There have been a number of reports showing that cytokines may influence central neurotransmitter activity and/or levels. Given the marked effects of cytokines on neuroendocrine factors, it is not surprising that particular attention was devoted to neurotransmitter variations within the hypothalamus. For instance, when systemically administered, both IL-ip and LPS increased the utilization of hypothalamic NE (Masana, Heyes, & Mefford, 1990; Mefford & Heyes, 1990; Mefford, Masters, Heyes, & Eskay, 1991; Terao, Oikawa, & Saito, 1993; Zalcman, Green-Johnson, Murray, Nance, Dyck, Anisman, & Greenberg, 1994). Similarly, like stressors, LPS increased NE turnover and tryptophan levels in hypothalamus, and these effects were modifiable by blocking sympathetic nervous system activity (Dunn & Welch, 1991). As in the case of the post-mortem amine changes, systemic IL-ip influences the in vivo release of hypothalamic NE (Smagin, Swiergel, & Dunn, 1996) and central administration of IL-lp enhanced the in vivo release of hypothalamic NE, DA, and 5-HT (Mohankumar & Quadri, 1993; Mohankumar, Thyagarajan, & Quadri, 1991, 1993; Shintani, Kanba, Nakaki, Nibuya, Kinoshita, Suzuki, Yagi, Kato, & Asai, 1993). In addition to the hypothalamic amine variations, LPS and endotoxin affect tryptophan levels and NE turnover in the prefrontal cortex, parietal cortex and brain stem (Dunn & Welch, 1991), as well as NE and 5-HT turnover in hippocampus, and DA turnover within the prefrontal cortex (Zalcman et al., 1994). In vivo studies have also revealed that systemic IL-1 P and LPS lead to marked hippocampal 5-HT release which coincided with changes of locomotor activity, and these effects were attenuated by icv administration of IL-lra (Linthorst et al., 1995). As will be described shortly, results from our laboratory have essentially confirmed these findings, and suggest that peripheral immune and cytokine challenges also influences NE and DA at several brain sites.
It has been demonstrated that IL-2 influences secretion of several hormones, but few studies have assessed the central neurochemical consequences of IL-2 (see Hanisch & Quirion, 1996). Intravenous administration of IL-2 was shown to provoke hypothalamic unit activity (Bartholomew & Hoffman, 1993), and intraperitoneal IL-2 treatment increased the turnover of hypothalamic NE (Zalcman et al., 1994). It was likewise observed that IL-2 increased DA release from striatal slices under basal conditions, as well as during potassium-evoked neuronal depolarization (Alonso, Chaudieu, Dorio, Krishnamurthy, Quirion, & Boksa, 1993; Lapchak, 1992), although biphasic effects were observed such that lower doses increased and higher doses inhibited veratrine-evoked DA release (Petitto, McCarthy, Rinker, Huang, & Getty, 1997). As in the case of DA, it has been demonstrated that in hippocampal slices the application of IL-2 in relatively low doses potentiated ACh release, but had an inhibitory effect in higher doses (Araujo, Lapchak, Collier, & Quirion, 1989; Seto, Kar, & Quirion, 1997). As well, when applied directly to the caudate or substantia nigra, IL-2 provoked ipsilateral turning, suggesting that the lymphokine may act through DA or opioid mechanisms, which induce circling when applied to nigrostriatal structures (Lapchak, 1992). Taken together, these data suggest a possible role for IL-2 in subserving both some of the central neurochemical and behavioral effects associated with antigenic challenge.
Stressors, like antigenic challenge, increase IL-ip mRNA expression, as well as that of IL-IRa (Suzuki, Shintani, Kanba, Asai, & Nakaki, 1997). Interestingly, in the absence of the inhibitory effects of glucocorticoids (i.e., in adrenalectomized animals), stressor exposure markedly increased IL-ip protein in hypothalamus, hippocampus, cerebellum, and nucleus tractis solitaris (Nguyen, Deak, Owens Kohn, Fleshner, Watkins, & Maier, 1998). Inasmuch as IL-ip strongly influences CRH release from the PVN, thus provoking HPA activity (Ericsson, Arias, & Sawchenko, 1997; Ericsson, Ek, Wahlstrom, Kovacs, Liu, Hart, & Sawchenko, 1996; Rivest, 1995; Rivest & Rivier, 1994), the possibility has been explored that this cytokine may play a pivotal role in the HPA activation associated with stressful events. Indeed, it was demonstrated (Shintani et al., 1993; Shintani, Nakaki, Kanba, Sato, Kato, & Asai, 1995; Shintani, Nakaki, Kanba, Sato, Yagi, Shiozawa, Aiso, Kato, & Asai, 1995) that the increased hypothalamic NE and DA activity and pituitary ACTH release associated with a stressor was prevented by icv pretreatment with IL-lra. If the cytokine receptor antagonist was applied once the stressor had been administered, then there was no effect on either HPA functioning or the central amines. In effect, IL-lra acted in a prophylactic fashion in affecting HPA functioning and amine changes engendered by stressors; however, once these processes were initiated, the cytokine antagonist did not operate in a therapeutic capacity in attenuating the stressor effects.
