The illustration shows events after ischemic injury of either the brain or the heart that lead to accelerated disease progression in atherosclerotic plaque (some events are experimentally proven in mice, while others are still hypothetical). The enlarged inset depicts processes in the bone marrow microenvironment after myocardial infarction (MI). Here, niche cells provide signals that regulate hematopoietic stem cell activity, retention, and leukocyte production. After MI, increased sympathetic nervous signaling releases noradrenaline in the bone marrow niche, which binds to β3 adrenoreceptors on niche cells. These withdraw the soluble factor stromal cell–derived factor 1 (SDF-1), which results in increased hematopoietic progenitor cell activity and emigration to extramedullary sites. Similar processes may be active after stroke. Increased production of leukocytes then feeds an expanded pool of circulating monocytes, which are recruited not only to the injured brain or myocardial infarct but also to the atherosclerotic plaque in higher numbers, accelerating plaque growth and vulnerability. This feedback loop may cause the high clinical reoccurrence rates of MI and stroke. CAR indicates CXCL12 abundant reticular cell; HSC, hematopoietic stem cell; HSPC, hematopoietic stem and progenitor cells; MSC, mesenchymal stem cell; MΦ, macrophage; and SCF, stem cell factor.
Courties G, Moskowitz MA, Nahrendorf M. The Innate Immune System After Ischemic InjuryLessons to Be Learned From the Heart and Brain. JAMA Neurol. 2014;71(2):233–236. doi:10.1001/jamaneurol.2013.5026
Innate immune cells are critically involved in ischemic complications of atherosclerosis. While new insight emerged on the origin and role of leukocytes in steady state, the knowledge about myeloid cells’ sources, functions, and fate after stroke is limited. In our review, we highlight open questions in this important area while examining potential parallels in the immune response after stroke and myocardial infarction. We stress the need to better understand systemic interactions between ischemic tissue, immunity, and hematopoiesis, as turnover of leukocytes in inflammatory sites can be rapid, and cell production and supply may serve as future therapeutic targets to modulate inflammation in the vessel wall, brain, and heart.
Stroke is the third most common cause of death in the United States, and most strokes are due to thrombotic or embolic complications of atherosclerosis. One in four strokes are recurrent events,1 highlighting that both primary and secondary prevention are currently insufficient.2 While it is increasingly agreed that innate immune cells importantly contribute to atherosclerosis and its ischemic complications, the role of leukocytes, their subsets, sources, and fates after stroke are incompletely understood. Ischemic stroke and myocardial infarction (MI) have in common that the sustained tissue injury is sterile and that it is caused by a lack of oxygen. Distinct differences include the phenotype of injured cells and tissues as well as the nature and timing of signals from ischemic brain and heart. Because both MI and ischemic stroke are caused by atherosclerosis, we suspect that there are many similarities. Therefore, the comparison of the immune response to these 2 most deadly complications of vascular disease may be useful. In our review, we highlight open questions regarding the systemic innate immune response after stroke, relating them to recent insight obtained after MI.
