Neuroinflammation as common denominator in heart failure associated mental dysfunction
Gouweleeuw, Leonie
DOI:
10.33612/diss.122192415
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Gouweleeuw, L. (2020). Neuroinflammation as common denominator in heart failure associated mental
dysfunction: Studies in animal models. University of Groningen. https://doi.org/10.33612/diss.122192415
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
CHAPTER 3
Evidence for neuroinflammation
after myocardial infarction in a
mouse model
Leonie Gouweleeuw1, Christine Pol2, Warner S. Simonides2, Dominique P.V. de Kleijn3,
Regien G. Schoemaker1,4
1 Dept. of Molecular Neurobiology, University of Groningen, Groningen, The Netherlands 2 Dept. of Physiology, VU Medical Center, Amsterdam, The Netherlands
3 Dept. of Vascular surgery and Cardiology, University Medical Center Utrecht, The
Netherlands
4 Dept. of Cardiology, University Medical Centre Groningen, University of Groningen,
Groningen, The Netherlands Heart Mind 2017, 1(4):134-140
Abstract
Background: The cardiovascular system and central nervous system are known to influence
each other. Accordingly, neurological changes may occur after myocardial infarction (MI), which may be mediated by neuroinflammation. We investigated Tumor Necrosis Factor alpha (TNF-α) and microglia activation in post-MI neuroinflammation.
Methods: MI or sham surgery was induced in 28 male mice. Two weeks later, we performed
ECHOcardiography and dissected the brains for western blot on TNF-α and its receptors (n=10) or for immunohistochemical stainings for microglia, doublecourtin X (DCX) and TNF-α (n=18). Plasma was collected for the measurement of circulatory cytokines.
Results: The MI mice had an average infarct size of 38% of the left ventricle, heart failure was
confirmed by decreased fractional shortening and increased lung weight. Plasma cytokine levels were unaltered. In brains of MI mice there was a higher expression of TNF-α precursor protein, with trends for higher TNF-R1 and lower TNF-R2 expression. Furthermore, MI mice had more activated microglia in the inner blade of the dentate gyrus of the hippocampus. The amount of neurogenesis measured by DCX staining was unaltered.
Conclusions: Our mouse model of MI showed signs of persistent neuroinflammation as
indicated by raised levels of TNF-α precursor protein and an increased number of activated microglia in the hippocampus. The extent to which these neuroinflammatory hallmarks influence central nervous system functioning remains to be determined.
Introduction
Acute myocardial infarction (MI) evokes an inflammatory response important for healing and scar formation. This inflammatory response, however, is not restricted to the heart itself, as increased levels of circulating cytokines are observed that can last for weeks following the initial MI [1]. It has been suggested that this prolonged inflammation after MI leads to neuroinflammation, influencing the central nervous system. An increase in central nervous system cytokines can influence several processes to alter neuronal function, including apoptosis, oxidative stress, and metabolic processes [2]. Furthermore, neuroinflammation following physical injury implicated in behavioral outcomes including anxiety, depression, and cognitive impairment [3]. One of the initial cytokines upregulated after MI is Tumor Necrosis Factor alpha (TNF-α). TNF-α is produced locally by resident cardiac mast cells [4]. Circulating levels of TNF-α are raised significantly for weeks after the MI, with a peak at 4 weeks post MI in rats [5]. There is accumulating evidence that TNF-α influences the permeability of the blood-brain barrier at specific locations, allowing leukocyte infiltration and inflammation in the brain, as extensively reviewed by Liu et al., 2013 [6]. TNF-α was also found to be upregulated in the paraventricular nucleus (PVN) of the hypothalamus in MI rats [7]. The PVN is well known for its role in connecting peripheral sympathetic outflow to the body to oxytocin/vasopressin systems in higher brain areas, including the hippocampus and amygdala. Besides an upregulation of TNF-α, prolonged microglia activation was also found in the hypothalamus of infarcted rats, supporting a process of neuroinflammation [8-10]. Persistently increased numbers of microglia were also found in the hippocampus of mice after cardiac ischemia-reperfusion injury; this was linked to worse performance of the cardiac injury mice in hippocampal dependent learning tasks [11]. In the present study, we investigated hallmarks of inflammation and neurogenesis in several brain regions. Mice were subjected to myocardial infarction and plasma cytokines were measured to detect signs of peripheral inflammation, while neuroinflammation and neurogenesis was examined using western blot and immunohistochemical staining of brain tissue.
