Splenic glucocorticoid resistance following psychosocial stress requires physical injury

Abstract

Mice exposed to chronic subordinate colony housing (CSC) stress show glucocorticoid (GC) resistance of in vitro lipopolysaccharide (LPS)-stimulated splenocytes, increased anxiety and colitis. Similar effects were reported in wounded mice exposed to social disruption (SDR). Here we show that CSC exposure induced GC resistance in isolated and in vitro LPS-stimulated, but not unstimulated, splenocytes, and these effects were absent when CD11b⁺ splenocytes were depleted. Moreover, re-active coping behaviour during CSC correlated with the attacks and bites received by the resident, which in turn highly correlated with the dimension of splenic GC resistance, as with basal and LPS-induced in vitro splenocyte viability. Importantly, social stress promoted spleen cell activation, independent of bite wounds or CD11b⁺/CD11b^- cell phenotype, whereas GC resistance was dependent on both bite wounds and the presence of CD11b⁺ cells. Together, our findings indicate that the mechanisms underlying splenic immune activation and GC resistance following social stress in male mice are paradigm independent and, to a large extent, dependent on wounding, which, in turn, is associated with a re-active coping style.

Figure 1

Assessment of individual behavioral coping patterns in chronic subordinate colony housing (CSC) mice. Pro-active, re-active and received offensive behavior shown during CSC exposure were assessed as detailed in (A), allowing calculation of a dominance index (DI) for each CSC mouse. Each behavioral pattern was analyzed in CSC mice (n = 46) on days 1, 8 and 15 for one hour in the morning (Supplementary Fig. 2A)  and one hour in the evening (Supplementary Fig. 2B) of the respective days and an overall DI was calculated as in detail shown in (B). CSC.

Figure 2

Effects of 19 days of chronic subordinate colony housing (CSC) on physiological parameters. On day 20 of CSC, CSC (n = 23) and respective single-housed control (SHC; n = 24) mice were weighed before decapitation. Following decapitation, adrenal glands as well as thymus were removed, pruned of fat and weighed. Despite an unchanged body weight gain [g] (A), CSC mice showed a significant increase in the absolute [mg] (B) and relative [mg/g] (C) adrenal weight and a significantly decreased absolute [mg] (D) and relative [mg/g] (E) thymus weight. SHC; CSC. **P ≤ 0.01, ***P ≤ 0.001 versus respective SHC.

Figure 3

Effects of 19 days of chronic subordinate colony housing (CSC) on the spleen, in the presence and absence of CD11b⁺-cells. Depicted is the absolute spleen weight [mg] (A) and the number of splenocytes per spleen (B) following 20 days of CSC (n = 48) and single-housing (single housed controls, SHC; n = 48). Splenocytes were cultured without (basal) and with lipopolysaccharide (LPS, 1 µg/ml) in presence of various corticosterone (CORT) concentrations. After 48 h of incubation, cell viability was measured. Depicted is the cell viability [optical density (OD)] under basal and LPS-stimulated conditions in the absence of CORT (C) and the basal cell viability (without LPS) in the presence of different CORT concentrations (CORT = 0 µM was set to 100%; D). Delta cell viability of LPS-stimulated minus respective basal wells at CORT = 0 µM is depicted in (E). Shown is further the delta cell viability [OD] (LPS-stimulated minus respective basal wells) at various CORT concentrations in percent of respective values at CORT = 0 µM (set to 100%; F). Depicted are further the absolute numbers of CD19⁺, CD3⁺CD4⁺, CD3⁺CD8⁺ and CD11b⁺ cells within total viable splenocytes (G) of another set of SHC (n = 8) and CSC (n = 8) mice. CD11b⁺ as well as CD11b⁻ cells from these mice were cultured without (basal) and with LPS (1 µg/µl) in absence or presence of various CORT concentrations. Depicted is the cell viability under basal and LPS-stimulated conditions (H) as well as delta cell viability of LPS-stimulated minus respective basal wells (I) of CD11b⁻ and CD11b⁺ splenocytes. Shown is further the delta cell viability (LPS-stimulated minus respective basal wells) of CD11b⁻ splenocytes at various CORT concentrations in percent of respective values at CORT = 0 µM (set to 100%; J). A third set of SHC (n = 11) and CSC (n = 10) mice was employed to determine relative protein expression of glucocorticoid receptor (GR; K) and FK506 binding protein 51 (FKBP51; L) [grey density] on day 20 of CSC using western blotting, all normalized to the loading control glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Representative images of bands detected for GR (~86 kDa) and FKBP51 (~51 kDa) and respective loading control GAPDH (~36 kDa) are shown for SHC and CSC mice below each subfigure. SHC; CSC. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 versus respective SHC; ^# P ≤ 0.05, ^## P ≤ 0.01, ^### P ≤ 0.001 versus respective basal (CORT = 0 µM) values.

Figure 4

Correlation analyses - Part I. Depicted are significant correlations between individual behavioral coping patterns shown in the morning (am) and/or in the evening (pm) on days 1, 8 and 15 with the bite scores in all CSC (n = 46) mice. The number of attacks received was positively correlated with the re-active coping behavior (A), flight (B), as well as with submissive upright posture (C). A negative correlation was found between the dominance index (DI) and the number of attacks received (D), as well as the bite score (E). The DI in the evening also correlated negatively with the bite score (F). Furthermore, the bite score was positively correlated with re-active coping behavior of CSC mice in the evening (G), whereas it correlated negatively with the amount of pro-active behavior (H). In detail, the bite score showed a positive correlation with submissive upright posture (I), flight (J), avoiding (K), all in the evening, as well as the total number of attacks received (L).

Figure 5

Correlation analyses – Part II. Depicted are significant correlations between individual behavioral coping patterns shown in the morning and in the evening on days 1, 8 and 15 with the bite scores as well as physiological parameters in all CSC (n = 46) mice. The absolute spleen weight (A), the cell viability of isolated splenocytes under basal (B) and LPS-stimulated (C) conditions (CORT = 0 µM), as well as the splenic glucocorticoid (GC) resistance (D) correlated positively with the established bite score of each CSC mouse. The developed GC resistance (E) was also positively correlated with the absolute spleen weight. The number of attacks received (F) and bite score (G) correlated negatively with the absolute thymus weight, whereas the absolute adrenal weight was positively correlated with the bite score (H).