Although stressors influence IL-lp mRNA or protein levels within the brain and pituitary (Minami, Kuraishi, Yamaguchi, Nakai, Hirai, & Satoh, 1991; Nguyen et al., 1998; Shintani et al., 1995a, b; Takao, Tojo, Nishioka et al., 1994; Yabuuchi, Maruta, Minami, & Satoh, 1996), stressors and IL-ip may act through independent inputs to CRH neurons (Harbuz & Lightman, 1992; Whitnall, Perlstein, Mougey, & Neta, 1992). Indeed, the view was offered that while processive stressors influence HPA activity via the central amygdala, systemic stressors (including IL-lp) do so via an amygdala-independent mechanism (Herman & Cullinan, 1997). This is not to say that cytokines are without effect on amygdaloid activity, as IL-lp increased Fos-immunoreactivity in the central amygdala and bed nucleus of the stria terminalis, as well as cetecholamin-ergic neurons of the nucleus tractis solitaris and ventrolateral medulla (the latter projecting to regions of the PVN) (Day & Akil, 1996; Ericsson, Kovacs, & Sawchenko, 1994; Sawchenko, Brown, Chan, Ericsson, Li, Roland, & Kovacs, 1996). Furthermore, as will be discussed shortly, we have observed that systemic IL-1(3 increased the in vivo release of NE from the central amygdala (Merali, Michaud, Mclntyre, & Anisman, unpublished report).
In addition to their immediate effects, both stressors and acute IL-1(3 may provoke persistent effects on HPA functioning (Schmidt, Janszen, Wouterlood, & Tilders, 1995; Tilders, Schmidt, & De Goeij, 1993). With the passage of time following stressor or IL-1(3 treatment, phenotypic variations occurred in CRH terminals within the external zone of the median eminence, such that they came to coexpress arginine vasopressin (AVP). As CRH and AVP synergistically promote ACTH release, phenotypic shifts of CRH and AVP coexpressing cells induced by IL-1(3 and by stressors may account for long-term and/or the sensitization effects (Bartanusz, Jezova, Bertini, Tilders, Aubry, & Kiss, 1993; Schmidt et al., 1995;Tilders et al., 1993). Interestingly, there is reason to suppose that chronic depressive illness, as well as the high rates of illness recurrence following successful pharmacotherapy may be related to enhanced coex-pression of CRH and AVP (Gold, Chrousos, Kellner, Post, Roy, Augerinas, Schulte, Oldgield, & Loriaux, 1984; Holsboer, 1995; Nemeroff, 1996). As will be discussed shortly, we have likewise observed time-dependent sensitization with respect to the actions of TNFa on plasma ACTH and corticosterone concentrations, as well as NE, DA, and 5-HT turnover and levels in several brain regions (Hayley, Brebner, Staines, Merali, & Anisman, 1998).
4.2. Anhedonic and Anxiogenic Actions of Cytokines: Stressor-Iike Effects
Since several cytokines have been shown to be potent stimulators of corti-cotropin-releasing hormone (CRH) release from the PVN (Dunn, 1995; Hopkins & Rothwell, 1995; Rivier, 1993), a peptide which induces anxiogenic effects (Gray, 1991; Heilig, Koob, Ekman, & Britton, 1994), we assessed whether various cytokines would provoke an anxiogenic response. Since the anxiogenic effects of various treatments may be evident in some test situations and not others, we have typically assessed anxiety in several different paradigms. In our experiments, which are still at early phase, we have assessed the effects of LPS and various cytokines on exploratory patterns, neophobia (as reflected by approach to novel stimuli), responding in an elevated plus-maze test and a light-dark box, and we are currently assessing startle and fear-potentiated startle.