Materials and methods
Animals
Male 12 week old C57Bl/6J mice were used for this study (28 total), the brains of 10 mice were used for western blotting and the brains of the other 18 animals for immunohistochemical staining. The animals were sacrificed at 14 days post myocardial infarction. Animals were housed individually with food and water ad libitum. All experiments complied with the Guide for Care and Use of Laboratory Animals of the National Institutes of Health (NIH Publication no. 86-23, revised 1996) and were approved by the institutional animal care and use committee of the Vrije Universiteit University Medical Center Amsterdam.
Myocardial infarction and sacrifice
Mice were randomly assigned to either sham-surgery or MI. MI was induced by permanent ligation of the left coronary artery in isoflurane anesthetized animals, as described earlier [12, 13]. Sham animals were anesthetized and underwent thoracotomy without induction of MI. At sacrifice 14 days later, cardiac function and dimensions were obtained by ECHOcardiography. Cardiac perfusion with saline was performed, after which the heart, lungs and brain were collected. Blood was collected in EDTA-tubes. Hearts and lungs were weighed and the left ventricle was divided into infarcted and non-infarcted area, which were weighed separately. Infarct size was determined as the weight of the infarcted area of the left ventricle as a percentage of the entire left ventricle weight. The brains were either fixated in 4% paraformaldehyde or frozen in liquid nitrogen and stored at -80°C.
Western blotting
Frozen whole brain tissue was homogenized in lysis buffer (150mM NaCl, 100mM β-glycerophosphate, 50 mM HEPES pH 7.4, 0.2% NP-40, 50mM sodium fluoride, 15mM sodium pyrophosphate, 10mM EDTA, 5mM orthovanadate, 4mM EGTA, 1mM DTT, 1mM PMSF) to which protease and phosphatase inhibitors were added. After homogenization, samples were centrifuged at 11.000g for 20 minutes at 4°C. The pellet and supernatant fractions were separated for the analysis of membrane bound and soluble proteins, respectively. Both samples were sonicated on ice for one minute. Protein concentration was determined using a Bradford assay and samples were diluted with homogenization buffer to a concentration of 2µg/ml, with 20% sample buffer (300 mM Tris pH 6.8, 26% glycerol, 100g/L DTT, 0.06% Bromophenol Blue, 0.4% SDS). Protein samples of 40µg were loaded on 10% SDS-Page gels and blotted onto polyvinylidinedifluoride blotting membranes (Millipore). Membranes were blocked in 5% non-fat milk to block aspecific binding. Blots were incubated overnight with primary antibodies against TNF-α, TNF receptor 1 or TNF receptor 2 (all antibodies from cell-signaling and used at 1:1 dilution in I-block from ThermoFisher Scientific). Β-actin was used as a housekeeping gene to control for equal loading (MP Biomedicals, 1:1,000,000). After several wash steps, blots were incubated for 1 hour with an appropriate horseradish peroxidase labeled secondary antibody. Bands were visualized with a commercially available enhanced chemiluminescence kit (Pierce Biotechnology). Pictures were taken using the Chemi Doc XRS system (Bio-Rad). Image Lab software was used for analysis of the bands (Bio-Rad).
