates the secretion of Cortisol from bovine adrenal cells (Winter et al., 1990) and corticosterone from rat adrenal cortex in vitro (Gwosdow et al., 1990; Andreis et al., 1991), but several other studies did not find such effects of IL-1 on adrenocortical cells in vitro (Harlin and Parker, 1991; Cambronero et al., 1992). The nature of the preparation may be critical, because Gwosdow et al. (1990) observed no effect of
IL-1 on cultured adrenocortical cells, but IL-1 was effective in hemisected adrenals. A direct effect on the adrenal cortex is unlikely to explain completely the normal in vivo elevation of plasma concentrations of corticosterone, because IL-1 administration also elevates plasma ACTH in both rats and mice (Besedovsky et al., 1986; Dunn, 1993a). Also, IL-1 failed to induce increases in plasma corticosterone in hypophysectomized rats (Gwosdow et al., 1990; Olsen et al., 1992) and mice (Dunn, 1993a), although a very modest increase in ACTH and corticosterone with high doses of IL-1 p was detected in one study in rats (Andreis et al., 1991). Moreover, the ACTH and corticosterone responses to IL-1 were largely prevented by in vivo pretreatment with an antibody to CRF in rats (Berkenbosch et al., 1987; Sapolsky et al., 1987; Uehara et al., 1987b) and mice (Dunn, 1993a). Consistent with this, mice lacking the gene for CRF exhibited minimal increases in plasma corticosterone in response to IL-1 (Dunn and Swiergiel, 1999).
A critical role for the pituitary in the HPA response to IL-1 was indicated by the effect of hypophysect-omy which prevented the ACTH and corticosterone responses to IL-1 (see above). A direct pituitary effect appears to be excluded as the primary physiological mechanism because lesions of the PVN largely prevented the ACTH and corticosterone responses to IL-1 (Rivest and Rivier, 1991; Kovacs and Elenkov, 1995), although Kovacs and Elenkov (1995) indicated that the block was not complete. Also, as indicated above, pretreatment with antibody to CRF prevented the IL-1-induced increases in plasma ACTH and glucocorticoids.
Nevertheless, a number of in vitro studies have indicated that IL-1 has the ability to elicit ACTH secretion from pituitary cells. Some early reports indicated that IL-1 induced secretion of ACTH (Woloski et al., 1985; Fukata et al., 1989) and (3-endorphin (Fagarasan et al., 1989) from AtT20 cells, although Sapolsky et al. (1987) found no such effect. Interestingly, in two of the studies, prolonged incubations of the cells were necessary to observe such effects (Fukata et al., 1989; Fagarasan et al.,
1990). AtT-20 cells are a tumor line, so that the mechanisms regulating the ACTH release may not be the same as in normal pituitaries. Nevertheless, several reports have indicated that IL-1 stimulates ACTH release from primary cultures of anterior pituitary cells in vitro (Bernton et al., 1987; Kehrer et al., 1987; Suda et al., 1989), although several other investigators failed to find such effects (Sapolsky et al., 1987; Uehara et al., 1987a; McGillis et al., 1988; Tsagarakis et al., 1989). It is important to bear in mind that in vitro studies cannot provide definitive answers regarding in vivo mechanisms.
It is interesting that most of the positive results required prolonged in vitro incubations, such that IL-1-induced ACTH secretion was not observed in the first several hours (Kehrer et al., 1988; Suda et al., 1989). It has been reported that prolonged incubation appears to increase the sensitivity of pituitary ACTH secretion to IL-1, while decreasing the response to CRF (Suda et al., 1989). It is intriguing that prolonged incubation of both adrenocortical and adenohypophyseal cells increases their sensitivity to IL-1. It is possible that the ability of IL-1 to elevate circulating glucocorticoids is so important for the organism that when higher components of the HPA axis fail to function properly, downstream organs can gain that function.
The hypothalamus appears to be important for the elevations of plasma ACTH and corticosterone induced by peripherally administered IL-1 because complete mediobasal hypothalamic deafferentation prevented the HPA response to intraperitoneally injected IL-1 (Ovadia et al., 1989). This conclusion was supported by the observation that lesions of the PVN prevented the increases in plasma ACTH and corticosterone induced by ip IL-1 (Rivest and Rivier, 1991; Kovacs and Elenkov, 1995). Also, IL-1 increased the electrophysiological activity of CRF neurons in vivo (Saphier, 1989; Besedovsky et al.,
1991). In support of a hypothalamic involvement, IL-1 stimulated CRF release from hypothalamic slabs in vitro (Tsagarakis et al., 1989), although such in vitro studies cannot be definitive. CRF is implicated by the observation that peripherally administered IL-1 elevates concentrations of CRF in portal blood (Sapolsky et al., 1987), and depletes CRF from the median eminence (Berkenbosch et al., 1987), both presumably reflecting increased release of CRF. Also, immunoneutralization of CRF in rats prevented the increases in plasma ACTH and corticosterone induced by IL-1 in rats (Berkenbosch et al., 1987; Sapolsky et al., 1987; Uehara et al., 1987b) and mice (Dunn, 1993a). Moreover, CRF-knockout mice (mice lacking the gene for CRF and thus unable to produce it) showed only a minuscule increase in plasma corticosterone after IL-1 administration (Dunn and Swiergiel, 1999).
The evidence for direct actions on the pituitary and adrenal glands derives largely from in vitro experiments and is therefore susceptible to artifact. The in vivo evidence summarized above strongly favors a role for hypothalamic CRF as the major mechanism for the action of peripheral IL-1 in normal healthy animals. However, in studies in mice treated with antibody to CRF, there were small increases in plasma corticosterone following intraperitoneal IL-1 (Dunn, 1993a), and similar small, but statistically significant, increases were also observed in CRF-knockout mice (Dunn and Swiergiel, 1999). This suggests that when the functions of higher levels of the HPA axis are impaired, the pituitary and/or adrenal cortex may gain the ability to respond to IL-1 and mount a modest glucocorticoid response. This may have pathological significance in that a glucocorticoid response may be conserved when the pituitary or hypothalamus is unable to (or fails to) respond. It should also be noted that IL-1- and LPS-induced secretion of ACTH and corticosterone can be observed in young rats at a time when HPA responses to stressors are minimal or absent (Levine et al., 1994). Thus in ontogeny, the ability of IL-1 to induce glucocorticoid secretion appears very early.
