Invited Symposium: Cytokines, Monoamines and Behavior
Similarities between the behavioral, endocrine, and neurochemical sequelae of exposure to psychological stressors and agents that activate peripheral immune cells (viruses, lipopolysaccharide, arthritis inducing agents, etc.) have often been noted. Central cytokines such as IL-1b play an important role in mediating many of these responses that follow immune activation, and the purpose of the present chapter is to summarize work that is directed at determining whether central cytokines also play a role in mediating the consequences of exposure to stressors. In addition, there will be a focus on whether central cytokines are involved in the production of a particular stress-related phenomenon, learned helplessness. Any discussion of the role of central IL-1b must begin with the issue of whether IL-1b is expressed in the “non-pathological” brain, and of whether any IL-1b expressed is neuronal.
Expression of Il-1b in the Brain
There is dispute concerning whether there is constitutive expression of IL-1b in brain, although there is increasing evidence that under basal conditions there is IL-1b bioactivity (Quan et al., 1996), immunoreactivity (Hagan, Poole, & Bristow, 1993), protein (Nguyen et al., 1998), and basal release (Tringali et al., 1997). However, it is clear that the administration of agents such as lipopolysacharide (LPS) which release IL-1 in the periphery and the peripheral administration of IL-1 itself are followed by regionally specific increases in brain IL-1b mRNA (Ban et al., 1992), immunoreactivity (Van Dam et al., 1992), bioactivity (Quan et al, 1994), and protein levels (Nguyen et al., 1998).
The cellular source(s) of IL-1 includes glial cells, but perhaps neurons as well (Tringali et al., 1996). This is not a critical issue for the present discussion since IL-1 released from either glia or neurons would be available to bind to IL-1 receptors on neurons, thereby altering neuronal activity.
The intracerebroventricular (ICV) administration of IL-1b produces a constellation of behavioral and physiological “sickness” responses including fever, increased pituitary-adrenal activity, decreased plasma carrier proteins, increased plasma acute phase proteins, reduced locomotion, reduced social, sexual, and exploratory behavior, and increased numbers of white blood cells. Thus a first question might be whether stressors duplicate this pattern. It is known that exposure to stressors leads to reduced activity, exploration, sexual, and social behavior, but does it also lead to other alterations characteristic of infection and brain IL-1 administration?
Fever is perhaps the key physiological adjustment to infection, and it has been shown that even relatively mild stressors such as placement in a novel environment produce increases in core body temperature (Morrow et al., 1993). However, fever has been examined only during or very shortly after a brief stressor exposure, and so whether there would be a prolonged fever such as observed during infection is unknown. Thus we (Deak et al., 1997) exposed loosely restrained rats to a single session of inescapable tailshock (IS) with parameters typical in our laboratory (100 5 sec ISs of 1.0 mA in intensity, occurring on the average of once a min) and measured core body temperature using implanted thermisters during IS or control treatment as well as for the periods 1-4 and 20-24 hr after IS. It was not surprising that core body temperature increased during IS. However, IS produced a fever still present 24 hr later. In subsequent work we have found this fever to persist for 4, but not for 7 days.
We have also measured white blood cells, the carrier protein corticosteroid binding globulin (CBG), seromucoids (a measure of acute phase proteins). and the acute phase protein haptoglobin in blood samples taken 24 hr after IS or control treatment. Exposure to IS led to the typical pattern produced by pathogens. We have also measured the acute phase protein a-2 glycoprotein, and it begins to rise 6 hr after IS and remains elevated for several days (Deak et al., 1997). A final aspect that we have studied involves basal levels of plasma corticosteroids. We have shown that the administration of inflammatory agents increases basal corticosteroid levels for a number of days and IS also increases basal cortiocsoteroids measured 24 and 48 hr after IS (Fleshner et al., 1995).
In sum, exposure to IS produces a pattern very similar to that produced by peripheral inflammatory agents and ICV IL-1. This suggests that IS might led to the production of IL-1b in brain, whether glial or neural. This issue is complex for a variety of reasons. One is that adrenal corticosteroids (CORT) destabilize IL-1b mRNA (Amano, Lee, & Allison, 1993) and inhibit IL-1b gene transcription (Lee et al., 1988) as well as a number of translational and post-translational processes (Kern et al., 1988) involved in the production of IL-1b protein. Because IS increases adrenal CORT levels, we sought to determine whether IS would increase brain IL-1b protein levels in both adrenalectomized and sham surgery subjects. Basal CORT was maintained at normal levels by the addition of CORT to the drinking water.