Parabiosis experiments revealed that in the steady state, microglia primarily derive from local progenitors rather than from circulating leukocytes.3 In response to stroke, microglia are rapidly activated and develop a proinflammatory phenotype.4 Once brain tissue is compromised due to ischemia, the injury also triggers a systemic inflammatory response that contributes to lesion maturation and the removal of dead or dying cells.5,6 In patients7,8 and mice9,10 with stroke, acutely elevated blood counts of innate immune cells such as neutrophils and monocytes parallel data obtained after MI (reviewed by Swirski and Nahrendorf11). These blood leukocytes are recruited in large numbers to the ischemic brain, where they have a critical role in wound healing but may also contribute to reperfusion injury.5,12 As seen in MI,11 the ischemic brain first recruits neutrophils and later monocytes.4 However, in contrast to the ischemic heart, microglia substantially contribute to the cellular inflammatory response in the brain.4 Limited data are available on the role of monocyte subsets in inflammation, healing, and resolution of inflammation after ischemic brain injury. In the infarcted mouse heart,13 inflammatory Ly6Chigh monocytes are recruited first via CCL2/CCR2 and dominate the first 3 days after injury. The sources of inflammatory cytokines are Ly6Chigh monocytes, which pursue proteolytic and phagocytic removal of necrotic tissue. Likely, these cells give rise to M1 macrophages with similar proinflammatory functions. Starting on day 4 after MI, an inflammation resolution phenotype emerges, as Ly6Clow monocytes and macrophages are recruited via CX3CR1 to orchestrate tissue repair. These cells regulate angiogenesis and extracellular matrix production but also continue phagocytosis of tissue debris. Blocking either monocytic phase impairs infarct healing and promotes heart failure in mice.13 Several lines of evidence suggest analogous roles for monocyte and macrophage subsets after stroke. A parallel temporal pattern of inflammatory and proresolution macrophage phenotype occurs in murine brain after middle cerebral artery occlusion.14 Contrasting MI data, a study14 reports that Ly6Clow cells are not recruited separately via CX3CR1 but rather derive from CCR2+ Ly6Chigh monocytes. Monocyte depletion increases hemorrhagic conversion of ischemic stroke, likely due to delayed myeloid cell repair functions.14 On the other hand, deletion of CCR2 or its CCL2 ligand in mice results in smaller brain infarcts, together with decreased infiltration of monocytes and macrophages and proinflammatory cytokine production.15,16 These data point to potentially harmful as well as helpful functions of monocytes and macrophages after stroke, highlighting the necessity to better understand the role of subsets, timing, inflammation resolution, and magnitude of innate immune responses in the injured brain.
In atherosclerotic plaques, leukocytes contribute decisively to growth, inflammation, instability, and rupture. In patients, ischemia often results from artery-occluding atherosclerotic plaque and thus occurs in a setting of chronic inflammation that generated the vulnerable, ruptured plaque in the first place. The immune response to ischemic heart and brain in the setting of atherosclerosis may be fundamentally different—we hypothesize exaggerated—when compared with external wounding after trauma in an otherwise healthy person. Excessive levels of circulating inflammatory monocytes hamper resolution of inflammation, impair infarct healing, and cause heart failure in ApoE−/− mice after coronary ligation.17 Thus, preexisting chronic inflammation alters critical signals compared with a healthy steady state, and the raised systemic immune activity associated with atherosclerosis may impair the resolution of inflammation after ischemic brain injury. If stroke results from atherosclerosis, inflammation resolution in the ischemic brain may be disturbed and harmful functions of inflammatory leukocytes may impair outcome. This hypothesis should be tested experimentally. Of note, increased leukocyte and monocyte counts in the blood correlate with post-MI heart failure progression in clinical studies,18 and the blood level of the CD14highCD16− monocyte subset is associated with poor outcome and increased mortality in patients with stroke.19
In sites of acute inflammation, the turnover of innate immune cells, especially neutrophils, monocytes, and macrophages, may be surprisingly rapid. Fate-mapping studies determined that even 5 days after ischemic myocardial injury, myeloid cells are recruited in high numbers to the healing tissue, accounting for their relatively short infarct residence time of only 20 hours.20 Similar numbers are not available for stroke; however, the leukocyte turnover may also be rapid. It will be important to understand local cell kinetics in the brain acutely after stroke, as high leukocyte turnover and ongoing recruitment necessitate increased leukocyte supply and production, all of which could be targeted therapeutically.
An interesting feature shared by ischemic brain and injured heart is a transient decrease in spleen size, likely reflecting the organ’s release of leukocytes.21,22 Studies in animals splenectomized prior to ischemic insult of the brain or heart showed a decreased infiltration of innate immune cells into inflamed tissues.20,23 More recently, a lower number of splenic monocytes was described in an autopsy study of patients with acute MI.24 A decrease in spleen size has been documented in patients with acute stroke.25 Contrary to the systemically increased levels of innate immune cells, poststroke inflammation is accompanied by a severe loss of lymphocytes in blood and spleen. Lymphopenia after stroke most likely results from lymphocyte apoptosis, which may contribute to the reduction in the spleen’s size9 and possibly to compromised immunity.