There is little evidence for a role of arginine vasopressin (AVP) in the HPA activation by IL-1. Most in vivo studies have found no effect of peripherally administered IL-1 on AVP secretion. Berkenbosch et al. (1989) found no evidence for increases in AVP turnover in the median eminence region following IL-1 (3 at a dose that maximally activated the HPA axis, even though CRF turnover was increased. However, using push-pull perfusion, Watanobe and Takebe (1994) observed increases in
CRF and AVP release from the median eminence and the PVN when IL-ip was injected iv. Harbuz et al. (1996) found no effect of peripheral or central injection of IL-1 (3 on plasma concentrations of AVP at doses that stimulated the HPA axis, as determined by increases in plasma corticosterone. However, in humans, Mastorakos et al. (1994) found that relatively high doses of IL-6 increased plasma AVP, along with ACTH.
In vitro, Nakatsuru et al. (1991) using a super-perfused hypothalamo-neurohypophyseal complex observed AVP secretion with low doses of hIL-la and hIL-ip (O.l-lOnM), but not at higher doses (100 nM). However, Spinedi et al. (1992) failed to find any effect of IL-1 P on AVP secretion from the excised medial basal hypothalamus, although an increase in CRF was observed. Nevertheless, Yasin et al. (1994) observed increased release of AVP and oxytocin from hypothalamic explants in response to IL-1, and to IL-6 at higher doses. These effects were prevented by the cyclooxygenase (COX) inhibitors, indomethacin and ibuprofen, but not by the lipoxygenase inhibitor, BW A4C. However, Zelazowski et al. (1993) observed decreases in AVP secretion from hypothalamic explants in response to IL-1 p.
Some effects have been observed following intracerebral administration of cytokines. Landgraf et al. (1995) observed increases in AVP and oxytocin in blood microdialysates after 200 ng IL-1 p was injected intracerebroventricularly (icv), and evidence for AVP release from the SON, but not the PVN, was obtained when IL-1 P was injected directly into those structures. Wilkinson et al. (1994) observed increases of AVP secretion from the bed nucleus of stria terminalis, but it is not known whether this altered HPA activity.
Routes by which IL-1 acts on the brain to activate the HPA axis
Cytokines are relatively large molecules, large enough that they will not readily penetrate the blood-brain barrier. Thus, it should not be assumed that the action of IL-1 is exerted directly on the hypothalamus even though intrahypothalamic injections of IL-1 can activate the HPA axis. So how do cytokines induce their effects on the brain? The answer is complex, because there are multiple mechanisms by which cytokines can affect the activity of the brain, some of which do not require cytokine penetration of the brain (see reviews by Ericsson et al, 1996; Dunn, 2002).
First of all, cytokines can act on brain cells at sites where there is no blood-brain barrier, specifically the circumventricular organs (CVOs). There is some evidence that IL-1 may act on the median eminence (Turnbull and Rivier, 1999), on the organum vasculosum laminae terminalis (OVLT), the preoptic area (Katsuura et al, 1990; Blatteis and Sehic, 1997), and the area postrema (Ericsson et al, 1996; Turnbull and Rivier, 1999; Dunn, 2002). Some of these regions are located in the hypothalamus (the OVLT and preoptic area), and others have direct connections to the hypothalamus. Thus, cytokines may be able to exert relatively direct effects on the PVN.
Second, cytokines can be transported into the brain to a limited extent using selective uptake systems (transporters), thus bypassing the blood-brain barrier (Banks et al, 1995). The capacity of these systems is quite limited, and their significance is unclear. The anatomical distribution of the uptake sites has revealed little, but clearly they may be important for certain specific functions (see, for example, Banks et al, 2001).
Third, cytokines may act directly or indirectly on peripheral nerves that send afferent signals to the brain. The hypothalamus can also be activated indirectly, for example by the vagus nerve. The vagus contains afferent neurons that project to the brainstem, and which can activate cell bodies of neurons that project to the hypothalamus, for example, noradrenergic neurons in the nucleus tractus solitar-ius. Numerous studies have indicated that IL-1 (and LPS) can signal the brain by activating such afferents, because lesions of the vagus nerve can prevent various physiological and behavioral responses to intraper-itoneally injected IL-1 (Watkins et al, 1995). Such lesions also affect HPA responses to IL-1 (Fleshner et al, 1995) and to TNFa (Fleshner et al, 1998).
Fourth, cytokines can act on peripheral tissues inducing the synthesis of molecules whose ability to penetrate the brain is not limited by the barrier. A major target appears to be brain endothelial cells which bear receptors for IL-1 (and LPS). Systemic treatments with LPS and IL-1 induce the inducible form of cyclooxygenase, COX2, in the endothelium (Cao et al., 1997; Quan et al., 1998). COX2 activation may result in the production of prostaglandin E2 (PGE2) which, because it is a lipid, can pass freely across the blood-brain barrier. PGE2 can induce fever in the anterior hypothalamus (Blatteis and Sehic, 1997), and can activate PVN-CRF cells, and thus the HPA axis (Rivest, 2001).
Fifth, cytokines can be synthesized by immune cells that infiltrate the brain. It is well established that peripheral LPS administration initiates a process that results in macrophages invading the brain (appearing as microglia), which appear to migrate through the brain parenchyma and appear to contain IL-1 (van Dam et al., 1995; Quan et al., 1999).
Something of the mechanism of IL-1-induced HPA activation has been revealed by the use of pharmacological antagonists. Below, the evidence for the involvement of COX, and of NA, is reviewed and some negative findings are summarized.
The cyclooxygenase (COX) involvement
It has long been known that COX is involved in the IL-1-induced activation of the HPA axis. Several early studies indicated that various COX inhibitors (the so-called nonsteroidal anti-inflammatory drugs, NSAIDS) inhibited the elevation of plasma ACTH and corticosterone following IL-1 administration. Thus, we were surprised when we failed to observe an inhibition by indomethacin of the elevation of plasma corticosterone when we injected IL-1 into mice intraperitoneally. A review of the literature indicated that all the positive reports had used iv injection of IL-1. Thus, we tested several different COX inhibitors on the plasma corticosterone responses to both iv and ip IL-1. The results showed clearly that when IL-1 (3 was injected iv, indomethacin blocked the increase in plasma corticosterone, whereas there was no inhibition when the IL-1 was injected ip (Dunn and Chuluyan, 1992). Because the peak response to iv IL-1 in mice occurred around 40min, compared to 120min with ip IL-1, we tested the response at different times. The results showed that the early phase of the response to ip IL-1 was inhibited by COX inhibitors, whereas the later phase was not. This was the first clear evidence that the effect of IL-1
depended upon its route of injection, and that more than one mechanism was involved in the HPA responses to peripherally administered IL-1.