Immunohistochemistry
Brain tissue was cryo-protected with 30% w/v sucrose solution, frozen in liquid nitrogen and stored at -80°C. Tissue was cut in 20µm sections on a cryostat and stored in 0.01M PBS, pH 7.4 with 0.1% Sodium Azide until use. The sections were blocked with a 0.1% H2O2 solution and incubated with primary antibody against Iba-1 (Wako pure chemical industries, 1:2500) for microglia, Doublecourtin X (DCX, Santa Cruz, 1:1000) for immature neurons or TNF-α (Abcam, 1:1000) to detect TNF-α expression. After 3 PBS washes, sections were incubated with biotinylated secondary antibodies (either rabbit-anti-goat or goat-anti-rabbit, both Jackson and used at 1:500 dilution). Staining was visualized by incubation with an avidine-biotine complex kit (Vector laboratories) followed by incubation with 3,3’ diaminobenzidine (Sigma) activated with hydrogen peroxide. After dehydration and mounting, pictures were taken with a Leica microscope at 100x magnification for DCX and Iba-1 staining and a 200x magnification for the TNF-α staining. The surface area of DCX stained cells was measured using ImagePro Plus 6.0.0.26 software (Media Cybernatics, Inc. Rockville, USA). Microglia activation was measured as the number of microglia profiles per high power field, as well as an increased ratio of cell body size compared to the surface area of the processes, according to methods described by Hovens et al.,[14] using ImagePro Plus 6.0.0.26 software. TNF-α positive cells were manually counted. Brain sections from TNFα-/- mice from our breeding
facility were used as a negative control.
Plasma cytokine and chemokine measurements
Mouse plasma cytokine levels of IL-1β, IL-2, IL-6, TNF-α, IFN-γ, KC, MCP-1, MiP-1β, RANTES were measured using a custom made 9-plex assay using a luminex multiplex assay (Bio-Rad). ICAM-1 was measured in a single-plex luminex assay (Biorad). The provided standard curves were used for determining cytokine and chemokine concentrations in plasma on a Bio-Plex 200 suspension array system (Bio-Rad).
Statistics
All statistical analyses were performed with SPSS for Windows version 20. Data of mice with small MI sizes (<20% of left ventricle) were excluded from analysis. Independent samples T-test was used to determine differences between the sham operated and large MI group. When appropriate, Welch’s correction was used in case of unequal variances. A p-value of <0.05 was considered significant. All data are presented as average ± SEM.
Results
Cardiac dysfunction in MI mice
Sham operated and MI mice were sacrificed 14 days after surgery. As a measure for cardiac function and contractility, ejection fraction and fractional shortening were measured, respectively. MI mice showed a decreased fractional shortening compared to sham operated mice. A summary of the heart failure data can be seen in Table 1. To further verify cardiac dysfunction in MI mice, heart and lung weight were determined, infarct size was measured, and fractional shortening was obtained. Increased heart weight and lung weight corrected for tibia length were apparent in MI animals, indicating cardiac hypertrophy and lung edema. Infarct size in the MI group was 38 ± 9% of the left ventricle.
Plasma cytokine levels
The plasma levels of 10 cytokines and chemokines were determined in a subset of animals with a luminex assay and shown in Figure 1. Plasma levels for IL-1β could not be detected for most of the samples and have not been analyzed. The assay showed no increases in cytokine levels at 14 days post-MI. A trend towards decreased expression may be seen for IL-6 and the chemokine KC (CXCL-1), however, these results were not statistically significant.
Table 1: characterization of experimental groups (mean + SEM)
Sham MI p-value
N 16 14
Heart weight/tibia length 61.4±3.38 87.0±2.03 0.006
Lung weight/tibia length 78.2±1.45 101.1±7.29 0.008
Lung wet/dry ratio 4.67±0.03 4.84±0.04 0.001
LV-EDD 3.4±0.12 5.2±0.13 <0.001
LV-ESD 1.9±0.09 4.3±0.20 <0.001
FS 43.9±1.15 17±2.06 <0.001
Infarct size - 38±9
-LV-EDD: left ventricle end diastolic diameter, LV-ESD: left ventricle end systolic diameter, FS: fractional shortening.
