LAUSR

Dear Editor, Strokes cause 5.8 million deaths each year. Among these victims, ~30% are from China. Acute ischemic stroke (AIS) is the most prevalent subtype of strokes. Although drugs can alleviate the symptoms, the recoveries of functional vessels within ischemic areas are the critical factor determining the prognosis of patients suffering from AIS. Nevertheless, the mechanisms involved in cerebral revascularization remain largely unknown. Myeloid cells are among the first cells arriving around and within the injured areas after the ischemic assault. A specific subgroup of Tie2expressing monocytes (TEMs) has demonstrated vessel-repairing properties in tumors and ischemic limbs. But, it is not clear whether TEMs participate in revascularization and neurological recovery in AIS. Hence, we explore the impacts of TEMs on the prognosis of AIS and the potential mechanism beneath with clinical samples and mouse models. Pre-therapeutic blood samples from patients within 24 h after onset of AIS and from age-matched controls (AMCs) were collected and analyzed to determine whether TEMs are upregulated in response to ischemic brain injury. The demographics of the enrolled participants are listed in Supplementary Table 1. Figure1a shows representative magnetic resonance imaging (MRI) from an enrolled patient taken before emergency treatment at the onset day. By flow cytometric analysis, we found the proportion of circulating monocytes relative to the total number of white blood cells in patients with AIS was higher than in the AMCs (Supplementary Figs. S1, 2). Similarly, the proportion and level of Tie2 expressing in CD14 monocytes were higher in the AIS group than in AMC (Fig. 1b). Within 24 h before the patient was discharged, the modified Rankin Scale (mRS) was used to estimate stroke prognosis (scores ≤2 consider better clinical outcome). Although the Tie2 expression was not correlated with the accurate mRS score of AIS patients (Supplementary Fig. S2d), patients with mRS scores ≤2 presented higher Tie2 expression in CD14 monocytes than scores >2 at baseline evaluation (Fig. 1b). Overall, our clinical study demonstrated the frequency of TEMs (CD45/CD14/ Tie2) in circulation positively related to AIS patients’ good prognosis as assessed by mRS score. Based on CD16 expression, we further divided monocytes into three main subsets: the classical subset (CD14/CD16), the non-classical subset (CD14/CD16), and the intermediate subset (CD14/CD16). Flow cytometric analysis showed the scales of both non-classical and intermediate monocytes were higher in patients with AIS than in AMCs, and of classical monocytes were lower accordingly (Supplementary Fig. S3a, b). Tie2 was expressed in all three subsets for AIS patients but most strongly within the intermediate one (Supplementary Fig. S3c). Intriguingly, as no statistically significant association was found between circulating CD14/CD16/Tie2 monocyte counts and AIS patient outcomes (Supplementary Fig. S3d), and as Tie2 expression did not always correspond with the three monocyte subsets in line with other reports, we suggest the current monocyte nomenclature may inadvertently conceal the remarkable roles of the Tie2 monocyte population. To validate the clinical data, we investigated circulating monocyte accumulation and their Tie2 expression in peripheral blood samples from shamand transient middle cerebral artery occlusion (tMCAO)-treated mice (Fig. 1c). The mouse model of tMCAO has been widely used to mimics the clinical context in patients with AIS. Compared with the sham group, Tie2 expression of the tMCAO-treated mice was remarkably upregulated at 24 h post surgery (Fig. 1d and Supplementary Fig. S4). Meanwhile, in contrast to transiently elevated in sham-treated mice, the circulating monocyte/macrophage (CD115/CD11b) subgroups increased significantly and stably expanded in the tMCAO group (Supplementary Fig. S4). Next, we examined whether TEM levels are raised in brain tissue after tMCAO surgery. Of note, the bloodbrain barrier (BBB) formed by continuous endothelial cell (EC) membrane acts as a regulating interface with adherens and tight junctions to protect the brain compartment as an immunoprivileged site. In this scenario, questioning whether myeloid-derived TEMs can penetrate the BBB into the brain parenchyma is reasonable. Here, we collected brain sections at 24 h post tMCAO induction, and immunofluorescence staining showed severely damaged blood vessels (CD31) accompanied by rapidly accumulating microglia and monocytes/macrophages (CD11b) in the occlusion areas of tMCAO mice (Fig. 1e, f). Notably, emerging TEMs (CD11b/Tie2) were identified around ruptured blood vessels, and they were not colocalized with the fractalkine receptor CX3CR1, a highly expressed marker on mature microglia (Supplementary Fig. S5). Our data suggest TEMs infiltrate the injured brain tissue from the peripheral blood after tMCAO induction. Next, we generated a transgenic mouse model of Tie2 Lyz mice (M) and used the littermates with a genotype Tie2Lyz as the control (CTR; Supplementary Fig. S6) to investigate the effects of TEM deficiency in the ischemic brain. As shown in Fig. 1g, Tie2 deficiency in monocytes/macrophages was associated with markedly increased infarction areas by MRI performance 24 h post tMCAO. Statistical analysis of the data from MRI screening indicated the infarct volumes were enlarged mainly in mice lacking Tie2 myeloid cells (Fig. 1g), which further confirms the effects of TEMs on infarction in vivo. Consistent with brain injury levels, the stepping speed and cadence were decreased in M mice compared with the CTR counterparts at 7 days after surgery (Fig. 1h, i), indicating that Tie2 deficiency delayed functional improvement. In the meantime, bodyweight loss was more frequently observed in M mice after tMCAO