There is substantial evidence for the involvement of brain noradrenergic systems in the IL-1-induced activation of the HPA axis. Peripheral administration of IL-lp activates brain noradrenergic neurons, especially in the hypothalamus (Dunn, 1988; Kabiersch et al., 1988). They may be activated in the nucleus tractus solitarius of the brainstem, the site of origin of ascending noradrenergic neurons that innervate the hypothalamus, including the PVN (Plotsky et al., 1989). The activation may be local via the area postrema, or indirectly via vagal afferents from the periphery. Over a very large number of experiments, we have observed a very high correlation between the increases of noradrenergic activity in the mouse hypothalamus (determined by increases in MHPG, the major catabolite of NA), and HPA activation induced by IL-1 (and other agents, such as LPS, Newcastle disease virus, and influenza virus) (Dunn et al., 1999). We have also observed similar close correlations between hypothalamic NA release and plasma corticosterone following iv and ip injection of IL-1 into freely moving rats implanted with in vivo microdialysis probes in the medial hypothalamus, and from which blood samples were collected from iv catheters (Smagin et al., 1996; Wieczorek and Dunn, 2003). Such a relationship between hypothalamic NA and the HPA axis fits well with the aforementioned evidence for noradrenergic activation of PVN-CRF neurons. The fact that COX inhibitors prevent the noradrenergic response to IL-1 (Dunn and Chuluyan, 1992; Wieczorek and Dunn, 2003) bolsters the argument that this noradrenergic activation drives the HPA activation.
Owing to the coactivation of hypothalamic NA release and the HPA axis, we and others tested the effects of adrenergic antagonists. In rats, Rivier (1995b) failed to find any effect of the (3-adrenergic antagonist, propranolol, or the aradrenergic antagonist, prazosin, or the combination of the two drugs on the HPA activation by IL-1 (confirmed by
Besedovsky and del Rey, personal communication). In mice, we observed no effect of propranolol at any dose, but did observe partial inhibition of the plasma corticosterone response at high doses (1 mg/kg) of prazosin (Dunn et al, 1999). The effect of prazosin was not enhanced by the addition of propranolol. However, when we lesioned the ventral noradrenergic ascending bundle or the PVN of rats with 6-hydroxy-dopamine (6-OHDA), the IL-1-induced increase in plasma corticosterone was markedly decreased when the PVN depletions of NA exceeded 70% (Chuluyan et al, 1992). This result was consistent with earlier observations of Weidenfeld et al. (1989), who noted that the HPA response to icv IL-1 (but not to restraint) was prevented by VNAB lesions with 6-OHDA or prazosin, but not by propranolol. Curiously however, in mice, depletion of whole brain NA by 96% or more, failed to decrease the plasma corticosterone response to ip IL-1 p (Swiergiel et al, 1996). (A small statistically significant decrease was observed in two of six experiments.)
We have very recently tested this relationship further, by injecting IL-1 p ip into rats with microdialysis probes in the medial hypothalamus, and intravenous catheters for sampling plasma (Wieczorek and Dunn, 2003, and in preparation). Such rats exhibit increased NA release over a period of about 3h after the ip IL-1, and parallel increases in plasma ACTH and corticosterone. However, pretreatment with indomethacin prevented the increases in body temperature and in dialysate NA, with rather modest reductions of the increases in plasma ACTH and corticosterone. Thus the noradrenergic and the HPA responses can be dissociated, and the noradrenergic activation does not appear to be essential for the HPA activation.
It is also notable that a subdiaphragmatic vagotomy which lesions the vagal afferents that project to the nucleus tractus solitarius also inhibits the decrease in hypothalamic NA observed in response to IL-1 (Fleshner et al, 1995). Others have shown that a similar vagotomy largely prevented the ACTH and corticosterone response to IL-1 (Gaykema et al, 1995; Kapcala et al, 1996). Very recently, we have shown that subdiaphragmatic vagotomy in rats prevented the IL-1-induced increase in medial hypothalamic NA, while also largely inhibiting the increase in plasma corticosterone (Wieczorek and Dunn, unpublished observations).
We have observed no inhibition of the HPA response to ip IL-1 P in mice by the nonselective nitric oxide (NO) synthase (NOS) inhibitor, l-co-nitro-1-arginine methylester (l-NAME), as well as the iNOS-selective inhibitors, A^l-iminoethyljlysine (l-NIL), 2-amino-5,6-dihydro-6-methyl-4/7-1,3-thiazine (AMT), 0N01714, 1400W, CN256 and CN257, and the nNOS-selective inhibitor, 7-nitroindazole (7-NI) (Dunn, 1993b, and unpublished observations). Moreover, the responses to IL-1 and LPS in nNOS-, iNOS-and e-NOS-knockout mice were normal. However, Rivier (1995a) found that NOS inhibitors actually enhance the response to ip IL-1 in rats, and that this effect occurred when NOS inhibitors were administered locally in the PVN.
No impairment of the HPA response to ip IL-1 has been observed with the cholinergic antagonist, scopolamine, the ganglionic blocker, chlorisond-amine, the histamine antagonists, pyrilamine (Hi), and cimetidine (H2) (see also Perlstein et al, 1994), the 5-HT2 antagonist, cinanserin, the opiate antagonist, naloxone, the lipoxygenase inhibitor, BWA4C, the NPY-Y1 antagonist, BIBP3226, or the NK1-receptor antagonists, L703,606 and 733,060, the NK2 antagonist, L659,877, and a platelet-activating factor (PAF) antagonist (Dunn, unpublished observations).