Thus the ADX subjects had normal basal CORT and a normal circadian rhythm of CORT (the rats drink much more during the dark part of the cycle), but could not increase CORT in response to IS. We measured IL-b by ELISA (Nguyen et al., 1998), and IS increased IL-1b protein in brain in a regionally selective manner, with the most robust increases in hippocampus and hypothalamus. These increases occurred only in adrenalectomized subjects, although technical improvements since these experiments were conducted have led to the observation of IL-1b increases in hypothalamus in non-adrenalectomized subjects.
ELISA uses an antibody (Ab) specifically directed against rat IL-1b to detect IL-1. However, there is always the possibility that the Ab that is used is capable of detecting another protein as well. We have used a number of approaches in an attempt to overcome this potential difficulty. a) Three different Abs directed against rat IL-1b have been used. Although absolute values have differed between Abs, the direction and magnitude of change has been the same for all Abs. b) Specificity of Ab binding was assessed by Western blot. Strong bands appeared at approximately 17 kDa, the molecular weight of mature IL-1b.
However, faint bands were also detected at roughly 33 kDa, the molecular weight of the precursor (inactive) form of IL-1b. Several laboratories have now reported that Ab directed against 17 kDa IL-1b recognizes 33 kDa pro-IL-1b with approximately 10-fold less affinity (Dinarello, 1992). c) Total protein from the brain sonication supernatants was separated on a Sephadex G-50 column, and fractions assayed with the ELISA.
The signal was restricted to the fraction in the molecular weight range of IL-1b. d) The rat IL-1b Ab was verified for its ability to recognize pure rat IL-1b by immunoprecipitation. Thus far the data suggest that IS and ICV IL-1 produce similar outcomes, and that IS may induce IL-1 in brain. If brain IL-1b is involved in the mediation of these effects of IS, then antagonism of IL-2 should reduce these effects of IS. We have investigated this possibility by administering ICV both the IL-1 receptor antagonist (IL-1ra) and alpha-melanocyte stimulating hormone (a-MSH). a-MSH has been employed because it is known to antagonize many of the effects of IL-1 (see Catania and Lipton, 1994, for a review), although the precise nature of the mechanism by which it does so is unknown. With the idea that ICV a-MSH can antagonize the effects of IL-1 in brain, and because it is more readily available than IL-1ra, we conducted a series of experiments designed to determine whether ICV a-MSH would block the effects of IS on sickness or acute phase responses.
Separate groups received either vehicle or 0.5 mg a-MSH injected ICV before the IS. a--MSH blocked the reductions in CBG, increased basal CORT, fever, and reduced food and water intake produced by IS (Milligan et al., 1998). We have also shown that ICV IL-1ra blocks some of the typical behavioral effects of IS (Maier & Watkins, 1995).
In sum the data to date encourage the argument that IS induces IL-1 in brain, and that the induced IL-1 is involved in mediating the consequences of IS. However, some sequelae of IS are dependent on the inescapability of the stressor, and fail to occur if the shock is escapable. These have been called learned helplessness effects, with debilitated escape learning consequent to exposure to IS being the most often studied example. Thus, it is natural to inquire whether the escapability/inescapability of the shock modulates brain IL-1, and whether consequences of IS such as changes in acute phase proteins, fever, etc., depend on shock escapability. To summarize a number of experiments that we have conducted, both IS and equal escapable shocks produced similar increases in brain IL-1b in all regions except the amygdala in which only IS produced an increase, and both IS and escapable shock produced equivalent fever, decreased CBG, etc. Thus, stressor controllability did not modulate brain IL-1 or the effects likely mediated by IL-1.
The conclusions to be drawn from the literature and the pattern of our own data are far from clear, and any suggestion can only be made in very tentative fashion. The data tend to suggest that brain IL-1b may be involved in mediating some consequences of exposure to stressors per se. Tailshock did increase brain IL-1b protein levels, most notably in the hypothalamus where rapid increases could be detected in non-adrenalectomized subjects. In addition, antagonizing IL-1 either with a-MSH or IL-1ra blocked a variety of effects of tailshock including fever, increased acute phase proteins, increased white blood cells, decreased CBG, basal CORT shifts, and a number of behavioral effects. Interestingly, restraint did not increase brain IL-1b protein, nor does it produce outcomes characteristic of the acute phase response such as increases in acute phase proteins, prolonged fever, and the like. Thus not all stressors are likely to engage the brain IL-1 machinery, and perhaps only intense stressors such as tailshock will do so with the physical characteristics of the stressor rather than its psychological nature being crucial.
| Discussion Board | Previous Page | Your Symposium |
|Maier, S. F.; Watkins, L. R.; (1998). Stress, Learned Helplessness, and Brain Interleukin-1. Presented at INABIS '98 - 5th Internet World Congress on Biomedical Sciences at McMaster University, Canada, Dec 7-16th. Invited Symposium. Available at URL http://www.mcmaster.ca/inabis98/anisman/maier0245/index.html|
|© 1998 Author(s) Hold Copyright|