Identification of an immune-responsive mesolimbocortical serotonergic system: potential role in regulation of emotional behavior

Abstract

Peripheral immune activation can have profound physiological and behavioral effects including induction of fever and sickness behavior. One mechanism through which immune activation or immunomodulation may affect physiology and behavior is via actions on brainstem neuromodulatory systems, such as serotonergic systems. We have found that peripheral immune activation with antigens derived from the nonpathogenic, saprophytic bacterium, Mycobacterium vaccae, activated a specific subset of serotonergic neurons in the interfascicular part of the dorsal raphe nucleus (DRI) of mice, as measured by quantification of c-Fos expression following intratracheal (12 h) or s.c. (6 h) administration of heat-killed, ultrasonically disrupted M. vaccae, or heat-killed, intact M. vaccae, respectively. These effects were apparent after immune activation by M. vaccae or its components but not by ovalbumin, which induces a qualitatively different immune response. The effects of immune activation were associated with increases in serotonin metabolism within the ventromedial prefrontal cortex, consistent with an effect of immune activation on mesolimbocortical serotonergic systems. The effects of M. vaccae administration on serotonergic systems were temporally associated with reductions in immobility in the forced swim test, consistent with the hypothesis that the stimulation of mesolimbocortical serotonergic systems by peripheral immune activation alters stress-related emotional behavior. These findings suggest that the immune-responsive subpopulation of serotonergic neurons in the DRI is likely to play an important role in the neural mechanisms underlying regulation of the physiological and pathophysiological responses to both acute and chronic immune activation, including regulation of mood during health and disease states. Together with previous studies, these findings also raise the possibility that immune stimulation activates a functionally and anatomically distinct subset of serotonergic neurons, different from the subset of serotonergic neurons activated by anxiogenic stimuli or uncontrollable stressors. Consequently, selective activation of specific subsets of serotonergic neurons may have distinct behavioral outcomes.

Fig. 1

Influence of M. vaccae or its derivatives on pulmonary cytokine mRNA expression, c-Fos expression in the nTS and c-Fos expression in serotonergic neurons in the dorsal raphe nucleus (DR). (a) Graphs illustrate mean levels, relative to β-actin, of pulmonary IL-1β (IL-1), TNF-α (TNF), and IL-6 mRNA expression 12 h, and 3, 6, 10, and 17 days following i.t. injection of Mv-NC in M. vaccae–preimmunized mice (•) relative to cytokine mRNA expression following i.t. injection of vehicle in M. vaccae–preimmunized mice (○) (12 h time point only), as well as, for comparison, cytokine mRNA expression at the same time points following i.t. injection of OVA-NC in OVA/alum-preimmunized mice (▾) or vehicle in OVA/alum-preimmunized mice (▿) (12 h time point only). (b) Bar graphs illustrate the mean number (±S.E.M.) of c-Fos-ir nuclei in the AP, SolDL, DRC, and DRI in experiment 1. (c) Photographs illustrate nuclear c-Fos immunostaining (blue–black) 12 h following i.t. injection of M. vaccae in M. vaccae–preimmunized mice in the AP and SolDL of the nTS (top two photographs) and serotonergic neurons in the DRC (middle two photographs) and DRI parts (bottom two photographs) of the DR. Tyrosine hydroxylase immunostaining (brown) was used to aid in identification of neuroanatomical subdivisions of the nTS. Tryptophan hydroxylase immunostaining (brown) was used to identify serotonergic neurons in the DRC and DRI. (⇒) c-Fos-immunonegative serotonergic neurons, () c-Fos-immunopositive serotonergic neurons. Small black boxes in (c) indicate regions shown at higher magnification in insets. Scale bar=100 μm (c) top row; (c) middle and bottom rows, 25 μm; (c) insets, 12.5 μm. Abbreviations: IL-1, interleukin-1β; (Mv), preimmunization with s.c. injections of heat-killed M. vaccae; Mv-NC, i.t. challenge with sonicated heat-killed Mv-NC; NC, i.t. challenge with NC. * P≤0.05, compared with M. vaccae–preimmunized, vehicle-injected controls, Fisher’s protected LSD test. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