Altered levels of TNF-α and receptors in the brain
Whole brain homogenates were used to determine the expression levels of TNF-α and its receptors with western blot. The pellet fraction and supernatant fraction were analyzed separately. All expression data are corrected for the expression of β-actin. Of 6 MI mice tested, 2 mice had smaller infarcts (<10% of left ventricle) and were excluded from analyses.
Results for the expression of TNF-α and receptors are shown in Figure 2. We found bands at two molecular weights for TNF-α, the precursor protein at 26 kDa and the secreted protein at 17 kDa. The 26 kDa TNF-α protein was significantly increased in mice with MI compared to sham in the pellet fraction, whereas not altered in the supernatant fraction. No significant differences in the expression of the 17 kDa TNF-α protein expression were observed. A three times higher expression of the proinflammatory TNF-R1 in the pellet fraction of the MI group was observed with unaltered expression levels in the supernatant fraction. For the TNF-R2 protein we found no differences in expression in the pellet fraction, but the supernatant had a statistically significant decrease in the expression of TNF-R2 for MI mice.
Figure 1: Expression levels of the cytokines and chemokines IL-2, IL-6, IFN-γ, KC, MCP, MIP-1β, RANTES, TNF-α and ICAM in sham animals (white bars) and MI animals (black bars) 14 days after MI. For the markers IL-2, IFN-γ, KC, MCP, MIP-1b and ICAM n=6 for both groups. For IL-6, n=10 and n=16 for sham and MI mice. For RANTES, n=12 and 17 for sham and MI mice and for TNF-alpha n=11 and 17 for sham and MI mice. Data are shown as relative expression levels where sham animals are set to 1. Error bars represent SEM.
To detect brain areas expressing TNF-α, staining of brain sections was performed. TNF-α positive cells were found in the piriform cortex, prefrontal cortex and PVN, but no positive signal was observed in the hippocampus. No positive cells were found in sections from TNF-α-/-
mice. When counting the number of cells per field of focus, no differences were found between sham and MI mice. Results of the TNF-α staining are shown in Figure 3.
Increase in microglia number and activation in the dentate gyrus of the hippocampus
Brain sections of sham operated and MI mice were stained for DCX and for Iba-1 to visualize immature neurons and microglia, respectively. No difference in the expression of DCX between sham and MI animals was observed, as shown in Figure 4A.
Parameters obtained from the microglia staining are presented in Figure 4 C-E. We found a significant increase in the number of microglia in the inner blade of the dentate gyrus (Figure 4C-D). Activation of microglia, as an increase in the ratio of cell body/processes appeared significantly higher in this area (Figure 4E). In the outer blade of the dentate gyrus, as well as in the amygdala microglia activation may occur, however this difference was not statistically significant (p=0.066 and 0.133, respectively).
Discussion
In this study we investigated inflammatory changes following MI. We focused on TNF-α expression in particular, as this cytokine shows upregulation after myocardial infarction [7], and was previously found upregulated in the hypothalamus following MI [7]. Neuroinflammation in different brain areas was measured as microglia activation and TNF-α expression, while neurogenesis was obtained from DCX staining of immature neurons in the dentate gyrus of the hippocampus.
Myocardial damage and plasma cytokines
MI surgery led to an average infarct size of 38% of the left ventricle. Compared to sham animals, MI mice had significantly higher heart weight/tibia length and lung weight/tibia length. Cardiac ECHO revealed higher end-diastolic and end-systolic diameters and lower fractional shortening in MI mice. Results support MI-induced cardiac dysfunction and heart failure.
Despite the considerable cardiac damage, induction of MI in this study failed to show alterations of plasma cytokine levels, including TNF-α, 14 days after surgery. This is, however, in line with a previous study where most cytokines were unaltered compared to sham 6 or 7 days after MI in mice [15], but does not exclude increased circulating inflammatory markers at different time points.
