Characterization of macrophages from schizophrenia patients

Abstract

Genetic, epidemiological and post mortem studies have described an association between schizophrenia (SCZ) and the immune system. Microglia, the tissue-resident macrophages of the brain, not only play an essential role in inflammatory processes, but also in neurodevelopment and synapse refinement. It has therefore been hypothesized that aberrant functioning of these myeloid immune cells is involved in SCZ pathogenesis. Until now cellular research into the role of myeloid cells in SCZ has been limited to monocytes and functional assays are lacking. In this study we used monocyte-derived macrophages (mo-MΦs) as a model for macrophages and microglia in the CNS and examined two main functions: Inflammatory responses and expression and regulation of synapse refinement molecules. The expression of 24 genes involved in these key functions was assessed. Mo-MΦs were generated from 15 SCZ patients and 15 healthy controls. The cells were exposed to pro-inflammatory and anti-inflammatory stimuli (LPS, R848, IL-4 and dexamethasone), and the response was measured by qPCR and ELISA analyses. One of the genes of interest, P2RX7 that is associated with psychiatric diseases, was significantly reduced in expression after LPS stimulation in SCZ patients. None of the other assessed characteristics were different in this functional screen between mo-MΦs from SCZ patients compared to controls. Although these data suggest that overall the function of macrophages in SCZ is not impaired, further studies with larger groups that enable the possibility to study clinical subgroups and perform additional screenings to asses the full phenotype of the mo-MΦs are needed to strengthen this conclusion.

Fig. 1

mRNA expression levels of inflammatory response genes and cytokines secreted after inducing a pro-inflammatory or anti-inflammatory phenotype in mo-MΦs of SCZ patients () and controls (). mRNA levels of inflammatory response genes after 6-h stimulation with LPS or R848 were not different between patients (n = 15) and controls (n = 15) a Five or one samples gave undetermined values for IL6 expression in the control and patient group, respectively. The basal cytokine levels (b) and fold change values after pro-inflammatory stimulation (c LPS and R848) of IL-6, TNFα and IL-10 in the medium were similar between SCZ patients and controls. The base level of IL-6 was undetermined for one patient sample. At base level and after LPS stimulation two and three undetermined values were present for TNFα secretion in controls and patients samples, respectively. After R848 stimulation four and three samples gave undetermined values for controls and patients, respectively. Both in the patient and the healthy control group, three samples gave undetermined values for IL-10 secretion. There was no difference measured between the patient (n = 14) and control group (n = 15) on mRNA levels of inflammatory response genes after 72-h anti-inflammatory stimulation with IL-4 or dexamethasone (d). After IL-4 stimulation three samples gave undetected values in CD200R1 expression in the control group and five in the patient group. CD200R1 mRNA values of four control and seven patient samples were undetermined after dexamethasone stimulation. One patient sample was undetected for CD163 expression after IL-4 stimulation. mRNA expression was analyzed with qPCR and cytokine secretion with ELISA. The mRNA values a, d were first normalized (2^Δct) with GAPDH as reference gene after which they were divided by the value of the sample without stimulation (dotted line) resulting in the fold change. The median is depicted with a black bar. Mann–Whitney U tests were used to compare the SCZ patients with the healthy controls. UND undetermined value

Fig. 2

mRNA expression levels of synapse refinement receptors ITGAM (a), TREM2 (b) and P2RX7 (c) in mo-MΦs was compared between a non-stimulated state and after pro-inflammatory (n = 30) or anti-inflammatory (n = 29) phenotype induction. The pro-inflammatory phenotype was induced by 6-h stimulation with LPS or R848 and the anti-inflammatory phenotype was induced by 72-h stimulation with IL-4 or dexamethasone. ITGAM (a) and TREM2 (b) mRNA levels were significantly decreased after LPS (p < 0.01) and R848 (p < 0.01), increased after IL-4 (p < 0.01) and not altered after dexamethasone stimulation. P2RX7 (c) mRNA levels were not altered after dexamethasone stimulation, but significantly decreased after IL-4 (p < 0.01) and increased after LPS (p < 0.01) and R848 stimulation (p < 0.01). We did not distinguish between the schizophrenia patients () and healthy controls () in the statistical analyses. One patient sample gave an undetermined value for P2RX7 expression after dexamethasone stimulation. mRNA expression was analyzed with qPCR. All values were normalized (2^Δct) with GAPDH as the reference gene after which the value was multiplied by 1000 to ease visualization. Friedman’s ANOVA tests were used to compare the pro-inflammatory or anti-inflammatory responses to the non-stimulated state. (*p < 0.05; **p < 0.01)

Fig. 3

Pro-inflammatory or anti-inflammatory induced changes in mRNA expression levels of proteins important for synapse refinement compared between SCZ patients () and controls (). There was no difference in expression levels of the genes between patients (n = 15) and controls (n = 15) after the pro-inflammatory phenotype was induced by 6-h stimulation with LPS or R848 (a), except for P2RX7 that was significant lower expressed after LPS stimulation in patients versus controls (U = 65, p = 0.049). The anti-inflammatory phenotype was induced by 72-h stimulation with IL-4 or dexamethasone and was also not different between patients (n = 14) and controls (n = 15) (b). One patient sample gave an undetermined value for P2RX7 expression after dexamethasone stimulation. mRNA expression was analyzed with qPCR. All values were first normalized (2^Δct) with GAPDH as the reference gene after which they were divided by the value of the sample without stimulation (dotted line) resulting in the fold change. Mann–Whitney U tests were used to compare the SCZ patients with the healthy controls. UND undetermined value (*p < 0.05)