Interleukin-6 is both necessary and sufficient to produce perioperative neurocognitive disorder in mice

Abstract

Background: Perioperative neurocognitive disorders (PND) result in long-term morbidity and mortality with no effective interventions available. Because interleukin-6 (IL-6), a pro-inflammatory cytokine, is consistently up-regulated by trauma, including after surgery, we determined whether IL-6 is a putative therapeutic target for PND in a mouse model.

Methods: Following institutional approval, adult (12-14 weeks) male C57/Bl6 mice were pretreated with the IL-6 receptor (IL6R) blocking antibody tocilizumab prior to open tibia fracture with internal fixation under isoflurane anaesthesia. Inflammatory and behavioural responses in a trace fear conditioning (TFC) paradigm were assessed postoperatively. Separately, the effects of IL-6 administration or of depletion of bone marrow-derived monocytes (BM-DMs) with clodrolip on the inflammatory and behavioural responses were assessed. Blood brain barrier disruption, hippocampal microglial activation, and infiltration of BM-DMs were each assessed following IL-6 administration.

Results: The surgery-induced decrement in freezing time in the TFC assay, indicative of cognitive decline, was attenuated by tocilizumab (P<0.01). The surgery-induced increase in pro-inflammatory mediators was significantly reduced by tocilizumab. Exogenously administered IL-6 significantly impaired freezing behaviour (P<0.05) and up-regulated pro-inflammatory cytokines; both responses were prevented by depletion of BM-DMs. IL-6 disrupted the blood brain barrier, and increased hippocampal activation of microglia and infiltration of BM-DMs.

Conclusions: IL-6 is both necessary and sufficient to produce cognitive decline. Following further preclinical testing of its perioperative safety, the IL6R blocker tocilizumab is a candidate for prevention and/or treatment of PND.

Keywords: cognitive dysfunction; cytokines; interleukin-6 (IL-6); surgery; tocilizumab.

Copyright © 2017 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.

Figures

Fig 1

Experimental protocol. (A) Four groups of randomly-assigned mice were treated with either saline or an intraperitoneal (i.p.) injection of tocilizumab, an antibody directed against the interleukin-6 (IL-6) receptor (IL6R) followed by tibial fracture or sham surgery 2 h later. Thirty min prior to tibia fracture/sham, the training session for trace fear conditioning (TFC) was performed and context testing was undertaken 72 h later. (B) Four groups of mice were prepared as for A. Six h after surgery or sham, mice were killed and blood and brain were harvested. (C) Mice were randomly assigned into four groups and treated with i.p. injection of clodrolip in a liposome emulsion or liposome emulsion only 1 h before TFC training session. This was followed by an i.p injection of IL-6 (IL6A) or saline control after training session and thereafter context testing was performed 72 h later. (D) Four groups of mice were prepared as for C. Six h after IL6A, mice were killed and blood and brain were harvested.

Fig 2

Pre-emptive blockade of interleukin-6 (IL-6) receptor (IL6R) prevents postoperative decrement in freezing behaviour. Four groups of randomly-assigned mice (n=10) were treated with either saline or an intraperitoneal injection of tocilizumab, an antibody directed against the IL-6 receptor (IL6R) followed by tibia fracture or sham surgery 2 h later. Thirty min prior to tibia fracture/sham, the training session for TFC was performed and context testing for freezing behaviour was undertaken 72 h later. Data are expressed as means (SEM) and compared by one-way analysis of variance and Student-Newman-Keuls test. *P<0.01.

Fig 3

IL-6 Receptor blocker reduces inflammatory response after surgery. Four groups of randomly-assigned mice (n=8) were treated with either saline or an intraperitoneal (i.p.) injection of tocilizumab, an antibody directed against the interleukin-6 (IL-6) receptor (IL6R) followed by tibia fracture or sham surgery 2 h later. Six h after surgery, mice were killed and blood and brain were harvested and assayed for circulating IL-6 (A), circulating MCP-1 (B), hippocampal IL-6 (C), and hippocampal monocyte chemoattractant protein-1 (MCP-1) (D). Data were analysed by one-way analysis of variance and Student-Newman-Keuls test. *P<0.05; **P<0.01.

Fig 4

Depletion of bone marrow-derived monocytes prevents interleukin-6 (IL-6) induced decrement in freezing behaviour. Four groups each of randomly-assigned mice (n=15) were treated with either saline, IL-6 (IL6A), liposomes containing clodronate (clodrolip)+IL6A, or liposomes+IL6A. One h later the training session for trace fear conditioning (TFC) was performed and context testing for freezing behaviour was undertaken 72 h later. Data are expressed as means (SEM) and compared by one-way analysis of variance and Student-Newman-Keuls test. *P<0.01.

Fig 5

Depletion of bone marrow-derived monocytes prevents interleukin-6 (IL-6) induced inflammatory response. Two groups of randomly-assigned mice (n=8) were treated with either saline or an intraperitoneal (i.p.) injection of IL-6 (IL6A). Two groups of randomly-assigned mice (n=8) were pre-treated with liposomes containing clodronate (clodrolip) or liposomes without clodronate followed by injection of IL-6 (IL6A) 1 h later. Six h after treatment, mice were killed and blood and brain were harvested and assayed for circulating IL-6 (A), circulating IL-1β (B), hippocampal IL-6 (C), and hippocampal IL-1β (D), circulating MCP-1 (E), and hippocampal monocyte chemoattractant protein-1 (MCP-1) (F). Data were analysed by one way analysis of variance followed by Student-Newman-Keuls test. *P<0.05; **P<0.01.

Fig 6

Peripheral injection of interleukin-6 (IL-6) induces disruption of the blood brain barrier. Two groups of randomly-assigned mice (n=7) were treated with vehicle (control) or 50 μg kg−1 IL-6. Six h after treatment, mice were sacrificed and the brains were harvested for immunoblotting of albumin expression. Data are expressed as mean (SD) relative to control and were analysed by two way t-test. *P<0.05.

Fig 7

Peripheral injection of interleukin-6 (IL-6) induces hippocampal microglia activation and recruitment of CCR2-expressing cells. (A) Representative immunofluorescence images taken from CX3CR1-GFP/+: CCR2-RFP/+ mice 6 h following intraperitoneal injection of 50 μg kg−1 IL-6, showing the matching of anatomical locations using nuclear counterstaining with 4-6-diamidino-2-phenylindole (DAPI), infiltration of CCR2-expressing cells (red), and increased Iba1 expression (grey) among hippocampal microglia in response to IL-6 injection (scale bars=50 μm). Iba1+ cells also exhibit a transition from a quiescent morphology to one indicative of inflammatory activation showing increased size and fluorescence intensity (inset). (B) Quantification of data from (A) showing CX3CR1+, CCR2+, and Iba1+ cell number. (C–D) Quantification of size (C), and fluorescence intensity (D) of Iba1- expressing cells in anatomically matched hippocampal sections from the same groups of mice represented in A (n=5). *P<0.05 vs vehicle.