Because the heightened demand for leukocytes in the ischemic lesion quickly exhausts ready-made cells in the body’s reservoirs (ie, blood, bone marrow, and spleen), the hematopoietic system likely increases cell production. In mice with MI, we observed transfer of leukocyte progenitors from the bone marrow to the spleen as well as splenic monocyte production20,26 (Figure). Despite the health burden caused by stroke and MI, surprisingly little is known about how the bone marrow compartment reacts to ischemic injury and how it is alerted after MI or stroke. The widespread systemic inflammatory response in both patients7 and animal models27 argues for an activation of the hematopoietic system after stroke. The recently reported increased hematopoietic stem and progenitor cell activation in mice and in patients with MI26 is likely generalizable, although its impacts on the brain, heart, and vasculature are probably organ and tissue specific. Future studies will investigate hematopoietic stem cell behavior and their progeny’s fate after ischemic brain injury and examine the pathways that regulate bone marrow activity after stroke.
Long-range signals transfer information from the site of injury to the site of innate immune cell production. These could be transmitted in the bloodstream in the form of intracellular danger signals released from dying cells in the wound, acting as receptor ligands on progenitor cells or on the hematopoietic niche (Figure). Alternatively, signals could be delivered via extravascular routes through the fibers of the sympathetic nervous system. Noradrenaline, released from sympathetic nerves in the bone marrow after MI, binds to β3 adrenergic receptors expressed on mesenchymal stromal cells, providing the microenvironmental cues in hematopoietic tissue after MI.26 Identifying and inhibiting those signals may help to curtail leukocyte overproduction and oversupply, as both are likely rate limiting for inflammation in the brain, heart, and artery wall.
The cells produced in response to ischemia not only travel to the injury site but also may be diverted to atherosclerotic lesions in the arterial wall. Similar cell types, especially neutrophils and inflammatory monocytes, are early responders after MI and stroke as well as major instigators of inflammatory atherosclerosis. In parallel with the high reoccurrence rates in patients with ischemic complications of atherosclerosis, we observed that MI as well as stroke accelerate atherosclerosis progression in ApoE−/− mice.26 If we improve our understanding of the putative cross talk between the immune and hematopoietic systems with injured brain and heart (Figure), we may expand our clinical ability to prevent recurrent ischemia. The development of new therapeutic strategies modulating the hematopoietic system with translatable pharmacological interventions may present a crucial step toward secondary prevention of stroke and MI. On this path, it will likely be helpful to compare the systemic immune response after MI and stroke and to determine the prevailing parallels and differences.
Corresponding Author: Matthias Nahrendorf, MD, PhD, Center for Systems Biology, 185 Cambridge St, Boston, MA 02114 (firstname.lastname@example.org).
Accepted for Publication: August 30, 2013.
Published Online: December 2, 2013. doi:10.1001/jamaneurol.2013.5026.
Author Contributions: Dr Nahrendorf had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: All authors.
Acquisition of data: Moskowitz, Nahrendorf.
Analysis and interpretation of data: Moskowitz, Nahrendorf.
Drafting of the manuscript: Courties, Nahrendorf.
Critical revision of the manuscript for important intellectual content: Moskowitz, Nahrendorf.
Obtained funding: All authors.
Administrative, technical, or material support: Moskowitz.
Study supervision: Nahrendorf.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported in part by an American Heart Association postdoctoral fellowship 13POST16580004 to Dr Courties and grants R01-HL114477 and R01-HL095629 from the National Heart, Lung, and Blood Institute.
Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: The figure was prepared using Servier Medical Art.