Some early reports indicated that IL-2 elicited ACTH secretion from the pituitary cells in vitro (Smith et al, 1989; Karanth and McCann, 1991). Karanth and McCann (1991) tested hIL-2 on rat hemipituitaries at doses from 10"17 to 10~9M and reported increased ACTH secretion at 10"11 and 10"12M at 1 h, but not at 2h. Fukata et al. (1989) found no effect of hIL-2 on AtT20 cells. It was also reported that rat IL-2 elevated plasma concentrations of ACTH in the rat, but human IL-2 did not (Naito et al, 1989). Acute icv administration of IL-2 (500 ng) elevated plasma corticosterone in rats (Pauli et al, 1998), while 14 days of icv IL-2 administration elevated plasma
ACTH and corticosterone (Hanisch et al, 1996). However, subsequent studies have failed to find any such effect of IL-2 on plasma ACTH and corticosterone in mice (Lacosta et al, 2000). Thus, IL-2 probably has some effects on the HPA axis, but it is a much less effective activator of the axis than IL-1, IL-6, and TNFa, and its effects may be indirect.
Interleukin-6 (IL-6) has long been known to have HPA-activating activity. Peripheral administration of human IL-6 increased plasma concentrations of ACTH and corticosterone in rats (Naitoh et al, 1988; Matta et al, 1992), mice (Perlstein et al, 1991; Wang and Dunn, 1998), and man (Mastorakos et al, 1993; Stouthard et al, 1995; Spath-Schwalbe et al, 1998), and mouse IL-6 was effective in rats (Matta et al, 1992) and mice (Wang and Dunn, 1998). Kovács and Elenkov (1995) found that the ACTH response to IL-6 in rats was delayed (about 1 h), but in mice the plasma ACTH and corticosterone responses to iv and ip IL-6 were fast and short-lived compared to those for IL-1 (Wang and Dunn, 1998). The short-lived effects of IL-6 on the HPA axis are consistent with pharmacokinetic studies of IL-6 which indicate a plasma half-life after iv injection of around 3 min in rats and 7 min in mice (Mulé et al, 1990; Bocci, 1991). mIL-6 and hIL-6 were significantly less potent in activating the HPA axis in mice (Dunn, 1992; Wang and Dunn, 1998; Silverman et al, 2003) than reported for hIL-6 or mIL-6 in rats (Naitoh et al, 1988; Lyson et al, 1991; Besedovsky and del Rey, 1992; Matta et al, 1992), or in man, in which plasma Cortisol is readily elevated by relatively small doses of IL-6 (Mastorakis et al, 1993). IL-6 is also significantly less potent and effective in activating the HPA axis than IL-1. In rats and mice, doses of IL-6 more than an order of magnitude higher than IL-1 were required to induce equivalent effects (Besedovsky and del Rey, 1991; Perlstein et al, 1991; Wang and Dunn, 1998; Silverman et al, 2003). Moreover, the maximum response is significantly less than that elicited by IL-1. Even doses as high as 2 or 5 ng per mouse failed to induce maximal plasma concentrations of corticosterone (Wang and Dunn,
1998), suggesting that the mechanisms of the responses may differ.
As is in the case of IL-1, the mechanism of action of IL-6 on the HPA axis may be complex. PVN lesions completely prevented the ACTH response to iv IL-6 in rats (Kovacs and Elenkov, 1995). Matta et al. (1992) showed that third ventricle infusion of IL-6 also elevated plasma ACTH. The HPA-activating effect of IL-6 in rats has been reported to be sensitive to antibodies to CRF (Naitoh et al, 1988; van der Meer et al, 1996; Ando et al, 1998). These results all suggest that, like IL-1, IL-6 works through hypothalamic CRF, although IL-1 (3 administration increased CRF mRNA in the PVN, but IL-6 did not (Harbuz et al, 1992). However, in CRF-knockout mice, IL-6 induced an exaggerated response in plasma ACTH, although corticosterone was less affected (Bethin et al, 2000). Mice that lacked the genes for both CRF and IL-6 showed a minimal response to restraint and to LPS (Bethin et al, 2000; Muglia et al, 2000). This observation led these authors to suggest that IL-6 can substitute for CRF to activate the HPA axis by immune stimuli. In this respect it is interesting that IL-1 can induce IL-6 secretion from the anterior pituitary (Spangelo et al, 1991). This effect of IL-1 was sensitive to lipoxygenase inhibitors, but not to COX inhibitors (Spangelo et al, 1991). IL-6 has been reported to have a direct stimulatory effect on CRF secretion from the median eminence (Spinedi et al, 1992) and on ACTH secretion from the pituitary (Lyson et al, 1991; Matt et al, 1992). In contrast with IL-1, IL-6 stimulates Fos expression in the pituitary gland of rats, but not in the PVN (Callahan and Piekut, 1997). These results suggest that IL-6 may act directly on the anterior pituitary to elevate plasma ACTH and corticosterone, consistent with the rapid time course.
IL-6 has also been reported to stimulate corticosterone secretion in cultured adrenocortical cells from various species (Salas et al, 1990; Weber et al, 1997; Barney et al, 2000; Path et al, 2000). In all cases, prolonged incubation was necessary to observe significant effects, suggesting that inactivity may induce sensitivity to cytokines, as discussed for IL-1 (see above).
The HPA response to IL-6 is not sensitive to COX inhibitors, such as indomethacin (Wang and Dunn, 1998). These cytokines are "promiscuous" in that administration of one cytokine can stimulate the synthesis and secretion of others. Thus, IL-1 administration induces IL-6, so that IL-6 could contribute to the HPA response to IL-1. Pretreatment of mice, with a neutralizing monoclonal antibody to mIL-6, indicated that IL-6 contributed to the late phase of the plasma ACTH and corticosterone responses to mIL-ip (the ACTH response was not attenuated at 2h, but was 4 h after IL-1). However, the antibody to IL-6 did not alter the tryptophan or 5-hydroxyindoleacetic acid (5-HIAA) responses to IL-1 (Wang and Dunn, 1999). The antibody had similar effects on the responses to LPS, except that, in this case, the tryptophan and 5-HIAA responses were attenuated (Wang and Dunn, 1999). IL-1 does not appear to be involved in the responses to IL-6, because iv infusion of the IL-1-receptor antagonist (IL-lra) which effectively reduced the plasma corticosterone response to IL-1, had no effect on the response to iv IL-6 (van der Meer et al., 1996).