TNF-α
In contrast to the absence of peripheral inflammation, we did find signs of neuroinflammation 14 days after MI. First, an increased expression of the 26 kDa precursor protein of TNF-α was found in MI animals. This was accompanied by trends towards a higher TNF-R1 expression in the membrane fraction and a lower expression of TNF-R2 in the supernatant fraction. A previous study in myocardial ischemia found that TNF-α promoted cardiac damage through TNF-R1 and protected against injury through TNF-R2 [16]. Similarly, in a model for traumatic brain injury, TNF-R1 promotes damage while TNF-R2 seems to serve a protective role [17].
Considering this information, a higher expression of the ligand (TNF-α), with a shift in receptor expression towards the pro-inflammatory R1 at the expense of the cytoprotective R2 at 14 days after MI, may indicate a chronic proinflammatory state in our model.
Figure 2: Western blot results of homogenized brain tissue from sham (white bars), and MI (black bars). N=4 per
group. *= p<0.05
Immunohistochemical staining against TNF-α revealed the presence of positive cells in the piriform cortex, PVN and prefrontal cortex (PFC), but not in the hippocampus. When analyzing the staining per area, no statistically significant differences were found between sham and MI, although in the PFC about twice as many TNF-α+ cells were observed. Analysis of gene expression by RNA sequencing in prefrontal cortex and hippocampus revealed changes in genes related to inflammation, including a PFC-specific signal of Nr4a1 [18]. This appeared the only gene upregulated in the heart as well as in PFC in MI mice, and may reflect the role of macrophages in these tissues [19].
Figure 3: TNF-α staining in piriform cortex (PIR), paraventricular nucleus of the hypothalamus (PVN) and prefrontal cortex (PFC) in sham, MI and TNF-α-/- mice.
Microglia
The altered TNF-α signaling observed could be attributed to either a general distributed proinflammatory state or, more likely, a localized neuroinflammatory response. To explore possible localized neuroinflammation, we analyzed microglia activation in relevant brain areas. Only in the hippocampus, we found an increase in the number and activation of microglia. The hippocampus has been reported as a brain area particularly sensitive to microglia activation [20, 21], therefore, it is not surprising that microglia activation was found specifically in this region. Microglia activation in the hippocampus after MI has been reported previously [11, 22]. Other brain regions where changes in microglia were observed following MI are the prefrontal cortex [22] and PVN [8]. However, we were not able to reproduce these findings in our mouse model.
Figure 4: Results from immunohistochemical stainings for doublecourtin X (a and b) and microglia (Iba-1, c-e). (a)
Doublecourtin staining for the dentate gyrus of the hippocampus (10 × 10). (c) Microglia in the inner blade of the dentate gyrus (10 × 40) * P < 0.05
Neurogenesis
DCX staining was performed to investigate neurogenesis in our MI model. Neurogenesis was previously found to be increased in a rat model 14 days after MI [23], whilst in a mouse model of cardiac ischemia reperfusion injury neurogenesis was decreased 72 hours post injury [11]. Other studies have found increased apoptosis in the hippocampus in experimental MI models [24, 25].
When staining for DCX as marker for neurogenesis, we were not able to show differences in expression of this marker in MI animals compared to sham, despite an increase found in microglia activation in this region. Whether microglia activation in the hippocampus is linked to either depression or neurogenesis after MI is unclear. It has previously been mentioned that IL-1β release from microglia contributes to decreased neurogenesis [26]. Also, a previous study in surgery rats showed that microglia activation was accompanied by a decrease in DCX [27].
Limitations
While our study yields interesting results, there are several limitations. First of all, the sample size is quite small, especially since the groups were halved in order to separate processing for western blot and immunohistochemistry. Further, we chose a mouse model for myocardial infarction. While this is a valid model previously described in cardiovascular studies, most of the previous results concerning neuroinflammation following myocardial infarction have been published in rat models, making it difficult to compare results. Lastly, we only investigated signs of inflammation at 14 days post-MI. Studies in rats showed a distinct time course of changes in plasma TNF-α. Evaluating more time points could have given us additional insight.