Fig. 2

Both i.t. Mv-NC and i.t. OVA-NC induced c-Fos expression in the nTS but only Mv-NC increased c-Fos expression in DRI serotonergic neurons. Immunostained products are the same as in Fig. 1. (a) Bar graphs illustrate the mean number (±S.E.M.) of c-Fos-ir nuclei in the AP and SolDL (left) and c-Fos-ir/tryptophan hydroxylase-ir neurons in the DRC and DRI (right) in experiment 2. (b, c) Photomicrographs illustrate c-Fos responses to Mv-NC in M. vaccae–preimmunized mice or to OVA-NC in OVA/alum-preimmunized mice in the SolDL and AP (b) or the DRI (c). (⇒) c-Fos-immunonegative serotonergic neurons, () c-Fos-immunopositive serotonergic neurons. Small black boxes indicate regions shown at higher magnification in insets. Scale bar=100 μm b; b (insets), 50 μm; c, 25 μm; c (insets) 12.5 μm. Abbreviations: (OVA), preimmunization with s.c. injections of OVA/alum; OVA-NC, i.t. challenge with OVA-NC. For additional abbreviations, see Fig. 1 legend. * P≤0.05, compared with the appropriate M. vaccae– or OVA/alum-preimmunized, vehicle-injected control group, Fisher’s protected LSD test.

Fig. 3

Activation of bronchopulmonary afferent vagal pathways was not necessary for the effects of Mv-NC on DRI serotonergic neurons. Immunostained products are the same as in Fig. 1. (a) Bar graphs illustrate the mean number (±S.E.M.) of c-Fos-ir nuclei in the AP and SolDL (left) and c-Fos-ir/tryptophan hydroxylase-ir neurons in the DRC and DRI (right) in experiment 4. (b, c) Photomicrographs illustrate c-Fos responses to i.t. Mv-NC or s.c. M. vaccae in M. vaccae–preimmunized mice in the SolDL and AP (b) or the DRI (c) at the 12 h time point. For abbreviations, see Fig. 1 legend. (⇒) c-Fos-immunonegative serotonergic neurons, () c-Fos-immunopositive serotonergic neurons. Scale bar=100 μm b; b (insets), 50 μm; c, 25 μm; c (insets), 12.5 μm. Abbreviations: Mv, s.c. challenge with heat-killed M. vaccae; Sal, s.c. challenge with saline vehicle. For additional abbreviations, see Fig. 1 legend. * P≤0.05, compared with the appropriate M. vaccae–preimmunized, vehicle-injected control group, Fisher’s protected LSD test.

Fig. 4

S.c. challenge with whole heat-killed M. vaccae in M. vaccae–preimmunized mice increased 5-HT and 5-HT metabolite concentrations in the medial prefrontal cortex. Graphs illustrate 5-HIAA, 5-HT, and l-tryptophan concentrations in the left and right hemispheres of each brain region 12 h following s.c. injections of vehicle or whole heat-killed M. vaccae in M. vaccae–preimmunized mice (n=8–9). Abbreviations: Mv, s.c. challenge with heat-killed M. vaccae, Sal, s.c. challenge with saline vehicle; TRP, l-tryptophan. For additional abbreviations, see Fig. 1 legend. * P≤0.05, compared with M. vaccae–preimmunized, vehicle-injected controls based on multifactor ANOVA with repeated measures analysis within each brain region.

Fig. 5

S.c. challenge with whole heat-killed M. vaccae in M. vaccae–preimmunized mice decreased immobility in the forced swim test measured 12 h following challenge. Abbreviations: (Sal), s.c. preimmunization with saline vehicle. For additional abbreviations, see Fig. 1 legend. ^# P<0.05 compared with M. vaccae–preimmunized, saline-challenged control group. * P<0.01 compared with saline-preimmunized, M. vaccae–challenged control group.