The results reviewed above suggest that IL-6 may be able to act at all three levels of the HPA axis, the hypothalamus, the pituitary, and the adrenal cortex. As in the case of IL-1, the pituitary and/or the adrenal cortex appears to be able to develop the ability to respond to IL-6 in the absence of stimulation from the next higher level. HPA responses to immune stimuli are intact in the absence of CRF or CRF1 receptors. For example, increases in plasma ACTH and corticosterone occur in response to LPS in CRF-knockout (Karalis et al., 1997; Bethin et al., 2000) and CRF 1-knockout mice (Turnbull et al., 1999), to turpentine in CRF 1-knockout mice (Turnbull et al., 1999), and to carrageenin in CRF-knockout mice (Karalis et al., 1997). Thus, IL-6 may be an important mediator of ACTH and glucocorticoid responses in the absence of CRF, especially in response to immune stimuli (Bethin et al., 2000; Silverman et al., 2003).
IL-10 may also play a role in the regulation of the HPA axis. IL-10 administration increased CRF concentrations in the median eminence, and elevated plasma ACTH and corticosterone (Di Santo et al., 1995). IL-10 is regarded as a specific anti-inflammatory cytokine, but it is not clear to what extent these
HPA axis effects contribute to the anti-inflammatory effects (Smith et al., 1999).
Lissoni et al. (1997) reported that IL-12 administration to renal cancer patients increased plasma Cortisol. We are not aware of other reports of HPA activation by IL-12 in animals or man.
Tumor necrosis factor a (TNFa)
Peripheral administration of TNFa to rats at doses that failed to affect blood pressure, food consumption, or plasma prolactin concentrations, caused significant elevations of plasma ACTH within 20 min (Darling et al., 1989; Sharp et al., 1989; Besedovsky and del Rey, 1991). Most reported studies have found TNFa, like IL-6, to be significantly less potent in activating the HPA axis than IL-1 in rats (Darling et al., 1989; Bernardini et al., 1990; Besedovsky et al., 1991) and mice (Dunn, 1992; Ando and Dunn, 1999), although Sharp et al. (1989) found human TNFa iv to be almost equipotent with human IL-lß in rats.
Although TNFa can induce the synthesis of other proinflammatory cytokines, the corticosterone response to iv TNFa was not altered by peripheral administration of IL-lra (van der Meer et al., 1996). Activation of the HPA axis by peripheral TNFa was abolished by lesions of the PVN in rats (Koväcs and Elenkov, 1995). Treatment with a CRF antibody also blocked the ACTH response (Bernardini et al., 1990), whereas the corticosterone response was only partially inhibited (Bernardini et al., 1990; van der Meer et al., 1996). These findings implicate hypothalamic CRF in the HPA response to TNFa. However, TNFa injected icv failed to elevate plasma corticosterone in two studies (Sharp et al., 1989; van der Meer et al., 1996), while a third study found a modest increase. Moreover, TNFa was able to elicit Cortisol secretion from human adrenocortical cells (Darling et al., 1989), suggesting that the adrenal cortex has (or can gain) the ability to respond to this cytokine as well as IL-1 and IL-6. However, other studies have found that TNFa inhibited glucocorticoid secretion from the adrenal cortex (van der Meer et al., 1996; Barney et al., 2000). Indomethacin dose dependency blocked the ACTH response to TNFa in rats (Sharp and Matta, 1993).
Several reports have indicated that interferon-a elevated plasma ACTH and Cortisol in man (Miiller et al., 1991; Gisslinger et al., 1993; Corssmit et al., 2000; Cassidy et al., 2002), but data from the rat are conflicting. Whereas, Menzies et al. (1996) found excitatory effects, Saphier et al. (1993) found that IFN-a had inhibitory effects that were mediated by (i-opioid receptors. Administration of human or mouse interferon-a (IFNa) to mice IP failed to alter plasma ACTH or corticosterone (Leenhouwers, Crnic, and Dunn, unpublished observations; see Dunn et al., 1999). IFN-a administration also elevates plasma IL-6 in man, so that IL-6 may mediate the elevation of plasma Cortisol (Cassidy et al., 2002).
IFN-y has been reported to elevate Cortisol, but not ACTH in man (Krishnan et al., 1987), although higher doses did elevate ACTH (Holsboer et al., 1988).
Leukemia inhibitory factor (LIF)
Leukemia inhibitory factor has been found in the hypothalamus and the anterior pituitary (Chesnokova et al., 2000). It can stimulate ACTH secretion from the pituitary in vivo and in vitro in mice and men. LIF appears to play a role in basal ACTH secretion (Reichlin, 1998). LIF-knockout mice show low-plasma ACTH and impaired HPA responses to stress and immune stimuli (Chesnokova et al., 1998), and these deficits can be reversed by LIF administration. LIF appears to be involved in the mediation of HPA responses to inflammation induced by Freund's adjuvant and turpentine because LIF-knockout mice show markedly diminished responses (Chesnokova et al., 2000). The relationships between LIF and IL-6 in this respect are not clear.
Synergism and antagonism among cytokines activating the HPA axis
There have been several reports of synergistic effects of cytokine activation of the HPA axis. Among the earliest was Perlstein et al. (1991, 1993) who proposed synergism between hIL-la and hIL-6 in activating the HPA axis. Another group failed to find synergism between hIL-lp and hIL-6, but did report synergism between hIL-ip and hTNFa (Brebner et al., 2000). Unfortunately, these claims of synergism have been made on the basis of single doses of each of the cytokines. Pharmacologists know very well that studies using such limited ranges of doses can be misleading, and that an isobolographic analysis involving multiple doses of each of the components is necessary to establish true synergism (Tallarida et al., 1989). Thus, assessment of the existence of such synergistic effects of the cytokines awaits more thorough analyses.
a-MSH appears to have the ability to antagonize IL-1-induced activation of the HPA axis. Daynes et al. (1987) showed that iv injection of 30 (ig a-MSH markedly reduced the corticosterone response to 36ng of hIL-ip in mice. In rats, icv a-MSH (10ng) abrogated the effects of icv IL-1 on plasma ACTH and corticosterone (Weiss et al., 1991). In mice, subcutaneous a-MSH (10 or 30 |xg) dose dependently attenuated the ACTH response to hIL-ip, but not to hIL-la (Rivier et al., 1989). Similar effects of a-MSH were observed on the responses to LPS. Shalts et al. (1992) reported that icv infusion of a-MSH (60|ig/h) inhibited the ACTH response to icv ILla (4.2 (xg/ 30min), in ovariectomized rhesus monkeys. However, a-MSH did not alter the response to CRF, suggesting that a-MSH acted at a level above the pituitary. A physiological role for a-MSH in this respect was indicated by the observation that icv infusion of an antibody to a-MSH enhanced the ACTH and corticosterone responses to icv hIL-ip (2ng, but not 20 ng) in rats (Papadopoulos and Wardlaw, 1999).