Conclusions
In this study we found signs of neuroinflammation 14 days after MI in mice, in the absence of altered plasma cytokine levels. TNF-α precursor protein was elevation in brains of MI mice, which was accompanied by a trend for higher TNF-R1 expression and lower TNF-R2 expression. This suggests a shift towards a more proinflammatory state. However, staining for TNF-α did not give a clear answer on localization within the brain. Microglia number and activation score were specifically increased in the hippocampus, but were not associated with decreased neurogenesis. More research is needed to determine whether anti-inflammatory treatment is neuroprotective in models of myocardial infarction and whether this could lead to new treatment options for patients.
References
1. Mizia-Stec K, Gasior Z, Zahorska-Markiewicz B, Janowska J, Szulc A, Jastrzebska-Maj E, et al. Serum tumour necrosis factor-alpha, interleukin-2 and interleukin-10 activation in stable angina and acute coronary syndromes. Coron Artery Dis. 2003 Sep;14(6):431-8.
2. Anisman H, Merali Z, Poulter MO, Hayley S. Cytokines as a precipitant of depressive illness: Animal and human studies. Curr Pharm Des. 2005;11(8):963-72.
3. Quan N. In-depth conversation: Spectrum and kinetics of neuroimmune afferent pathways. Brain Behav Immun. 2014 Aug;40:1-8.
4. Frangogiannis NG, Lindsey ML, Michael LH, Youker KA, Bressler RB, Mendoza LH, et al. Resident cardiac mast cells degranulate and release preformed TNF-alpha, initiating the cytokine cascade in experimental canine myocardial ischemia/reperfusion. Circulation. 1998 Aug 18;98(7):699-710.
5. Francis J, Zhang ZH, Weiss RM, Felder RB. Neural regulation of the proinflammatory cytokine response to acute myocardial infarction. Am J Physiol Heart Circ Physiol. 2004 Aug;287(2):H791-7.
6. Liu H, Luiten PG, Eisel UL, Dejongste MJ, Schoemaker RG. Depression after myocardial infarction: TNF-alpha-induced alterations of the blood-brain barrier and its putative therapeutic implications. Neurosci Biobehav Rev. 2013 May;37(4):561-72.
7. Francis J, Chu Y, Johnson AK, Weiss RM, Felder RB. Acute myocardial infarction induces hypothalamic cytokine synthesis. Am J Physiol Heart Circ Physiol. 2004 Jun;286(6):H2264-71.
8. Badoer E. Microglia: Activation in acute and chronic inflammatory states and in response to cardiovascular dysfunction. Int J Biochem Cell Biol. 2010 Oct;42(10):1580-5.
9. Dworak M, Stebbing M, Kompa AR, Rana I, Krum H, Badoer E. Sustained activation of microglia in the hypothalamic PVN following myocardial infarction. Auton Neurosci. 2012 Aug 16;169(2):70-6.
10. Rana I, Stebbing M, Kompa A, Kelly DJ, Krum H, Badoer E. Microglia activation in the hypothalamic PVN following myocardial infarction. Brain Res. 2010 Apr 22;1326:96-104.
11. Evonuk KS, Prabhu SD, Young ME, DeSilva TM. Myocardial ischemia/reperfusion impairs neurogenesis and hippocampal-dependent learning and memory. Brain Behav Immun. 2017 Mar;61:266-73.
12. Pol CJ, Muller A, Zuidwijk MJ, van Deel ED, Kaptein E, Saba A, et al. Left-ventricular remodeling after myocardial infarction is associated with a cardiomyocyte-specific hypothyroid condition. Endocrinology. 2011 Feb;152(2):669-79.