As indicated earlier, there is reason to believe that immune activation and cytokine changes may be associated with depressive illness. Furthermore, as anhedo-nia is a characteristic feature of depression, it was of interest to assess whether LPS or cytokine challenge would engender anhedonic-like effects. While consumption of palatable substances have been used to assess anhedonia, this type paradigm does not readily distinguish anorexic from anhedonic effects of a given treatment. This is not to disparage the use of feeding paradigms, as they certainly have been used effectively. Rather, it was felt that in this instance it would be advantageous to employ a test devoid of appetitive motivation. Accordingly, we opted to assess the effects of LPS and cytokine treatments on responding for rewarding brain stimulation in rats, a behavioral test previously shown to be sensitive to the effects of stressors (Anisman, Zalcman, Shanks, & Zacharko, 1991).
Typically, when electrodes are implanted in the medial forebrain bundle, or any number of other DA rich brain regions (prefrontal cortex, nucleus accumbens, sub-
stantia nigra, ventral tegmentum), animals will emit operant responses to receive electrical stimulation. In view of the large number of responses animals emit, it is assumed that responding is rewarding. Treatments that reduce the apparent rewarding value of the stimulation may, however, provoke an increase of response rate. It seems that animals will compensate for the reduced value of the stimulation by increasing its response frequency. Further, a decline in responding may be secondary to motoric effects of a given treatment, independent of effects on the variations in the rewarding properties of the brain stimulation. Thus, rather than employing rate-dependent measures (e.g., number of responses emitted), we employ a titration paradigm in which animals are permitted to respond for rewarding brain stimulation (where the operant is a head dip response into a hole located on the floor of the test chamber) over the course of a descending series of current intensities (or current frequencies), after which responding is measured over an ascending series. The two curves are then combined to provide a single titration curve, unless the ascending and descending series are very different (in which case the two curves are both presented). In addition, in some of our experiments a simultaneous discrimination paradigm is employed, wherein animals are required to respond to an illuminated hole (+) and not to a nonilluminated hole (-). In this way it is possible to determine whether the change in the titration curve is accompanied by variations in the animals discrimination ability (i.e., altered responses to the + stimulus, without affecting responding to the— stimulus, or conversely high responses being maintained but in a less discriminating fashion).
Predictably, in nonstressed animals responding declines as the current intensity is reduced, reflecting the diminished reward obtained for an operant response. Ordinarily, stressors provoke a marked reduction of responding for rewarding brain stimulation, such that responding declines at virtually all of the current intensities. That is, the stressor induces a right-ward shift of the dose response curve, so that higher current intensities are needed to elicit the same number of responses. However, it is important to note that with sufficiently high current intensities the blunted responding in stressed animals may be overcome. In effect, it seems that stressed animals are capable of responding, and motoric factors are likely not responsible for behavioral impairments. Parenthetically, the effects of the stressor are pronounced soon after stressor exposure, and are still evident at lengthy intervals afterward (e.g., 168hr), provided that the initial test session was administered soon after stressor exposure. Moreover, the stressor-induced disruption of responding for rewarding brain stimulation may be attenuated by a regimen of chronic antidepressant treatment (Zacharko & Anisman, 1991). It is important to emphasize, as well, that the effects of stressors on rewarding stimulation vary across brain regions. For instance, we observed that footshock (and other stressors) reliably disrupted responding from the medial forebrain bundle, as well as the prefrontal cortex, nucleus accumbens, and dorsal aspects of the ventral tegmentum, brain regions in which stressors influence DA activity. In contrast, responding for rewarding stimulation was unaffected from those brain regions where DA was not influenced by stressors (i.e., substantia nigra and caudate) (Zacharko & Anisman, 1991). The latter findings are consistent with the supposition that DA mechanisms, among others, contribute to the stressor-related performance changes (Fibiger & Phillips, 1988; Wise, 1985), and also indicate that the altered behavioral responsivity is not attributable to general malaise or motoric changes attributable to the stressor. After all, if such factors were pertinent, then behavioral disturbances should have been evident irrespective of the site of electrode placements.
Was this article helpful?
Have you ever been envious of people who seem to have no end of clever ideas, who are able to think quickly in any situation, or who seem to have flawless memories? Could it be that they're just born smarter or quicker than the rest of us? Or are there some secrets that they might know that we don't?