The biological significance of the activating effects of cytokines on the HPA axis
As indicated in the Introduction, Besedovsky et al. (1986) immediately appreciated the significance of IL-l's ability to activate the HPA axis. Knowing very well the ability of glucocorticoids to inhibit immune responses, they postulated that this would provide negative feedback on immune responses, limiting the immune activation, important to prevent over-active immune responses that might result in damaging autoimmune responses. This attractive hypothesis has been widely accepted by most investigators in the field. Glucocorticoids also have the ability to inhibit cytokine synthesis (Bertini et al, 1988), so that the HPA activation also provides negative feedback on the synthesis of the cytokines themselves. However, this is not the only mechanism by which glucocorticoids inhibit immune system function.
The ability of IL-1 (and perhaps that of other cytokines) to activate the HPA axis provides a mechanism for the activation of a classical physiological stress response to environmental threats recognized by the immune system. Thus the immune system may indeed function as a sixth sensory system, as proposed by Blalock (1984). The stress response acts by diverting energy and resources to organs and systems that need to address the environmental threat, and focuses the attention of the brain on the sources of the stress. It may be that the ability of the glucocorticoids to restrain the immune system focuses its activities in areas of concern, such as local infections and tissue damage, thus conserving its resources for the critical pathologies.
IL-1 may be the major messenger signalling threats to the organism detected by the immune system. It is interesting in this respect that IL-6, whose expression frequently appears along with IL-1, may have the ability to function as a CRF (see above). It is striking that in mice that lack the gene for CRF, IL-6 appears to assume the functions of CRF, and that CRF-knockout mice are hyper-responsive to immune challenges (Karalis et al, 1997). Thus, it appears that the IL-1-IL-6-CRF system is very important in host defense.
It is clear that several cytokines have the ability to affect the HPA axis, most of them causing an activation, resulting in elevated plasma concentrations of ACTH and corticosterone. However to date, no cytokine has proven to be as potent or effective as IL-1. It is particularly interesting that the cytokines whose HPA-activating activity is best known (i.e., IL-1, IL-6, and TNFa) all appear to have the ability to act at multiple levels of the axis: the hypothalamus, the anterior pituitary, and the adrenal cortex. In many cases, significant activation at the pituitary and adrenocortical levels follows a compromised activity at higher levels, suggesting that the ability to respond at the pituitary and adrenocortical levels may be induced when the activity at the hypothalamic or pituitary level is ineffective. Such adaptations suggest that the ability of cytokines to elevate circulating concentrations of glucocorticoids is critical to the survival of the organism. It is relevant that the ability of IL-1 (and perhaps other cytokines) to stimulate adrenocortical secretions appears developmentally before ACTH, and hence HPA axis function, which may enable coping with environmental and other perceived stressors. Thus it is reasonable to speculate that the ability to induce corticosteroid production in response to environmental threats that involve immune activation, and hence cytokine production, is especially important early in life.
Besedovsky's hypothesis that the IL-1-induced induction of glucocorticoids is important to provide negative feedback to the immune system is a compelling one. Clearly, a hyperactive immune response has the potential to damage the organism, especially during development when immune memory is poorly developed. The axis may also assist the immune system in identifying self from nonself, and in preventing autoimmune responses in the adult. Presumably effective attack on invading pathogens, and on cancer cells requires a sophisticated and delicate balance of the systems involved. The elevation of glucocorticoids by IL-1, IL-6, and TNFa provides feedback limiting inflammatory responses. However, because in many situations, these cytokines are produced locally, the systemic elevation of glucocorticoids may serve to limit inflammatory responses by confining them to the damaged or infected sites in the body.
Another interesting aspect of these effects of cytokines is the extent to which the major (so-called) proinflammatory cytokines (IL-1, IL-6, and TNFot) have similar and redundant activities. IL-1 induces IL-6, and TNFa induces IL-1, and LPS induces all three cytokines. However, mice with any one of these cytokine genes "knocked out" are viable, suggesting that none of them is essential for survival, reinforcing the idea of redundancy. It is particularly interesting that IL-6 may be able to assume the role of CRF in certain circumstances, most notably in CRF-knock-out mice. This suggests that despite the existence of multiple CRFs (CRF and the three urocortins), a major redundancy occurs with respect to immune stress.
It will clearly take many more years to unravel the intricacies of the cytokine network, so as to understand not only the complex coordination of the immune response, but also the ways in which the immune system interacts with the other bodily systems, especially the nervous system. Nevertheless, it is also clear that the intimate relationship between cytokines and the HPA axis suggests that the immune system is a critical component of the sensory system signaling stress to the CNS, and that the HPA response is critical for effective immune surveillance.
The author's work described in this chapter was supported by grants from the National Institutes of Health (MH45270, MH46261, and NS35370).
Akil, H., Shiomi, H. and Matthews, J. (1985) Induction of the intermediate pituitary by stress: synthesis and release of a nonopioid form of (3-endorphin. Science, 227: 424-426. Al-Damluji, S. (1988) Adrenergic mechanisms in the control of corticotrophin secretion. J. Endocrinol., 119: 5-14. Ando, T. and Dunn, A.J. (1999) Mouse tumor necrosis factor-a increases brain tryptophan concentrations and norepinephrine metabolism while activating the HPA axis in mice. Neuroimmunomodulation, 6: 319-329. Ando, T., Rivier, J., Yanaihara, H. and Arimura, A. (1998) Peripheral corticotropin-releasing factor mediates the elevation of plasma IL-6 by immobilization stress in rats. Am. J. Physiol., 275: R1461-R1467. Andreis, P.G., Neri, G. and Nussdorfer, G.G. (1991) Corticotropin-releasing hormone (CRH) directly stimulates corticosterone secretion by the rat adrenal gland. Endocrinol., 128: 1198-1200. Antoni, F.A. (1993) Vasopressinergic control of pituitary adrenocorticotropin secretion comes of age. Front. Neuroendocrinol., 14: 76-122. Banks, W.A., Kastin, A.J. and Broadwell, R.D. (1995) Passage of cytokines across the blood-brain barrier. Neuroimmunomodulation, 2: 241-248.