13. de Waard MC, van der Velden J, Bito V, Ozdemir S, Biesmans L, Boontje NM, et al. Early exercise training normalizes myofilament function and attenuates left ventricular pump dysfunction in mice with a large myocardial infarction. Circ Res. 2007 Apr 13;100(7):1079-88.
14. Hovens, I.B., Nyakas, C., Schoemaker, R.G. A novel method for evaluating microglial activation using ionized calcium-binding adaptor protein-1 staining: Cell body to cell size ratio. NN journal. 2014;1:82-8
15. Vistnes M, Waehre A, Nygard S, Sjaastad I, Andersson KB, Husberg C, et al. Circulating cytokine levels in mice with heart failure are etiology dependent. J Appl Physiol (1985). 2010 May;108(5):1357-64.
16. Zhang Y, Zhao J, Lau WB, Jiao LY, Liu B, Yuan Y, et al. Tumor necrosis factor-alpha and lymphotoxin-alpha mediate myocardial ischemic injury via TNF receptor 1, but are cardioprotective when activating TNF receptor 2. PLoS One. 2013 May 21;8(5):e60227.
17. Longhi L, Perego C, Ortolano F, Aresi S, Fumagalli S, Zanier ER, et al. Tumor necrosis factor in traumatic brain injury: Effects of genetic deletion of p55 or p75 receptor. J Cereb Blood Flow Metab. 2013 Aug;33(8):1182-9. 18. Frey A, Popp S, Post A, Langer S, Lehmann M, Hofmann U, et al. Experimental heart failure causes depression-like
behavior together with differential regulation of inflammatory and structural genes in the brain. Front Behav Neurosci. 2014 Oct 31;8:376.
19. Pei L, Castrillo A, Tontonoz P. Regulation of macrophage inflammatory gene expression by the orphan nuclear receptor Nur77. Mol Endocrinol. 2006 Apr;20(4):786-94.
20. Yirmiya R, Goshen I. Immune modulation of learning, memory, neural plasticity and neurogenesis. Brain Behav Immun. 2011 Feb;25(2):181-213.
21. Vinet J, Weering HR, Heinrich A, Kalin RE, Wegner A, Brouwer N, et al. Neuroprotective function for ramified microglia in hippocampal excitotoxicity. J Neuroinflammation. 2012 Jan 31;9:27,2094-9-27.
22. Rinaldi B, Guida F, Furiano A, Donniacuo M, Luongo L, Gritti G, et al. Effect of prolonged moderate exercise on the changes of nonneuronal cells in early myocardial infarction. Neural Plast. 2015;2015:265967. 23. Hwang IK, Yoo KY, Han TH, Lee CH, Choi JH, Yi SS, et al. Enhanced cell proliferation and neuroblast
differentiation in the rat hippocampal dentate gyrus following myocardial infarction. Neurosci Lett. 2009 Feb 6;450(3):275-80.
24. Kaloustian S, Bah TM, Rondeau I, Mathieu S, Lada-Moldovan L, Ryvlin P, et al. Tumor necrosis factor-alpha participates in apoptosis in the limbic system after myocardial infarction. Apoptosis. 2009 Nov;14(11):1308-16. 25. Wann BP, Bah TM, Kaloustian S, Boucher M, Dufort AM, Le Marec N, et al. Behavioural signs of depression and apoptosis in the limbic system following myocardial infarction: Effects of sertraline. J Psychopharmacol. 2009 Jun;23(4):451-9.
26. Gemma C, Bachstetter AD, Cole MJ, Fister M, Hudson C, Bickford PC. Blockade of caspase-1 increases neurogenesis in the aged hippocampus. Eur J Neurosci. 2007 Nov;26(10):2795-803.
27. Hovens IB, Schoemaker RG, van der Zee EA, Absalom AR, Heineman E, van Leeuwen BL. Postoperative cognitive dysfunction: Involvement of neuroinflammation and neuronal functioning. Brain Behav Immun. 2014 May;38:202-10.