Banks, W.A., Farr, S.A., La Scola, M.E. and Morley, J.E. (2001) Intravenous human interleukin-1 a impairs memory processing in mice: dependence on blood-brain barrier transport into posterior division of the septum. J. Pharmacol. Exptl Therap., 299: 1-6.
Barney, M„ Call, G.B., Mcllmoil, C.J., Husein, O.F., Adams, A., Balls, A.G., Oliveira, G.K., Miner, E.C., Richards, T.A., Crawford, B.K., Heckmann, R.A., Bell, J.D. and Judd, A.M. (2000) Stimulation by interleukin-6 and inhibition by tumour necrosis factor of Cortisol release from bovine adrenal zona fasciculata cells through their receptors. Endocrin. J., 13: 369-377.
Berkenbosch, F., van Oers, J., del Rey, A., Tilders, F. and Besedovsky, H. (1987) Corticotropin-releasing factor-producing neurons in the rat activated by interleukin-1. Science, 238: 524-526.
Berkenbosch, F., De Goeij, D.E.C., del Rey, A. and Besedovsky, H.O. (1989) Neuroendocrine, sympathetic and metabolic responses induced by interleukin-1. Neuroendocrinol., 50: 570-576.
Bernardini, B., Kamilaris, T.C., Calogero, A.E., Johnson, E.O., Gomez, M.T., Gold, P.W. and Chrousos, G.P. (1990) Interactions between tumor necrosis factor-a, hypothalamic corticotropin-releasing hormone, and adrenocorticotropin secretion in the rat. Endocrinol., 126: 2876-2881.
Bernton, E.W., Beach, J.E., Holaday, J.W., Smallridge, R.C. and Fein, H.G. (1987) Release of multiple hormones by a direct action of interleukin-1 on pituitary cells. Science, 238: 519-521.
Bertini, R., Bianchi, M. and Ghezzi, P. (1988). Adrenalectomy sensitizes mice to the lethal effects of interleukin 1 and tumor necrosis factor. J. Exptl. Med., 167: 1708-1712.
Besedovsky, H.O. and del Rey, A. (1991) Physiological implications of the immune-neuro-endocrine network. In: Ader, R„ Felten, D. and Cohen, N. (Eds.), Pyschoneuroimmunology, 2nd ed. Academic Press, New York, pp. 589-608.
Besedovsky, H.O. and del Rey, A. (1992) Immune-neuro-endocrine circuits: integrative role of cytokines. Front. Neuroendocrinol, 13: 61-94.
Besedovsky, H.O. and del Rey, A. (1996) Immune-neuro-endocrine interactions: facts and hypotheses. Endocrine Rev., 17: 64-102.
Besedovsky, H.O., del Rey, A., Sorkin, E. and Dinarello, C.A. (1986) Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones. Science, 233: 652-654.
Besedovsky, H.O., del Rey, A., Klusman, I., Furukawa, H., Monge Arditi, G. and Kabiersch, A. (1991) Cytokines as modulators of the hypothalamus-pituitary-adrenal axis. J. Steroid Biochem. Molec. Biol., 40: 613-618.
Bethin, K.E., Vogt, S.K. and Muglia, L.J. (2000) IL-6 is an essential, corticotropin-releasing hormone-independent, stimulator of the adrenal axis during immune system activation. Proc. Natl Acad. Sci. USA, 97: 9317-9322.
Blalock, J.E. (1984) The immune system as a sensory organ. J. Immunol, 132: 1067-1070.
Blatteis, C.M. and Sehic, E. (1997) Fever: how may circulating pyrogens signal the brain. News Physiol. Sei, 12: 1-9.
Bocci, V. (1991) Interleukins: clinical pharmacokinetics and practical implications. Clin. Pharmacokinet, 21: 274-284.
Brebner, K, Hayley, S, Zacharko, R, Merali, Z. and Anisman, H. (2000). Synergistic effects of interleukin-lß, interleukin-6, and tumor necrosis factor-a: central monamine, corticosterone, and behavioral variations. Neuropsychopharmacol, 22: 566-580.
Brouckaert, P.G.G, Leroux-Roels, G.G., Guisez, Y, Tavernier, J. and Fiers, W. (1986) In vivo anti-tumour activity of recombinant human and murine TNF, alone and in combination with murine IFN-y, on a syngeneic murine melanoma. Int. J. Cancer, 38: 763-769.
Brouckaert, P, Libert, C, Everaerdt, B. and Fiers, W. (1992) Selective species specificity of tumor necrosis factor for toxicity in the mouse. Lymphokine Cytokine Res, 11: 193-196.
Callahan, T.A. and Piekut, D.T. (1997) Differential Fos expression induced by IL-lß and IL-6 in rat hypothalamus and pituitary gland. J. Neuroimmunol, 73: 207-211.
Cambronero, J.C, Rivas, F.J, Borrell, J. and Guaza, C. (1992) Is the adrenal cortex a putative site for the action of interleukin-1? Horm. Metab. Res, 24: 48^19.
Cao, C, Matsumu, K. and Watanabe, Y. (1997) Induction of cyclooxygenase-2 in the brain by cytokines. Ann. N.Y. Acad. Sei, 813: 307-309.
Cassidy, E.M, Manning, D, Byrne, S, Böiger, E, Murray, F, Sharifi, N, Wallace, E, Keogan, M. and O'Keane, V. (2002) Acute effects of low-dose interferon-alpha on serum Cortisol and plasma interleukin-6. J. Psychopharmacol, 16: 230-234.
Chesnokova, V. and Melmed, S. (2000) Leukemia inhibitory factor mediates the hypothalamic pituitary adrenal axis response to inflammation. Endocrinol, 141: 4032^1040.
Chesnokova, V, Auernhammer, C.J. and Melmed, S. (1998) Murine leukemia inhibitory factor gene disruption attenuates the hypothalamo-pituitary-adrenal axis stress response. Endocrinol, 139: 2209-2216.
Chowers, I, Hammel, H.T, Eisenman, J, Abrams, R.M. and McCann, S.M. (1966) Comparison of effect of environmental and preoptic heating and pyrogen on plasma Cortisol. Amer. J. Physiol, 210: 606-610.
Chuluyan, H, Saphier, D, Röhn, W.M. and Dunn, A.J. (1992) Noradrenergic innervation of the hypothalamus participates in the adrenocortical responses to interleukin-1. Neuroendocrinol, 56: 106-111.
Corssmit, E.P, de Metz, J, Sauerwein, H.P. and Romijn, J.A. (2000) Biologic responses to IFN-a administration in humans. J. Interferon Cytokine Res, 20: 1039-1047.
Darling, G, Goldstein, D.S., Stull, R, Gorschboth, C.M. and Norton, J.A. (1989) Tumor necrosis factor: immune endocrine interaction. Surgery, 106: 1155-1160.
Daynes, R.A, Robertson, B.A, Cho, B.-H, Burnham, D.K. and Newton, R. (1987) a-Melanocyte-stimuiating hormone exhibits target cell selectivity in its capacity to affect interleukin 1-inducible responses in vivo and in vitro. J. Immunol, 139: 103-109.
Di Santo, E, Sironi, M, Pozzi, P, Gnocchi, P, Isetta, A.M., Delvaux, A, Goldman, M, Marchant, A. and Ghezzi, P.
(1995) Interleukin-10 inhibits lipopolysaccharide-induced tumor necrosis factor and interleukin-1 beta production in the brain without affecting the activation of the hypothala-mus-pituitary-adrenal axis. Neuroimmunomod, 2: 149-154.
Donaldson, C.J, Sutton, S.W, Perrin, M.H, Corrigan, A.Z, Lewis, K, Rivier, J.E, Vaughan, J.M. and Vale, W.W.
(1996) Cloning and characterization of human urocortin. Endocrinol, 137: 2167-2170.
Dunn, A.J. (1988) Systemic interleukin-1 administration stimulates hypothalamic norepinephrine metabolism parallelling the increased plasma corticosterone. Life Sci, 43: 429-435.
Dunn, A.J. (1990) Interleukin-1 as a stimulator of hormone secretion. Prog. NeuroEndocrinlmmunol, 3: 26-34.
Dunn, A.J. (1992) The role of interleukin-1 and tumor necrosis factor a in the neurochemical and neuroendocrine responses to endotoxin. Brain Res. Bull, 29: 807-812.
Dunn, A.J. (1993a) Role of cytokines in infection-induced stress. Ann. N.Y. Acad. Sci, 697: 189-202.
Dunn, A.J. (1993b) Nitric oxide synthase inhibitors prevent the cerebral tryptophan and serotonergic responses to endotoxin and interleukin-1. Neurosci. Res. Commun, 13: 149-156.
Dunn, A.J. (2000) Cytokine activation of the HPA axis. Ann. N.Y. Acad. Sci, 917: 608-617.
Dunn, A.J. (2002) Mechanisms by which cytokines signal the brain. In: Clow, A. and Hucklebridge, F. (Eds.), Neurobiology of the Immune System, International Review of Biology, Vol. 52. Academic Press, San Diego, pp. 43-65.
Dunn, A.J. (2004) Stress neurochemistry. In: Adelman, G. (Ed.), Encyclopedia of Neuroscience, 3rd ed. Birkhauser Boston, Cambridge, MA.
Dunn, A.J. and Chuluyan, H. (1992) The role of cyclo-oxygenase and lipoxygenase in the interleukin-1-induced activation of the HPA axis: dependence on the route of injection. Life Sci, 51: 219-225.
Dunn, A.J. and Swiergiel, A.H. (1999) Behavioral responses to stress are intact in CRF-deficient mice. Brain Res, 845: 14-20.
Dunn, A.J, Wang, J.-P. and Ando, T. (1999) Effects of cytokines on central neurotransmission: Comparison with the effects of stress. Adv. Exptl. Med. Biol, 461: 117-127.
Ericsson, A, Ek, M, Wahlstrom, I, Kovacs, K, Liu, C, Hart, R. and Sawchenko, P.E. (1996) Pathways and mechanisms for interleukin-1 mediated regulation of the hypothalamo-pituitary-adrenal axis. In: McCarty, R, Aguilera, G, Sabban, E. and Kvetnansky, R. (Eds.), Stress:
Molecular Genetica and Neurobiological Advances. Gordon and Breach Science Publishers, New York, pp. 101-120.
Fagarasan, M.O, Eskay, R. and Axelrod, J. (1989) Interleukin 1 potentiates the secretion of p-endorphin induced by secretagogues in a mouse pituitary cell line (AtT-20). Proc. Natl. Acad. Sci. USA, 86: 2070-2073.
Fagarasan, M.O, Aiello. F, Muegge, K, Durum, S. and Axelrod, J. (1990) Interleukin 1 induces P-endorphin secretion via Fos and Jun in AtT-20 pituitary cells. Proc. Natl. Acad. Sci. USA, 87: 7871-7874.
Fleshner, M, Goehler, L.E, Hermann, J, Relton, J.K, Maier, S.F. and Watkins, L.R. (1995) Interleukin-1 p induced corticosterone elevation and hypothalamic NE depletion is vagally mediated. Brain Res. Bull.. 37: 605-610.
Fleshner, M, Goehler. L.E., Schwartz, B.A, McGorry, M, Martin, D„ Maier. S.F. and Watkins, L.R. (1998) Thermogenic and corticosterone responses to intravenous cytokines (IL-lp and TNF-a) are attenuated by subdiaphragmatic vagotomy. J. Neuroimmunol, 86: 134-141.
Fransen, L, Miiller, R, Marmenout, A., Tavernier, J., Van der Heyden, J, Kawashima, E, Chollet, A, Tizard, R., Van Heuverswyn, H, Van Vliet, A., Ruysschaert, M.-R. and Fiers, W. (1985) Molecular cloning of mouse tumor necrosis factor cDNA and its eukaryotic expression. Nucleic Acids Res, 13: 4417-4429.
Fukata, J, Usui, Y., Naitoh, Y, Nakai, Y. and Imura, H. (1989) Effects of recombinant human interleukin-la, -IP, 2 and 6 on ACTH synthesis and release in the mouse pituitary tumor line AtT-20. J. Endocrinol., 122: 33-39.
Gaykema, R.P.A, Dijkstra, I. and Tilders, F.J.H. (1995) Subdiaphragmatic vagotomy suppresses endotoxin-induced activation of hypothalamic corticotropin-releasing hormone neurons and ACTH secretion.
Was this article helpful?
All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.